CME

Laboratory tests are often ordered inappropriately for patients in whom a rheumatologic illness is suspected; this occurs in both primary and secondary care.¹ Some tests are available both singly and as part of a battery of tests screening healthy people without symptoms.

The problem: negative test results are by no means always reassuring, and false-positive results raise the risks of unnecessary anxiety for patients and clinicians, needless referrals, and potential morbidity due to further unnecessary testing and exposure to wrong treatments.² Clinicians should be aware of the pitfalls of these tests in order to choose them wisely and interpret the results correctly.

This article provides practical guidance on requesting and interpreting some common tests in rheumatology, with the aid of case vignettes.

RHEUMATOID FACTOR AND ANTICITRULLINATED PEPTIDE ANTIBODY

A 41-year-old woman, previously in good health, presents to her primary care practitioner with a 6-week history of pain and swelling in her hands and early morning stiffness lasting about 2 hours. She denies having any extraarticular symptoms. Physical examination reveals synovitis across her right metacarpophalangeal joints, proximal interphalangeal joint of the left middle finger, and left wrist. The primary care physician is concerned that her symptoms might be due to rheumatoid arthritis.

Would testing for rheumatoid factor and anticitrullinated peptide antibody be useful in this patient?

Rheumatoid factor is an antibody (immunoglobulin M, IgG, or IgA) targeted against the Fc fragment of IgG.³ It was so named because it was originally detected in patients with rheumatoid arthritis, but it is neither sensitive nor specific for this condition. A meta-analysis of more than 5,000 patients with rheumatoid arthritis reported that rheumatoid factor testing had a sensitivity of 69% and specificity of 85%.⁴

Numerous other conditions can be associated with a positive test for rheumatoid factor (Table 1). Hence, a diagnosis of rheumatoid arthritis cannot be confirmed with a positive result alone, nor can it be excluded with a negative result.

Anticitrullinated peptide antibody, on the other hand, is much more specific for rheumatoid arthritis (95%), as it is seldom seen in other conditions, but its sensitivity is similar to that of rheumatoid factor (68%).^4–6 A positive result would thus lend strength to the diagnosis of rheumatoid arthritis, but a negative result would not exclude it.

Approach to early arthritis

When faced with a patient with early arthritis, some key questions to ask include^7,8:

Is this an inflammatory or a mechanical problem? Inflammatory arthritis is suggested by joint swelling that is not due to trauma or bony hypertrophy, early morning stiffness lasting longer than 30 minutes, and elevated inflammatory markers (erythrocyte sedimentation rate or C-reactive protein). Involvement of the small joints of the hands and feet may be suggested by pain on compression of the metacarpophalangeal and metatarsophalangeal joints, respectively.

Is there a definite identifiable underlying cause for the inflammatory arthritis? The pattern of development of joint symptoms or the presence of extraarticular symptoms may suggest an underlying problem such as gout, psoriatic arthritis, systemic lupus erythematosus, or sarcoidosis.

If the arthritis is undifferentiated (ie, there is no definite identifiable cause), is it likely to remit or persist? This is perhaps the most important question to ask in order to prognosticate. Patients with risk factors for persistent disease, ie, for development of rheumatoid arthritis, should be referred to a rheumatologist early for timely institution of disease-modifying antirheumatic drug therapy.⁹ Multiple studies have shown that patients in whom this therapy is started early have much better clinical, functional, and radiologic outcomes than those in whom it is delayed.^10–12

The revised American College of Rheumatology and European League Against Rheumatism criteria¹³ include the following factors as predictors of persistence:

• Number of involved joints (with greater weight given to involvement of small joints)
• Duration of symptoms 6 weeks or longer
• Elevated acute-phase response (erythrocyte sedimentation rate or C-reactive protein level)
• A positive serologic test (either rheumatoid factor or anticitrullinated peptide antibody).

If both rheumatoid factor and anticitrullinated peptide antibody are positive in a patient with early undifferentiated arthritis, the risk of progression to rheumatoid arthritis is almost 100%, thus underscoring the importance of testing for these antibodies.^5,6 Referral to a rheumatologist should, however, not be delayed in patients with negative test results (more than one-third of patients with rheumatoid arthritis may be negative for both), and should be considered in those with inflammatory joint symptoms persisting longer than 6 weeks, especially with involvement of the small joints (sparing the distal interphalangeals) and elevated acute-phase response.

Rheumatoid factor in healthy people without symptoms

In some countries, testing for rheumatoid factor is offered as part of a battery of screening tests in healthy people who have no symptoms, a practice that should be strongly discouraged.

Multiple studies, both prospective and retrospective, have demonstrated that both rheumatoid factor and anticitrullinated peptide antibody may be present several years before the clinical diagnosis of rheumatoid arthritis.^6,14–16 But the risk of developing rheumatoid arthritis for asymptomatic individuals who are rheumatoid factor-positive depends on the rheumatoid factor titer, positive family history of rheumatoid arthritis in first-degree relatives, and copresence of anticitrullinated peptide antibody. The absolute risk, nevertheless, is still very small. In some, there might be an alternative explanation such as undiagnosed Sjögren syndrome or hepatitis C.

In any event, no strategy is currently available that is proven to prevent the development of rheumatoid arthritis, and there is no role for disease-modifying therapy during the preclinical phase.¹⁶

Back to our patient

Blood testing in our patient reveals normal complete blood cell counts, aminotransferase levels, and serum creatinine concentration; findings on urinalysis are normal. Her erythrocyte sedimentation rate is 56 mm/hour (reference range 0–15), and her C-reactive protein level is 26 mg/dL (normal < 3). Testing is negative for rheumatoid factor and anticitrullinated peptide antibody.

Although her rheumatoid factor and anticitrullinated peptide antibody tests are negative, she is referred to a rheumatologist because she has predictors of persistent disease, ie, symptom duration of 6 weeks, involvement of the small joints of the hands, and elevated erythrocyte sedimentation rate and C-reactive protein. The rheumatologist checks her parvovirus serology, which is negative.

The patient is given parenteral depot corticosteroid therapy, to which she responds briefly. Because her symptoms persist and continue to worsen, methotrexate treatment is started after an additional 6 weeks.

 

 

ANTINUCLEAR ANTIBODY

A 37-year-old woman presents to her primary care physician with the complaint of tiredness. She has a family history of systemic lupus erythematosus in her sister and maternal aunt. She is understandably worried about lupus because of the family history and is asking to be tested for it.

Would testing for antinuclear antibody be reasonable?

Antinuclear antibody is not a single antibody but rather a family of autoantibodies that are directed against nuclear constituents such as single- or double-stranded deoxyribonucleic acid (dsDNA), histones, centromeres, proteins complexed with ribonucleic acid (RNA), and enzymes such as topoisomerase.^17,18

Protein antigens complexed with RNA and some enzymes in the nucleus are also known as extractable nuclear antigens (ENAs). They include Ro, La, Sm, Jo-1, RNP, and ScL-70 and are named after the patient in whom they were first discovered (Robert, Lavine, Smith, and John), the antigen that is targeted (ribonucleoprotein or RNP), and the disease with which they are associated (anti-ScL-70 or antitopoisomerase in diffuse cutaneous scleroderma).

Antinuclear antibody testing is commonly requested to exclude connective tissue diseases such as lupus, but the clinician needs to be aware of the following points:

Antinuclear antibody may be encountered in conditions other than lupus

These include¹⁹:

• Other autoimmune diseases such as rheumatoid arthritis, primary Sjögren syndrome, systemic sclerosis, autoimmune thyroid disease, and myasthenia gravis
• Infection with organisms that share the epitope with self-antigens (molecular mimicry)
• Cancers
• Drugs such as hydralazine, procainamide, and minocycline.

Antinuclear antibody might also be produced by the healthy immune system from time to time to clear the nuclear debris that is extruded from aging cells.

A study in healthy individuals²⁰ reported a prevalence of positive antinuclear antibody of 32% at a titer of 1/40, 15% at a titer of 1/80, 7% at a titer of 1/160, and 3% at a titer of 1/320. Importantly, a positive result was more common among family members of patients with autoimmune connective tissue diseases.²¹ Hence, a positive antinuclear antibody result does not always mean lupus.

Antinuclear antibody testing is highly sensitive for lupus

With current laboratory methods, antinuclear antibody testing has a sensitivity close to 100%. Hence, a negative result virtually rules out lupus.

Two methods are commonly used to test for antinuclear antibody: indirect immunofluorescence and enzyme-linked immunosorbent assay (ELISA).²² While human epithelial (Hep2) cells are used as the source of antigen in immunofluorescence, purified nuclear antigens coated on multiple-well plates are used in ELISA.

Although ELISA is simpler to perform, immunofluorescence has a slightly better sensitivity (because the Hep2 cells express a wide range of antigens) and is still considered the gold standard. As expected, the higher sensitivity occurs at the cost of reduced specificity (about 60%), so antinuclear antibody will also be detected in all the other conditions listed above.²³

To improve the specificity of antinuclear antibody testing, laboratories report titers (the highest dilution of the test serum that tested positive); a cutoff of greater than 1/80 is generally considered significant.

Do not order antinuclear antibody testing indiscriminately

If the antinuclear antibody test is requested indiscriminately, the positive predictive value for the diagnosis of lupus is only 11%.²⁴ The test should be requested only when the pretest probability of lupus or other connective tissue disease is high. The positive predictive value is much higher in patients presenting with clinical or laboratory manifestations involving 2 or more organ systems (Table 2).^18,25

Categorization of the specific antigen target improves disease specificity. The antinuclear antibody in patients with lupus may be targeted against single- or double-stranded DNA, histones, or 1 or more of the ENAs. Among these, the presence of anti-dsDNA or anti-Sm is highly specific for a diagnosis of lupus (close to 100%). Neither is sensitive for lupus, however, with anti-dsDNA present in only 60% of patients with lupus and anti-Sm in about 30%.¹⁷ Hence, patients with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may continue to pose a diagnostic challenge. Other examples of specific disease associations are listed in Table 3.

To sum up, the antinuclear antibody test should be requested only in patients with involvement of multiple organ systems. Although a negative result would make it extremely unlikely that the clinical presentation is due to lupus, a positive result is insufficient on its own to make a diagnosis of lupus.

Diagnosing lupus is straightforward when patients present with a specific manifestation such as inflammatory arthritis, photosensitive skin rash, hemolytic anemia, thrombocytopenia, or nephritis, or with specific antibodies such as those against dsDNA or Sm. Patients who present with nonspecific symptoms such as arthralgia or tiredness with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may present difficulties even for the specialist.^25–27

Back to our patient

Our patient denies arthralgia. She has no extraarticular symptoms such as skin rashes, oral ulcers, sicca symptoms, muscle weakness, Raynaud phenomenon, pleuritic chest pain, or breathlessness. Findings on physical examination and urinalysis are unremarkable.

Her primary care physician decides to check her complete blood cell count, erythrocyte sedimentation rate, and thyroid-stimulating hormone level. Although she is reassured that her tiredness is not due to lupus, she insists on getting an antinuclear antibody test.

Her complete blood cell counts are normal. Her erythrocyte sedimentation rate is 6 mm/hour. However, her thyroid-stimulating hormone level is elevated, and subsequent testing shows low free thyroxine and positive thyroid peroxidase antibodies. The antinuclear antibody is positive in a titer of 1/80 and negative for anti-dsDNA and anti-ENA.

We explain to her that the positive antinuclear antibody is most likely related to her autoimmune thyroid disease. She is referred to an endocrinologist.

 

 

ANTIPHOSPHOLIPID ANTIBODIES

A 24-year-old woman presents to the emergency department with acute unprovoked deep vein thrombosis in her right leg, confirmed by ultrasonography. She has no history of previous thrombosis, and the relevant family history is unremarkable. She has never been pregnant. Her platelet count is 84 × 10⁹/L (reference range 150–400), and her baseline activated partial thromboplastin time is prolonged at 62 seconds (reference range 23.0–32.4). The rest of her blood counts and her prothrombin time, liver enzyme levels, and serum creatinine level are normal.

Should this patient be tested for antiphospholipid antibodies?

Antiphospholipid antibodies are important because of their association with thrombotic risk (both venous and arterial) and pregnancy morbidity. The name is a misnomer, as these antibodies are targeted against some proteins that are bound to phospholipids and not only to the phospholipids themselves.

According to the modified Sapporo criteria for the classification of antiphospholipid syndrome,²⁸ antiphospholipid antibodies should remain persistently positive on at least 2 separate occasions at least 12 weeks apart for the result to be considered significant because some infections and drugs may be associated with the transient presence of antiphospholipid antibodies.

Screening for antiphospholipid antibodies should include testing for IgM and IgG anticardiolipin antibodies, lupus anticoagulant, and IgM and IgG beta-2 glycoprotein I antibodies.^29,30

Anticardiolipin antibodies

Anticardiolipin (aCL) antibodies may be targeted either against beta-2 glycoprotein I (beta-2GPI) that is bound to cardiolipin (a phospholipid) or against cardiolipin alone; the former is more specific. Antibodies directed against cardiolipin alone are usually transient and are associated with infections and drugs. The result is considered significant only when anticardiolipin antibodies are present in a medium to high titer (> 40 IgG phospholipid units or IgM phospholipid units, or > 99th percentile).

Lupus anticoagulant

The antibody with “lupus anticoagulant activity” is targeted against prothrombin plus phospholipid or beta-2GPI plus phospholipid. The test for it is a functional assay involving 3 steps:

Demonstrating the prolongation of a phospholipid-dependent coagulation assay like the activated partial thromboplastin time (aPTT). (This may explain the prolongation of aPTT in the patient described in the vignette.) Although the presence of lupus anticoagulant is associated with thrombosis, it is called an “anticoagulant” because of this in vitro prolongation of phospholipid-dependent coagulation assays.

Mixing study. The phospholipid-dependent coagulation assay could be prolonged because of either the deficiency of a coagulation factor or the presence of the antiphospholipid antibodies. This can be differentiated by mixing the patient’s plasma with normal plasma (which will have all the clotting factors) in a 1:1 ratio. If the coagulation assay remains prolonged after the addition of normal plasma, clotting factor deficiency can be excluded.

Addition of a phospholipid. If the prolongation of the coagulation assay is due to the presence of an antiphospholipid antibody, addition of extra phospholipid will correct this.

Beta-2 glycoprotein I antibody (anti-beta-2GPI)

The beta-2GPI that is not bound to the cardiolipin can be detected by separately testing for beta-2GPI (the anticardiolipin test only detects the beta-2GPI that is bound to the cardiolipin). The result is considered significant if beta-2GPI is present in a medium to high titer (> 99th percentile).

Studies have shown that antiphospholipid antibodies may be present in 1% to 5% of apparently healthy people in the general population.³¹ These are usually low-titer anticardiolipin or anti-beta-GPI IgM antibodies that are not associated with thrombosis or adverse pregnancy outcomes. Hence, the term antiphospholipid syndrome should be reserved for those who have had at least 1 episode of thrombosis or pregnancy morbidity and persistent antiphospholipid antibodies, and not those who have asymptomatic or transient antiphospholipid antibodies.

Triple positivity (positive anticardiolipin, lupus anticoagulant, and anti-beta-2GPI) seems to be associated with the highest risk of thrombosis, with a 10-year cumulative incidence of 37.1% (95% confidence interval [CI] 19.9–54.3) for a first thrombotic event,³² and 44.2% (95% CI 38.6–49.8) for recurrent thrombosis.³³

The association with thrombosis is stronger for lupus anticoagulant than with the other 2 antibodies, with different studies³⁴ finding an odds ratio ranging from 5 to 16. A positive lupus anticoagulant test with or without a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a high-risk profile, while a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a moderate-risk profile. A low titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a low-risk profile that may not be associated with thrombosis.³⁵

Antiphospholipid syndrome is important to recognize because of the need for long-term anticoagulation to prevent recurrence.³⁶ It may be primary, when it occurs on its own, or secondary, when it occurs in association with another autoimmune disease such as lupus.

Venous events in antiphospholipid syndrome most commonly manifest as lower-limb deep vein thrombosis or pulmonary embolism, while arterial events most commonly manifest as stroke or transient ischemic attack.³⁷ Obstetric manifestations may include not only miscarriage and stillbirth, but also preterm delivery, intrauterine growth retardation, and preeclampsia, all occurring due to placental insufficiency.

The frequency of antiphospholipid antibodies has been estimated as 13.5% in patients with stroke, 11% with myocardial infarction, 9.5% with deep vein thrombosis, and 6% for those with pregnancy morbidity.³⁸

Some noncriteria manifestations have also been recognized in antiphospholipid syndrome, such as thrombocytopenia, cardiac vegetations (Libman-Sachs endocarditis), livedo reticularis, and nephropathy.

The indications for antiphospholipid antibody testing are listed in Table 4.²⁹ For the patient described in the vignette, it would be appropriate to test for antiphospholipid antibodies because of her unprovoked thrombosis, thrombocytopenia, and prolonged aPTT. Anticoagulant treatment is known to be associated with false-positive lupus anticoagulant, so any blood samples should be drawn before such treatment is commenced.

Back to our patient

Our patient’s anticardiolipin IgG test is negative, while her lupus anticoagulant and beta-2GPI IgG are positive. She has no clinical or laboratory features suggesting lupus.

She is started on warfarin. After 3 months, the warfarin is interrupted for several days, and she is retested for all 3 antiphospholipid antibodies. Her beta-2GPI I IgG and lupus anticoagulant tests are again positive. Because of the persistent antiphospholipid antibody positivity and clinical history of deep vein thrombosis, her condition is diagnosed as primary antiphospholipid syndrome. She is advised to continue anticoagulant therapy indefinitely.

 

 

ANTINEUTROPHIL CYTOPLASMIC ANTIBODY

A 34-year-old man who is an injecting drug user presents with a 2-week history of fever, malaise, and generalized arthralgia. There are no localizing symptoms of infection. Notable findings on examination include a temperature of 38.0°C (100.4°F), needle track marks in his arms, nonblanching vasculitic rash in his legs, and a systolic murmur over the precordium.

His white blood cell count is 15.3 × 10⁹/L (reference range 3.7–11.0), and his C-reactive protein level is 234 mg/dL (normal < 3). Otherwise, results of blood cell counts, liver enzyme tests, renal function tests, urinalysis, and chest radiography are normal.

Two sets of blood cultures are drawn. Transthoracic echocardiography and the antineutrophil cytoplasmic antibody (ANCA) test are requested, as are screening tests for human immunodeficiency virus, hepatitis B, and hepatitis C.

Was the ANCA test indicated in this patient?

ANCAs are autoantibodies against antigens located in the cytoplasmic granules of neutrophils and monocytes. They are associated with small-vessel vasculitides such as granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (EGPA), and isolated pauciimmune crescentic glomerulonephritis, all collectively known as ANCA-associated vasculitis (AAV).³⁹

Laboratory methods to detect ANCA include indirect immunofluorescence and antigen-specific enzyme immunoassays. Indirect immunofluorescence only tells us whether or not an antibody that is targeting a cytoplasmic antigen is present. Based on the indirect immunofluorescent pattern, ANCA can be classified as follows:

• Perinuclear or p-ANCA (if the targeted antigen is located just around the nucleus and extends into it)
• Cytoplasmic or c-ANCA (if the targeted antigen is located farther away from the nucleus)
• Atypical ANCA (if the indirect immunofluorescent pattern does not fit with either p-ANCA or c-ANCA).

Indirect immunofluorescence does not give information about the exact antigen that is targeted; this can only be obtained by performing 1 of the antigen-specific immunoassays. The target antigen for c-ANCA is usually proteinase-3 (PR3), while that for p-ANCA could be myeloperoxidase (MPO), cathepsin, lysozyme, lactoferrin, or bactericidal permeability inhibitor. Anti-PR3 is highly specific for GPA, while anti-MPO is usually associated with MPA and EGPA. Less commonly, anti-PR3 may be seen in patients with MPA and anti-MPO in those with GPA. Hence, there is an increasing trend toward classifying ANCA-associated vasculitis into PR3-associated or MPO-associated vasculitis rather than as GPA, MPA, EGPA, or renal-limited vasculitis.⁴⁰

Several audits have shown that the ANCA test is widely misused and requested indiscriminately to rule out vasculitis. This results in a lower positive predictive value, possible harm to patients due to increased false-positive rates, and increased burden on the laboratory.^41–43 At least 2 separate groups have demonstrated that a gating policy that refuses ANCA testing in patients without clinical evidence of systemic vasculitis can reduce the number of inappropriate requests, improve the diagnostic yield, and make it more clinically relevant and cost-effective.^44,45

The clinician should bear in mind that:

ANCA testing should be requested only if the pretest probability of ANCA-associated vasculitis is high. The indications proposed by the International Consensus Statement on ANCA testing⁴⁶ are listed in Table 5. These criteria have been clinically validated, with 1 study even demonstrating that no cases of ANCA-associated vasculitis would be missed if these guidelines are followed.⁴⁷

Current guidelines recommend using one of the antigen-specific assays for PR3 and MPO as the primary screening method.⁴⁸ Until recently, indirect immunofluorescence was used to screen for ANCA-associated vasculitis, and positive results were confirmed by ELISA to detect ANCAs specific for PR3 and MPO,⁴⁹ but this is no longer recommended because of recent evidence suggesting a large variability between the different indirect immunofluorescent methods and improved diagnostic performance of the antigen-specific assays.

In a large multicenter study by Damoiseaux et al, the specificity with the different antigen-specific immunoassays was 98% to 99% for PR3-ANCA and 96% to 99% for MPO-ANCA.⁵⁰

ANCA-associated vasculitis should not be considered excluded if the PR3 and MPO-ANCA are negative. In the Damoiseaux study, about 11% to 15% of patients with GPA and 8% to 24% of patients with MPA tested negative for both PR3 and MPO-ANCA.⁵⁰

If the ANCA result is negative and clinical suspicion for ANCA-associated vasculitis is high, the clinician may wish to consider requesting another immunoassay method or indirect immunofluorescence. Results of indirect immunofluorescent testing results may be positive in those with a negative immunoassay, and vice versa.

A positive ANCA result is not diagnostic of ANCA-associated vasculitis. Numerous other conditions are associated with ANCA, usually p-ANCA or atypical ANCA (Table 6). The antigens targeted by these ANCAs are usually cathepsin, lysozyme, lactoferrin, and bactericidal permeability inhibitor.

Thus, the ANCA result should always be interpreted in the context of the whole clinical picture.⁵¹ Biopsy should still be considered the gold standard for the diagnosis of ANCA-associated vasculitis. The ANCA titer can help to improve clinical interpretation, because the likelihood of ANCA-associated vasculitis increases with higher levels of PR3 and MPO-ANCA.⁵²

Back to our patient

Our patient’s blood cultures grow methicillin-sensitive Staphylococcus aureus in both sets after 48 hours. Transthoracic echocardiography reveals vegetations around the tricuspid valve, with no evidence of valvular regurgitation. The diagnosis is right-sided infective endocarditis. He is started on appropriate antibiotics.

Tests for human immunodeficiency virus, hepatitis B, and hepatitis C are negative. The ANCA test is positive for MPO-ANCA at 28 IU/mL (normal < 10).

The positive ANCA is thought to be related to the infective endocarditis. His vasculitis is most likely secondary to infective endocarditis and not ANCA-associated vasculitis. The ANCA test need not have been requested in the first place.

 

 

HUMAN LEUKOCYTE ANTIGEN-B27

A 22-year-old man presents to his primary care physician with a 4-month history of gradually worsening low back pain associated with early morning stiffness lasting more than 2 hours. He has no peripheral joint symptoms.

In the last 2 years, he has had 2 separate episodes of uveitis. There is a family history of ankylosing spondylitis in his father. Examination reveals global restriction of lumbar movements but is otherwise unremarkable. Magnetic resonance imaging (MRI) of the lumbar spine and sacroiliac joints is normal.

Should this patient be tested for human leukocyte antigen-B27 (HLA-B27)?

The major histocompatibility complex (MHC) is a gene complex that is present in all animals. It encodes proteins that help with immunologic tolerance. HLA simply refers to the human version of the MHC.⁵³ The HLA gene complex, located on chromosome 6, is categorized into class I, class II, and class III. HLA-B is one of the 3 class I genes. Thus, a positive HLA-B27 result simply means that the particular gene is present in that person.

HLA-B27 is strongly associated with ankylosing spondylitis, also known as axial spondyloarthropathy.⁵⁴ Other genes also contribute to the pathogenesis of ankylosing spondylitis, but HLA-B27 is present in more than 90% of patients with this disease and is by far considered the most important. The association is not as strong for peripheral spondyloarthropathy, with studies reporting a frequency of up to 75% for reactive arthritis and inflammatory bowel disease-associated arthritis, and up to 50% for psoriatic arthritis and uveitis.⁵⁵

About 9% of healthy, asymptomatic individuals may have HLA-B27, so the mere presence of this gene is not evidence of disease.⁵⁶ There may be up to a 20-fold increased risk of ankylosing spondylitis among those who are HLA-B27-positive.⁵⁷

Some HLA genes have many different alleles, each of which is given a number (explaining the number 27 that follows the B). Closely related alleles that differ from one another by only a few amino-acid substitutions are then categorized together, thus accounting for more than 100 subtypes of HLA-B27 (designated from HLA-B*2701 to HLA-B*27106). These subtypes vary in frequency among different racial groups, and the population prevalence of ankylosing spondylitis parallels the frequency of HLA-B27.⁵⁸ The most common subtype seen in white people and American Indians is B*2705. HLA-B27 is rare in blacks, explaining the rarity of ankylosing spondylitis in this population. Further examples include HLA-B*2704, which is seen in Asians, and HLA-B*2702, seen in Mediterranean populations. Not all subtypes of HLA-B27 are associated with disease, and some, like HLA-B*2706, may also be protective.

When should the clinician consider testing for HLA-B27?

Not all patients with low back pain need an HLA-B27 test. First, it is important to look for clinical features of axial spondyloarthropathy (Table 7). The unifying feature of spondyloarthropathy is enthesitis (inflammation at the sites of insertion of tendons or ligaments on the skeleton). Inflammation of axial entheses causes spondylitis and sacroiliitis, manifesting as inflammatory back pain. Clinical clues to inflammatory back pain include insidious onset, aggravation with rest or inactivity, prolonged early morning stiffness, disturbed sleep during the second half of the night, relief with movement or activity, alternating gluteal pain (due to sacroiliitis), and good response to anti-inflammatory medication (although nonspecific).

Peripheral spondyloarthropathy may present with arthritis, enthesitis (eg, heel pain due to inflammation at the site of insertion of the Achilles tendon or plantar fascia), or dactylitis (“sausage” swelling of the whole finger or toe due to extension of inflammation beyond the margins of the joint). Other clues may include psoriasis, inflammatory bowel disease, history of preceding gastrointestinal or genitourinary infection, family history of similar conditions, and history of recurrent uveitis.

For the initial assessment of patients who have inflammatory back pain, plain radiography of the sacroiliac joints is considered the gold standard.⁵⁹ If plain radiography does not show evidence of sacroiliitis, MRI of the sacroiliac joints should be considered. While plain radiography can reveal only structural changes such as sclerosis, erosions, and ankylosis, MRI is useful to evaluate for early inflammatory changes such as bone marrow edema. Imaging the lumbar spine is not necessary, as the sacroiliac joints are almost invariably involved in axial spondyloarthropathy, and lesions seldom occur in the lumbar spine in isolation.⁶⁰

The diagnosis of ankylosing spondylitis previously relied on confirmatory imaging features, but based on the new International Society classification criteria,^61–63 which can be applied to patients with more than 3 months of back pain and age of onset of symptoms before age 45, patients can be classified as having 1 of the following:

• Radiographic axial spondyloarthropathy, if they have evidence of sacroiliitis on imaging plus 1 other feature of spondyloarthropathy
• Nonradiographic axial spondyloarthropathy, if they have a positive HLA-B27 plus 2 other features of spondyloarthropathy (Table 7).

These new criteria have a sensitivity of 82.9% and specificity of 84.4%.^62,63 The disease burden of radiographic and nonradiographic axial spondyloarthropathy has been shown to be similar, suggesting that they are part of the same disease spectrum. Thus, the HLA-B27 test is useful to make a diagnosis of axial spondyloarthropathy even in the absence of imaging features and could be requested in patients with 2 or more features of spondyloarthropathy. In the absence of imaging features and a negative HLA-B27 result, however, the patient cannot be classified as having axial spondyloarthropathy.

Back to our patient

The absence of radiographic evidence would not exclude axial spondyloarthropathy in our patient. The HLA-B27 test is requested because of the inflammatory back pain and the presence of 2 spondyloarthropathy features (uveitis and the family history) and is reported to be positive. His disease is classified as nonradiographic axial spondyloarthropathy.

He is started on regular naproxen and is referred to a physiotherapist. After 1 month, he reports significant symptomatic improvement. He asks if he can be retested for HLA-B27 to see if it has become negative. We tell him that there is no point in repeating it, as it is a gene and will not disappear.

SUMMARY: CONSIDER THE CLINICAL PICTURE

When approaching a patient suspected of having a rheumatologic disease, a clinician should first consider the clinical presentation and the intended purpose of each test. The tests, in general, might serve several purposes. They might help to:

Increase the likelihood of the diagnosis in question. For example, a positive rheumatoid factor or anticitrullinated peptide antibody can help diagnose rheumatoid arthritis in a patient with early polyarthritis, a positive HLA-B27 can help diagnose ankylosing spondylitis in patients with inflammatory back pain and normal imaging, and a positive ANCA can help diagnose ANCA-associated vasculitis in a patient with glomerulonephritis.

Reduce the likelihood of the diagnosis in question. For example, a negative antinuclear antibody test reduces the likelihood of lupus in a patient with joint pains.

Monitor the condition. For example DNA antibodies can be used to monitor the activity of lupus.

Plan the treatment strategy. For example, one might consider lifelong anticoagulation if antiphospholipid antibodies are persistently positive in a patient with thrombosis.

Prognosticate. For example, positive rheumatoid factor and anticitrullinated peptide antibody increase the risk of erosive rheumatoid arthritis.

If the test was requested in the absence of a clear indication and the result is positive, it is important to bear in mind the potential pitfalls associated with that test and not attach a diagnostic label prematurely. None of the tests can confirm or exclude a condition, so the results should always be interpreted in the context of the whole clinical picture.   

Laboratory tests are often ordered inappropriately for patients in whom a rheumatologic illness is suspected; this occurs in both primary and secondary care.¹ Some tests are available both singly and as part of a battery of tests screening healthy people without symptoms.

The problem: negative test results are by no means always reassuring, and false-positive results raise the risks of unnecessary anxiety for patients and clinicians, needless referrals, and potential morbidity due to further unnecessary testing and exposure to wrong treatments.² Clinicians should be aware of the pitfalls of these tests in order to choose them wisely and interpret the results correctly.

This article provides practical guidance on requesting and interpreting some common tests in rheumatology, with the aid of case vignettes.

RHEUMATOID FACTOR AND ANTICITRULLINATED PEPTIDE ANTIBODY

A 41-year-old woman, previously in good health, presents to her primary care practitioner with a 6-week history of pain and swelling in her hands and early morning stiffness lasting about 2 hours. She denies having any extraarticular symptoms. Physical examination reveals synovitis across her right metacarpophalangeal joints, proximal interphalangeal joint of the left middle finger, and left wrist. The primary care physician is concerned that her symptoms might be due to rheumatoid arthritis.

Would testing for rheumatoid factor and anticitrullinated peptide antibody be useful in this patient?

Rheumatoid factor is an antibody (immunoglobulin M, IgG, or IgA) targeted against the Fc fragment of IgG.³ It was so named because it was originally detected in patients with rheumatoid arthritis, but it is neither sensitive nor specific for this condition. A meta-analysis of more than 5,000 patients with rheumatoid arthritis reported that rheumatoid factor testing had a sensitivity of 69% and specificity of 85%.⁴

Numerous other conditions can be associated with a positive test for rheumatoid factor (Table 1). Hence, a diagnosis of rheumatoid arthritis cannot be confirmed with a positive result alone, nor can it be excluded with a negative result.

Anticitrullinated peptide antibody, on the other hand, is much more specific for rheumatoid arthritis (95%), as it is seldom seen in other conditions, but its sensitivity is similar to that of rheumatoid factor (68%).^4–6 A positive result would thus lend strength to the diagnosis of rheumatoid arthritis, but a negative result would not exclude it.

Approach to early arthritis

When faced with a patient with early arthritis, some key questions to ask include^7,8:

Is this an inflammatory or a mechanical problem? Inflammatory arthritis is suggested by joint swelling that is not due to trauma or bony hypertrophy, early morning stiffness lasting longer than 30 minutes, and elevated inflammatory markers (erythrocyte sedimentation rate or C-reactive protein). Involvement of the small joints of the hands and feet may be suggested by pain on compression of the metacarpophalangeal and metatarsophalangeal joints, respectively.

Is there a definite identifiable underlying cause for the inflammatory arthritis? The pattern of development of joint symptoms or the presence of extraarticular symptoms may suggest an underlying problem such as gout, psoriatic arthritis, systemic lupus erythematosus, or sarcoidosis.

If the arthritis is undifferentiated (ie, there is no definite identifiable cause), is it likely to remit or persist? This is perhaps the most important question to ask in order to prognosticate. Patients with risk factors for persistent disease, ie, for development of rheumatoid arthritis, should be referred to a rheumatologist early for timely institution of disease-modifying antirheumatic drug therapy.⁹ Multiple studies have shown that patients in whom this therapy is started early have much better clinical, functional, and radiologic outcomes than those in whom it is delayed.^10–12

The revised American College of Rheumatology and European League Against Rheumatism criteria¹³ include the following factors as predictors of persistence:

• Number of involved joints (with greater weight given to involvement of small joints)
• Duration of symptoms 6 weeks or longer
• Elevated acute-phase response (erythrocyte sedimentation rate or C-reactive protein level)
• A positive serologic test (either rheumatoid factor or anticitrullinated peptide antibody).

If both rheumatoid factor and anticitrullinated peptide antibody are positive in a patient with early undifferentiated arthritis, the risk of progression to rheumatoid arthritis is almost 100%, thus underscoring the importance of testing for these antibodies.^5,6 Referral to a rheumatologist should, however, not be delayed in patients with negative test results (more than one-third of patients with rheumatoid arthritis may be negative for both), and should be considered in those with inflammatory joint symptoms persisting longer than 6 weeks, especially with involvement of the small joints (sparing the distal interphalangeals) and elevated acute-phase response.

Rheumatoid factor in healthy people without symptoms

In some countries, testing for rheumatoid factor is offered as part of a battery of screening tests in healthy people who have no symptoms, a practice that should be strongly discouraged.

Multiple studies, both prospective and retrospective, have demonstrated that both rheumatoid factor and anticitrullinated peptide antibody may be present several years before the clinical diagnosis of rheumatoid arthritis.^6,14–16 But the risk of developing rheumatoid arthritis for asymptomatic individuals who are rheumatoid factor-positive depends on the rheumatoid factor titer, positive family history of rheumatoid arthritis in first-degree relatives, and copresence of anticitrullinated peptide antibody. The absolute risk, nevertheless, is still very small. In some, there might be an alternative explanation such as undiagnosed Sjögren syndrome or hepatitis C.

In any event, no strategy is currently available that is proven to prevent the development of rheumatoid arthritis, and there is no role for disease-modifying therapy during the preclinical phase.¹⁶

Back to our patient

Blood testing in our patient reveals normal complete blood cell counts, aminotransferase levels, and serum creatinine concentration; findings on urinalysis are normal. Her erythrocyte sedimentation rate is 56 mm/hour (reference range 0–15), and her C-reactive protein level is 26 mg/dL (normal < 3). Testing is negative for rheumatoid factor and anticitrullinated peptide antibody.

Although her rheumatoid factor and anticitrullinated peptide antibody tests are negative, she is referred to a rheumatologist because she has predictors of persistent disease, ie, symptom duration of 6 weeks, involvement of the small joints of the hands, and elevated erythrocyte sedimentation rate and C-reactive protein. The rheumatologist checks her parvovirus serology, which is negative.

The patient is given parenteral depot corticosteroid therapy, to which she responds briefly. Because her symptoms persist and continue to worsen, methotrexate treatment is started after an additional 6 weeks.

 

 

ANTINUCLEAR ANTIBODY

A 37-year-old woman presents to her primary care physician with the complaint of tiredness. She has a family history of systemic lupus erythematosus in her sister and maternal aunt. She is understandably worried about lupus because of the family history and is asking to be tested for it.

Would testing for antinuclear antibody be reasonable?

Antinuclear antibody is not a single antibody but rather a family of autoantibodies that are directed against nuclear constituents such as single- or double-stranded deoxyribonucleic acid (dsDNA), histones, centromeres, proteins complexed with ribonucleic acid (RNA), and enzymes such as topoisomerase.^17,18

Protein antigens complexed with RNA and some enzymes in the nucleus are also known as extractable nuclear antigens (ENAs). They include Ro, La, Sm, Jo-1, RNP, and ScL-70 and are named after the patient in whom they were first discovered (Robert, Lavine, Smith, and John), the antigen that is targeted (ribonucleoprotein or RNP), and the disease with which they are associated (anti-ScL-70 or antitopoisomerase in diffuse cutaneous scleroderma).

Antinuclear antibody testing is commonly requested to exclude connective tissue diseases such as lupus, but the clinician needs to be aware of the following points:

Antinuclear antibody may be encountered in conditions other than lupus

These include¹⁹:

• Other autoimmune diseases such as rheumatoid arthritis, primary Sjögren syndrome, systemic sclerosis, autoimmune thyroid disease, and myasthenia gravis
• Infection with organisms that share the epitope with self-antigens (molecular mimicry)
• Cancers
• Drugs such as hydralazine, procainamide, and minocycline.

Antinuclear antibody might also be produced by the healthy immune system from time to time to clear the nuclear debris that is extruded from aging cells.

A study in healthy individuals²⁰ reported a prevalence of positive antinuclear antibody of 32% at a titer of 1/40, 15% at a titer of 1/80, 7% at a titer of 1/160, and 3% at a titer of 1/320. Importantly, a positive result was more common among family members of patients with autoimmune connective tissue diseases.²¹ Hence, a positive antinuclear antibody result does not always mean lupus.

Antinuclear antibody testing is highly sensitive for lupus

With current laboratory methods, antinuclear antibody testing has a sensitivity close to 100%. Hence, a negative result virtually rules out lupus.

Two methods are commonly used to test for antinuclear antibody: indirect immunofluorescence and enzyme-linked immunosorbent assay (ELISA).²² While human epithelial (Hep2) cells are used as the source of antigen in immunofluorescence, purified nuclear antigens coated on multiple-well plates are used in ELISA.

Although ELISA is simpler to perform, immunofluorescence has a slightly better sensitivity (because the Hep2 cells express a wide range of antigens) and is still considered the gold standard. As expected, the higher sensitivity occurs at the cost of reduced specificity (about 60%), so antinuclear antibody will also be detected in all the other conditions listed above.²³

To improve the specificity of antinuclear antibody testing, laboratories report titers (the highest dilution of the test serum that tested positive); a cutoff of greater than 1/80 is generally considered significant.

Do not order antinuclear antibody testing indiscriminately

If the antinuclear antibody test is requested indiscriminately, the positive predictive value for the diagnosis of lupus is only 11%.²⁴ The test should be requested only when the pretest probability of lupus or other connective tissue disease is high. The positive predictive value is much higher in patients presenting with clinical or laboratory manifestations involving 2 or more organ systems (Table 2).^18,25

Categorization of the specific antigen target improves disease specificity. The antinuclear antibody in patients with lupus may be targeted against single- or double-stranded DNA, histones, or 1 or more of the ENAs. Among these, the presence of anti-dsDNA or anti-Sm is highly specific for a diagnosis of lupus (close to 100%). Neither is sensitive for lupus, however, with anti-dsDNA present in only 60% of patients with lupus and anti-Sm in about 30%.¹⁷ Hence, patients with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may continue to pose a diagnostic challenge. Other examples of specific disease associations are listed in Table 3.

To sum up, the antinuclear antibody test should be requested only in patients with involvement of multiple organ systems. Although a negative result would make it extremely unlikely that the clinical presentation is due to lupus, a positive result is insufficient on its own to make a diagnosis of lupus.

Diagnosing lupus is straightforward when patients present with a specific manifestation such as inflammatory arthritis, photosensitive skin rash, hemolytic anemia, thrombocytopenia, or nephritis, or with specific antibodies such as those against dsDNA or Sm. Patients who present with nonspecific symptoms such as arthralgia or tiredness with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may present difficulties even for the specialist.^25–27

Back to our patient

Our patient denies arthralgia. She has no extraarticular symptoms such as skin rashes, oral ulcers, sicca symptoms, muscle weakness, Raynaud phenomenon, pleuritic chest pain, or breathlessness. Findings on physical examination and urinalysis are unremarkable.

Her primary care physician decides to check her complete blood cell count, erythrocyte sedimentation rate, and thyroid-stimulating hormone level. Although she is reassured that her tiredness is not due to lupus, she insists on getting an antinuclear antibody test.

Her complete blood cell counts are normal. Her erythrocyte sedimentation rate is 6 mm/hour. However, her thyroid-stimulating hormone level is elevated, and subsequent testing shows low free thyroxine and positive thyroid peroxidase antibodies. The antinuclear antibody is positive in a titer of 1/80 and negative for anti-dsDNA and anti-ENA.

We explain to her that the positive antinuclear antibody is most likely related to her autoimmune thyroid disease. She is referred to an endocrinologist.

 

 

ANTIPHOSPHOLIPID ANTIBODIES

A 24-year-old woman presents to the emergency department with acute unprovoked deep vein thrombosis in her right leg, confirmed by ultrasonography. She has no history of previous thrombosis, and the relevant family history is unremarkable. She has never been pregnant. Her platelet count is 84 × 10⁹/L (reference range 150–400), and her baseline activated partial thromboplastin time is prolonged at 62 seconds (reference range 23.0–32.4). The rest of her blood counts and her prothrombin time, liver enzyme levels, and serum creatinine level are normal.

Should this patient be tested for antiphospholipid antibodies?

Antiphospholipid antibodies are important because of their association with thrombotic risk (both venous and arterial) and pregnancy morbidity. The name is a misnomer, as these antibodies are targeted against some proteins that are bound to phospholipids and not only to the phospholipids themselves.

According to the modified Sapporo criteria for the classification of antiphospholipid syndrome,²⁸ antiphospholipid antibodies should remain persistently positive on at least 2 separate occasions at least 12 weeks apart for the result to be considered significant because some infections and drugs may be associated with the transient presence of antiphospholipid antibodies.

Screening for antiphospholipid antibodies should include testing for IgM and IgG anticardiolipin antibodies, lupus anticoagulant, and IgM and IgG beta-2 glycoprotein I antibodies.^29,30

Anticardiolipin antibodies

Anticardiolipin (aCL) antibodies may be targeted either against beta-2 glycoprotein I (beta-2GPI) that is bound to cardiolipin (a phospholipid) or against cardiolipin alone; the former is more specific. Antibodies directed against cardiolipin alone are usually transient and are associated with infections and drugs. The result is considered significant only when anticardiolipin antibodies are present in a medium to high titer (> 40 IgG phospholipid units or IgM phospholipid units, or > 99th percentile).

Lupus anticoagulant

The antibody with “lupus anticoagulant activity” is targeted against prothrombin plus phospholipid or beta-2GPI plus phospholipid. The test for it is a functional assay involving 3 steps:

Demonstrating the prolongation of a phospholipid-dependent coagulation assay like the activated partial thromboplastin time (aPTT). (This may explain the prolongation of aPTT in the patient described in the vignette.) Although the presence of lupus anticoagulant is associated with thrombosis, it is called an “anticoagulant” because of this in vitro prolongation of phospholipid-dependent coagulation assays.

Mixing study. The phospholipid-dependent coagulation assay could be prolonged because of either the deficiency of a coagulation factor or the presence of the antiphospholipid antibodies. This can be differentiated by mixing the patient’s plasma with normal plasma (which will have all the clotting factors) in a 1:1 ratio. If the coagulation assay remains prolonged after the addition of normal plasma, clotting factor deficiency can be excluded.

Addition of a phospholipid. If the prolongation of the coagulation assay is due to the presence of an antiphospholipid antibody, addition of extra phospholipid will correct this.

Beta-2 glycoprotein I antibody (anti-beta-2GPI)

The beta-2GPI that is not bound to the cardiolipin can be detected by separately testing for beta-2GPI (the anticardiolipin test only detects the beta-2GPI that is bound to the cardiolipin). The result is considered significant if beta-2GPI is present in a medium to high titer (> 99th percentile).

Studies have shown that antiphospholipid antibodies may be present in 1% to 5% of apparently healthy people in the general population.³¹ These are usually low-titer anticardiolipin or anti-beta-GPI IgM antibodies that are not associated with thrombosis or adverse pregnancy outcomes. Hence, the term antiphospholipid syndrome should be reserved for those who have had at least 1 episode of thrombosis or pregnancy morbidity and persistent antiphospholipid antibodies, and not those who have asymptomatic or transient antiphospholipid antibodies.

Triple positivity (positive anticardiolipin, lupus anticoagulant, and anti-beta-2GPI) seems to be associated with the highest risk of thrombosis, with a 10-year cumulative incidence of 37.1% (95% confidence interval [CI] 19.9–54.3) for a first thrombotic event,³² and 44.2% (95% CI 38.6–49.8) for recurrent thrombosis.³³

The association with thrombosis is stronger for lupus anticoagulant than with the other 2 antibodies, with different studies³⁴ finding an odds ratio ranging from 5 to 16. A positive lupus anticoagulant test with or without a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a high-risk profile, while a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a moderate-risk profile. A low titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a low-risk profile that may not be associated with thrombosis.³⁵

Antiphospholipid syndrome is important to recognize because of the need for long-term anticoagulation to prevent recurrence.³⁶ It may be primary, when it occurs on its own, or secondary, when it occurs in association with another autoimmune disease such as lupus.

Venous events in antiphospholipid syndrome most commonly manifest as lower-limb deep vein thrombosis or pulmonary embolism, while arterial events most commonly manifest as stroke or transient ischemic attack.³⁷ Obstetric manifestations may include not only miscarriage and stillbirth, but also preterm delivery, intrauterine growth retardation, and preeclampsia, all occurring due to placental insufficiency.

The frequency of antiphospholipid antibodies has been estimated as 13.5% in patients with stroke, 11% with myocardial infarction, 9.5% with deep vein thrombosis, and 6% for those with pregnancy morbidity.³⁸

Some noncriteria manifestations have also been recognized in antiphospholipid syndrome, such as thrombocytopenia, cardiac vegetations (Libman-Sachs endocarditis), livedo reticularis, and nephropathy.

The indications for antiphospholipid antibody testing are listed in Table 4.²⁹ For the patient described in the vignette, it would be appropriate to test for antiphospholipid antibodies because of her unprovoked thrombosis, thrombocytopenia, and prolonged aPTT. Anticoagulant treatment is known to be associated with false-positive lupus anticoagulant, so any blood samples should be drawn before such treatment is commenced.

Back to our patient

Our patient’s anticardiolipin IgG test is negative, while her lupus anticoagulant and beta-2GPI IgG are positive. She has no clinical or laboratory features suggesting lupus.

She is started on warfarin. After 3 months, the warfarin is interrupted for several days, and she is retested for all 3 antiphospholipid antibodies. Her beta-2GPI I IgG and lupus anticoagulant tests are again positive. Because of the persistent antiphospholipid antibody positivity and clinical history of deep vein thrombosis, her condition is diagnosed as primary antiphospholipid syndrome. She is advised to continue anticoagulant therapy indefinitely.

 

 

ANTINEUTROPHIL CYTOPLASMIC ANTIBODY

A 34-year-old man who is an injecting drug user presents with a 2-week history of fever, malaise, and generalized arthralgia. There are no localizing symptoms of infection. Notable findings on examination include a temperature of 38.0°C (100.4°F), needle track marks in his arms, nonblanching vasculitic rash in his legs, and a systolic murmur over the precordium.

His white blood cell count is 15.3 × 10⁹/L (reference range 3.7–11.0), and his C-reactive protein level is 234 mg/dL (normal < 3). Otherwise, results of blood cell counts, liver enzyme tests, renal function tests, urinalysis, and chest radiography are normal.

Two sets of blood cultures are drawn. Transthoracic echocardiography and the antineutrophil cytoplasmic antibody (ANCA) test are requested, as are screening tests for human immunodeficiency virus, hepatitis B, and hepatitis C.

Was the ANCA test indicated in this patient?

ANCAs are autoantibodies against antigens located in the cytoplasmic granules of neutrophils and monocytes. They are associated with small-vessel vasculitides such as granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (EGPA), and isolated pauciimmune crescentic glomerulonephritis, all collectively known as ANCA-associated vasculitis (AAV).³⁹

Laboratory methods to detect ANCA include indirect immunofluorescence and antigen-specific enzyme immunoassays. Indirect immunofluorescence only tells us whether or not an antibody that is targeting a cytoplasmic antigen is present. Based on the indirect immunofluorescent pattern, ANCA can be classified as follows:

• Perinuclear or p-ANCA (if the targeted antigen is located just around the nucleus and extends into it)
• Cytoplasmic or c-ANCA (if the targeted antigen is located farther away from the nucleus)
• Atypical ANCA (if the indirect immunofluorescent pattern does not fit with either p-ANCA or c-ANCA).

Indirect immunofluorescence does not give information about the exact antigen that is targeted; this can only be obtained by performing 1 of the antigen-specific immunoassays. The target antigen for c-ANCA is usually proteinase-3 (PR3), while that for p-ANCA could be myeloperoxidase (MPO), cathepsin, lysozyme, lactoferrin, or bactericidal permeability inhibitor. Anti-PR3 is highly specific for GPA, while anti-MPO is usually associated with MPA and EGPA. Less commonly, anti-PR3 may be seen in patients with MPA and anti-MPO in those with GPA. Hence, there is an increasing trend toward classifying ANCA-associated vasculitis into PR3-associated or MPO-associated vasculitis rather than as GPA, MPA, EGPA, or renal-limited vasculitis.⁴⁰

Several audits have shown that the ANCA test is widely misused and requested indiscriminately to rule out vasculitis. This results in a lower positive predictive value, possible harm to patients due to increased false-positive rates, and increased burden on the laboratory.^41–43 At least 2 separate groups have demonstrated that a gating policy that refuses ANCA testing in patients without clinical evidence of systemic vasculitis can reduce the number of inappropriate requests, improve the diagnostic yield, and make it more clinically relevant and cost-effective.^44,45

The clinician should bear in mind that:

ANCA testing should be requested only if the pretest probability of ANCA-associated vasculitis is high. The indications proposed by the International Consensus Statement on ANCA testing⁴⁶ are listed in Table 5. These criteria have been clinically validated, with 1 study even demonstrating that no cases of ANCA-associated vasculitis would be missed if these guidelines are followed.⁴⁷

Current guidelines recommend using one of the antigen-specific assays for PR3 and MPO as the primary screening method.⁴⁸ Until recently, indirect immunofluorescence was used to screen for ANCA-associated vasculitis, and positive results were confirmed by ELISA to detect ANCAs specific for PR3 and MPO,⁴⁹ but this is no longer recommended because of recent evidence suggesting a large variability between the different indirect immunofluorescent methods and improved diagnostic performance of the antigen-specific assays.

In a large multicenter study by Damoiseaux et al, the specificity with the different antigen-specific immunoassays was 98% to 99% for PR3-ANCA and 96% to 99% for MPO-ANCA.⁵⁰

ANCA-associated vasculitis should not be considered excluded if the PR3 and MPO-ANCA are negative. In the Damoiseaux study, about 11% to 15% of patients with GPA and 8% to 24% of patients with MPA tested negative for both PR3 and MPO-ANCA.⁵⁰

If the ANCA result is negative and clinical suspicion for ANCA-associated vasculitis is high, the clinician may wish to consider requesting another immunoassay method or indirect immunofluorescence. Results of indirect immunofluorescent testing results may be positive in those with a negative immunoassay, and vice versa.

A positive ANCA result is not diagnostic of ANCA-associated vasculitis. Numerous other conditions are associated with ANCA, usually p-ANCA or atypical ANCA (Table 6). The antigens targeted by these ANCAs are usually cathepsin, lysozyme, lactoferrin, and bactericidal permeability inhibitor.

Thus, the ANCA result should always be interpreted in the context of the whole clinical picture.⁵¹ Biopsy should still be considered the gold standard for the diagnosis of ANCA-associated vasculitis. The ANCA titer can help to improve clinical interpretation, because the likelihood of ANCA-associated vasculitis increases with higher levels of PR3 and MPO-ANCA.⁵²

Back to our patient

Our patient’s blood cultures grow methicillin-sensitive Staphylococcus aureus in both sets after 48 hours. Transthoracic echocardiography reveals vegetations around the tricuspid valve, with no evidence of valvular regurgitation. The diagnosis is right-sided infective endocarditis. He is started on appropriate antibiotics.

Tests for human immunodeficiency virus, hepatitis B, and hepatitis C are negative. The ANCA test is positive for MPO-ANCA at 28 IU/mL (normal < 10).

The positive ANCA is thought to be related to the infective endocarditis. His vasculitis is most likely secondary to infective endocarditis and not ANCA-associated vasculitis. The ANCA test need not have been requested in the first place.

 

 

HUMAN LEUKOCYTE ANTIGEN-B27

A 22-year-old man presents to his primary care physician with a 4-month history of gradually worsening low back pain associated with early morning stiffness lasting more than 2 hours. He has no peripheral joint symptoms.

In the last 2 years, he has had 2 separate episodes of uveitis. There is a family history of ankylosing spondylitis in his father. Examination reveals global restriction of lumbar movements but is otherwise unremarkable. Magnetic resonance imaging (MRI) of the lumbar spine and sacroiliac joints is normal.

Should this patient be tested for human leukocyte antigen-B27 (HLA-B27)?

The major histocompatibility complex (MHC) is a gene complex that is present in all animals. It encodes proteins that help with immunologic tolerance. HLA simply refers to the human version of the MHC.⁵³ The HLA gene complex, located on chromosome 6, is categorized into class I, class II, and class III. HLA-B is one of the 3 class I genes. Thus, a positive HLA-B27 result simply means that the particular gene is present in that person.

HLA-B27 is strongly associated with ankylosing spondylitis, also known as axial spondyloarthropathy.⁵⁴ Other genes also contribute to the pathogenesis of ankylosing spondylitis, but HLA-B27 is present in more than 90% of patients with this disease and is by far considered the most important. The association is not as strong for peripheral spondyloarthropathy, with studies reporting a frequency of up to 75% for reactive arthritis and inflammatory bowel disease-associated arthritis, and up to 50% for psoriatic arthritis and uveitis.⁵⁵

About 9% of healthy, asymptomatic individuals may have HLA-B27, so the mere presence of this gene is not evidence of disease.⁵⁶ There may be up to a 20-fold increased risk of ankylosing spondylitis among those who are HLA-B27-positive.⁵⁷

Some HLA genes have many different alleles, each of which is given a number (explaining the number 27 that follows the B). Closely related alleles that differ from one another by only a few amino-acid substitutions are then categorized together, thus accounting for more than 100 subtypes of HLA-B27 (designated from HLA-B*2701 to HLA-B*27106). These subtypes vary in frequency among different racial groups, and the population prevalence of ankylosing spondylitis parallels the frequency of HLA-B27.⁵⁸ The most common subtype seen in white people and American Indians is B*2705. HLA-B27 is rare in blacks, explaining the rarity of ankylosing spondylitis in this population. Further examples include HLA-B*2704, which is seen in Asians, and HLA-B*2702, seen in Mediterranean populations. Not all subtypes of HLA-B27 are associated with disease, and some, like HLA-B*2706, may also be protective.

When should the clinician consider testing for HLA-B27?

Not all patients with low back pain need an HLA-B27 test. First, it is important to look for clinical features of axial spondyloarthropathy (Table 7). The unifying feature of spondyloarthropathy is enthesitis (inflammation at the sites of insertion of tendons or ligaments on the skeleton). Inflammation of axial entheses causes spondylitis and sacroiliitis, manifesting as inflammatory back pain. Clinical clues to inflammatory back pain include insidious onset, aggravation with rest or inactivity, prolonged early morning stiffness, disturbed sleep during the second half of the night, relief with movement or activity, alternating gluteal pain (due to sacroiliitis), and good response to anti-inflammatory medication (although nonspecific).

Peripheral spondyloarthropathy may present with arthritis, enthesitis (eg, heel pain due to inflammation at the site of insertion of the Achilles tendon or plantar fascia), or dactylitis (“sausage” swelling of the whole finger or toe due to extension of inflammation beyond the margins of the joint). Other clues may include psoriasis, inflammatory bowel disease, history of preceding gastrointestinal or genitourinary infection, family history of similar conditions, and history of recurrent uveitis.

For the initial assessment of patients who have inflammatory back pain, plain radiography of the sacroiliac joints is considered the gold standard.⁵⁹ If plain radiography does not show evidence of sacroiliitis, MRI of the sacroiliac joints should be considered. While plain radiography can reveal only structural changes such as sclerosis, erosions, and ankylosis, MRI is useful to evaluate for early inflammatory changes such as bone marrow edema. Imaging the lumbar spine is not necessary, as the sacroiliac joints are almost invariably involved in axial spondyloarthropathy, and lesions seldom occur in the lumbar spine in isolation.⁶⁰

The diagnosis of ankylosing spondylitis previously relied on confirmatory imaging features, but based on the new International Society classification criteria,^61–63 which can be applied to patients with more than 3 months of back pain and age of onset of symptoms before age 45, patients can be classified as having 1 of the following:

• Radiographic axial spondyloarthropathy, if they have evidence of sacroiliitis on imaging plus 1 other feature of spondyloarthropathy
• Nonradiographic axial spondyloarthropathy, if they have a positive HLA-B27 plus 2 other features of spondyloarthropathy (Table 7).

These new criteria have a sensitivity of 82.9% and specificity of 84.4%.^62,63 The disease burden of radiographic and nonradiographic axial spondyloarthropathy has been shown to be similar, suggesting that they are part of the same disease spectrum. Thus, the HLA-B27 test is useful to make a diagnosis of axial spondyloarthropathy even in the absence of imaging features and could be requested in patients with 2 or more features of spondyloarthropathy. In the absence of imaging features and a negative HLA-B27 result, however, the patient cannot be classified as having axial spondyloarthropathy.

Back to our patient

The absence of radiographic evidence would not exclude axial spondyloarthropathy in our patient. The HLA-B27 test is requested because of the inflammatory back pain and the presence of 2 spondyloarthropathy features (uveitis and the family history) and is reported to be positive. His disease is classified as nonradiographic axial spondyloarthropathy.

He is started on regular naproxen and is referred to a physiotherapist. After 1 month, he reports significant symptomatic improvement. He asks if he can be retested for HLA-B27 to see if it has become negative. We tell him that there is no point in repeating it, as it is a gene and will not disappear.

SUMMARY: CONSIDER THE CLINICAL PICTURE

When approaching a patient suspected of having a rheumatologic disease, a clinician should first consider the clinical presentation and the intended purpose of each test. The tests, in general, might serve several purposes. They might help to:

Increase the likelihood of the diagnosis in question. For example, a positive rheumatoid factor or anticitrullinated peptide antibody can help diagnose rheumatoid arthritis in a patient with early polyarthritis, a positive HLA-B27 can help diagnose ankylosing spondylitis in patients with inflammatory back pain and normal imaging, and a positive ANCA can help diagnose ANCA-associated vasculitis in a patient with glomerulonephritis.

Reduce the likelihood of the diagnosis in question. For example, a negative antinuclear antibody test reduces the likelihood of lupus in a patient with joint pains.

Monitor the condition. For example DNA antibodies can be used to monitor the activity of lupus.

Plan the treatment strategy. For example, one might consider lifelong anticoagulation if antiphospholipid antibodies are persistently positive in a patient with thrombosis.

Prognosticate. For example, positive rheumatoid factor and anticitrullinated peptide antibody increase the risk of erosive rheumatoid arthritis.

If the test was requested in the absence of a clear indication and the result is positive, it is important to bear in mind the potential pitfalls associated with that test and not attach a diagnostic label prematurely. None of the tests can confirm or exclude a condition, so the results should always be interpreted in the context of the whole clinical picture.   

Laboratory tests are often ordered inappropriately for patients in whom a rheumatologic illness is suspected; this occurs in both primary and secondary care.¹ Some tests are available both singly and as part of a battery of tests screening healthy people without symptoms.

The problem: negative test results are by no means always reassuring, and false-positive results raise the risks of unnecessary anxiety for patients and clinicians, needless referrals, and potential morbidity due to further unnecessary testing and exposure to wrong treatments.² Clinicians should be aware of the pitfalls of these tests in order to choose them wisely and interpret the results correctly.

This article provides practical guidance on requesting and interpreting some common tests in rheumatology, with the aid of case vignettes.

RHEUMATOID FACTOR AND ANTICITRULLINATED PEPTIDE ANTIBODY

A 41-year-old woman, previously in good health, presents to her primary care practitioner with a 6-week history of pain and swelling in her hands and early morning stiffness lasting about 2 hours. She denies having any extraarticular symptoms. Physical examination reveals synovitis across her right metacarpophalangeal joints, proximal interphalangeal joint of the left middle finger, and left wrist. The primary care physician is concerned that her symptoms might be due to rheumatoid arthritis.

Would testing for rheumatoid factor and anticitrullinated peptide antibody be useful in this patient?

Rheumatoid factor is an antibody (immunoglobulin M, IgG, or IgA) targeted against the Fc fragment of IgG.³ It was so named because it was originally detected in patients with rheumatoid arthritis, but it is neither sensitive nor specific for this condition. A meta-analysis of more than 5,000 patients with rheumatoid arthritis reported that rheumatoid factor testing had a sensitivity of 69% and specificity of 85%.⁴

Numerous other conditions can be associated with a positive test for rheumatoid factor (Table 1). Hence, a diagnosis of rheumatoid arthritis cannot be confirmed with a positive result alone, nor can it be excluded with a negative result.

Anticitrullinated peptide antibody, on the other hand, is much more specific for rheumatoid arthritis (95%), as it is seldom seen in other conditions, but its sensitivity is similar to that of rheumatoid factor (68%).^4–6 A positive result would thus lend strength to the diagnosis of rheumatoid arthritis, but a negative result would not exclude it.

Approach to early arthritis

When faced with a patient with early arthritis, some key questions to ask include^7,8:

Is this an inflammatory or a mechanical problem? Inflammatory arthritis is suggested by joint swelling that is not due to trauma or bony hypertrophy, early morning stiffness lasting longer than 30 minutes, and elevated inflammatory markers (erythrocyte sedimentation rate or C-reactive protein). Involvement of the small joints of the hands and feet may be suggested by pain on compression of the metacarpophalangeal and metatarsophalangeal joints, respectively.

Is there a definite identifiable underlying cause for the inflammatory arthritis? The pattern of development of joint symptoms or the presence of extraarticular symptoms may suggest an underlying problem such as gout, psoriatic arthritis, systemic lupus erythematosus, or sarcoidosis.

If the arthritis is undifferentiated (ie, there is no definite identifiable cause), is it likely to remit or persist? This is perhaps the most important question to ask in order to prognosticate. Patients with risk factors for persistent disease, ie, for development of rheumatoid arthritis, should be referred to a rheumatologist early for timely institution of disease-modifying antirheumatic drug therapy.⁹ Multiple studies have shown that patients in whom this therapy is started early have much better clinical, functional, and radiologic outcomes than those in whom it is delayed.^10–12

The revised American College of Rheumatology and European League Against Rheumatism criteria¹³ include the following factors as predictors of persistence:

• Number of involved joints (with greater weight given to involvement of small joints)
• Duration of symptoms 6 weeks or longer
• Elevated acute-phase response (erythrocyte sedimentation rate or C-reactive protein level)
• A positive serologic test (either rheumatoid factor or anticitrullinated peptide antibody).

If both rheumatoid factor and anticitrullinated peptide antibody are positive in a patient with early undifferentiated arthritis, the risk of progression to rheumatoid arthritis is almost 100%, thus underscoring the importance of testing for these antibodies.^5,6 Referral to a rheumatologist should, however, not be delayed in patients with negative test results (more than one-third of patients with rheumatoid arthritis may be negative for both), and should be considered in those with inflammatory joint symptoms persisting longer than 6 weeks, especially with involvement of the small joints (sparing the distal interphalangeals) and elevated acute-phase response.

Rheumatoid factor in healthy people without symptoms

In some countries, testing for rheumatoid factor is offered as part of a battery of screening tests in healthy people who have no symptoms, a practice that should be strongly discouraged.

Multiple studies, both prospective and retrospective, have demonstrated that both rheumatoid factor and anticitrullinated peptide antibody may be present several years before the clinical diagnosis of rheumatoid arthritis.^6,14–16 But the risk of developing rheumatoid arthritis for asymptomatic individuals who are rheumatoid factor-positive depends on the rheumatoid factor titer, positive family history of rheumatoid arthritis in first-degree relatives, and copresence of anticitrullinated peptide antibody. The absolute risk, nevertheless, is still very small. In some, there might be an alternative explanation such as undiagnosed Sjögren syndrome or hepatitis C.

In any event, no strategy is currently available that is proven to prevent the development of rheumatoid arthritis, and there is no role for disease-modifying therapy during the preclinical phase.¹⁶

Back to our patient

Blood testing in our patient reveals normal complete blood cell counts, aminotransferase levels, and serum creatinine concentration; findings on urinalysis are normal. Her erythrocyte sedimentation rate is 56 mm/hour (reference range 0–15), and her C-reactive protein level is 26 mg/dL (normal < 3). Testing is negative for rheumatoid factor and anticitrullinated peptide antibody.

Although her rheumatoid factor and anticitrullinated peptide antibody tests are negative, she is referred to a rheumatologist because she has predictors of persistent disease, ie, symptom duration of 6 weeks, involvement of the small joints of the hands, and elevated erythrocyte sedimentation rate and C-reactive protein. The rheumatologist checks her parvovirus serology, which is negative.

The patient is given parenteral depot corticosteroid therapy, to which she responds briefly. Because her symptoms persist and continue to worsen, methotrexate treatment is started after an additional 6 weeks.

 

 

ANTINUCLEAR ANTIBODY

A 37-year-old woman presents to her primary care physician with the complaint of tiredness. She has a family history of systemic lupus erythematosus in her sister and maternal aunt. She is understandably worried about lupus because of the family history and is asking to be tested for it.

Would testing for antinuclear antibody be reasonable?

Antinuclear antibody is not a single antibody but rather a family of autoantibodies that are directed against nuclear constituents such as single- or double-stranded deoxyribonucleic acid (dsDNA), histones, centromeres, proteins complexed with ribonucleic acid (RNA), and enzymes such as topoisomerase.^17,18

Protein antigens complexed with RNA and some enzymes in the nucleus are also known as extractable nuclear antigens (ENAs). They include Ro, La, Sm, Jo-1, RNP, and ScL-70 and are named after the patient in whom they were first discovered (Robert, Lavine, Smith, and John), the antigen that is targeted (ribonucleoprotein or RNP), and the disease with which they are associated (anti-ScL-70 or antitopoisomerase in diffuse cutaneous scleroderma).

Antinuclear antibody testing is commonly requested to exclude connective tissue diseases such as lupus, but the clinician needs to be aware of the following points:

Antinuclear antibody may be encountered in conditions other than lupus

These include¹⁹:

• Other autoimmune diseases such as rheumatoid arthritis, primary Sjögren syndrome, systemic sclerosis, autoimmune thyroid disease, and myasthenia gravis
• Infection with organisms that share the epitope with self-antigens (molecular mimicry)
• Cancers
• Drugs such as hydralazine, procainamide, and minocycline.

Antinuclear antibody might also be produced by the healthy immune system from time to time to clear the nuclear debris that is extruded from aging cells.

A study in healthy individuals²⁰ reported a prevalence of positive antinuclear antibody of 32% at a titer of 1/40, 15% at a titer of 1/80, 7% at a titer of 1/160, and 3% at a titer of 1/320. Importantly, a positive result was more common among family members of patients with autoimmune connective tissue diseases.²¹ Hence, a positive antinuclear antibody result does not always mean lupus.

Antinuclear antibody testing is highly sensitive for lupus

With current laboratory methods, antinuclear antibody testing has a sensitivity close to 100%. Hence, a negative result virtually rules out lupus.

Two methods are commonly used to test for antinuclear antibody: indirect immunofluorescence and enzyme-linked immunosorbent assay (ELISA).²² While human epithelial (Hep2) cells are used as the source of antigen in immunofluorescence, purified nuclear antigens coated on multiple-well plates are used in ELISA.

Although ELISA is simpler to perform, immunofluorescence has a slightly better sensitivity (because the Hep2 cells express a wide range of antigens) and is still considered the gold standard. As expected, the higher sensitivity occurs at the cost of reduced specificity (about 60%), so antinuclear antibody will also be detected in all the other conditions listed above.²³

To improve the specificity of antinuclear antibody testing, laboratories report titers (the highest dilution of the test serum that tested positive); a cutoff of greater than 1/80 is generally considered significant.

Do not order antinuclear antibody testing indiscriminately

If the antinuclear antibody test is requested indiscriminately, the positive predictive value for the diagnosis of lupus is only 11%.²⁴ The test should be requested only when the pretest probability of lupus or other connective tissue disease is high. The positive predictive value is much higher in patients presenting with clinical or laboratory manifestations involving 2 or more organ systems (Table 2).^18,25

Categorization of the specific antigen target improves disease specificity. The antinuclear antibody in patients with lupus may be targeted against single- or double-stranded DNA, histones, or 1 or more of the ENAs. Among these, the presence of anti-dsDNA or anti-Sm is highly specific for a diagnosis of lupus (close to 100%). Neither is sensitive for lupus, however, with anti-dsDNA present in only 60% of patients with lupus and anti-Sm in about 30%.¹⁷ Hence, patients with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may continue to pose a diagnostic challenge. Other examples of specific disease associations are listed in Table 3.

To sum up, the antinuclear antibody test should be requested only in patients with involvement of multiple organ systems. Although a negative result would make it extremely unlikely that the clinical presentation is due to lupus, a positive result is insufficient on its own to make a diagnosis of lupus.

Diagnosing lupus is straightforward when patients present with a specific manifestation such as inflammatory arthritis, photosensitive skin rash, hemolytic anemia, thrombocytopenia, or nephritis, or with specific antibodies such as those against dsDNA or Sm. Patients who present with nonspecific symptoms such as arthralgia or tiredness with a positive antinuclear antibody and negative anti-dsDNA and anti-Sm may present difficulties even for the specialist.^25–27

Back to our patient

Our patient denies arthralgia. She has no extraarticular symptoms such as skin rashes, oral ulcers, sicca symptoms, muscle weakness, Raynaud phenomenon, pleuritic chest pain, or breathlessness. Findings on physical examination and urinalysis are unremarkable.

Her primary care physician decides to check her complete blood cell count, erythrocyte sedimentation rate, and thyroid-stimulating hormone level. Although she is reassured that her tiredness is not due to lupus, she insists on getting an antinuclear antibody test.

Her complete blood cell counts are normal. Her erythrocyte sedimentation rate is 6 mm/hour. However, her thyroid-stimulating hormone level is elevated, and subsequent testing shows low free thyroxine and positive thyroid peroxidase antibodies. The antinuclear antibody is positive in a titer of 1/80 and negative for anti-dsDNA and anti-ENA.

We explain to her that the positive antinuclear antibody is most likely related to her autoimmune thyroid disease. She is referred to an endocrinologist.

 

 

ANTIPHOSPHOLIPID ANTIBODIES

A 24-year-old woman presents to the emergency department with acute unprovoked deep vein thrombosis in her right leg, confirmed by ultrasonography. She has no history of previous thrombosis, and the relevant family history is unremarkable. She has never been pregnant. Her platelet count is 84 × 10⁹/L (reference range 150–400), and her baseline activated partial thromboplastin time is prolonged at 62 seconds (reference range 23.0–32.4). The rest of her blood counts and her prothrombin time, liver enzyme levels, and serum creatinine level are normal.

Should this patient be tested for antiphospholipid antibodies?

Antiphospholipid antibodies are important because of their association with thrombotic risk (both venous and arterial) and pregnancy morbidity. The name is a misnomer, as these antibodies are targeted against some proteins that are bound to phospholipids and not only to the phospholipids themselves.

According to the modified Sapporo criteria for the classification of antiphospholipid syndrome,²⁸ antiphospholipid antibodies should remain persistently positive on at least 2 separate occasions at least 12 weeks apart for the result to be considered significant because some infections and drugs may be associated with the transient presence of antiphospholipid antibodies.

Screening for antiphospholipid antibodies should include testing for IgM and IgG anticardiolipin antibodies, lupus anticoagulant, and IgM and IgG beta-2 glycoprotein I antibodies.^29,30

Anticardiolipin antibodies

Anticardiolipin (aCL) antibodies may be targeted either against beta-2 glycoprotein I (beta-2GPI) that is bound to cardiolipin (a phospholipid) or against cardiolipin alone; the former is more specific. Antibodies directed against cardiolipin alone are usually transient and are associated with infections and drugs. The result is considered significant only when anticardiolipin antibodies are present in a medium to high titer (> 40 IgG phospholipid units or IgM phospholipid units, or > 99th percentile).

Lupus anticoagulant

The antibody with “lupus anticoagulant activity” is targeted against prothrombin plus phospholipid or beta-2GPI plus phospholipid. The test for it is a functional assay involving 3 steps:

Demonstrating the prolongation of a phospholipid-dependent coagulation assay like the activated partial thromboplastin time (aPTT). (This may explain the prolongation of aPTT in the patient described in the vignette.) Although the presence of lupus anticoagulant is associated with thrombosis, it is called an “anticoagulant” because of this in vitro prolongation of phospholipid-dependent coagulation assays.

Mixing study. The phospholipid-dependent coagulation assay could be prolonged because of either the deficiency of a coagulation factor or the presence of the antiphospholipid antibodies. This can be differentiated by mixing the patient’s plasma with normal plasma (which will have all the clotting factors) in a 1:1 ratio. If the coagulation assay remains prolonged after the addition of normal plasma, clotting factor deficiency can be excluded.

Addition of a phospholipid. If the prolongation of the coagulation assay is due to the presence of an antiphospholipid antibody, addition of extra phospholipid will correct this.

Beta-2 glycoprotein I antibody (anti-beta-2GPI)

The beta-2GPI that is not bound to the cardiolipin can be detected by separately testing for beta-2GPI (the anticardiolipin test only detects the beta-2GPI that is bound to the cardiolipin). The result is considered significant if beta-2GPI is present in a medium to high titer (> 99th percentile).

Studies have shown that antiphospholipid antibodies may be present in 1% to 5% of apparently healthy people in the general population.³¹ These are usually low-titer anticardiolipin or anti-beta-GPI IgM antibodies that are not associated with thrombosis or adverse pregnancy outcomes. Hence, the term antiphospholipid syndrome should be reserved for those who have had at least 1 episode of thrombosis or pregnancy morbidity and persistent antiphospholipid antibodies, and not those who have asymptomatic or transient antiphospholipid antibodies.

Triple positivity (positive anticardiolipin, lupus anticoagulant, and anti-beta-2GPI) seems to be associated with the highest risk of thrombosis, with a 10-year cumulative incidence of 37.1% (95% confidence interval [CI] 19.9–54.3) for a first thrombotic event,³² and 44.2% (95% CI 38.6–49.8) for recurrent thrombosis.³³

The association with thrombosis is stronger for lupus anticoagulant than with the other 2 antibodies, with different studies³⁴ finding an odds ratio ranging from 5 to 16. A positive lupus anticoagulant test with or without a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a high-risk profile, while a moderate to high titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a moderate-risk profile. A low titer of anticardiolipin or anti-beta-2GPI IgM or IgG constitutes a low-risk profile that may not be associated with thrombosis.³⁵

Antiphospholipid syndrome is important to recognize because of the need for long-term anticoagulation to prevent recurrence.³⁶ It may be primary, when it occurs on its own, or secondary, when it occurs in association with another autoimmune disease such as lupus.

Venous events in antiphospholipid syndrome most commonly manifest as lower-limb deep vein thrombosis or pulmonary embolism, while arterial events most commonly manifest as stroke or transient ischemic attack.³⁷ Obstetric manifestations may include not only miscarriage and stillbirth, but also preterm delivery, intrauterine growth retardation, and preeclampsia, all occurring due to placental insufficiency.

The frequency of antiphospholipid antibodies has been estimated as 13.5% in patients with stroke, 11% with myocardial infarction, 9.5% with deep vein thrombosis, and 6% for those with pregnancy morbidity.³⁸

Some noncriteria manifestations have also been recognized in antiphospholipid syndrome, such as thrombocytopenia, cardiac vegetations (Libman-Sachs endocarditis), livedo reticularis, and nephropathy.

The indications for antiphospholipid antibody testing are listed in Table 4.²⁹ For the patient described in the vignette, it would be appropriate to test for antiphospholipid antibodies because of her unprovoked thrombosis, thrombocytopenia, and prolonged aPTT. Anticoagulant treatment is known to be associated with false-positive lupus anticoagulant, so any blood samples should be drawn before such treatment is commenced.

Back to our patient

Our patient’s anticardiolipin IgG test is negative, while her lupus anticoagulant and beta-2GPI IgG are positive. She has no clinical or laboratory features suggesting lupus.

She is started on warfarin. After 3 months, the warfarin is interrupted for several days, and she is retested for all 3 antiphospholipid antibodies. Her beta-2GPI I IgG and lupus anticoagulant tests are again positive. Because of the persistent antiphospholipid antibody positivity and clinical history of deep vein thrombosis, her condition is diagnosed as primary antiphospholipid syndrome. She is advised to continue anticoagulant therapy indefinitely.

 

 

ANTINEUTROPHIL CYTOPLASMIC ANTIBODY

A 34-year-old man who is an injecting drug user presents with a 2-week history of fever, malaise, and generalized arthralgia. There are no localizing symptoms of infection. Notable findings on examination include a temperature of 38.0°C (100.4°F), needle track marks in his arms, nonblanching vasculitic rash in his legs, and a systolic murmur over the precordium.

His white blood cell count is 15.3 × 10⁹/L (reference range 3.7–11.0), and his C-reactive protein level is 234 mg/dL (normal < 3). Otherwise, results of blood cell counts, liver enzyme tests, renal function tests, urinalysis, and chest radiography are normal.

Two sets of blood cultures are drawn. Transthoracic echocardiography and the antineutrophil cytoplasmic antibody (ANCA) test are requested, as are screening tests for human immunodeficiency virus, hepatitis B, and hepatitis C.

Was the ANCA test indicated in this patient?

ANCAs are autoantibodies against antigens located in the cytoplasmic granules of neutrophils and monocytes. They are associated with small-vessel vasculitides such as granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (EGPA), and isolated pauciimmune crescentic glomerulonephritis, all collectively known as ANCA-associated vasculitis (AAV).³⁹

Laboratory methods to detect ANCA include indirect immunofluorescence and antigen-specific enzyme immunoassays. Indirect immunofluorescence only tells us whether or not an antibody that is targeting a cytoplasmic antigen is present. Based on the indirect immunofluorescent pattern, ANCA can be classified as follows:

• Perinuclear or p-ANCA (if the targeted antigen is located just around the nucleus and extends into it)
• Cytoplasmic or c-ANCA (if the targeted antigen is located farther away from the nucleus)
• Atypical ANCA (if the indirect immunofluorescent pattern does not fit with either p-ANCA or c-ANCA).

Indirect immunofluorescence does not give information about the exact antigen that is targeted; this can only be obtained by performing 1 of the antigen-specific immunoassays. The target antigen for c-ANCA is usually proteinase-3 (PR3), while that for p-ANCA could be myeloperoxidase (MPO), cathepsin, lysozyme, lactoferrin, or bactericidal permeability inhibitor. Anti-PR3 is highly specific for GPA, while anti-MPO is usually associated with MPA and EGPA. Less commonly, anti-PR3 may be seen in patients with MPA and anti-MPO in those with GPA. Hence, there is an increasing trend toward classifying ANCA-associated vasculitis into PR3-associated or MPO-associated vasculitis rather than as GPA, MPA, EGPA, or renal-limited vasculitis.⁴⁰

Several audits have shown that the ANCA test is widely misused and requested indiscriminately to rule out vasculitis. This results in a lower positive predictive value, possible harm to patients due to increased false-positive rates, and increased burden on the laboratory.^41–43 At least 2 separate groups have demonstrated that a gating policy that refuses ANCA testing in patients without clinical evidence of systemic vasculitis can reduce the number of inappropriate requests, improve the diagnostic yield, and make it more clinically relevant and cost-effective.^44,45

The clinician should bear in mind that:

ANCA testing should be requested only if the pretest probability of ANCA-associated vasculitis is high. The indications proposed by the International Consensus Statement on ANCA testing⁴⁶ are listed in Table 5. These criteria have been clinically validated, with 1 study even demonstrating that no cases of ANCA-associated vasculitis would be missed if these guidelines are followed.⁴⁷

Current guidelines recommend using one of the antigen-specific assays for PR3 and MPO as the primary screening method.⁴⁸ Until recently, indirect immunofluorescence was used to screen for ANCA-associated vasculitis, and positive results were confirmed by ELISA to detect ANCAs specific for PR3 and MPO,⁴⁹ but this is no longer recommended because of recent evidence suggesting a large variability between the different indirect immunofluorescent methods and improved diagnostic performance of the antigen-specific assays.

In a large multicenter study by Damoiseaux et al, the specificity with the different antigen-specific immunoassays was 98% to 99% for PR3-ANCA and 96% to 99% for MPO-ANCA.⁵⁰

ANCA-associated vasculitis should not be considered excluded if the PR3 and MPO-ANCA are negative. In the Damoiseaux study, about 11% to 15% of patients with GPA and 8% to 24% of patients with MPA tested negative for both PR3 and MPO-ANCA.⁵⁰

If the ANCA result is negative and clinical suspicion for ANCA-associated vasculitis is high, the clinician may wish to consider requesting another immunoassay method or indirect immunofluorescence. Results of indirect immunofluorescent testing results may be positive in those with a negative immunoassay, and vice versa.

A positive ANCA result is not diagnostic of ANCA-associated vasculitis. Numerous other conditions are associated with ANCA, usually p-ANCA or atypical ANCA (Table 6). The antigens targeted by these ANCAs are usually cathepsin, lysozyme, lactoferrin, and bactericidal permeability inhibitor.

Thus, the ANCA result should always be interpreted in the context of the whole clinical picture.⁵¹ Biopsy should still be considered the gold standard for the diagnosis of ANCA-associated vasculitis. The ANCA titer can help to improve clinical interpretation, because the likelihood of ANCA-associated vasculitis increases with higher levels of PR3 and MPO-ANCA.⁵²

Back to our patient

Our patient’s blood cultures grow methicillin-sensitive Staphylococcus aureus in both sets after 48 hours. Transthoracic echocardiography reveals vegetations around the tricuspid valve, with no evidence of valvular regurgitation. The diagnosis is right-sided infective endocarditis. He is started on appropriate antibiotics.

Tests for human immunodeficiency virus, hepatitis B, and hepatitis C are negative. The ANCA test is positive for MPO-ANCA at 28 IU/mL (normal < 10).

The positive ANCA is thought to be related to the infective endocarditis. His vasculitis is most likely secondary to infective endocarditis and not ANCA-associated vasculitis. The ANCA test need not have been requested in the first place.

 

 

HUMAN LEUKOCYTE ANTIGEN-B27

A 22-year-old man presents to his primary care physician with a 4-month history of gradually worsening low back pain associated with early morning stiffness lasting more than 2 hours. He has no peripheral joint symptoms.

In the last 2 years, he has had 2 separate episodes of uveitis. There is a family history of ankylosing spondylitis in his father. Examination reveals global restriction of lumbar movements but is otherwise unremarkable. Magnetic resonance imaging (MRI) of the lumbar spine and sacroiliac joints is normal.

Should this patient be tested for human leukocyte antigen-B27 (HLA-B27)?

The major histocompatibility complex (MHC) is a gene complex that is present in all animals. It encodes proteins that help with immunologic tolerance. HLA simply refers to the human version of the MHC.⁵³ The HLA gene complex, located on chromosome 6, is categorized into class I, class II, and class III. HLA-B is one of the 3 class I genes. Thus, a positive HLA-B27 result simply means that the particular gene is present in that person.

HLA-B27 is strongly associated with ankylosing spondylitis, also known as axial spondyloarthropathy.⁵⁴ Other genes also contribute to the pathogenesis of ankylosing spondylitis, but HLA-B27 is present in more than 90% of patients with this disease and is by far considered the most important. The association is not as strong for peripheral spondyloarthropathy, with studies reporting a frequency of up to 75% for reactive arthritis and inflammatory bowel disease-associated arthritis, and up to 50% for psoriatic arthritis and uveitis.⁵⁵

About 9% of healthy, asymptomatic individuals may have HLA-B27, so the mere presence of this gene is not evidence of disease.⁵⁶ There may be up to a 20-fold increased risk of ankylosing spondylitis among those who are HLA-B27-positive.⁵⁷

Some HLA genes have many different alleles, each of which is given a number (explaining the number 27 that follows the B). Closely related alleles that differ from one another by only a few amino-acid substitutions are then categorized together, thus accounting for more than 100 subtypes of HLA-B27 (designated from HLA-B*2701 to HLA-B*27106). These subtypes vary in frequency among different racial groups, and the population prevalence of ankylosing spondylitis parallels the frequency of HLA-B27.⁵⁸ The most common subtype seen in white people and American Indians is B*2705. HLA-B27 is rare in blacks, explaining the rarity of ankylosing spondylitis in this population. Further examples include HLA-B*2704, which is seen in Asians, and HLA-B*2702, seen in Mediterranean populations. Not all subtypes of HLA-B27 are associated with disease, and some, like HLA-B*2706, may also be protective.

When should the clinician consider testing for HLA-B27?

Not all patients with low back pain need an HLA-B27 test. First, it is important to look for clinical features of axial spondyloarthropathy (Table 7). The unifying feature of spondyloarthropathy is enthesitis (inflammation at the sites of insertion of tendons or ligaments on the skeleton). Inflammation of axial entheses causes spondylitis and sacroiliitis, manifesting as inflammatory back pain. Clinical clues to inflammatory back pain include insidious onset, aggravation with rest or inactivity, prolonged early morning stiffness, disturbed sleep during the second half of the night, relief with movement or activity, alternating gluteal pain (due to sacroiliitis), and good response to anti-inflammatory medication (although nonspecific).

Peripheral spondyloarthropathy may present with arthritis, enthesitis (eg, heel pain due to inflammation at the site of insertion of the Achilles tendon or plantar fascia), or dactylitis (“sausage” swelling of the whole finger or toe due to extension of inflammation beyond the margins of the joint). Other clues may include psoriasis, inflammatory bowel disease, history of preceding gastrointestinal or genitourinary infection, family history of similar conditions, and history of recurrent uveitis.

For the initial assessment of patients who have inflammatory back pain, plain radiography of the sacroiliac joints is considered the gold standard.⁵⁹ If plain radiography does not show evidence of sacroiliitis, MRI of the sacroiliac joints should be considered. While plain radiography can reveal only structural changes such as sclerosis, erosions, and ankylosis, MRI is useful to evaluate for early inflammatory changes such as bone marrow edema. Imaging the lumbar spine is not necessary, as the sacroiliac joints are almost invariably involved in axial spondyloarthropathy, and lesions seldom occur in the lumbar spine in isolation.⁶⁰

The diagnosis of ankylosing spondylitis previously relied on confirmatory imaging features, but based on the new International Society classification criteria,^61–63 which can be applied to patients with more than 3 months of back pain and age of onset of symptoms before age 45, patients can be classified as having 1 of the following:

• Radiographic axial spondyloarthropathy, if they have evidence of sacroiliitis on imaging plus 1 other feature of spondyloarthropathy
• Nonradiographic axial spondyloarthropathy, if they have a positive HLA-B27 plus 2 other features of spondyloarthropathy (Table 7).

These new criteria have a sensitivity of 82.9% and specificity of 84.4%.^62,63 The disease burden of radiographic and nonradiographic axial spondyloarthropathy has been shown to be similar, suggesting that they are part of the same disease spectrum. Thus, the HLA-B27 test is useful to make a diagnosis of axial spondyloarthropathy even in the absence of imaging features and could be requested in patients with 2 or more features of spondyloarthropathy. In the absence of imaging features and a negative HLA-B27 result, however, the patient cannot be classified as having axial spondyloarthropathy.

Back to our patient

The absence of radiographic evidence would not exclude axial spondyloarthropathy in our patient. The HLA-B27 test is requested because of the inflammatory back pain and the presence of 2 spondyloarthropathy features (uveitis and the family history) and is reported to be positive. His disease is classified as nonradiographic axial spondyloarthropathy.

He is started on regular naproxen and is referred to a physiotherapist. After 1 month, he reports significant symptomatic improvement. He asks if he can be retested for HLA-B27 to see if it has become negative. We tell him that there is no point in repeating it, as it is a gene and will not disappear.

SUMMARY: CONSIDER THE CLINICAL PICTURE

When approaching a patient suspected of having a rheumatologic disease, a clinician should first consider the clinical presentation and the intended purpose of each test. The tests, in general, might serve several purposes. They might help to:

Increase the likelihood of the diagnosis in question. For example, a positive rheumatoid factor or anticitrullinated peptide antibody can help diagnose rheumatoid arthritis in a patient with early polyarthritis, a positive HLA-B27 can help diagnose ankylosing spondylitis in patients with inflammatory back pain and normal imaging, and a positive ANCA can help diagnose ANCA-associated vasculitis in a patient with glomerulonephritis.

Reduce the likelihood of the diagnosis in question. For example, a negative antinuclear antibody test reduces the likelihood of lupus in a patient with joint pains.

Monitor the condition. For example DNA antibodies can be used to monitor the activity of lupus.

Plan the treatment strategy. For example, one might consider lifelong anticoagulation if antiphospholipid antibodies are persistently positive in a patient with thrombosis.

Prognosticate. For example, positive rheumatoid factor and anticitrullinated peptide antibody increase the risk of erosive rheumatoid arthritis.

If the test was requested in the absence of a clear indication and the result is positive, it is important to bear in mind the potential pitfalls associated with that test and not attach a diagnostic label prematurely. None of the tests can confirm or exclude a condition, so the results should always be interpreted in the context of the whole clinical picture.   

NOTE: The scenario presented here is partly based on cases reported elsewhere by Martínez-Valles et al¹ and Fernández-Rodríguez et al.²

A 55-year-old man is admitted to the hospital with generalized malaise, paresthesias, and severe hypertension. He says he had experienced agitation along with weakness on exertion 24 hours before presentation to the emergency department, with subsequent onset of paresthesias in his lower extremities and perioral area.

He is already known to have mild chronic obstructive pulmonary disease, with a ratio of forced expiratory volume in 1 second (FEV₁)to forced vital capacity (FVC) of less than 70% and an FEV₁ 85% of predicted. In addition, he was recently diagnosed with diabetes, resistant hypertension requiring maximum doses of 3 agents (a calcium channel blocker, an angiotensin-converting enzyme inhibitor, and a loop diuretic), and hyperlipidemia.

He is a current smoker with a 30-pack-year smoking history. He does not use alcohol. His family history is noncontributory.

His blood pressure is 190/110 mm Hg despite adherence to his 3-drug regimen. His oxygen saturation is 94% on room air, respiratory rate in the low 30s, and pulse 110 beats/minute. He has normal breath sounds, normal S1 and S2 with an S4 gallop, bilateral lower-extremity edema, truncal obesity, and abdominal striae. Electrocardiography shows tachycardia with first-degree atrioventicular block. Chest radiography shows an opacity in the right middle lung field. Initial laboratory results and those from 1 year ago are shown in Table 1.

ASSESSING ACID-BASE DISORDERS

1. What type of acid-base disorder does this patient have?

• Metabolic acidosis
• Respiratory acidosis
• Metabolic alkalosis
• Respiratory alkalosis

The patient has metabolic alkalosis.

A 5-step approach

If a patient has an acid-base disorder, one should use a 5-step process to characterize it (Table 2).³

1. Acidosis or alkalosis? The patient’s arterial pH is 7.5, which is alkalemic because it is higher than 7.44.

2. Metabolic or respiratory? The primary process in our patient is overwhelmingly metabolic, as his partial pressure of carbon dioxide (Pco₂) is slightly elevated, a direction that would cause acidosis, not alkalosis.

3. The anion gap (the serum sodium concentration minus the sum of the chloride and bicarbonate concentrations) is normal at 8 mmol/L (DRG:HYBRiD-XL Immunoassay and Clinical Chemistry Analyzer, reference range 8–16).

4. Is the disturbance compensated? We have determined that this patient has a metabolic alkalemia; the question now is whether there is any compensation for the primary disturbance.

In metabolic alkalosis, the Pco₂ may increase by approximately 0.6 mm Hg (range 0.5–0.8) above the nominal normal level of 40 mm Hg for each 1-mmol/L increase in bicarbonate above the nominal normal level of 25 mmol/L.⁴ If the patient requires oxygen, the calculation may be unreliable, however, as hypoxemia may have an overriding influence on respiratory drive.

Patients with chronically high Pco₂ levels such as those with chronic obstructive pulmonary disease can become accustomed to high carbon dioxide levels and lose their hyper-
capnic respiratory drive. Giving oxygen supplementation is thought to decrease respiratory drive in these patients, so that they will breathe slower and retain more carbon dioxide. There is some degree of respiratory compensation for metabolic alkalosis that occurs by breathing less, though it is limited overall—even in very alkalotic patients, breathing less results in CO₂ retention, which, by displacing O₂ molecules in the alveoli, will in turn result in hypoxia. The brain then senses the hypoxia and makes one breathe faster, thereby limiting this compensation. 

This patient’s serum bicarbonate level is 40 mmol/L, or 15 mmol/L higher than the nominal normal level. If he is compensating, his Pco₂ should be 40 + (15 × 0.6) = 49 mm Hg, and in fact it is 51 mm Hg, which is within the normal range of expected compensation (47.5–52 mm Hg). Therefore, yes, he is compensating for the primary disturbance.  

5. In metabolic acidosis, is there a delta gap? As our patient has metabolic alkalosis, not acidosis, this question does not apply in this case.

 

 

WHICH TEST TO FIND THE CAUSE?

2. Which is the best test to order next to determine the cause of this patient’s hypokalemic metabolic alkalosis?

• Serum magnesium level
• Spot urine chloride
• Renal ultrasonography
• 24-hour urine collection for sodium, potassium, and chloride

The first step in the algorithm for hypokalemic metabolic alkalosis (Figure 1) is to obtain a spot urine chloride measurement. If this value is low, the hypokalemic metabolic alkalosis is volume-responsive; if it is high, the disturbance is volume-independent.

The patient’s loop diuretic is withheld for 12 hours and a spot urine chloride is obtained, which is reported as 44 mmol/L. This high value suggests that a volume-independent hypo­kalemic metabolic alkalosis is present with potassium depletion.

As for the other answer choices:

Serum magnesium. Though hypomagnesemia can cause hypokalemia due to lack of inhibition of renal outer medullary potassium channels and subsequent increased excretion of potassium in the apical tubular membrane, it is not independently associated with acid-base disturbances.⁵

Renal ultrasonography gives information about structural kidney disease but is of limited utility in identifying the cause of hypokalemic metabolic alkalosis.

A 24-hour urine collection is unnecessary in this setting and would ultimately result in delay in diagnosis, as spot urine chloride is a more efficient means of rapidly distinguishing volume-responsive vs volume-independent causes of hypokalemic metabolic alkalosis.⁶

IS HIS HYPERTENSION SECONDARY? IF SO, WHAT IS THE CAUSE?

Several features of this case suggest that the patient’s hypertension is secondary rather than primary. It is of recent onset. The patient’s family history is noncontributory, and his hypertension is resistant to the use of maximum doses of 3 antihypertensive agents.

3. Which of the following causes of secondary hypertension is not commonly associated with hypokalemia and metabolic alkalosis?

• Hyperaldosteronism
• Liddle syndrome
• Cushing syndrome
• Renal parenchymal disease
• Chronic licorice ingestion

Renal parenchymal disease is a cause of resistant hypertension, but it is not characterized by metabolic alkalosis, hypokalemia, and  elevated urine chloride,⁷ while the others listed here—hyperaldosteronism, Liddle syndrome, Cushing syndrome, and chronic licorice ingestion­—are. Other common causes of resistant hypertension without these metabolic abnormalities include obstructive sleep apnea, alcohol abuse, and nonadherence to treatment.

While treatment of hypertension with loop diuretics can result in hypokalemia and metabolic alkalosis due to the effect of these drugs on potassium reabsorption in the loop of Henle, the patient’s hypokalemia persisted after this agent was withdrawn.⁸

Causes of hypokalemic metabolic alkalosis with and without hypertension are further delineated in Figure 1.

Additional diagnostic testing: Plasma renin and plasma aldosterone

At this juncture, the differential diagnosis for this patient’s potassium depletion, metabolic alkalosis, high urine chloride, and hypertension has been narrowed to primary or secondary hyperaldosteronism, surreptitious mineralocorticoid ingestion, Cushing syndrome, licorice ingestion, Liddle syndrome, or one of the 3 hydroxylase deficiencies (11-, 17-, and 21-) (Figure 1).

Although clues in the history, physical examination, and imaging may suggest a specific cause of his abnormal laboratory values, the next step in the diagnostic workup is to measure the plasma renin and aldosterone levels (Table 3).

 

HYPERALDOSTERONISM

4. Hyperaldosteronism is associated with which of the following patterns of renin and aldosterone values?

• High renin, high aldosterone, normal ratio of plasma aldosterone concentration (PAC) to plasma renin activity (PRA)
• Low renin, low aldosterone, normal PAC–PRA ratio
• Low renin, high aldosterone, high PAC–PRA ratio
• High renin, low aldosterone, low PAC–PRA ratio

The pattern of low renin, high aldosterone, and high PAC–PRA ratio is associated with hyperaldosteronism.

Primary hyperaldosteronism

Primary hyperaldosteronism is one of the most common causes of resistant hypertension and is underappreciated, being diagnosed in up to 20% of patients referred to hypertension specialty clinics.⁷ Potassium levels may be normal, likely contributing to its lack of recognition in this target population.

Primary hyperaldosteronism should be suspected in patients who have a plasma aldosterone PAC–PRA ratio greater than 20 with elevated plasma aldosterone concentrations
(> 15 ng/dL).

Persistently elevated aldosterone levels in the setting of elevated plasma volume is proof that aldosterone secretion is independent of the renin-angiotensin-aldosterone axis, and therefore is autonomous (secondary to adrenal tumor or hyperplasia). Further testing in the form of oral salt loading, saline infusion, or fludrocortisone (a sodium-retaining steroid) administration is thus required to confirm inappropriate, autonomous aldosterone secretion.⁹

After establishing the diagnosis of primary hyperaldosteronism, one should determine the subtype (ie, due to an adrenal carcinoma, unilateral hypersecreting adenoma, or unilateral or bilateral hyperplasia). Further testing includes adrenal computed tomography (CT) to rule out adrenal carcinomas, which are suspected with adenomas larger than 4 cm. Though part of the diagnostic workup, CT as a means of confirmational testing alone does not preclude the possibility of bilateral adrenal hyperplasia in some patients, even in the presence of an adrenal adenoma. For this reason, adrenal venous sampling is required to definitively determine whether the condition is due to a hypersecreting adrenal adenoma or unilateral or bilateral hyperplasia.^9,10

Treatment of primary hyperaldosteronism depends on the subtype of the disease and involves salt restriction in addition to an aldosterone antagonist (spironolactone or eplerenone in the case of bilateral disease) or surgery (unilateral disease).^9,11,12

Secondary hyperaldosteronism

Secondary hyperaldosteronism should be suspected when plasma renin and aldosterone levels are both elevated with a PAC–PRA ratio less than 10.

This pattern is most commonly seen with diuretic use but can also be a consequence of renal artery stenosis or, rarely, a renin-secreting tumor.¹³ Renal artery stenosis is a common finding in patients with hypertension undergoing cardiac catheterization, which is not surprising as more than 90% of such stenoses are atherosclerotic.⁷ Renin-secreting tumors are exceedingly rare, with fewer than 100 cases reported in the literature, and are more common in younger individuals.¹³

Our patient has low-normal aldosterone and plasma renin

On further testing, this patient’s plasma aldosterone level is 2.55 ng/dL (normal < 15 ng/dL), his plasma renin activity is 0.53 ng/mL/hour (normal 0.2–2.8 ng/mL/hour), and his PAC–PRA ratio is therefore 4.81.

The categories discussed thus far have included primary and secondary hyperaldosteronism, which typically do not present with low to normal levels of both renin and aldosterone. Surreptitious mineralocorticoid use could present in this manner, but is unlikely in this patient, whose medications do not include fludrocortisone.

The low-normal values thus lead to consideration of a third category: apparent mineralocorticoid excess. Diseases in this category such as Cushing disease or adrenocorticotropic hormone (ACTH) excess are characterized by increases in corticosteroids so that the potassium depletion, metabolic alkalosis, and hypertension are not a consequence of renin and aldosterone but rather the excess corticosteroids.¹⁴

Causes of apparent mineralocorticoid excess

There are several possible causes of mineralocorticoid excess associated with hypertension and hypokalemic metabolic alkalosis not due to renin and aldosterone.

Chronic licorice ingestion in high volumes is one such cause and is thought to result in inhibition of 11B-hydroxysteroid dehydrogenase or possibly cortisol oxidase by licorice’s active component, glycyrrhetinic acid. This inhibition results in an inability to convert cortisol to cortisone. The cortisol excess binds to mineralocorticoid receptors, and acting like aldosterone, results in hypertension and hypokalemic metabolic alkalosis as well as feedback inhibition of renin and aldosterone levels.¹⁵

Partial hydroxylase deficiencies, though rare, should also be considered as a cause of hypokalemic metabolic alkalosis, hypertension, and, potentially, hirsutism and clitoromegaly in women. They can be diagnosed with elevated levels of 17-ketosteroids and dehydroepiandrosterone sulfate, both of which, in excess, may act on aldosterone receptors in a manner similar to cortisol.¹⁶

Liddle syndrome, a rare autosomal dominant condition, may also present with suppressed levels of both renin and aldosterone. In contrast to the disorders of nonaldosterone mineralocorticoid excess, however, the sodium channel defect in Liddle syndrome is characterized by a primary increase in sodium reabsorption in the collecting tubule and potassium wasting. The resultant volume expansion leads to suppressed renin and aldosterone levels and hypertension with low potassium and elevated bicarbonate concentrations.¹⁷

Liddle syndrome is commonly diagnosed in childhood but may go unrecognized due to occasional absence of hypokalemia at presentation. Potassium-sparing diuretics such as amiloride or triamterene are the mainstays of treatment.¹⁸

Hypercortisolism results in hypokalemic metabolic alkalosis through the effect of excess cortisol on mineralocorticoid receptors, similar to what occurs in chronic licorice ingestion. Under normal conditions, 11B-hydroxysteroid dehydrogenase converts cortisol to cortisone and is the rate-limiting step in the mineralocorticoid action of cortisol. When plasma cortisol levels are in excess, however, the enzyme is saturated so that its action is insufficient, resulting in cortisol binding to mineralocorticoid receptors to produce effects similar to that of aldosterone on acid-base and electrolyte balance and blood pressure.¹⁹

The increase in blood pressure that is associated with elevated plasma levels of cortisol is not attributable solely to its effect on mineralocorticoid receptors, however. The pathogenesis is multifactorial and not fully understood, but it also is thought to involve increased peripheral vascular sensitivity to adrenergic agonists, increased hepatic production of angiotensinogen, as well as direct and indirect cardiotoxic effects via metabolic and electrolyte aberrations.²⁰ Other common effects and manifestations of hypercortisolism are listed in Table 4.

Rates of cardiovascular and all-cause mortality are increased in patients with long-term hypercortisolism, even after plasma concentrations of cortisol are normalized.²¹

Figure 2 shows the cascade of the hypothalamic-pituitary-adrenal axis.

 

 

TESTING FOR HYPERCORTISOLISM IN OUR PATIENT

Given the patient’s clinical presentation and laboratory and imaging findings with normal plasma renin and aldosterone levels, a workup for suspected hypercortisolism is initiated.

Initial diagnostic testing for hypercortisolism depends on the degree of clinical suspicion. In those with low probability of the disease, testing should consist of 1 of the following, as a single negative test may be sufficient to rule out the disease:

• 24-hour urinary cortisol levels
• Overnight dexamethasone suppression testing
• Late-night salivary cortisol measurements.

In those with a high index of suspicion, 2 of the aforementioned tests should be performed, as 1 normal result may not be sufficient to exclude the diagnosis.^22,23

A 24-hour urinary cortisol collection and overnight dexamethasone suppression test are obtained. His 24-hour urinary free cortisol level is elevated at 6,600 µg (normal 4–100), and suppression testing with 8 mg of dexamethasone (a form of “high-dose” testing)demonstrates only an 8% decline in serum cortisol levels. Cortisol should generally drop more than 90%.

Morning serum cortisol concentration is less than 5 µg/dL (140 nmol/L) in most patients with Cushing disease (ie, a pituitary tumor), and is usually undetectable in normal subjects. Only about 50% of neuroendocrine ACTH-secreting tumors will suppress with this test.

The patient’s clinical presentation, in conjunction with his diagnostic testing, are thus consistent with Cushing syndrome.

CUSHING SYNDROME

Cushing syndrome is most often exogenous or iatrogenic, ie, a result of supraphysiologic doses of glucocorticoids used to treat a variety of inflammatory, autoimmune, and neoplastic conditions.

Endogenous Cushing syndrome, on the other hand, is rare, with an estimated prevalence of 0.7 to 2.4 cases per million per year. ACTH-dependent causes account for 80% of endogenous Cushing syndrome cases, with ACTH-secreting pituitary adenomas (Cushing disease) accounting for 75% to 80% and ectopic ACTH secretion accounting for 15% to 20%. Less than 1% of cases are due to tumors that produce corticotropin-releasing hormone (CRH).

ACTH-independent Cushing syndrome is diagnosed in 20% of endogenous cases and is most commonly caused by a unilateral adrenal tumor. Rare causes of ACTH-independent disease include adrenal carcinoma, McCune-Albright syndrome, and adrenal hyperplasia.²⁴

The patient’s ACTH is high

To determine whether this is an ACTH-dependent or independent process, the next step is to order an ACTH level. His ACTH level is high at 107 pg/mL (normal < 46 pg/mL), confirming the diagnosis of ACTH-dependent Cushing syndrome.

To find out if this ACTH-dependent process is due to a pituitary adenoma, magnetic resonance imaging (MRI) of the pituitary is obtained but is normal.

Large masses (> 6 mm) strongly suggest Cushing disease, but these tumors are often small and may be missed even with more advanced imaging techniques. Corticotropin-secreting adenomas arising from normal cells in the pituitary retain some sensitivity to glucocorticoid negative feedback and CRH stimulation, and thus high-dose dexamethasone suppression testing in conjunction with CRH stimulation testing can be used to differentiate Cushing disease from ectopic ACTH secretion.^24,25 Both of these tests have poor diagnostic accuracy, however, and thus inferior petrosal sampling remains the gold standard for the diagnosis of Cushing disease.

Given this patient’s history of smoking and a right hilar pulmonary opacity on chest radiography, inferior petrosal sampling was deferred in favor of CT of the chest, which showed a right consolidative lung lesion (Figure 3). Subsequent CT-guided fine-needle biopsy demonstrated a small-cell carcinoma.

ACTH-SECRETING TUMORS

5. Cushing syndrome due to ectopic ACTH secretion is most commonly attributed to which of the following tumors?

• Small-cell lung carcinoma
• Pancreatic carcinoma
• Medullary thyroid carcinoma
• Gastrinoma

Severe cases of Cushing syndrome are often attributable to ectopic ACTH secretion due to an underlying malignancy, most commonly small-cell lung carcinoma or neuroendocrine tumors of pulmonary origin. Other causes include pancreatic and thymic neuroendocrine tumors, gastrinomas, and medullary thyroid carcinoma.^25,26

Because most ACTH-producing tumors are intrathoracic, initial imaging in cases of suspected ectopic ACTH secretion should focus on the chest, with CT the usual first choice. Octreotide scintigraphy can also be useful in localizing disease, as many neuroendocrine tumors express somatostatin receptors. Specialized positron-emission tomography scans may also be helpful in tumor identification.²⁴

 

 

TREATMENT OF CUSHING SYNDROME DUE TO ECTOPIC ACTH SECRETION

6. Which of the following is most appropriate medical therapy for suppression of cortisol secretion in Cushing syndrome due to ectopic ACTH secretion?

• Spironolactone
• Dexamethasone
• Somatostatin
• Estrogen
• Ketoconazole

Hyperglycemia, hypokalemia, hypertension, psychiatric disturbances, venous thromboembolism, and systemic infections appear to be common in ectopic ACTH syndrome and often correlate with the degree of hypercortisolemia. Severe Cushing syndrome due to ectopic ACTH secretion is an emergency requiring prompt control of cortisol secretion.

First-line treatments include steroidogenesis inhibitors (ketoconazole, metyrapone, etomidate, mitotane) and glucocorticoid receptor antagonists (mifepristone). High-dose spironolactone and eplerenone can also be used to treat the hypertension and hypokalemia associated with mineralocorticoid receptor stimulation. Definitive treatment involves surgical resection, chemotherapy, or radiotherapy when applicable.^24,25

After confirmation of the diagnosis, the patient is prescribed ketoconazole and spironolactone, with substantial improvement. He subsequently is started on combination chemotherapy and radiation therapy for his small-cell lung carcinoma.

DISCUSSION

The differential diagnosis for hypokalemia is broad and relies on information obtained during the history and physical examination, followed by interpretation of selected laboratory results. Myriad pathologies in diverse organ systems, eg, diarrhea, renal tubular acidosis, and adrenal disease, may be responsible for a low serum potassium. Further categorizing potassium depletion on the basis of an associated acid-base disturbance, such as metabolic alkalosis, allows one to use an algorithmic approach that can identify specific etiologies responsible for both the potassium and the acid-base disturbances.

Using the spot urine chloride in the setting of hypokalemic metabolic alkalosis with or without hypertension can narrow the differential diagnosis and allow additional clinical findings to guide clinical problem-solving and decision-making, even for conditions not commonly encountered in routine medical practice.

Obtaining renin and aldosterone measurements in patients with potassium depletion, metabolic alkalosis, high urine chloride excretion, and hypertension permits further categorization into 3 clinical groups: elevated aldosterone and renin (secondary hyperaldosteronism), elevated aldosterone and low renin (primary hyperaldosteronism), or apparent mineralocorticoid excess wherein neither renin nor aldosterone are responsible for the syndrome.

The patient in our case had apparent mineralocorticoid excess as a consequence of an ACTH-producing small-cell carcinoma.

NOTE: The scenario presented here is partly based on cases reported elsewhere by Martínez-Valles et al¹ and Fernández-Rodríguez et al.²

A 55-year-old man is admitted to the hospital with generalized malaise, paresthesias, and severe hypertension. He says he had experienced agitation along with weakness on exertion 24 hours before presentation to the emergency department, with subsequent onset of paresthesias in his lower extremities and perioral area.

He is already known to have mild chronic obstructive pulmonary disease, with a ratio of forced expiratory volume in 1 second (FEV₁)to forced vital capacity (FVC) of less than 70% and an FEV₁ 85% of predicted. In addition, he was recently diagnosed with diabetes, resistant hypertension requiring maximum doses of 3 agents (a calcium channel blocker, an angiotensin-converting enzyme inhibitor, and a loop diuretic), and hyperlipidemia.

He is a current smoker with a 30-pack-year smoking history. He does not use alcohol. His family history is noncontributory.

His blood pressure is 190/110 mm Hg despite adherence to his 3-drug regimen. His oxygen saturation is 94% on room air, respiratory rate in the low 30s, and pulse 110 beats/minute. He has normal breath sounds, normal S1 and S2 with an S4 gallop, bilateral lower-extremity edema, truncal obesity, and abdominal striae. Electrocardiography shows tachycardia with first-degree atrioventicular block. Chest radiography shows an opacity in the right middle lung field. Initial laboratory results and those from 1 year ago are shown in Table 1.

ASSESSING ACID-BASE DISORDERS

1. What type of acid-base disorder does this patient have?

• Metabolic acidosis
• Respiratory acidosis
• Metabolic alkalosis
• Respiratory alkalosis

The patient has metabolic alkalosis.

A 5-step approach

If a patient has an acid-base disorder, one should use a 5-step process to characterize it (Table 2).³

1. Acidosis or alkalosis? The patient’s arterial pH is 7.5, which is alkalemic because it is higher than 7.44.

2. Metabolic or respiratory? The primary process in our patient is overwhelmingly metabolic, as his partial pressure of carbon dioxide (Pco₂) is slightly elevated, a direction that would cause acidosis, not alkalosis.

3. The anion gap (the serum sodium concentration minus the sum of the chloride and bicarbonate concentrations) is normal at 8 mmol/L (DRG:HYBRiD-XL Immunoassay and Clinical Chemistry Analyzer, reference range 8–16).

4. Is the disturbance compensated? We have determined that this patient has a metabolic alkalemia; the question now is whether there is any compensation for the primary disturbance.

In metabolic alkalosis, the Pco₂ may increase by approximately 0.6 mm Hg (range 0.5–0.8) above the nominal normal level of 40 mm Hg for each 1-mmol/L increase in bicarbonate above the nominal normal level of 25 mmol/L.⁴ If the patient requires oxygen, the calculation may be unreliable, however, as hypoxemia may have an overriding influence on respiratory drive.

Patients with chronically high Pco₂ levels such as those with chronic obstructive pulmonary disease can become accustomed to high carbon dioxide levels and lose their hyper-
capnic respiratory drive. Giving oxygen supplementation is thought to decrease respiratory drive in these patients, so that they will breathe slower and retain more carbon dioxide. There is some degree of respiratory compensation for metabolic alkalosis that occurs by breathing less, though it is limited overall—even in very alkalotic patients, breathing less results in CO₂ retention, which, by displacing O₂ molecules in the alveoli, will in turn result in hypoxia. The brain then senses the hypoxia and makes one breathe faster, thereby limiting this compensation. 

This patient’s serum bicarbonate level is 40 mmol/L, or 15 mmol/L higher than the nominal normal level. If he is compensating, his Pco₂ should be 40 + (15 × 0.6) = 49 mm Hg, and in fact it is 51 mm Hg, which is within the normal range of expected compensation (47.5–52 mm Hg). Therefore, yes, he is compensating for the primary disturbance.  

5. In metabolic acidosis, is there a delta gap? As our patient has metabolic alkalosis, not acidosis, this question does not apply in this case.

 

 

WHICH TEST TO FIND THE CAUSE?

2. Which is the best test to order next to determine the cause of this patient’s hypokalemic metabolic alkalosis?

• Serum magnesium level
• Spot urine chloride
• Renal ultrasonography
• 24-hour urine collection for sodium, potassium, and chloride

The first step in the algorithm for hypokalemic metabolic alkalosis (Figure 1) is to obtain a spot urine chloride measurement. If this value is low, the hypokalemic metabolic alkalosis is volume-responsive; if it is high, the disturbance is volume-independent.

The patient’s loop diuretic is withheld for 12 hours and a spot urine chloride is obtained, which is reported as 44 mmol/L. This high value suggests that a volume-independent hypo­kalemic metabolic alkalosis is present with potassium depletion.

As for the other answer choices:

Serum magnesium. Though hypomagnesemia can cause hypokalemia due to lack of inhibition of renal outer medullary potassium channels and subsequent increased excretion of potassium in the apical tubular membrane, it is not independently associated with acid-base disturbances.⁵

Renal ultrasonography gives information about structural kidney disease but is of limited utility in identifying the cause of hypokalemic metabolic alkalosis.

A 24-hour urine collection is unnecessary in this setting and would ultimately result in delay in diagnosis, as spot urine chloride is a more efficient means of rapidly distinguishing volume-responsive vs volume-independent causes of hypokalemic metabolic alkalosis.⁶

IS HIS HYPERTENSION SECONDARY? IF SO, WHAT IS THE CAUSE?

Several features of this case suggest that the patient’s hypertension is secondary rather than primary. It is of recent onset. The patient’s family history is noncontributory, and his hypertension is resistant to the use of maximum doses of 3 antihypertensive agents.

3. Which of the following causes of secondary hypertension is not commonly associated with hypokalemia and metabolic alkalosis?

• Hyperaldosteronism
• Liddle syndrome
• Cushing syndrome
• Renal parenchymal disease
• Chronic licorice ingestion

Renal parenchymal disease is a cause of resistant hypertension, but it is not characterized by metabolic alkalosis, hypokalemia, and  elevated urine chloride,⁷ while the others listed here—hyperaldosteronism, Liddle syndrome, Cushing syndrome, and chronic licorice ingestion­—are. Other common causes of resistant hypertension without these metabolic abnormalities include obstructive sleep apnea, alcohol abuse, and nonadherence to treatment.

While treatment of hypertension with loop diuretics can result in hypokalemia and metabolic alkalosis due to the effect of these drugs on potassium reabsorption in the loop of Henle, the patient’s hypokalemia persisted after this agent was withdrawn.⁸

Causes of hypokalemic metabolic alkalosis with and without hypertension are further delineated in Figure 1.

Additional diagnostic testing: Plasma renin and plasma aldosterone

At this juncture, the differential diagnosis for this patient’s potassium depletion, metabolic alkalosis, high urine chloride, and hypertension has been narrowed to primary or secondary hyperaldosteronism, surreptitious mineralocorticoid ingestion, Cushing syndrome, licorice ingestion, Liddle syndrome, or one of the 3 hydroxylase deficiencies (11-, 17-, and 21-) (Figure 1).

Although clues in the history, physical examination, and imaging may suggest a specific cause of his abnormal laboratory values, the next step in the diagnostic workup is to measure the plasma renin and aldosterone levels (Table 3).

 

HYPERALDOSTERONISM

4. Hyperaldosteronism is associated with which of the following patterns of renin and aldosterone values?

• High renin, high aldosterone, normal ratio of plasma aldosterone concentration (PAC) to plasma renin activity (PRA)
• Low renin, low aldosterone, normal PAC–PRA ratio
• Low renin, high aldosterone, high PAC–PRA ratio
• High renin, low aldosterone, low PAC–PRA ratio

The pattern of low renin, high aldosterone, and high PAC–PRA ratio is associated with hyperaldosteronism.

Primary hyperaldosteronism

Primary hyperaldosteronism is one of the most common causes of resistant hypertension and is underappreciated, being diagnosed in up to 20% of patients referred to hypertension specialty clinics.⁷ Potassium levels may be normal, likely contributing to its lack of recognition in this target population.

Primary hyperaldosteronism should be suspected in patients who have a plasma aldosterone PAC–PRA ratio greater than 20 with elevated plasma aldosterone concentrations
(> 15 ng/dL).

Persistently elevated aldosterone levels in the setting of elevated plasma volume is proof that aldosterone secretion is independent of the renin-angiotensin-aldosterone axis, and therefore is autonomous (secondary to adrenal tumor or hyperplasia). Further testing in the form of oral salt loading, saline infusion, or fludrocortisone (a sodium-retaining steroid) administration is thus required to confirm inappropriate, autonomous aldosterone secretion.⁹

After establishing the diagnosis of primary hyperaldosteronism, one should determine the subtype (ie, due to an adrenal carcinoma, unilateral hypersecreting adenoma, or unilateral or bilateral hyperplasia). Further testing includes adrenal computed tomography (CT) to rule out adrenal carcinomas, which are suspected with adenomas larger than 4 cm. Though part of the diagnostic workup, CT as a means of confirmational testing alone does not preclude the possibility of bilateral adrenal hyperplasia in some patients, even in the presence of an adrenal adenoma. For this reason, adrenal venous sampling is required to definitively determine whether the condition is due to a hypersecreting adrenal adenoma or unilateral or bilateral hyperplasia.^9,10

Treatment of primary hyperaldosteronism depends on the subtype of the disease and involves salt restriction in addition to an aldosterone antagonist (spironolactone or eplerenone in the case of bilateral disease) or surgery (unilateral disease).^9,11,12

Secondary hyperaldosteronism

Secondary hyperaldosteronism should be suspected when plasma renin and aldosterone levels are both elevated with a PAC–PRA ratio less than 10.

This pattern is most commonly seen with diuretic use but can also be a consequence of renal artery stenosis or, rarely, a renin-secreting tumor.¹³ Renal artery stenosis is a common finding in patients with hypertension undergoing cardiac catheterization, which is not surprising as more than 90% of such stenoses are atherosclerotic.⁷ Renin-secreting tumors are exceedingly rare, with fewer than 100 cases reported in the literature, and are more common in younger individuals.¹³

Our patient has low-normal aldosterone and plasma renin

On further testing, this patient’s plasma aldosterone level is 2.55 ng/dL (normal < 15 ng/dL), his plasma renin activity is 0.53 ng/mL/hour (normal 0.2–2.8 ng/mL/hour), and his PAC–PRA ratio is therefore 4.81.

The categories discussed thus far have included primary and secondary hyperaldosteronism, which typically do not present with low to normal levels of both renin and aldosterone. Surreptitious mineralocorticoid use could present in this manner, but is unlikely in this patient, whose medications do not include fludrocortisone.

The low-normal values thus lead to consideration of a third category: apparent mineralocorticoid excess. Diseases in this category such as Cushing disease or adrenocorticotropic hormone (ACTH) excess are characterized by increases in corticosteroids so that the potassium depletion, metabolic alkalosis, and hypertension are not a consequence of renin and aldosterone but rather the excess corticosteroids.¹⁴

Causes of apparent mineralocorticoid excess

There are several possible causes of mineralocorticoid excess associated with hypertension and hypokalemic metabolic alkalosis not due to renin and aldosterone.

Chronic licorice ingestion in high volumes is one such cause and is thought to result in inhibition of 11B-hydroxysteroid dehydrogenase or possibly cortisol oxidase by licorice’s active component, glycyrrhetinic acid. This inhibition results in an inability to convert cortisol to cortisone. The cortisol excess binds to mineralocorticoid receptors, and acting like aldosterone, results in hypertension and hypokalemic metabolic alkalosis as well as feedback inhibition of renin and aldosterone levels.¹⁵

Partial hydroxylase deficiencies, though rare, should also be considered as a cause of hypokalemic metabolic alkalosis, hypertension, and, potentially, hirsutism and clitoromegaly in women. They can be diagnosed with elevated levels of 17-ketosteroids and dehydroepiandrosterone sulfate, both of which, in excess, may act on aldosterone receptors in a manner similar to cortisol.¹⁶

Liddle syndrome, a rare autosomal dominant condition, may also present with suppressed levels of both renin and aldosterone. In contrast to the disorders of nonaldosterone mineralocorticoid excess, however, the sodium channel defect in Liddle syndrome is characterized by a primary increase in sodium reabsorption in the collecting tubule and potassium wasting. The resultant volume expansion leads to suppressed renin and aldosterone levels and hypertension with low potassium and elevated bicarbonate concentrations.¹⁷

Liddle syndrome is commonly diagnosed in childhood but may go unrecognized due to occasional absence of hypokalemia at presentation. Potassium-sparing diuretics such as amiloride or triamterene are the mainstays of treatment.¹⁸

Hypercortisolism results in hypokalemic metabolic alkalosis through the effect of excess cortisol on mineralocorticoid receptors, similar to what occurs in chronic licorice ingestion. Under normal conditions, 11B-hydroxysteroid dehydrogenase converts cortisol to cortisone and is the rate-limiting step in the mineralocorticoid action of cortisol. When plasma cortisol levels are in excess, however, the enzyme is saturated so that its action is insufficient, resulting in cortisol binding to mineralocorticoid receptors to produce effects similar to that of aldosterone on acid-base and electrolyte balance and blood pressure.¹⁹

The increase in blood pressure that is associated with elevated plasma levels of cortisol is not attributable solely to its effect on mineralocorticoid receptors, however. The pathogenesis is multifactorial and not fully understood, but it also is thought to involve increased peripheral vascular sensitivity to adrenergic agonists, increased hepatic production of angiotensinogen, as well as direct and indirect cardiotoxic effects via metabolic and electrolyte aberrations.²⁰ Other common effects and manifestations of hypercortisolism are listed in Table 4.

Rates of cardiovascular and all-cause mortality are increased in patients with long-term hypercortisolism, even after plasma concentrations of cortisol are normalized.²¹

Figure 2 shows the cascade of the hypothalamic-pituitary-adrenal axis.

 

 

TESTING FOR HYPERCORTISOLISM IN OUR PATIENT

Given the patient’s clinical presentation and laboratory and imaging findings with normal plasma renin and aldosterone levels, a workup for suspected hypercortisolism is initiated.

Initial diagnostic testing for hypercortisolism depends on the degree of clinical suspicion. In those with low probability of the disease, testing should consist of 1 of the following, as a single negative test may be sufficient to rule out the disease:

• 24-hour urinary cortisol levels
• Overnight dexamethasone suppression testing
• Late-night salivary cortisol measurements.

In those with a high index of suspicion, 2 of the aforementioned tests should be performed, as 1 normal result may not be sufficient to exclude the diagnosis.^22,23

A 24-hour urinary cortisol collection and overnight dexamethasone suppression test are obtained. His 24-hour urinary free cortisol level is elevated at 6,600 µg (normal 4–100), and suppression testing with 8 mg of dexamethasone (a form of “high-dose” testing)demonstrates only an 8% decline in serum cortisol levels. Cortisol should generally drop more than 90%.

Morning serum cortisol concentration is less than 5 µg/dL (140 nmol/L) in most patients with Cushing disease (ie, a pituitary tumor), and is usually undetectable in normal subjects. Only about 50% of neuroendocrine ACTH-secreting tumors will suppress with this test.

The patient’s clinical presentation, in conjunction with his diagnostic testing, are thus consistent with Cushing syndrome.

CUSHING SYNDROME

Cushing syndrome is most often exogenous or iatrogenic, ie, a result of supraphysiologic doses of glucocorticoids used to treat a variety of inflammatory, autoimmune, and neoplastic conditions.

Endogenous Cushing syndrome, on the other hand, is rare, with an estimated prevalence of 0.7 to 2.4 cases per million per year. ACTH-dependent causes account for 80% of endogenous Cushing syndrome cases, with ACTH-secreting pituitary adenomas (Cushing disease) accounting for 75% to 80% and ectopic ACTH secretion accounting for 15% to 20%. Less than 1% of cases are due to tumors that produce corticotropin-releasing hormone (CRH).

ACTH-independent Cushing syndrome is diagnosed in 20% of endogenous cases and is most commonly caused by a unilateral adrenal tumor. Rare causes of ACTH-independent disease include adrenal carcinoma, McCune-Albright syndrome, and adrenal hyperplasia.²⁴

The patient’s ACTH is high

To determine whether this is an ACTH-dependent or independent process, the next step is to order an ACTH level. His ACTH level is high at 107 pg/mL (normal < 46 pg/mL), confirming the diagnosis of ACTH-dependent Cushing syndrome.

To find out if this ACTH-dependent process is due to a pituitary adenoma, magnetic resonance imaging (MRI) of the pituitary is obtained but is normal.

Large masses (> 6 mm) strongly suggest Cushing disease, but these tumors are often small and may be missed even with more advanced imaging techniques. Corticotropin-secreting adenomas arising from normal cells in the pituitary retain some sensitivity to glucocorticoid negative feedback and CRH stimulation, and thus high-dose dexamethasone suppression testing in conjunction with CRH stimulation testing can be used to differentiate Cushing disease from ectopic ACTH secretion.^24,25 Both of these tests have poor diagnostic accuracy, however, and thus inferior petrosal sampling remains the gold standard for the diagnosis of Cushing disease.

Given this patient’s history of smoking and a right hilar pulmonary opacity on chest radiography, inferior petrosal sampling was deferred in favor of CT of the chest, which showed a right consolidative lung lesion (Figure 3). Subsequent CT-guided fine-needle biopsy demonstrated a small-cell carcinoma.

ACTH-SECRETING TUMORS

5. Cushing syndrome due to ectopic ACTH secretion is most commonly attributed to which of the following tumors?

• Small-cell lung carcinoma
• Pancreatic carcinoma
• Medullary thyroid carcinoma
• Gastrinoma

Severe cases of Cushing syndrome are often attributable to ectopic ACTH secretion due to an underlying malignancy, most commonly small-cell lung carcinoma or neuroendocrine tumors of pulmonary origin. Other causes include pancreatic and thymic neuroendocrine tumors, gastrinomas, and medullary thyroid carcinoma.^25,26

Because most ACTH-producing tumors are intrathoracic, initial imaging in cases of suspected ectopic ACTH secretion should focus on the chest, with CT the usual first choice. Octreotide scintigraphy can also be useful in localizing disease, as many neuroendocrine tumors express somatostatin receptors. Specialized positron-emission tomography scans may also be helpful in tumor identification.²⁴

 

 

TREATMENT OF CUSHING SYNDROME DUE TO ECTOPIC ACTH SECRETION

6. Which of the following is most appropriate medical therapy for suppression of cortisol secretion in Cushing syndrome due to ectopic ACTH secretion?

• Spironolactone
• Dexamethasone
• Somatostatin
• Estrogen
• Ketoconazole

Hyperglycemia, hypokalemia, hypertension, psychiatric disturbances, venous thromboembolism, and systemic infections appear to be common in ectopic ACTH syndrome and often correlate with the degree of hypercortisolemia. Severe Cushing syndrome due to ectopic ACTH secretion is an emergency requiring prompt control of cortisol secretion.

First-line treatments include steroidogenesis inhibitors (ketoconazole, metyrapone, etomidate, mitotane) and glucocorticoid receptor antagonists (mifepristone). High-dose spironolactone and eplerenone can also be used to treat the hypertension and hypokalemia associated with mineralocorticoid receptor stimulation. Definitive treatment involves surgical resection, chemotherapy, or radiotherapy when applicable.^24,25

After confirmation of the diagnosis, the patient is prescribed ketoconazole and spironolactone, with substantial improvement. He subsequently is started on combination chemotherapy and radiation therapy for his small-cell lung carcinoma.

DISCUSSION

The differential diagnosis for hypokalemia is broad and relies on information obtained during the history and physical examination, followed by interpretation of selected laboratory results. Myriad pathologies in diverse organ systems, eg, diarrhea, renal tubular acidosis, and adrenal disease, may be responsible for a low serum potassium. Further categorizing potassium depletion on the basis of an associated acid-base disturbance, such as metabolic alkalosis, allows one to use an algorithmic approach that can identify specific etiologies responsible for both the potassium and the acid-base disturbances.

Using the spot urine chloride in the setting of hypokalemic metabolic alkalosis with or without hypertension can narrow the differential diagnosis and allow additional clinical findings to guide clinical problem-solving and decision-making, even for conditions not commonly encountered in routine medical practice.

Obtaining renin and aldosterone measurements in patients with potassium depletion, metabolic alkalosis, high urine chloride excretion, and hypertension permits further categorization into 3 clinical groups: elevated aldosterone and renin (secondary hyperaldosteronism), elevated aldosterone and low renin (primary hyperaldosteronism), or apparent mineralocorticoid excess wherein neither renin nor aldosterone are responsible for the syndrome.

The patient in our case had apparent mineralocorticoid excess as a consequence of an ACTH-producing small-cell carcinoma.

NOTE: The scenario presented here is partly based on cases reported elsewhere by Martínez-Valles et al¹ and Fernández-Rodríguez et al.²

A 55-year-old man is admitted to the hospital with generalized malaise, paresthesias, and severe hypertension. He says he had experienced agitation along with weakness on exertion 24 hours before presentation to the emergency department, with subsequent onset of paresthesias in his lower extremities and perioral area.

He is already known to have mild chronic obstructive pulmonary disease, with a ratio of forced expiratory volume in 1 second (FEV₁)to forced vital capacity (FVC) of less than 70% and an FEV₁ 85% of predicted. In addition, he was recently diagnosed with diabetes, resistant hypertension requiring maximum doses of 3 agents (a calcium channel blocker, an angiotensin-converting enzyme inhibitor, and a loop diuretic), and hyperlipidemia.

He is a current smoker with a 30-pack-year smoking history. He does not use alcohol. His family history is noncontributory.

His blood pressure is 190/110 mm Hg despite adherence to his 3-drug regimen. His oxygen saturation is 94% on room air, respiratory rate in the low 30s, and pulse 110 beats/minute. He has normal breath sounds, normal S1 and S2 with an S4 gallop, bilateral lower-extremity edema, truncal obesity, and abdominal striae. Electrocardiography shows tachycardia with first-degree atrioventicular block. Chest radiography shows an opacity in the right middle lung field. Initial laboratory results and those from 1 year ago are shown in Table 1.

ASSESSING ACID-BASE DISORDERS

1. What type of acid-base disorder does this patient have?

• Metabolic acidosis
• Respiratory acidosis
• Metabolic alkalosis
• Respiratory alkalosis

The patient has metabolic alkalosis.

A 5-step approach

If a patient has an acid-base disorder, one should use a 5-step process to characterize it (Table 2).³

1. Acidosis or alkalosis? The patient’s arterial pH is 7.5, which is alkalemic because it is higher than 7.44.

2. Metabolic or respiratory? The primary process in our patient is overwhelmingly metabolic, as his partial pressure of carbon dioxide (Pco₂) is slightly elevated, a direction that would cause acidosis, not alkalosis.

3. The anion gap (the serum sodium concentration minus the sum of the chloride and bicarbonate concentrations) is normal at 8 mmol/L (DRG:HYBRiD-XL Immunoassay and Clinical Chemistry Analyzer, reference range 8–16).

4. Is the disturbance compensated? We have determined that this patient has a metabolic alkalemia; the question now is whether there is any compensation for the primary disturbance.

In metabolic alkalosis, the Pco₂ may increase by approximately 0.6 mm Hg (range 0.5–0.8) above the nominal normal level of 40 mm Hg for each 1-mmol/L increase in bicarbonate above the nominal normal level of 25 mmol/L.⁴ If the patient requires oxygen, the calculation may be unreliable, however, as hypoxemia may have an overriding influence on respiratory drive.

Patients with chronically high Pco₂ levels such as those with chronic obstructive pulmonary disease can become accustomed to high carbon dioxide levels and lose their hyper-
capnic respiratory drive. Giving oxygen supplementation is thought to decrease respiratory drive in these patients, so that they will breathe slower and retain more carbon dioxide. There is some degree of respiratory compensation for metabolic alkalosis that occurs by breathing less, though it is limited overall—even in very alkalotic patients, breathing less results in CO₂ retention, which, by displacing O₂ molecules in the alveoli, will in turn result in hypoxia. The brain then senses the hypoxia and makes one breathe faster, thereby limiting this compensation. 

This patient’s serum bicarbonate level is 40 mmol/L, or 15 mmol/L higher than the nominal normal level. If he is compensating, his Pco₂ should be 40 + (15 × 0.6) = 49 mm Hg, and in fact it is 51 mm Hg, which is within the normal range of expected compensation (47.5–52 mm Hg). Therefore, yes, he is compensating for the primary disturbance.  

5. In metabolic acidosis, is there a delta gap? As our patient has metabolic alkalosis, not acidosis, this question does not apply in this case.

 

 

WHICH TEST TO FIND THE CAUSE?

2. Which is the best test to order next to determine the cause of this patient’s hypokalemic metabolic alkalosis?

• Serum magnesium level
• Spot urine chloride
• Renal ultrasonography
• 24-hour urine collection for sodium, potassium, and chloride

The first step in the algorithm for hypokalemic metabolic alkalosis (Figure 1) is to obtain a spot urine chloride measurement. If this value is low, the hypokalemic metabolic alkalosis is volume-responsive; if it is high, the disturbance is volume-independent.

The patient’s loop diuretic is withheld for 12 hours and a spot urine chloride is obtained, which is reported as 44 mmol/L. This high value suggests that a volume-independent hypo­kalemic metabolic alkalosis is present with potassium depletion.

As for the other answer choices:

Serum magnesium. Though hypomagnesemia can cause hypokalemia due to lack of inhibition of renal outer medullary potassium channels and subsequent increased excretion of potassium in the apical tubular membrane, it is not independently associated with acid-base disturbances.⁵

Renal ultrasonography gives information about structural kidney disease but is of limited utility in identifying the cause of hypokalemic metabolic alkalosis.

A 24-hour urine collection is unnecessary in this setting and would ultimately result in delay in diagnosis, as spot urine chloride is a more efficient means of rapidly distinguishing volume-responsive vs volume-independent causes of hypokalemic metabolic alkalosis.⁶

IS HIS HYPERTENSION SECONDARY? IF SO, WHAT IS THE CAUSE?

Several features of this case suggest that the patient’s hypertension is secondary rather than primary. It is of recent onset. The patient’s family history is noncontributory, and his hypertension is resistant to the use of maximum doses of 3 antihypertensive agents.

3. Which of the following causes of secondary hypertension is not commonly associated with hypokalemia and metabolic alkalosis?

• Hyperaldosteronism
• Liddle syndrome
• Cushing syndrome
• Renal parenchymal disease
• Chronic licorice ingestion

Renal parenchymal disease is a cause of resistant hypertension, but it is not characterized by metabolic alkalosis, hypokalemia, and  elevated urine chloride,⁷ while the others listed here—hyperaldosteronism, Liddle syndrome, Cushing syndrome, and chronic licorice ingestion­—are. Other common causes of resistant hypertension without these metabolic abnormalities include obstructive sleep apnea, alcohol abuse, and nonadherence to treatment.

While treatment of hypertension with loop diuretics can result in hypokalemia and metabolic alkalosis due to the effect of these drugs on potassium reabsorption in the loop of Henle, the patient’s hypokalemia persisted after this agent was withdrawn.⁸

Causes of hypokalemic metabolic alkalosis with and without hypertension are further delineated in Figure 1.

Additional diagnostic testing: Plasma renin and plasma aldosterone

At this juncture, the differential diagnosis for this patient’s potassium depletion, metabolic alkalosis, high urine chloride, and hypertension has been narrowed to primary or secondary hyperaldosteronism, surreptitious mineralocorticoid ingestion, Cushing syndrome, licorice ingestion, Liddle syndrome, or one of the 3 hydroxylase deficiencies (11-, 17-, and 21-) (Figure 1).

Although clues in the history, physical examination, and imaging may suggest a specific cause of his abnormal laboratory values, the next step in the diagnostic workup is to measure the plasma renin and aldosterone levels (Table 3).

 

HYPERALDOSTERONISM

4. Hyperaldosteronism is associated with which of the following patterns of renin and aldosterone values?

• High renin, high aldosterone, normal ratio of plasma aldosterone concentration (PAC) to plasma renin activity (PRA)
• Low renin, low aldosterone, normal PAC–PRA ratio
• Low renin, high aldosterone, high PAC–PRA ratio
• High renin, low aldosterone, low PAC–PRA ratio

The pattern of low renin, high aldosterone, and high PAC–PRA ratio is associated with hyperaldosteronism.

Primary hyperaldosteronism

Primary hyperaldosteronism is one of the most common causes of resistant hypertension and is underappreciated, being diagnosed in up to 20% of patients referred to hypertension specialty clinics.⁷ Potassium levels may be normal, likely contributing to its lack of recognition in this target population.

Primary hyperaldosteronism should be suspected in patients who have a plasma aldosterone PAC–PRA ratio greater than 20 with elevated plasma aldosterone concentrations
(> 15 ng/dL).

Persistently elevated aldosterone levels in the setting of elevated plasma volume is proof that aldosterone secretion is independent of the renin-angiotensin-aldosterone axis, and therefore is autonomous (secondary to adrenal tumor or hyperplasia). Further testing in the form of oral salt loading, saline infusion, or fludrocortisone (a sodium-retaining steroid) administration is thus required to confirm inappropriate, autonomous aldosterone secretion.⁹

After establishing the diagnosis of primary hyperaldosteronism, one should determine the subtype (ie, due to an adrenal carcinoma, unilateral hypersecreting adenoma, or unilateral or bilateral hyperplasia). Further testing includes adrenal computed tomography (CT) to rule out adrenal carcinomas, which are suspected with adenomas larger than 4 cm. Though part of the diagnostic workup, CT as a means of confirmational testing alone does not preclude the possibility of bilateral adrenal hyperplasia in some patients, even in the presence of an adrenal adenoma. For this reason, adrenal venous sampling is required to definitively determine whether the condition is due to a hypersecreting adrenal adenoma or unilateral or bilateral hyperplasia.^9,10

Treatment of primary hyperaldosteronism depends on the subtype of the disease and involves salt restriction in addition to an aldosterone antagonist (spironolactone or eplerenone in the case of bilateral disease) or surgery (unilateral disease).^9,11,12

Secondary hyperaldosteronism

Secondary hyperaldosteronism should be suspected when plasma renin and aldosterone levels are both elevated with a PAC–PRA ratio less than 10.

This pattern is most commonly seen with diuretic use but can also be a consequence of renal artery stenosis or, rarely, a renin-secreting tumor.¹³ Renal artery stenosis is a common finding in patients with hypertension undergoing cardiac catheterization, which is not surprising as more than 90% of such stenoses are atherosclerotic.⁷ Renin-secreting tumors are exceedingly rare, with fewer than 100 cases reported in the literature, and are more common in younger individuals.¹³

Our patient has low-normal aldosterone and plasma renin

On further testing, this patient’s plasma aldosterone level is 2.55 ng/dL (normal < 15 ng/dL), his plasma renin activity is 0.53 ng/mL/hour (normal 0.2–2.8 ng/mL/hour), and his PAC–PRA ratio is therefore 4.81.

The categories discussed thus far have included primary and secondary hyperaldosteronism, which typically do not present with low to normal levels of both renin and aldosterone. Surreptitious mineralocorticoid use could present in this manner, but is unlikely in this patient, whose medications do not include fludrocortisone.

The low-normal values thus lead to consideration of a third category: apparent mineralocorticoid excess. Diseases in this category such as Cushing disease or adrenocorticotropic hormone (ACTH) excess are characterized by increases in corticosteroids so that the potassium depletion, metabolic alkalosis, and hypertension are not a consequence of renin and aldosterone but rather the excess corticosteroids.¹⁴

Causes of apparent mineralocorticoid excess

There are several possible causes of mineralocorticoid excess associated with hypertension and hypokalemic metabolic alkalosis not due to renin and aldosterone.

Chronic licorice ingestion in high volumes is one such cause and is thought to result in inhibition of 11B-hydroxysteroid dehydrogenase or possibly cortisol oxidase by licorice’s active component, glycyrrhetinic acid. This inhibition results in an inability to convert cortisol to cortisone. The cortisol excess binds to mineralocorticoid receptors, and acting like aldosterone, results in hypertension and hypokalemic metabolic alkalosis as well as feedback inhibition of renin and aldosterone levels.¹⁵

Partial hydroxylase deficiencies, though rare, should also be considered as a cause of hypokalemic metabolic alkalosis, hypertension, and, potentially, hirsutism and clitoromegaly in women. They can be diagnosed with elevated levels of 17-ketosteroids and dehydroepiandrosterone sulfate, both of which, in excess, may act on aldosterone receptors in a manner similar to cortisol.¹⁶

Liddle syndrome, a rare autosomal dominant condition, may also present with suppressed levels of both renin and aldosterone. In contrast to the disorders of nonaldosterone mineralocorticoid excess, however, the sodium channel defect in Liddle syndrome is characterized by a primary increase in sodium reabsorption in the collecting tubule and potassium wasting. The resultant volume expansion leads to suppressed renin and aldosterone levels and hypertension with low potassium and elevated bicarbonate concentrations.¹⁷

Liddle syndrome is commonly diagnosed in childhood but may go unrecognized due to occasional absence of hypokalemia at presentation. Potassium-sparing diuretics such as amiloride or triamterene are the mainstays of treatment.¹⁸

Hypercortisolism results in hypokalemic metabolic alkalosis through the effect of excess cortisol on mineralocorticoid receptors, similar to what occurs in chronic licorice ingestion. Under normal conditions, 11B-hydroxysteroid dehydrogenase converts cortisol to cortisone and is the rate-limiting step in the mineralocorticoid action of cortisol. When plasma cortisol levels are in excess, however, the enzyme is saturated so that its action is insufficient, resulting in cortisol binding to mineralocorticoid receptors to produce effects similar to that of aldosterone on acid-base and electrolyte balance and blood pressure.¹⁹

The increase in blood pressure that is associated with elevated plasma levels of cortisol is not attributable solely to its effect on mineralocorticoid receptors, however. The pathogenesis is multifactorial and not fully understood, but it also is thought to involve increased peripheral vascular sensitivity to adrenergic agonists, increased hepatic production of angiotensinogen, as well as direct and indirect cardiotoxic effects via metabolic and electrolyte aberrations.²⁰ Other common effects and manifestations of hypercortisolism are listed in Table 4.

Rates of cardiovascular and all-cause mortality are increased in patients with long-term hypercortisolism, even after plasma concentrations of cortisol are normalized.²¹

Figure 2 shows the cascade of the hypothalamic-pituitary-adrenal axis.

 

 

TESTING FOR HYPERCORTISOLISM IN OUR PATIENT

Given the patient’s clinical presentation and laboratory and imaging findings with normal plasma renin and aldosterone levels, a workup for suspected hypercortisolism is initiated.

Initial diagnostic testing for hypercortisolism depends on the degree of clinical suspicion. In those with low probability of the disease, testing should consist of 1 of the following, as a single negative test may be sufficient to rule out the disease:

• 24-hour urinary cortisol levels
• Overnight dexamethasone suppression testing
• Late-night salivary cortisol measurements.

In those with a high index of suspicion, 2 of the aforementioned tests should be performed, as 1 normal result may not be sufficient to exclude the diagnosis.^22,23

A 24-hour urinary cortisol collection and overnight dexamethasone suppression test are obtained. His 24-hour urinary free cortisol level is elevated at 6,600 µg (normal 4–100), and suppression testing with 8 mg of dexamethasone (a form of “high-dose” testing)demonstrates only an 8% decline in serum cortisol levels. Cortisol should generally drop more than 90%.

Morning serum cortisol concentration is less than 5 µg/dL (140 nmol/L) in most patients with Cushing disease (ie, a pituitary tumor), and is usually undetectable in normal subjects. Only about 50% of neuroendocrine ACTH-secreting tumors will suppress with this test.

The patient’s clinical presentation, in conjunction with his diagnostic testing, are thus consistent with Cushing syndrome.

CUSHING SYNDROME

Cushing syndrome is most often exogenous or iatrogenic, ie, a result of supraphysiologic doses of glucocorticoids used to treat a variety of inflammatory, autoimmune, and neoplastic conditions.

Endogenous Cushing syndrome, on the other hand, is rare, with an estimated prevalence of 0.7 to 2.4 cases per million per year. ACTH-dependent causes account for 80% of endogenous Cushing syndrome cases, with ACTH-secreting pituitary adenomas (Cushing disease) accounting for 75% to 80% and ectopic ACTH secretion accounting for 15% to 20%. Less than 1% of cases are due to tumors that produce corticotropin-releasing hormone (CRH).

ACTH-independent Cushing syndrome is diagnosed in 20% of endogenous cases and is most commonly caused by a unilateral adrenal tumor. Rare causes of ACTH-independent disease include adrenal carcinoma, McCune-Albright syndrome, and adrenal hyperplasia.²⁴

The patient’s ACTH is high

To determine whether this is an ACTH-dependent or independent process, the next step is to order an ACTH level. His ACTH level is high at 107 pg/mL (normal < 46 pg/mL), confirming the diagnosis of ACTH-dependent Cushing syndrome.

To find out if this ACTH-dependent process is due to a pituitary adenoma, magnetic resonance imaging (MRI) of the pituitary is obtained but is normal.

Large masses (> 6 mm) strongly suggest Cushing disease, but these tumors are often small and may be missed even with more advanced imaging techniques. Corticotropin-secreting adenomas arising from normal cells in the pituitary retain some sensitivity to glucocorticoid negative feedback and CRH stimulation, and thus high-dose dexamethasone suppression testing in conjunction with CRH stimulation testing can be used to differentiate Cushing disease from ectopic ACTH secretion.^24,25 Both of these tests have poor diagnostic accuracy, however, and thus inferior petrosal sampling remains the gold standard for the diagnosis of Cushing disease.

Given this patient’s history of smoking and a right hilar pulmonary opacity on chest radiography, inferior petrosal sampling was deferred in favor of CT of the chest, which showed a right consolidative lung lesion (Figure 3). Subsequent CT-guided fine-needle biopsy demonstrated a small-cell carcinoma.

ACTH-SECRETING TUMORS

5. Cushing syndrome due to ectopic ACTH secretion is most commonly attributed to which of the following tumors?

• Small-cell lung carcinoma
• Pancreatic carcinoma
• Medullary thyroid carcinoma
• Gastrinoma

Severe cases of Cushing syndrome are often attributable to ectopic ACTH secretion due to an underlying malignancy, most commonly small-cell lung carcinoma or neuroendocrine tumors of pulmonary origin. Other causes include pancreatic and thymic neuroendocrine tumors, gastrinomas, and medullary thyroid carcinoma.^25,26

Because most ACTH-producing tumors are intrathoracic, initial imaging in cases of suspected ectopic ACTH secretion should focus on the chest, with CT the usual first choice. Octreotide scintigraphy can also be useful in localizing disease, as many neuroendocrine tumors express somatostatin receptors. Specialized positron-emission tomography scans may also be helpful in tumor identification.²⁴

 

 

TREATMENT OF CUSHING SYNDROME DUE TO ECTOPIC ACTH SECRETION

6. Which of the following is most appropriate medical therapy for suppression of cortisol secretion in Cushing syndrome due to ectopic ACTH secretion?

• Spironolactone
• Dexamethasone
• Somatostatin
• Estrogen
• Ketoconazole

Hyperglycemia, hypokalemia, hypertension, psychiatric disturbances, venous thromboembolism, and systemic infections appear to be common in ectopic ACTH syndrome and often correlate with the degree of hypercortisolemia. Severe Cushing syndrome due to ectopic ACTH secretion is an emergency requiring prompt control of cortisol secretion.

First-line treatments include steroidogenesis inhibitors (ketoconazole, metyrapone, etomidate, mitotane) and glucocorticoid receptor antagonists (mifepristone). High-dose spironolactone and eplerenone can also be used to treat the hypertension and hypokalemia associated with mineralocorticoid receptor stimulation. Definitive treatment involves surgical resection, chemotherapy, or radiotherapy when applicable.^24,25

After confirmation of the diagnosis, the patient is prescribed ketoconazole and spironolactone, with substantial improvement. He subsequently is started on combination chemotherapy and radiation therapy for his small-cell lung carcinoma.

DISCUSSION

The differential diagnosis for hypokalemia is broad and relies on information obtained during the history and physical examination, followed by interpretation of selected laboratory results. Myriad pathologies in diverse organ systems, eg, diarrhea, renal tubular acidosis, and adrenal disease, may be responsible for a low serum potassium. Further categorizing potassium depletion on the basis of an associated acid-base disturbance, such as metabolic alkalosis, allows one to use an algorithmic approach that can identify specific etiologies responsible for both the potassium and the acid-base disturbances.

Using the spot urine chloride in the setting of hypokalemic metabolic alkalosis with or without hypertension can narrow the differential diagnosis and allow additional clinical findings to guide clinical problem-solving and decision-making, even for conditions not commonly encountered in routine medical practice.

Obtaining renin and aldosterone measurements in patients with potassium depletion, metabolic alkalosis, high urine chloride excretion, and hypertension permits further categorization into 3 clinical groups: elevated aldosterone and renin (secondary hyperaldosteronism), elevated aldosterone and low renin (primary hyperaldosteronism), or apparent mineralocorticoid excess wherein neither renin nor aldosterone are responsible for the syndrome.

The patient in our case had apparent mineralocorticoid excess as a consequence of an ACTH-producing small-cell carcinoma.

Whether subclinical hypothyroidism is clinically important and should be treated remains controversial. Studies have differed in their findings, and although most have found this condition to be associated with a variety of adverse outcomes, large randomized controlled trials are needed to clearly demonstrate its clinical impact in various age groups and the benefit of levothyroxine therapy.

Currently, the best practical approach is to base treatment decisions on the magnitude of elevation of thyroid-stimulating hormone (TSH) and whether the patient has thyroid autoantibodies and associated comorbid conditions.

HIGH TSH, NORMAL FREE T₄ LEVELS

Subclinical hypothyroidism is defined by elevated TSH along with a normal free thyroxine (T₄).¹

The hypothalamic-pituitary-thyroid axis is a balanced homeostatic system, and TSH and thyroid hormone levels have an inverse log-linear relation: if free T₄ and triiodothyronine (T₃) levels go down even a little, TSH levels go up a lot.²

TSH secretion is pulsatile and has a circadian rhythm: serum TSH levels are 50% higher at night and early in the morning than during the rest of the day. Thus, repeated measurements in the same patient can vary by as much as half of the reference range.³

WHAT IS THE UPPER LIMIT OF NORMAL FOR TSH?

The upper limit of normal for TSH, defined as the 97.5th percentile, is approximately 4 or 5 mIU/L depending on the laboratory and the population, but some experts believe it should be lower.³

In favor of a lower upper limit: the distribution of serum TSH levels in the healthy general population does not seem to be a typical bell-shaped Gaussian curve, but rather has a tail at the high end. Some argue that some of the individuals with values in the upper end of the normal range may actually have undiagnosed hypothyroidism and that the upper 97.5th percentile cutoff would be 2.5 mIU/L if these people were excluded.⁴ Also, TSH levels higher than 2.5 mIU/L have been associated with a higher prevalence of antithyroid antibodies and a higher risk of clinical hypothyroidism.⁵

On the other hand, lowering the upper limit of normal to 2.5 mIU/L would result in 4 times as many people receiving a diagnosis of subclinical hypothyroidism, or 22 to 28 million people in the United States.^4,6 Thus, lowering the cutoff may lead to unnecessary therapy and could even harm from overtreatment.

Another argument against lowering the upper limit of normal for TSH is that, with age, serum TSH levels shift higher.⁷ The third National Health and Nutrition Education Survey (NHANES III) found that the 97.5th percentile for serum TSH was 3.56 mIU/L for age group 20 to 29 but 7.49 mIU/L for octogenarians.^7,8

It has been suggested that the upper limit of normal for TSH be adjusted for age, race, sex, and iodine intake.³ Currently available TSH reference ranges are not adjusted for these variables, and there is not enough evidence to suggest age-appropriate ranges,⁹ although higher TSH cutoffs for treatment are advised in elderly patients.¹⁰ Interestingly, higher TSH in older people has been linked to lower mortality rates in some studies.¹¹

Authors of the NHANES III⁸ and Hanford Thyroid Disease study¹² have proposed a cutoff of 4.1 mIU/L for the upper limit of normal for serum TSH in patients with negative antithyroid antibodies and normal findings on thyroid ultrasonography.

SUBCLINICAL HYPOTHYROIDISM IS COMMON

In different studies, the prevalence of subclinical hypothyroidism has been as low as 4% and as high as 20%.^1,8,13 The prevalence is higher in women and increases with age.⁸ It is higher in iodine-sufficient areas, and it increases in iodine-deficient areas with iodine supplementation.¹⁴ Genetics also plays a role, as subclinical hypothyroidism is more common in white people than in African Americans.⁸

A difficulty in estimating the prevalence is the disagreement about the cutoff for TSH, which may differ from that in the general population in certain subgroups such as adolescents, the elderly, and pregnant women.^10,15

A VARIETY OF CAUSES

The most common cause of subclinical hypothyroidism, accounting for 60% to 80% of cases, is Hashimoto (autoimmune) thyroiditis,² in which thyroid peroxidase antibodies are usually present.^2,16

Other causes include suboptimal treatment of hypothyroidism due to other reasons such as thyroidectomy, radioactive iodine treatment, external radiation, infiltrative diseases (eg, amyloidosis, sarcoidosis, hemochromatosis), and drugs (eg, iodinated contrast, amiodarone, lithium, tyrosine kinase inhibitors) (Table 1).^1,2,16

Also important to rule out are false-positive elevations due to substances that interfere with TSH assays (eg, heterophile antibodies, rheumatoid factor, biotin, macro-TSH); reversible causes such as the recovery phase of euthyroid sick syndrome; subacute, painless, or postpartum thyroiditis; central hypo- or hyperthyroidism; and thyroid hormone resistance.

 

SUBCLINICAL HYPOTHYROIDISM CAN RESOLVE OR PROGRESS 


“Subclinical” suggests that the disease is in its early stage, with changes in TSH already apparent but decreases in thyroid hormone levels yet to come.¹⁷ And indeed, subclinical hypothyroidism can progress to overt hypothyroidism,¹⁸ although it has been reported to resolve spontaneously in half of cases within 2 years,¹⁹ typically in patients with TSH values of 4 to 6 mIU/L.²⁰ The rate of progression to overt hypothyroidism is estimated to be 33% to 55% over 10 to 20 years of follow-up.¹⁸

The risk of progression to clinical disease is higher in patients with thyroid peroxidase antibody, reported as 4.3% per year compared with 2.6% per year in those without this antibody.^20,21 In one study, the risk of developing overt hypothyroidism in those with subclinical hypothyroidism increased from 1% to 4% with doubling of the TSH.²¹ Other risk factors for progression to hypothyroidism include female sex, older age, goiter, neck irradiation or radioactive iodine exposure, and high iodine intake.^18,22

Figure 1 shows the natural history of subclinical hypothyroidism.¹

GUIDELINES FOR SCREENING DIFFER

Guidelines differ on screening for thyroid disease in the general population, owing to lack of large-scale randomized controlled trials showing treatment benefit in otherwise-healthy people with mildly elevated TSH values.

Various professional societies have adopted different criteria for aggressive case-finding in patients at risk of thyroid disease. Risk factors include family history of thyroid disease, neck irradiation, partial thyroidectomy, dyslipidemia, atrial fibrillation, unexplained weight loss, hyperprolactinemia, autoimmune disorders, and use of medications affecting thyroid function.²³

The US Preventive Services Task Force in 2014 found insufficient evidence on the benefits and harms of screening.²⁴

The American Thyroid Association (ATA) recommends screening adults starting at age 35, with repeat testing every 5 years in patients who have no signs or symptoms of hypothyroidism, and more frequently in those who do.²⁵

The American Association of Clinical Endocrinologists recommends screening in women and older patients. Their guidelines and those of the ATA also suggest screening people at high risk of thyroid disease due to risk factors such as history of autoimmune diseases, neck irradiation, or medications affecting thyroid function.²⁶

The American Academy of Family Physicians recommends screening after age 60.¹⁸

The American College of Physicians recommends screening patients over age 50 who have symptoms.¹⁸

Our approach. Although evidence is lacking to recommend routine screening in adults, aggressive case-finding and treatment in patients at risk of thyroid disease can, we believe, offset the risks associated with subclinical hypothyroidism.²⁴

CLINICAL PRESENTATION

About 70% of patients with subclinical hypothyroidism have no symptoms.¹³

Tiredness was more common in subclinical hypothyroid patients with TSH levels lower than 10 mIU/L compared with euthyroid controls in 1 study, but other studies have been unable to replicate this finding.^27,28

Other frequently reported symptoms include dry skin, cognitive slowing, poor memory, muscle weakness, cold intolerance, constipation, puffy eyes, and hoarseness.¹³

The evidence in favor of levothyroxine therapy to improve symptoms in subclinical hypothyroidism has varied, with some studies showing an improvement in symptom scores compared with placebo, while others have not shown any benefit.^29–31

In one study, the average TSH value for patients whose symptoms did not improve with therapy was 4.6 mIU/L.³¹ An explanation for the lack of effect in this group may be that the TSH values for these patients were in the high-normal range. Also, because most subclinical hypothyroid patients have no symptoms, it is difficult to ascertain symptomatic improvement. Though it is possible to conclude that levothyroxine therapy has a limited role in this group, it is important to also consider the suggestive evidence that untreated subclinical hypothyroidism may lead to increased morbidity and mortality.

 

ADVERSE EFFECTS OF SUBCLINICAL HYPOTHYROIDISM, EFFECTS OF THERAPY

Subclinical hypothyroidism has been associated with adverse metabolic, cardiovascular, neuromuscular, and cognitive effects and has been shown to have a detrimental impact on quality of life. However, studies of levothyroxine therapy in subclinical hypothyroidism have yielded mixed results.¹⁶ Subclinical hypothyroidism affects many biologic systems, and levothyroxine may have a role (Table 2).^32–117

INDIVIDUALIZED MANAGEMENT AND SHARED DECISION-MAKING

The management of subclinical hypothyroidism should be individualized on the basis of extent of thyroid dysfunction, comorbid conditions, risk factors, and patient preference.¹¹⁸ Shared decision-making is key, weighing the risks and benefits of levothyroxine treatment and the patient’s goals.

There is some evidence to support levothyroxine treatment in nonpregnant patients with overt hypothyroidism (TSH > 10 mIU/L) or in patients with TSH 5 to 10 mIU/L with symptoms or hyperlipidemia and in younger patients at risk of cardiovascular disease.¹¹⁸ Table 3 describes various patient factors that should be considered during clinical evaluation and decisions about levothyroxine treatment in subclinical hypothyroidism.

The risks of treatment should be kept in mind and explained to the patient. Levothyroxine has a narrow therapeutic range, causing a possibility of overreplacement, and a half-life of 7 days that can cause dosing errors to have longer effect.^118,119

Adherence can be a challenge. The drug needs to be taken on an empty stomach because foods and supplements interfere with its absorption.^118,120 In addition, the cost of medication, frequent biochemical monitoring, and possible need for titration can add to financial burden.

When choosing the dose, one should consider the degree of hypothyroidism or TSH elevation and the patient’s weight, and adjust the dose gently.

If the TSH is high-normal

It is proposed that a TSH range of 3 to 5 mIU/L overlaps with normal thyroid function in a great segment of the population, and at this level it is probably not associated with clinically significant consequences. For these reasons, levothyroxine therapy is not thought to be beneficial for those with TSH in this range.

Pollock et al¹²¹ found that, in patients with symptoms suggesting hypothyroidism and TSH values in the upper end of the normal range, there was no improvement in cognitive function or psychological well-being after 12 weeks of levothyroxine therapy.

However, due to the concern for possible adverse maternal and fetal outcomes and low IQ in children of pregnant patients with subclinical hypothyroidism, levothyroxine therapy is advised in those who are pregnant or planning pregnancy who have TSH levels higher than 2.5 mIU/L, especially if they have thyroid peroxidase antibody. Levothyroxine therapy is not recommended for pregnant patients with negative thyroid peroxidase antibody and TSH within the pregnancy-specific range or less than 4 mIU/L if the reference ranges are unavailable.

Keep in mind that, even at these TSH values, there is risk of progression to overt hypothyroidism, especially in the presence of thyroid peroxidase antibody, so patients in this group should be monitored closely.

If TSH is mildly elevated

The evidence to support levothyroxine therapy in patients with subclinical hypothyroidism with TSH levels less than 10 mIU/L remains inconclusive, and the decision to treat should be based on clinical judgment.² The studies that have looked at the benefit of treating subclinical hypothyroidism in terms of cardiac, neuromuscular, cognitive, and neuropsychiatric outcomes have included patients with a wide range of TSH levels, and some of these studies were not stratified on the basis of degree of TSH elevation.

The risk that subclinical hypothyroidism will progress to overt hypothyroidism in patients with TSH higher than 8 mIU/L is high, and in 70% of these patients, the TSH level rises to more than 10 mIU/L within 4 years. Early treatment should be considered if the TSH is higher than 7 or 8 mIU/L.

If TSH is higher than 10 mIU/L

The strongest evidence in favor of treating subclinical hypothyroidism is in patients with TSH levels higher than 10 mIU/L.² Thyroid dysfunction with this degree of TSH elevation has been associated with adverse cardiometabolic, neuromuscular, cognitive, and psychiatric effects as described above, and has been shown to improve with levothyroxine therapy.

Figure 2 outlines an algorithmic approach to subclinical hypothyroidism in nonpregnant patients as suggested by Peeters.¹²²

Whether subclinical hypothyroidism is clinically important and should be treated remains controversial. Studies have differed in their findings, and although most have found this condition to be associated with a variety of adverse outcomes, large randomized controlled trials are needed to clearly demonstrate its clinical impact in various age groups and the benefit of levothyroxine therapy.

Currently, the best practical approach is to base treatment decisions on the magnitude of elevation of thyroid-stimulating hormone (TSH) and whether the patient has thyroid autoantibodies and associated comorbid conditions.

HIGH TSH, NORMAL FREE T₄ LEVELS

Subclinical hypothyroidism is defined by elevated TSH along with a normal free thyroxine (T₄).¹

The hypothalamic-pituitary-thyroid axis is a balanced homeostatic system, and TSH and thyroid hormone levels have an inverse log-linear relation: if free T₄ and triiodothyronine (T₃) levels go down even a little, TSH levels go up a lot.²

TSH secretion is pulsatile and has a circadian rhythm: serum TSH levels are 50% higher at night and early in the morning than during the rest of the day. Thus, repeated measurements in the same patient can vary by as much as half of the reference range.³

WHAT IS THE UPPER LIMIT OF NORMAL FOR TSH?

The upper limit of normal for TSH, defined as the 97.5th percentile, is approximately 4 or 5 mIU/L depending on the laboratory and the population, but some experts believe it should be lower.³

In favor of a lower upper limit: the distribution of serum TSH levels in the healthy general population does not seem to be a typical bell-shaped Gaussian curve, but rather has a tail at the high end. Some argue that some of the individuals with values in the upper end of the normal range may actually have undiagnosed hypothyroidism and that the upper 97.5th percentile cutoff would be 2.5 mIU/L if these people were excluded.⁴ Also, TSH levels higher than 2.5 mIU/L have been associated with a higher prevalence of antithyroid antibodies and a higher risk of clinical hypothyroidism.⁵

On the other hand, lowering the upper limit of normal to 2.5 mIU/L would result in 4 times as many people receiving a diagnosis of subclinical hypothyroidism, or 22 to 28 million people in the United States.^4,6 Thus, lowering the cutoff may lead to unnecessary therapy and could even harm from overtreatment.

Another argument against lowering the upper limit of normal for TSH is that, with age, serum TSH levels shift higher.⁷ The third National Health and Nutrition Education Survey (NHANES III) found that the 97.5th percentile for serum TSH was 3.56 mIU/L for age group 20 to 29 but 7.49 mIU/L for octogenarians.^7,8

It has been suggested that the upper limit of normal for TSH be adjusted for age, race, sex, and iodine intake.³ Currently available TSH reference ranges are not adjusted for these variables, and there is not enough evidence to suggest age-appropriate ranges,⁹ although higher TSH cutoffs for treatment are advised in elderly patients.¹⁰ Interestingly, higher TSH in older people has been linked to lower mortality rates in some studies.¹¹

Authors of the NHANES III⁸ and Hanford Thyroid Disease study¹² have proposed a cutoff of 4.1 mIU/L for the upper limit of normal for serum TSH in patients with negative antithyroid antibodies and normal findings on thyroid ultrasonography.

SUBCLINICAL HYPOTHYROIDISM IS COMMON

In different studies, the prevalence of subclinical hypothyroidism has been as low as 4% and as high as 20%.^1,8,13 The prevalence is higher in women and increases with age.⁸ It is higher in iodine-sufficient areas, and it increases in iodine-deficient areas with iodine supplementation.¹⁴ Genetics also plays a role, as subclinical hypothyroidism is more common in white people than in African Americans.⁸

A difficulty in estimating the prevalence is the disagreement about the cutoff for TSH, which may differ from that in the general population in certain subgroups such as adolescents, the elderly, and pregnant women.^10,15

A VARIETY OF CAUSES

The most common cause of subclinical hypothyroidism, accounting for 60% to 80% of cases, is Hashimoto (autoimmune) thyroiditis,² in which thyroid peroxidase antibodies are usually present.^2,16

Other causes include suboptimal treatment of hypothyroidism due to other reasons such as thyroidectomy, radioactive iodine treatment, external radiation, infiltrative diseases (eg, amyloidosis, sarcoidosis, hemochromatosis), and drugs (eg, iodinated contrast, amiodarone, lithium, tyrosine kinase inhibitors) (Table 1).^1,2,16

Also important to rule out are false-positive elevations due to substances that interfere with TSH assays (eg, heterophile antibodies, rheumatoid factor, biotin, macro-TSH); reversible causes such as the recovery phase of euthyroid sick syndrome; subacute, painless, or postpartum thyroiditis; central hypo- or hyperthyroidism; and thyroid hormone resistance.

 

SUBCLINICAL HYPOTHYROIDISM CAN RESOLVE OR PROGRESS 


“Subclinical” suggests that the disease is in its early stage, with changes in TSH already apparent but decreases in thyroid hormone levels yet to come.¹⁷ And indeed, subclinical hypothyroidism can progress to overt hypothyroidism,¹⁸ although it has been reported to resolve spontaneously in half of cases within 2 years,¹⁹ typically in patients with TSH values of 4 to 6 mIU/L.²⁰ The rate of progression to overt hypothyroidism is estimated to be 33% to 55% over 10 to 20 years of follow-up.¹⁸

The risk of progression to clinical disease is higher in patients with thyroid peroxidase antibody, reported as 4.3% per year compared with 2.6% per year in those without this antibody.^20,21 In one study, the risk of developing overt hypothyroidism in those with subclinical hypothyroidism increased from 1% to 4% with doubling of the TSH.²¹ Other risk factors for progression to hypothyroidism include female sex, older age, goiter, neck irradiation or radioactive iodine exposure, and high iodine intake.^18,22

Figure 1 shows the natural history of subclinical hypothyroidism.¹

GUIDELINES FOR SCREENING DIFFER

Guidelines differ on screening for thyroid disease in the general population, owing to lack of large-scale randomized controlled trials showing treatment benefit in otherwise-healthy people with mildly elevated TSH values.

Various professional societies have adopted different criteria for aggressive case-finding in patients at risk of thyroid disease. Risk factors include family history of thyroid disease, neck irradiation, partial thyroidectomy, dyslipidemia, atrial fibrillation, unexplained weight loss, hyperprolactinemia, autoimmune disorders, and use of medications affecting thyroid function.²³

The US Preventive Services Task Force in 2014 found insufficient evidence on the benefits and harms of screening.²⁴

The American Thyroid Association (ATA) recommends screening adults starting at age 35, with repeat testing every 5 years in patients who have no signs or symptoms of hypothyroidism, and more frequently in those who do.²⁵

The American Association of Clinical Endocrinologists recommends screening in women and older patients. Their guidelines and those of the ATA also suggest screening people at high risk of thyroid disease due to risk factors such as history of autoimmune diseases, neck irradiation, or medications affecting thyroid function.²⁶

The American Academy of Family Physicians recommends screening after age 60.¹⁸

The American College of Physicians recommends screening patients over age 50 who have symptoms.¹⁸

Our approach. Although evidence is lacking to recommend routine screening in adults, aggressive case-finding and treatment in patients at risk of thyroid disease can, we believe, offset the risks associated with subclinical hypothyroidism.²⁴

CLINICAL PRESENTATION

About 70% of patients with subclinical hypothyroidism have no symptoms.¹³

Tiredness was more common in subclinical hypothyroid patients with TSH levels lower than 10 mIU/L compared with euthyroid controls in 1 study, but other studies have been unable to replicate this finding.^27,28

Other frequently reported symptoms include dry skin, cognitive slowing, poor memory, muscle weakness, cold intolerance, constipation, puffy eyes, and hoarseness.¹³

The evidence in favor of levothyroxine therapy to improve symptoms in subclinical hypothyroidism has varied, with some studies showing an improvement in symptom scores compared with placebo, while others have not shown any benefit.^29–31

In one study, the average TSH value for patients whose symptoms did not improve with therapy was 4.6 mIU/L.³¹ An explanation for the lack of effect in this group may be that the TSH values for these patients were in the high-normal range. Also, because most subclinical hypothyroid patients have no symptoms, it is difficult to ascertain symptomatic improvement. Though it is possible to conclude that levothyroxine therapy has a limited role in this group, it is important to also consider the suggestive evidence that untreated subclinical hypothyroidism may lead to increased morbidity and mortality.

 

ADVERSE EFFECTS OF SUBCLINICAL HYPOTHYROIDISM, EFFECTS OF THERAPY

Subclinical hypothyroidism has been associated with adverse metabolic, cardiovascular, neuromuscular, and cognitive effects and has been shown to have a detrimental impact on quality of life. However, studies of levothyroxine therapy in subclinical hypothyroidism have yielded mixed results.¹⁶ Subclinical hypothyroidism affects many biologic systems, and levothyroxine may have a role (Table 2).^32–117

INDIVIDUALIZED MANAGEMENT AND SHARED DECISION-MAKING

The management of subclinical hypothyroidism should be individualized on the basis of extent of thyroid dysfunction, comorbid conditions, risk factors, and patient preference.¹¹⁸ Shared decision-making is key, weighing the risks and benefits of levothyroxine treatment and the patient’s goals.

There is some evidence to support levothyroxine treatment in nonpregnant patients with overt hypothyroidism (TSH > 10 mIU/L) or in patients with TSH 5 to 10 mIU/L with symptoms or hyperlipidemia and in younger patients at risk of cardiovascular disease.¹¹⁸ Table 3 describes various patient factors that should be considered during clinical evaluation and decisions about levothyroxine treatment in subclinical hypothyroidism.

The risks of treatment should be kept in mind and explained to the patient. Levothyroxine has a narrow therapeutic range, causing a possibility of overreplacement, and a half-life of 7 days that can cause dosing errors to have longer effect.^118,119

Adherence can be a challenge. The drug needs to be taken on an empty stomach because foods and supplements interfere with its absorption.^118,120 In addition, the cost of medication, frequent biochemical monitoring, and possible need for titration can add to financial burden.

When choosing the dose, one should consider the degree of hypothyroidism or TSH elevation and the patient’s weight, and adjust the dose gently.

If the TSH is high-normal

It is proposed that a TSH range of 3 to 5 mIU/L overlaps with normal thyroid function in a great segment of the population, and at this level it is probably not associated with clinically significant consequences. For these reasons, levothyroxine therapy is not thought to be beneficial for those with TSH in this range.

Pollock et al¹²¹ found that, in patients with symptoms suggesting hypothyroidism and TSH values in the upper end of the normal range, there was no improvement in cognitive function or psychological well-being after 12 weeks of levothyroxine therapy.

However, due to the concern for possible adverse maternal and fetal outcomes and low IQ in children of pregnant patients with subclinical hypothyroidism, levothyroxine therapy is advised in those who are pregnant or planning pregnancy who have TSH levels higher than 2.5 mIU/L, especially if they have thyroid peroxidase antibody. Levothyroxine therapy is not recommended for pregnant patients with negative thyroid peroxidase antibody and TSH within the pregnancy-specific range or less than 4 mIU/L if the reference ranges are unavailable.

Keep in mind that, even at these TSH values, there is risk of progression to overt hypothyroidism, especially in the presence of thyroid peroxidase antibody, so patients in this group should be monitored closely.

If TSH is mildly elevated

The evidence to support levothyroxine therapy in patients with subclinical hypothyroidism with TSH levels less than 10 mIU/L remains inconclusive, and the decision to treat should be based on clinical judgment.² The studies that have looked at the benefit of treating subclinical hypothyroidism in terms of cardiac, neuromuscular, cognitive, and neuropsychiatric outcomes have included patients with a wide range of TSH levels, and some of these studies were not stratified on the basis of degree of TSH elevation.

The risk that subclinical hypothyroidism will progress to overt hypothyroidism in patients with TSH higher than 8 mIU/L is high, and in 70% of these patients, the TSH level rises to more than 10 mIU/L within 4 years. Early treatment should be considered if the TSH is higher than 7 or 8 mIU/L.

If TSH is higher than 10 mIU/L

The strongest evidence in favor of treating subclinical hypothyroidism is in patients with TSH levels higher than 10 mIU/L.² Thyroid dysfunction with this degree of TSH elevation has been associated with adverse cardiometabolic, neuromuscular, cognitive, and psychiatric effects as described above, and has been shown to improve with levothyroxine therapy.

Figure 2 outlines an algorithmic approach to subclinical hypothyroidism in nonpregnant patients as suggested by Peeters.¹²²

Whether subclinical hypothyroidism is clinically important and should be treated remains controversial. Studies have differed in their findings, and although most have found this condition to be associated with a variety of adverse outcomes, large randomized controlled trials are needed to clearly demonstrate its clinical impact in various age groups and the benefit of levothyroxine therapy.

Currently, the best practical approach is to base treatment decisions on the magnitude of elevation of thyroid-stimulating hormone (TSH) and whether the patient has thyroid autoantibodies and associated comorbid conditions.

HIGH TSH, NORMAL FREE T₄ LEVELS

Subclinical hypothyroidism is defined by elevated TSH along with a normal free thyroxine (T₄).¹

The hypothalamic-pituitary-thyroid axis is a balanced homeostatic system, and TSH and thyroid hormone levels have an inverse log-linear relation: if free T₄ and triiodothyronine (T₃) levels go down even a little, TSH levels go up a lot.²

TSH secretion is pulsatile and has a circadian rhythm: serum TSH levels are 50% higher at night and early in the morning than during the rest of the day. Thus, repeated measurements in the same patient can vary by as much as half of the reference range.³

WHAT IS THE UPPER LIMIT OF NORMAL FOR TSH?

The upper limit of normal for TSH, defined as the 97.5th percentile, is approximately 4 or 5 mIU/L depending on the laboratory and the population, but some experts believe it should be lower.³

In favor of a lower upper limit: the distribution of serum TSH levels in the healthy general population does not seem to be a typical bell-shaped Gaussian curve, but rather has a tail at the high end. Some argue that some of the individuals with values in the upper end of the normal range may actually have undiagnosed hypothyroidism and that the upper 97.5th percentile cutoff would be 2.5 mIU/L if these people were excluded.⁴ Also, TSH levels higher than 2.5 mIU/L have been associated with a higher prevalence of antithyroid antibodies and a higher risk of clinical hypothyroidism.⁵

On the other hand, lowering the upper limit of normal to 2.5 mIU/L would result in 4 times as many people receiving a diagnosis of subclinical hypothyroidism, or 22 to 28 million people in the United States.^4,6 Thus, lowering the cutoff may lead to unnecessary therapy and could even harm from overtreatment.

Another argument against lowering the upper limit of normal for TSH is that, with age, serum TSH levels shift higher.⁷ The third National Health and Nutrition Education Survey (NHANES III) found that the 97.5th percentile for serum TSH was 3.56 mIU/L for age group 20 to 29 but 7.49 mIU/L for octogenarians.^7,8

It has been suggested that the upper limit of normal for TSH be adjusted for age, race, sex, and iodine intake.³ Currently available TSH reference ranges are not adjusted for these variables, and there is not enough evidence to suggest age-appropriate ranges,⁹ although higher TSH cutoffs for treatment are advised in elderly patients.¹⁰ Interestingly, higher TSH in older people has been linked to lower mortality rates in some studies.¹¹

Authors of the NHANES III⁸ and Hanford Thyroid Disease study¹² have proposed a cutoff of 4.1 mIU/L for the upper limit of normal for serum TSH in patients with negative antithyroid antibodies and normal findings on thyroid ultrasonography.

SUBCLINICAL HYPOTHYROIDISM IS COMMON

In different studies, the prevalence of subclinical hypothyroidism has been as low as 4% and as high as 20%.^1,8,13 The prevalence is higher in women and increases with age.⁸ It is higher in iodine-sufficient areas, and it increases in iodine-deficient areas with iodine supplementation.¹⁴ Genetics also plays a role, as subclinical hypothyroidism is more common in white people than in African Americans.⁸

A difficulty in estimating the prevalence is the disagreement about the cutoff for TSH, which may differ from that in the general population in certain subgroups such as adolescents, the elderly, and pregnant women.^10,15

A VARIETY OF CAUSES

The most common cause of subclinical hypothyroidism, accounting for 60% to 80% of cases, is Hashimoto (autoimmune) thyroiditis,² in which thyroid peroxidase antibodies are usually present.^2,16

Other causes include suboptimal treatment of hypothyroidism due to other reasons such as thyroidectomy, radioactive iodine treatment, external radiation, infiltrative diseases (eg, amyloidosis, sarcoidosis, hemochromatosis), and drugs (eg, iodinated contrast, amiodarone, lithium, tyrosine kinase inhibitors) (Table 1).^1,2,16

Also important to rule out are false-positive elevations due to substances that interfere with TSH assays (eg, heterophile antibodies, rheumatoid factor, biotin, macro-TSH); reversible causes such as the recovery phase of euthyroid sick syndrome; subacute, painless, or postpartum thyroiditis; central hypo- or hyperthyroidism; and thyroid hormone resistance.

 

SUBCLINICAL HYPOTHYROIDISM CAN RESOLVE OR PROGRESS 


“Subclinical” suggests that the disease is in its early stage, with changes in TSH already apparent but decreases in thyroid hormone levels yet to come.¹⁷ And indeed, subclinical hypothyroidism can progress to overt hypothyroidism,¹⁸ although it has been reported to resolve spontaneously in half of cases within 2 years,¹⁹ typically in patients with TSH values of 4 to 6 mIU/L.²⁰ The rate of progression to overt hypothyroidism is estimated to be 33% to 55% over 10 to 20 years of follow-up.¹⁸

The risk of progression to clinical disease is higher in patients with thyroid peroxidase antibody, reported as 4.3% per year compared with 2.6% per year in those without this antibody.^20,21 In one study, the risk of developing overt hypothyroidism in those with subclinical hypothyroidism increased from 1% to 4% with doubling of the TSH.²¹ Other risk factors for progression to hypothyroidism include female sex, older age, goiter, neck irradiation or radioactive iodine exposure, and high iodine intake.^18,22

Figure 1 shows the natural history of subclinical hypothyroidism.¹

GUIDELINES FOR SCREENING DIFFER

Guidelines differ on screening for thyroid disease in the general population, owing to lack of large-scale randomized controlled trials showing treatment benefit in otherwise-healthy people with mildly elevated TSH values.

Various professional societies have adopted different criteria for aggressive case-finding in patients at risk of thyroid disease. Risk factors include family history of thyroid disease, neck irradiation, partial thyroidectomy, dyslipidemia, atrial fibrillation, unexplained weight loss, hyperprolactinemia, autoimmune disorders, and use of medications affecting thyroid function.²³

The US Preventive Services Task Force in 2014 found insufficient evidence on the benefits and harms of screening.²⁴

The American Thyroid Association (ATA) recommends screening adults starting at age 35, with repeat testing every 5 years in patients who have no signs or symptoms of hypothyroidism, and more frequently in those who do.²⁵

The American Association of Clinical Endocrinologists recommends screening in women and older patients. Their guidelines and those of the ATA also suggest screening people at high risk of thyroid disease due to risk factors such as history of autoimmune diseases, neck irradiation, or medications affecting thyroid function.²⁶

The American Academy of Family Physicians recommends screening after age 60.¹⁸

The American College of Physicians recommends screening patients over age 50 who have symptoms.¹⁸

Our approach. Although evidence is lacking to recommend routine screening in adults, aggressive case-finding and treatment in patients at risk of thyroid disease can, we believe, offset the risks associated with subclinical hypothyroidism.²⁴

CLINICAL PRESENTATION

About 70% of patients with subclinical hypothyroidism have no symptoms.¹³

Tiredness was more common in subclinical hypothyroid patients with TSH levels lower than 10 mIU/L compared with euthyroid controls in 1 study, but other studies have been unable to replicate this finding.^27,28

Other frequently reported symptoms include dry skin, cognitive slowing, poor memory, muscle weakness, cold intolerance, constipation, puffy eyes, and hoarseness.¹³

The evidence in favor of levothyroxine therapy to improve symptoms in subclinical hypothyroidism has varied, with some studies showing an improvement in symptom scores compared with placebo, while others have not shown any benefit.^29–31

In one study, the average TSH value for patients whose symptoms did not improve with therapy was 4.6 mIU/L.³¹ An explanation for the lack of effect in this group may be that the TSH values for these patients were in the high-normal range. Also, because most subclinical hypothyroid patients have no symptoms, it is difficult to ascertain symptomatic improvement. Though it is possible to conclude that levothyroxine therapy has a limited role in this group, it is important to also consider the suggestive evidence that untreated subclinical hypothyroidism may lead to increased morbidity and mortality.

 

ADVERSE EFFECTS OF SUBCLINICAL HYPOTHYROIDISM, EFFECTS OF THERAPY

Subclinical hypothyroidism has been associated with adverse metabolic, cardiovascular, neuromuscular, and cognitive effects and has been shown to have a detrimental impact on quality of life. However, studies of levothyroxine therapy in subclinical hypothyroidism have yielded mixed results.¹⁶ Subclinical hypothyroidism affects many biologic systems, and levothyroxine may have a role (Table 2).^32–117

INDIVIDUALIZED MANAGEMENT AND SHARED DECISION-MAKING

The management of subclinical hypothyroidism should be individualized on the basis of extent of thyroid dysfunction, comorbid conditions, risk factors, and patient preference.¹¹⁸ Shared decision-making is key, weighing the risks and benefits of levothyroxine treatment and the patient’s goals.

There is some evidence to support levothyroxine treatment in nonpregnant patients with overt hypothyroidism (TSH > 10 mIU/L) or in patients with TSH 5 to 10 mIU/L with symptoms or hyperlipidemia and in younger patients at risk of cardiovascular disease.¹¹⁸ Table 3 describes various patient factors that should be considered during clinical evaluation and decisions about levothyroxine treatment in subclinical hypothyroidism.

The risks of treatment should be kept in mind and explained to the patient. Levothyroxine has a narrow therapeutic range, causing a possibility of overreplacement, and a half-life of 7 days that can cause dosing errors to have longer effect.^118,119

Adherence can be a challenge. The drug needs to be taken on an empty stomach because foods and supplements interfere with its absorption.^118,120 In addition, the cost of medication, frequent biochemical monitoring, and possible need for titration can add to financial burden.

When choosing the dose, one should consider the degree of hypothyroidism or TSH elevation and the patient’s weight, and adjust the dose gently.

If the TSH is high-normal

It is proposed that a TSH range of 3 to 5 mIU/L overlaps with normal thyroid function in a great segment of the population, and at this level it is probably not associated with clinically significant consequences. For these reasons, levothyroxine therapy is not thought to be beneficial for those with TSH in this range.

Pollock et al¹²¹ found that, in patients with symptoms suggesting hypothyroidism and TSH values in the upper end of the normal range, there was no improvement in cognitive function or psychological well-being after 12 weeks of levothyroxine therapy.

However, due to the concern for possible adverse maternal and fetal outcomes and low IQ in children of pregnant patients with subclinical hypothyroidism, levothyroxine therapy is advised in those who are pregnant or planning pregnancy who have TSH levels higher than 2.5 mIU/L, especially if they have thyroid peroxidase antibody. Levothyroxine therapy is not recommended for pregnant patients with negative thyroid peroxidase antibody and TSH within the pregnancy-specific range or less than 4 mIU/L if the reference ranges are unavailable.

Keep in mind that, even at these TSH values, there is risk of progression to overt hypothyroidism, especially in the presence of thyroid peroxidase antibody, so patients in this group should be monitored closely.

If TSH is mildly elevated

The evidence to support levothyroxine therapy in patients with subclinical hypothyroidism with TSH levels less than 10 mIU/L remains inconclusive, and the decision to treat should be based on clinical judgment.² The studies that have looked at the benefit of treating subclinical hypothyroidism in terms of cardiac, neuromuscular, cognitive, and neuropsychiatric outcomes have included patients with a wide range of TSH levels, and some of these studies were not stratified on the basis of degree of TSH elevation.

The risk that subclinical hypothyroidism will progress to overt hypothyroidism in patients with TSH higher than 8 mIU/L is high, and in 70% of these patients, the TSH level rises to more than 10 mIU/L within 4 years. Early treatment should be considered if the TSH is higher than 7 or 8 mIU/L.

If TSH is higher than 10 mIU/L

The strongest evidence in favor of treating subclinical hypothyroidism is in patients with TSH levels higher than 10 mIU/L.² Thyroid dysfunction with this degree of TSH elevation has been associated with adverse cardiometabolic, neuromuscular, cognitive, and psychiatric effects as described above, and has been shown to improve with levothyroxine therapy.

Figure 2 outlines an algorithmic approach to subclinical hypothyroidism in nonpregnant patients as suggested by Peeters.¹²²

Managing patients with malignant pleural effusion can be challenging. Symptoms are often distressing, and its presence signifies advanced disease. Median survival after diagnosis is 4 to 9 months,^1–3 although prognosis varies considerably depending on the type and stage of the malignancy.

How patients are best managed depends on clinical circumstances. Physicians should consider the risks and benefits of each option while keeping in mind realistic goals of care.

This article uses brief case presentations to review management strategies for malignant pleural effusion.

CANCER IS A COMMON CAUSE OF PLEURAL EFFUSION

Physicians and surgeons, especially in tertiary care hospitals, must often manage malignant pleural effusion.⁴ Malignancy is the third leading cause of pleural effusion after heart failure and pneumonia, accounting for 44% to 77% of exudates.⁵ Although pleural effusion can arise secondary to many different malignancies, the most common causes are lung cancer in men and breast cancer in women; these cancers account for about 75% of all cases of malignant pleural effusion.^6,7

A WOMAN ON CHEMOTHERAPY WITH ASYMPTOMATIC PLEURAL EFFUSION

An 18-year-old woman with non-Hodgkin lymphoma has received her first cycle of chemotherapy and is now admitted to the hospital for diarrhea. A routine chest radiograph reveals a left-sided pleural effusion covering one-third of the thoracic cavity. She is asymptomatic and reports no shortness of breath at rest or with exertion. Her oxygen saturation level is above 92% on room air without supplemental oxygen.

Thoracentesis reveals an exudative effusion, and cytologic study shows malignant lymphoid cells, consistent with a malignant pleural effusion. Cultures are negative.

What is the appropriate next step to manage this patient’s effusion?

Observation is reasonable

This patient is experiencing no symptoms and has just begun chemotherapy for her lymphoma. Malignant pleural effusion associated with lymphoma, small-cell lung cancer, and breast cancer is most sensitive to chemotherapy.⁵ For patients who do not have symptoms from the pleural effusion and who are scheduled to receive further chemotherapy, a watch-and-wait approach is reasonable.

It is important to follow the patient for developing symptoms and obtain serial imaging to evaluate for an increase in the effusion size. We recommend repeat imaging at 2- to 4-week intervals, and sooner if symptoms develop.

If progression is evident or if the patient’s oncologist indicates that the cancer is unresponsive to systemic therapy, further intervention may be necessary with one of the options discussed below.

A MAN WITH LUNG CANCER WITH PLEURAL EFFUSION, LUNG COLLAPSE

A 42-year-old man with a history of lung cancer is admitted for worsening shortness of breath. Chest radiography reveals a large left-sided pleural effusion with complete collapse of the left lung and contralateral shift of midline structures (Figure 1). Large-volume thoracentesis improves his symptoms. Pleural fluid cytology is positive for malignant cells. A repeat chest radiograph shows incomplete expansion of the left lung, thick pleura, and pneumothorax, indicating a trapped lung (ie, one unable to expand fully). Two weeks later, his symptoms recur, and chest radiography reveals a recurrent effusion.

How should this effusion be managed?

Indwelling pleural catheter placement

In a retrospective cohort study,⁸ malignant pleural effusion recurred in 97% of patients within 1 month (mean, 4.2 days) of therapeutic aspiration, highlighting the need for definitive treatment.

In the absence of lung expansion, pleuro­desis is rarely successful, and placing an indwelling pleural catheter in symptomatic patients is the preferred strategy. The US Food and Drug Administration approved this use in 1997.⁹

Indwelling pleural catheters are narrow (15.5 French, or about 5 mm in diameter) and soft (made of silicone), with distal fenestrations. The distal end remains positioned in the pleural cavity to enable drainage of pleural fluid. The middle portion passes through subcutaneous tissue, where a polyester cuff prevents dislodgement and infection. The proximal end of the catheter remains outside the patient’s skin and is connected to a 1-way valve that prevents air or fluid flow into the pleural cavity.

Pleural fluid is typically drained every 2 or 3 days for palliation. Patients must be educated about home drainage and proper catheter care.

 

 

Indwelling pleural catheters are now initial therapy for many

Although indwelling pleural catheters were first used for patients who were not candidates for pleurodesis, they are now increasingly used as first-line therapy.

Since these devices were introduced, several clinical series including more than 800 patients have found that their use for malignant pleural infusion led to symptomatic improvement in 89% to 100% of cases, with 90% of patients needing no subsequent pleural procedures after catheter insertion.^10–13

Davies et al¹⁴ randomized 106 patients with malignant pleural effusion to either receive an indwelling pleural catheter or undergo pleurodesis. In the first 6 weeks, the 2 groups had about the same incidence of dyspnea, but the catheter group had less dyspnea at 6 months, shorter index hospitalization (0 vs 4 days), fewer hospital days in the first year for treatment-related complications (1 vs 4.5 days), and fewer patients needing follow-up pleural procedures (6% vs 22%). On the other hand, adverse events were more frequent in the indwelling pleural catheter group (40% vs 13%). The most frequent events were pleural infection, cellulitis, and catheter blockage.

Fysh et al¹⁵ also compared indwelling pleural catheter insertion and pleurodesis (based on patient choice) in patients with malignant pleural effusion. As in the previous trial, those who received a catheter required significantly fewer days in the hospital and fewer additional pleural procedures than those who received pleurodesis. Safety profiles and symptom control were comparable.

Indwelling pleural catheters have several other advantages. They have been found to be more cost-effective than talc pleurodesis in patients not expected to live long (survival < 14 weeks).¹⁶ Patients with an indwelling pleural catheter can receive chemotherapy, and concurrent treatment does not increase risk of infection.¹⁷ And a systematic review¹⁸ found a 46% rate of autopleurodesis at a median of 52 days after insertion of an indwelling pleural catheter.

Drainage rate may need to be moderated

Chest pain has been reported with the use of indwelling pleural catheters, related to rapid drainage of the effusion in the setting of failed reexpansion of the trapped lung due to thickened pleura. Drainage schedules may need to be adjusted, with more frequent draining of smaller volumes, to control dyspnea without causing significant pain.

A WOMAN WITH RECURRENT PLEURAL EFFUSION, GOOD PROGNOSIS

A 55-year-old woman with a history of breast cancer presents with shortness of breath. Chest radiography reveals a right-sided effusion, which on thoracentesis is found to be malignant. After fluid removal, repeat chest radiography shows complete lung expansion.

One month later, she returns with symptoms and recurrence of the effusion. Ultrasonography does not reveal any adhesions in the pleural space. Her oncologist informs you that her expected survival is in years.

What is the next step?

Chemical pleurodesis

Chemical pleurodesis involves introducing a sclerosant into the pleural space to provoke an intense inflammatory response, creating adhesions and fibrosis that will obliterate the space. The sclerosing agent (typically talc) can be delivered by tube thoracostomy, video-assisted thoracic surgery (VATS), or medical pleuroscopy. Although the latter 2 methods allow direct visualization of the pleural space and, in theory, a more even distribution of the sclerosing agent, current evidence does not favor 1 option over the other,¹⁹ and practice patterns vary between institutions.

Tube thoracostomy. Typically, the sclerosing agent is administered once a chest radiograph shows lung reexpansion, and tube output of pleural fluid is less than 150 mL/day.¹⁹ However, some studies indicate that if pleural apposition can be confirmed using ultrasonography, then sclerosant administration at that time leads to optimal pleurodesis efficacy and shorter hospitalization.^20,21

VATS is usually done in the operating room with the patient under general anesthesia. A double-lumen endotracheal tube allows for single-lung ventilation; a camera is then inserted into the pleural space of the collapsed lung. Multiple ports of entry are usually employed, and the entire pleural space can be visualized and the sclerosing agent instilled uniformly. The surgeon may alternatively choose to perform mechanical pleurodesis, which entails abrading the visceral and parietal pleura with dry gauze to provoke diffuse petechial hemorrhage and an inflammatory reaction. VATS can also be used to perform biopsy, lobectomy, and pneumonectomy.

Medical pleuroscopy. Medical pleuroscopy is usually done using local anesthesia with the patient awake, moderately sedated, and not intubated. Because no double-lumen endotracheal tube is used, lung collapse may not be complete, making it difficult to completely visualize the entire pleural surfaces.

Although no randomized study of VATS vs medical pleuroscopy exists, a retrospective case-matched study²² comparing VATS (under general anesthesia) to single-port VATS (under local anesthesia) noted equivalent rates of pleurodesis. However, the local anesthesia group had a lower perioperative mortality rate (0% vs 2.3%), a lower postoperative major morbidity rate (5.2% vs 9%), earlier improvement in quality of life, and shorter hospitalization (3 vs 5 days).²² In general, the diagnostic sensitivity of pleuroscopy for pleural malignancy is similar to that of VATS (93% vs 97%).^23,24

A MAN WITH PLEURAL EFFUSION AND A POOR PROGNOSIS

A 60-year-old man with metastatic pancreatic cancer is brought to the clinic for worsening shortness of breath over the past 2 months. During that time, he has lost 6 kg and has become bedridden.

On examination, he has severe cachexia and is significantly short of breath at rest with associated hypoxia. His oncologist expects him to survive less than 3 months.

His laboratory investigations reveal hypoalbuminemia and leukocytosis. A chest radiograph shows a large left-sided pleural effusion that was not present 2 months ago.

What should be done for him?

Thoracentesis, repeat as needed

Malignant pleural effusion causing dyspnea is not uncommon in certain advanced malignancies and may contribute to significant suffering at the end of life. A study of 298 patients with malignant pleural effusion noted that the presence of leukocytosis, hypoalbuminemia, and hypoxemia was associated with a poorer prognosis. Patients having all 3 factors had a median survival of 42 days.²⁵

Thoracentesis, the least invasive option that may improve dyspnea, can be done in the clinic setting and is a reasonable strategy for patients with advanced cancer and an expected survival of less than 3 months.²⁶ Although recurrence is expected, it may take up to a few weeks, and repeat thoracentesis can be performed as needed.

Managing patients with malignant pleural effusion can be challenging. Symptoms are often distressing, and its presence signifies advanced disease. Median survival after diagnosis is 4 to 9 months,^1–3 although prognosis varies considerably depending on the type and stage of the malignancy.

How patients are best managed depends on clinical circumstances. Physicians should consider the risks and benefits of each option while keeping in mind realistic goals of care.

This article uses brief case presentations to review management strategies for malignant pleural effusion.

CANCER IS A COMMON CAUSE OF PLEURAL EFFUSION

Physicians and surgeons, especially in tertiary care hospitals, must often manage malignant pleural effusion.⁴ Malignancy is the third leading cause of pleural effusion after heart failure and pneumonia, accounting for 44% to 77% of exudates.⁵ Although pleural effusion can arise secondary to many different malignancies, the most common causes are lung cancer in men and breast cancer in women; these cancers account for about 75% of all cases of malignant pleural effusion.^6,7

A WOMAN ON CHEMOTHERAPY WITH ASYMPTOMATIC PLEURAL EFFUSION

An 18-year-old woman with non-Hodgkin lymphoma has received her first cycle of chemotherapy and is now admitted to the hospital for diarrhea. A routine chest radiograph reveals a left-sided pleural effusion covering one-third of the thoracic cavity. She is asymptomatic and reports no shortness of breath at rest or with exertion. Her oxygen saturation level is above 92% on room air without supplemental oxygen.

Thoracentesis reveals an exudative effusion, and cytologic study shows malignant lymphoid cells, consistent with a malignant pleural effusion. Cultures are negative.

What is the appropriate next step to manage this patient’s effusion?

Observation is reasonable

This patient is experiencing no symptoms and has just begun chemotherapy for her lymphoma. Malignant pleural effusion associated with lymphoma, small-cell lung cancer, and breast cancer is most sensitive to chemotherapy.⁵ For patients who do not have symptoms from the pleural effusion and who are scheduled to receive further chemotherapy, a watch-and-wait approach is reasonable.

It is important to follow the patient for developing symptoms and obtain serial imaging to evaluate for an increase in the effusion size. We recommend repeat imaging at 2- to 4-week intervals, and sooner if symptoms develop.

If progression is evident or if the patient’s oncologist indicates that the cancer is unresponsive to systemic therapy, further intervention may be necessary with one of the options discussed below.

A MAN WITH LUNG CANCER WITH PLEURAL EFFUSION, LUNG COLLAPSE

A 42-year-old man with a history of lung cancer is admitted for worsening shortness of breath. Chest radiography reveals a large left-sided pleural effusion with complete collapse of the left lung and contralateral shift of midline structures (Figure 1). Large-volume thoracentesis improves his symptoms. Pleural fluid cytology is positive for malignant cells. A repeat chest radiograph shows incomplete expansion of the left lung, thick pleura, and pneumothorax, indicating a trapped lung (ie, one unable to expand fully). Two weeks later, his symptoms recur, and chest radiography reveals a recurrent effusion.

How should this effusion be managed?

Indwelling pleural catheter placement

In a retrospective cohort study,⁸ malignant pleural effusion recurred in 97% of patients within 1 month (mean, 4.2 days) of therapeutic aspiration, highlighting the need for definitive treatment.

In the absence of lung expansion, pleuro­desis is rarely successful, and placing an indwelling pleural catheter in symptomatic patients is the preferred strategy. The US Food and Drug Administration approved this use in 1997.⁹

Indwelling pleural catheters are narrow (15.5 French, or about 5 mm in diameter) and soft (made of silicone), with distal fenestrations. The distal end remains positioned in the pleural cavity to enable drainage of pleural fluid. The middle portion passes through subcutaneous tissue, where a polyester cuff prevents dislodgement and infection. The proximal end of the catheter remains outside the patient’s skin and is connected to a 1-way valve that prevents air or fluid flow into the pleural cavity.

Pleural fluid is typically drained every 2 or 3 days for palliation. Patients must be educated about home drainage and proper catheter care.

 

 

Indwelling pleural catheters are now initial therapy for many

Although indwelling pleural catheters were first used for patients who were not candidates for pleurodesis, they are now increasingly used as first-line therapy.

Since these devices were introduced, several clinical series including more than 800 patients have found that their use for malignant pleural infusion led to symptomatic improvement in 89% to 100% of cases, with 90% of patients needing no subsequent pleural procedures after catheter insertion.^10–13

Davies et al¹⁴ randomized 106 patients with malignant pleural effusion to either receive an indwelling pleural catheter or undergo pleurodesis. In the first 6 weeks, the 2 groups had about the same incidence of dyspnea, but the catheter group had less dyspnea at 6 months, shorter index hospitalization (0 vs 4 days), fewer hospital days in the first year for treatment-related complications (1 vs 4.5 days), and fewer patients needing follow-up pleural procedures (6% vs 22%). On the other hand, adverse events were more frequent in the indwelling pleural catheter group (40% vs 13%). The most frequent events were pleural infection, cellulitis, and catheter blockage.

Fysh et al¹⁵ also compared indwelling pleural catheter insertion and pleurodesis (based on patient choice) in patients with malignant pleural effusion. As in the previous trial, those who received a catheter required significantly fewer days in the hospital and fewer additional pleural procedures than those who received pleurodesis. Safety profiles and symptom control were comparable.

Indwelling pleural catheters have several other advantages. They have been found to be more cost-effective than talc pleurodesis in patients not expected to live long (survival < 14 weeks).¹⁶ Patients with an indwelling pleural catheter can receive chemotherapy, and concurrent treatment does not increase risk of infection.¹⁷ And a systematic review¹⁸ found a 46% rate of autopleurodesis at a median of 52 days after insertion of an indwelling pleural catheter.

Drainage rate may need to be moderated

Chest pain has been reported with the use of indwelling pleural catheters, related to rapid drainage of the effusion in the setting of failed reexpansion of the trapped lung due to thickened pleura. Drainage schedules may need to be adjusted, with more frequent draining of smaller volumes, to control dyspnea without causing significant pain.

A WOMAN WITH RECURRENT PLEURAL EFFUSION, GOOD PROGNOSIS

A 55-year-old woman with a history of breast cancer presents with shortness of breath. Chest radiography reveals a right-sided effusion, which on thoracentesis is found to be malignant. After fluid removal, repeat chest radiography shows complete lung expansion.

One month later, she returns with symptoms and recurrence of the effusion. Ultrasonography does not reveal any adhesions in the pleural space. Her oncologist informs you that her expected survival is in years.

What is the next step?

Chemical pleurodesis

Chemical pleurodesis involves introducing a sclerosant into the pleural space to provoke an intense inflammatory response, creating adhesions and fibrosis that will obliterate the space. The sclerosing agent (typically talc) can be delivered by tube thoracostomy, video-assisted thoracic surgery (VATS), or medical pleuroscopy. Although the latter 2 methods allow direct visualization of the pleural space and, in theory, a more even distribution of the sclerosing agent, current evidence does not favor 1 option over the other,¹⁹ and practice patterns vary between institutions.

Tube thoracostomy. Typically, the sclerosing agent is administered once a chest radiograph shows lung reexpansion, and tube output of pleural fluid is less than 150 mL/day.¹⁹ However, some studies indicate that if pleural apposition can be confirmed using ultrasonography, then sclerosant administration at that time leads to optimal pleurodesis efficacy and shorter hospitalization.^20,21

VATS is usually done in the operating room with the patient under general anesthesia. A double-lumen endotracheal tube allows for single-lung ventilation; a camera is then inserted into the pleural space of the collapsed lung. Multiple ports of entry are usually employed, and the entire pleural space can be visualized and the sclerosing agent instilled uniformly. The surgeon may alternatively choose to perform mechanical pleurodesis, which entails abrading the visceral and parietal pleura with dry gauze to provoke diffuse petechial hemorrhage and an inflammatory reaction. VATS can also be used to perform biopsy, lobectomy, and pneumonectomy.

Medical pleuroscopy. Medical pleuroscopy is usually done using local anesthesia with the patient awake, moderately sedated, and not intubated. Because no double-lumen endotracheal tube is used, lung collapse may not be complete, making it difficult to completely visualize the entire pleural surfaces.

Although no randomized study of VATS vs medical pleuroscopy exists, a retrospective case-matched study²² comparing VATS (under general anesthesia) to single-port VATS (under local anesthesia) noted equivalent rates of pleurodesis. However, the local anesthesia group had a lower perioperative mortality rate (0% vs 2.3%), a lower postoperative major morbidity rate (5.2% vs 9%), earlier improvement in quality of life, and shorter hospitalization (3 vs 5 days).²² In general, the diagnostic sensitivity of pleuroscopy for pleural malignancy is similar to that of VATS (93% vs 97%).^23,24

A MAN WITH PLEURAL EFFUSION AND A POOR PROGNOSIS

A 60-year-old man with metastatic pancreatic cancer is brought to the clinic for worsening shortness of breath over the past 2 months. During that time, he has lost 6 kg and has become bedridden.

On examination, he has severe cachexia and is significantly short of breath at rest with associated hypoxia. His oncologist expects him to survive less than 3 months.

His laboratory investigations reveal hypoalbuminemia and leukocytosis. A chest radiograph shows a large left-sided pleural effusion that was not present 2 months ago.

What should be done for him?

Thoracentesis, repeat as needed

Malignant pleural effusion causing dyspnea is not uncommon in certain advanced malignancies and may contribute to significant suffering at the end of life. A study of 298 patients with malignant pleural effusion noted that the presence of leukocytosis, hypoalbuminemia, and hypoxemia was associated with a poorer prognosis. Patients having all 3 factors had a median survival of 42 days.²⁵

Thoracentesis, the least invasive option that may improve dyspnea, can be done in the clinic setting and is a reasonable strategy for patients with advanced cancer and an expected survival of less than 3 months.²⁶ Although recurrence is expected, it may take up to a few weeks, and repeat thoracentesis can be performed as needed.

Managing patients with malignant pleural effusion can be challenging. Symptoms are often distressing, and its presence signifies advanced disease. Median survival after diagnosis is 4 to 9 months,^1–3 although prognosis varies considerably depending on the type and stage of the malignancy.

How patients are best managed depends on clinical circumstances. Physicians should consider the risks and benefits of each option while keeping in mind realistic goals of care.

This article uses brief case presentations to review management strategies for malignant pleural effusion.

CANCER IS A COMMON CAUSE OF PLEURAL EFFUSION

Physicians and surgeons, especially in tertiary care hospitals, must often manage malignant pleural effusion.⁴ Malignancy is the third leading cause of pleural effusion after heart failure and pneumonia, accounting for 44% to 77% of exudates.⁵ Although pleural effusion can arise secondary to many different malignancies, the most common causes are lung cancer in men and breast cancer in women; these cancers account for about 75% of all cases of malignant pleural effusion.^6,7

A WOMAN ON CHEMOTHERAPY WITH ASYMPTOMATIC PLEURAL EFFUSION

An 18-year-old woman with non-Hodgkin lymphoma has received her first cycle of chemotherapy and is now admitted to the hospital for diarrhea. A routine chest radiograph reveals a left-sided pleural effusion covering one-third of the thoracic cavity. She is asymptomatic and reports no shortness of breath at rest or with exertion. Her oxygen saturation level is above 92% on room air without supplemental oxygen.

Thoracentesis reveals an exudative effusion, and cytologic study shows malignant lymphoid cells, consistent with a malignant pleural effusion. Cultures are negative.

What is the appropriate next step to manage this patient’s effusion?

Observation is reasonable

This patient is experiencing no symptoms and has just begun chemotherapy for her lymphoma. Malignant pleural effusion associated with lymphoma, small-cell lung cancer, and breast cancer is most sensitive to chemotherapy.⁵ For patients who do not have symptoms from the pleural effusion and who are scheduled to receive further chemotherapy, a watch-and-wait approach is reasonable.

It is important to follow the patient for developing symptoms and obtain serial imaging to evaluate for an increase in the effusion size. We recommend repeat imaging at 2- to 4-week intervals, and sooner if symptoms develop.

If progression is evident or if the patient’s oncologist indicates that the cancer is unresponsive to systemic therapy, further intervention may be necessary with one of the options discussed below.

A MAN WITH LUNG CANCER WITH PLEURAL EFFUSION, LUNG COLLAPSE

A 42-year-old man with a history of lung cancer is admitted for worsening shortness of breath. Chest radiography reveals a large left-sided pleural effusion with complete collapse of the left lung and contralateral shift of midline structures (Figure 1). Large-volume thoracentesis improves his symptoms. Pleural fluid cytology is positive for malignant cells. A repeat chest radiograph shows incomplete expansion of the left lung, thick pleura, and pneumothorax, indicating a trapped lung (ie, one unable to expand fully). Two weeks later, his symptoms recur, and chest radiography reveals a recurrent effusion.

How should this effusion be managed?

Indwelling pleural catheter placement

In a retrospective cohort study,⁸ malignant pleural effusion recurred in 97% of patients within 1 month (mean, 4.2 days) of therapeutic aspiration, highlighting the need for definitive treatment.

In the absence of lung expansion, pleuro­desis is rarely successful, and placing an indwelling pleural catheter in symptomatic patients is the preferred strategy. The US Food and Drug Administration approved this use in 1997.⁹

Indwelling pleural catheters are narrow (15.5 French, or about 5 mm in diameter) and soft (made of silicone), with distal fenestrations. The distal end remains positioned in the pleural cavity to enable drainage of pleural fluid. The middle portion passes through subcutaneous tissue, where a polyester cuff prevents dislodgement and infection. The proximal end of the catheter remains outside the patient’s skin and is connected to a 1-way valve that prevents air or fluid flow into the pleural cavity.

Pleural fluid is typically drained every 2 or 3 days for palliation. Patients must be educated about home drainage and proper catheter care.

 

 

Indwelling pleural catheters are now initial therapy for many

Although indwelling pleural catheters were first used for patients who were not candidates for pleurodesis, they are now increasingly used as first-line therapy.

Since these devices were introduced, several clinical series including more than 800 patients have found that their use for malignant pleural infusion led to symptomatic improvement in 89% to 100% of cases, with 90% of patients needing no subsequent pleural procedures after catheter insertion.^10–13

Davies et al¹⁴ randomized 106 patients with malignant pleural effusion to either receive an indwelling pleural catheter or undergo pleurodesis. In the first 6 weeks, the 2 groups had about the same incidence of dyspnea, but the catheter group had less dyspnea at 6 months, shorter index hospitalization (0 vs 4 days), fewer hospital days in the first year for treatment-related complications (1 vs 4.5 days), and fewer patients needing follow-up pleural procedures (6% vs 22%). On the other hand, adverse events were more frequent in the indwelling pleural catheter group (40% vs 13%). The most frequent events were pleural infection, cellulitis, and catheter blockage.

Fysh et al¹⁵ also compared indwelling pleural catheter insertion and pleurodesis (based on patient choice) in patients with malignant pleural effusion. As in the previous trial, those who received a catheter required significantly fewer days in the hospital and fewer additional pleural procedures than those who received pleurodesis. Safety profiles and symptom control were comparable.

Indwelling pleural catheters have several other advantages. They have been found to be more cost-effective than talc pleurodesis in patients not expected to live long (survival < 14 weeks).¹⁶ Patients with an indwelling pleural catheter can receive chemotherapy, and concurrent treatment does not increase risk of infection.¹⁷ And a systematic review¹⁸ found a 46% rate of autopleurodesis at a median of 52 days after insertion of an indwelling pleural catheter.

Drainage rate may need to be moderated

Chest pain has been reported with the use of indwelling pleural catheters, related to rapid drainage of the effusion in the setting of failed reexpansion of the trapped lung due to thickened pleura. Drainage schedules may need to be adjusted, with more frequent draining of smaller volumes, to control dyspnea without causing significant pain.

A WOMAN WITH RECURRENT PLEURAL EFFUSION, GOOD PROGNOSIS

A 55-year-old woman with a history of breast cancer presents with shortness of breath. Chest radiography reveals a right-sided effusion, which on thoracentesis is found to be malignant. After fluid removal, repeat chest radiography shows complete lung expansion.

One month later, she returns with symptoms and recurrence of the effusion. Ultrasonography does not reveal any adhesions in the pleural space. Her oncologist informs you that her expected survival is in years.

What is the next step?

Chemical pleurodesis

Chemical pleurodesis involves introducing a sclerosant into the pleural space to provoke an intense inflammatory response, creating adhesions and fibrosis that will obliterate the space. The sclerosing agent (typically talc) can be delivered by tube thoracostomy, video-assisted thoracic surgery (VATS), or medical pleuroscopy. Although the latter 2 methods allow direct visualization of the pleural space and, in theory, a more even distribution of the sclerosing agent, current evidence does not favor 1 option over the other,¹⁹ and practice patterns vary between institutions.

Tube thoracostomy. Typically, the sclerosing agent is administered once a chest radiograph shows lung reexpansion, and tube output of pleural fluid is less than 150 mL/day.¹⁹ However, some studies indicate that if pleural apposition can be confirmed using ultrasonography, then sclerosant administration at that time leads to optimal pleurodesis efficacy and shorter hospitalization.^20,21

VATS is usually done in the operating room with the patient under general anesthesia. A double-lumen endotracheal tube allows for single-lung ventilation; a camera is then inserted into the pleural space of the collapsed lung. Multiple ports of entry are usually employed, and the entire pleural space can be visualized and the sclerosing agent instilled uniformly. The surgeon may alternatively choose to perform mechanical pleurodesis, which entails abrading the visceral and parietal pleura with dry gauze to provoke diffuse petechial hemorrhage and an inflammatory reaction. VATS can also be used to perform biopsy, lobectomy, and pneumonectomy.

Medical pleuroscopy. Medical pleuroscopy is usually done using local anesthesia with the patient awake, moderately sedated, and not intubated. Because no double-lumen endotracheal tube is used, lung collapse may not be complete, making it difficult to completely visualize the entire pleural surfaces.

Although no randomized study of VATS vs medical pleuroscopy exists, a retrospective case-matched study²² comparing VATS (under general anesthesia) to single-port VATS (under local anesthesia) noted equivalent rates of pleurodesis. However, the local anesthesia group had a lower perioperative mortality rate (0% vs 2.3%), a lower postoperative major morbidity rate (5.2% vs 9%), earlier improvement in quality of life, and shorter hospitalization (3 vs 5 days).²² In general, the diagnostic sensitivity of pleuroscopy for pleural malignancy is similar to that of VATS (93% vs 97%).^23,24

A MAN WITH PLEURAL EFFUSION AND A POOR PROGNOSIS

A 60-year-old man with metastatic pancreatic cancer is brought to the clinic for worsening shortness of breath over the past 2 months. During that time, he has lost 6 kg and has become bedridden.

On examination, he has severe cachexia and is significantly short of breath at rest with associated hypoxia. His oncologist expects him to survive less than 3 months.

His laboratory investigations reveal hypoalbuminemia and leukocytosis. A chest radiograph shows a large left-sided pleural effusion that was not present 2 months ago.

What should be done for him?

Thoracentesis, repeat as needed

Malignant pleural effusion causing dyspnea is not uncommon in certain advanced malignancies and may contribute to significant suffering at the end of life. A study of 298 patients with malignant pleural effusion noted that the presence of leukocytosis, hypoalbuminemia, and hypoxemia was associated with a poorer prognosis. Patients having all 3 factors had a median survival of 42 days.²⁵

Thoracentesis, the least invasive option that may improve dyspnea, can be done in the clinic setting and is a reasonable strategy for patients with advanced cancer and an expected survival of less than 3 months.²⁶ Although recurrence is expected, it may take up to a few weeks, and repeat thoracentesis can be performed as needed.

At present, 300,000 US women undergo breast augmentation surgery each year,¹ making this the second most common aesthetic procedure in women (after liposuction),^2–4 and making it extremely likely that clinicians will encounter women who have breast implants. In addition, approximately 110,000 women undergo breast reconstructive surgery after mastectomy, of whom more than 88,000 (81%) receive implants (2016 data).⁵

This review discusses the evolution of breast implants, their complications, and key considerations with regard to aesthetic and reconstructive breast surgery, as the principles are similar.

EVOLUTION OF IMPLANTS

Reports of breast augmentation surgery, also known as augmentation mammoplasty, date back to 1895, when a fatty tumor (lipoma) was successfully transplanted from a patient’s back to a breast defect in a mastectomy patient.^2,3,6,7 In the 1930s, implantation of a glass ball into a patient’s breast marked the first implant-based breast augmentation.⁶ By 1954, attempts at breast augmentation using local dermal-fat flaps, adipose tissue, and even omentum were described.

Alloplastic materials gained popularity throughout the 1950s and 1960s and included polyurethane, polytetrafluoroethylene (Teflon), and other synthetics. Adverse reactions associated with alloplastic materials were plentiful: local tissue reactions, distortion of the breast mound, increased firmness, and discomfort all contributed to the eventual discontinuation of their use. The history of alloplastic breast augmentation also included epoxy resin, shellac, beeswax, paraffin, rubber, petroleum jelly, and liquefied silicone. Outcomes were not good, and many patients ultimately needed mastectomy.⁷

The first modern breast prosthesis was developed in 1961, and since then, implant composition and design have evolved significantly.⁸

From silicone to saline, and back again

The first silicone gel implants, introduced in the early 1960s,^8–19 had high complication rates—some centers reported an incidence of capsular contracture of up to 70%.^8,11 This is a foreign body reaction in which pathologic scar tissue encases the implant, causing it to distort, appear misshapen, harden, and even become painful.¹¹ Attempts to minimize this reaction led to later generations of silicone implants with polyurethane shells.¹²

Inflatable implants filled with sterile saline solution were originally developed in France in 1965. Unlike silicone implants, saline implants have undergone minimal changes since their inception, and grew in popularity during the 1970s in view of the high rates of capsular contracture with silicone implants.⁸ However, saline implants have their own problems, and as they became increasingly popular, deflation and the unnatural feel of saline sparked a renewed interest in silicone gel.

By the late 1980s, the thinner-shelled generation of silicone implants displayed its own frustrating complications including implant rupture, capsular contracture, infection, and possible systemic and disseminated granulomatous disease. From 1992 to 2006, the US Food and Drug Administration (FDA) placed a moratorium on silicone implants due to concerns about a possible link with autoimmune and connective tissue diseases and the possible carcinogenic nature of silicone.

While silicone implants were prohibited in the United States, development continued abroad, and eventually the moratorium was lifted after several meta-analyses failed to reveal any link regarding the aforementioned concerns.¹³

Today, silicone gel implants dominate the world market.¹⁴ In the United States, approximately 60% of implants contain silicone gel filler, and trends are similar in Europe.⁷

Table 1 summarizes the evolution of silicone breast implants over the last 50 years.^2,6,11,12Table 2 lists the advantages and disadvantages of silicone and saline breast implants.^2,6,8,15

CURRENT IMPLANT OPTIONS

Currently, 3 companies (Allergan, Mentor, Sientra) manufacture and distribute breast implants and implant-associated products such as tissue expanders and sizers in the US market.⁶

Another company, Motiva, makes an implant that is available in Europe, Asia, and Australia, and the device is currently undergoing a 10-year clinical trial in the United States that began recruiting patients in 16 centers in April 2018.¹⁶ Pending final approval, the Cleveland Clinic Department of Plastic Surgery may be among the centers involved in the clinical trial of the Motiva implant. Innovations in the Motiva implant include a high-performance shell that maintains consistent strength and includes a proprietary barrier layer, improved silicone gel filler, 3-D imprinted surface texturing, and an implant shape that adapts with vertical and horizontal movement. It also contains radio-frequency identification transponders that can transmit data about the implant wirelessly.^17–19

 

 

Surface (textured vs smooth)

Developed in the 1980s, texturing of the implant surface disrupts capsule formation around the prosthesis. Additionally, texturing stabilizes an anatomically shaped (teardrop) implant within the breast pocket, reducing malrotation.^20,21

The first textured implants were covered with polyurethane foam, but they were ultimately withdrawn from the US market because of concern for in vivo degradation to carcinogenic compounds. The focus subsequently turned to texturing implant shells by mechanically creating pores of different sizes. Smooth implants, by contrast, are manufactured by repeatedly dipping the implant shell into liquid silicone.²

The capsular contraction rate has been shown to be lower with textured silicone than with smooth silicone (number needed to treat = 7–9), and evidence suggests a lower risk of needing a secondary procedure.²¹

Form-stable vs fluid-form

Silicone is a polymer. The physical properties of polymers vary greatly and depend on the length of the individual chains and the degree to which those chains are cross-linked. Liquid silicone contains short chains and sparse cross-linking, resulting in an oily compound well suited for lubrication. Silicone gel contains longer chains and more cross-linking and is therefore more viscous.

In “form-stable” implants, the silicone interior has sufficient chain length and cross-linking to retain the designed shape even at rest,² but they require slightly larger incisions.⁷ “Fluid-form” refers to an implant with silicone filler with shorter chain length, less cross-linking, and more fluidity.⁶

Shell

As with silicone fillers, the properties of silicone implant shells also depend on chain length and cross-linking within the polymer. Silicone elastomer shells (Table 1) contain extensively cross-linked chains that impart a flexible yet rubbery character. Silicone elastomers can also be found in facial implants and tissue expanders.²

Implant shape (round vs anatomic)

The shape of an implant is determined by the gel distribution inside of it. To understand gel distribution and implant shape, one must understand the gel-shell ratio. This ratio increases as cohesivity of the filler increases, and it represents increased bonding of the gel filler to the shell and a preserved implant shape at rest.

The gel-shell ratio varies among manufacturers, and a less-viscous filler may be more prone to rippling or loss of upper pole fullness in some patients. For this reason, careful analysis, patient and implant selection, and discussion of complications remain paramount.²

No anatomically shaped implant is manufactured with a smooth shell, but rather with a textured shell that resists malrotation.^6,15 However, in the United States, 95% of patients receive round implants.¹⁶

PATIENT ASSESSMENT

Before breast augmentation surgery, the surgeon assesses a number of factors—physical and psychosocial—and helps the patient choose a type and size of implant. The surgeon and patient also plan where the implants will be placed—ie, above or beneath the chest wall muscle—and where the incisions will be made. Every decision is made in close consultation with the patient, taking into account the patient’s desires and expectations, as well as what the patient’s anatomy allows. An integral component of this shared decision-making process is a discussion of the possible complications, and often photographs to better illustrate what to expect postoperatively. 

Psychosocial factors

One must consider the patient’s psychology, motivations for surgery, and emotional stability. Here, we look for underlying body dysmorphic disorder; excessive or unusual encouragement to undergo the procedure by a spouse, friends, or others; a history of other aesthetic procedures; unrealistic expectations; and other factors influencing the desire to undergo this surgery.

Choosing an implant

Implant selection must take into account the patient’s height, weight,⁷ and overall body morphology: taller patients and those with wider hips or shoulders usually require larger implants. A reliable method for determining the appropriate implant must include the current breast shape, dimensions, volume, skin elasticity, soft-tissue thickness, and overall body habitus. Ultimately, the most important considerations include breast base diameter, implant volume,²⁰ and soft-tissue envelope.

Filler type, followed by shape (round or anatomically shaped), anterior-posterior profile, and shell type (smooth or textured) are subsequent considerations (Figure 1).

Preoperative sizing can involve placing sample implants within a brassiere so that the patient can preview possible outcomes. This method is particularly effective in minimizing dissatisfaction because it shares ownership of the decision-making process.¹⁵

A computerized implant selection program available in Europe suggests a “best-fit” implant based on a clinician’s measurements.⁷

 

 

Anatomic placement

Traditionally, plastic surgeons place breast implants either beneath the pectoralis major muscle (submuscular placement) or over the pectoralis⁸ but beneath the glandular breast parenchyma (subglandular placement) (Figure 2).⁷

Advantages of submuscular placement are a smoother transition of the upper breast pole from the chest wall and less rippling visible through the skin, due to the additional muscular coverage of the implant. Another advantage is that capsular contraction rates are lower with submuscular placement, likely due to possible contamination of implants by lactiferous ductal microbes when accessing the subglandular plane.^14,20 Disadvantages are pronounced discomfort after surgery and animation deformities with muscle contraction, particularly in young, highly active patients.

A popular modification of submuscular placement involves creating a surgical dissection plane between the subglandular tissue and the pectoralis major fascia. This “dual­­plane” approach allows the parenchyma to retract superiorly and reduce breast ptosis.⁷

Incisions

The incision is most commonly made along the inframammary fold (Figure 3), but it can also be done around the areola, in the axilla, or even through the umbilicus, although this approach is less commonly used.

Table 3 highlights important considerations with regard to incision location.^15,20,21

ANTIBIOTICS

Many surgeons give a single prophylactic dose of antibiotic before surgery, a practice that some studies have shown to be effective in reducing the risk of infection.¹⁵ However, the benefit of routine postoperative use of antibiotics remains unsubstantiated¹⁵: postoperative antibiotic use does not appear to protect against infection, capsular contracture, or overall complications in primary or secondary breast augmentation surgery.²⁰

PERIOPERATIVE PERIOD

At our institution, breast augmentation surgery is an ambulatory procedure—the patient goes home the same day unless circumstances such as pain control warrant admission. This is, however, according to surgeon preference, and differs on a case-by-case basis. General anesthesia is the standard of care.¹⁵

POSTOPERATIVE PERIOD

In the immediate postoperative period, patients are instructed to wear a surgical bra for up to 6 weeks to allow stable scarring. Early mobilization is encouraged.^7,15 Depending on the patient’s situation, recovery, and healing, she may be out of work for about 1 week, sometimes more, sometimes less. 

Additional instructions are surgeon-specific. However, the patient is instructed to avoid bathing, swimming, immersion in water, and wearing underwire brassieres that could impair healing of an inferior incision; instead, patients are often instructed to wear a surgical bra provided on the day of surgery until cleared in the clinic.

Showering is allowed the next day or the second day after surgery, and of course there is no driving while on narcotics. Additionally, patients are counseled extensively regarding hematoma formation and the signs and symptoms of infection.

Patients are typically seen in clinic 1 week after surgery.

The cost of surgery may be $5,000 to $6,000 but can vary significantly from center to center depending on who the patient sees and where, and whether the patient presents for breast reconstruction after cancer or repair of congenital anomalies, or in certain cases of transgender surgery. The patient is typically responsible for the fee, but again this depends on the patient, indications, and particular insurance concerns.

IMPLANT LONGEVITY AND RUPTURE

In the United States, implant rupture rates range from 1.1% to 17.7% at 6 to 10 years after primary augmentation, 2.9% to 14.7% after revision augmentation, 1.5% to 35.4% after primary breast reconstruction, and 0% to 19.6% after revision reconstruction.¹¹

Unfortunately, the existence of multiple implant manufacturers, numerous implant generations, and poorly standardized screening protocols and reporting systems make the true rate of implant rupture difficult to assess without definitive imaging or implant retrieval.¹¹

Damage from surgical instrumentation during implantation is the most common cause of silicone breast implant rupture (50% to 64% of cases).²² Other causes include underfilling and fold flaw from capsular contracture.

Leakage of silicone gel filler may be confined to the periprosthetic capsule (intracapsular rupture) or extend beyond and into the breast parenchyma (extracapsular rupture). One study reported that only 10% of intracapsular ruptures progressed extracapsularly, while 84% of patients with extracapsular involvement remained stable for up to 2 years,²³ indicating that intracapsular rupture may not portend worsening disease.¹¹

Implant rupture occurs silently in most cases, with no clinically detectable signs or symptoms. In other cases, patients may present with alterations in breast shape and size, sudden asymmetry, firmness, pronounced capsular contracture, contour irregularity, or pain.

Aside from physical examination, comprehensive diagnostic testing includes imaging—ultrasonography, mammography, computed tomography, and magnetic resonance imaging (MRI). Of these, MRI is the method of choice, with sensitivity and specificity exceeding 90% for detecting implant rupture.¹¹ Classic findings on MRI include the “linguine” sign from a deflating implant shell, or the teardrop sign from implant sagging. Classic findings on ultrasonography include the “snowstorm” sign of extracapsular rupture and the “stepladder” sign of intracapsular rupture.

Mammography effectively detects free silicone in breast tissue with extracapsular rupture (25% of ruptures according to some studies)²³; however, it cannot detect rupture within the implant capsule. As an aside, submuscular implant placement may interfere less with screening mammography than subglandular implants do.^14,24

Current FDA recommendations to detect implant rupture encourage women with silicone breast implants to undergo screening 3 years after implantation and then every 2 years thereafter; no long-term monitoring is suggested for saline implants.¹⁵ Many plastic surgeons evaluate silicone breast implant patients every 1 to 2 years for contracture and rupture.⁸ Of note, capsular contracture impairs the effectiveness of ultrasonography and may require MRI confirmation.¹¹

If implant rupture is confirmed, the current recommendation is to remove the implant and the capsule. Another implant may be placed depending on the patient’s preference. Rigorous washout remains a key feature of any surgical intervention for ruptured breast implants; however, in the event of extracapsular rupture, resection of silicone granulomas may also be required.¹¹

Reoperation rates for primary breast augmentation surgery approach 20% and are even higher for secondary augmentation over a patient’s lifetime—the highest rate of all aesthetic procedures.^7,14

 

 

CAPSULAR CONTRACTURE

Capsular contracture is the most common complication of breast augmentation,²⁵ typically presenting within the first postoperative year,^26,27 and the risk increases over time.²⁸ It occurs with both silicone and saline breast implants.

In some studies, the incidence exceeded 4% in the first 2 years after surgery,²⁹ and nearly 50% by 10 years.³⁰ Other studies found rates of 0% to 20% over 13 years.²⁰

The etiology is not well understood and is presumed to be multifactorial, with proposed mechanisms and factors that include bacterial contamination, surface texturing, the implant pocket selected, the incision type, drain placement, antibiotic use, and smoking.²⁵

A meta-analysis from 17,000 implants found that the risk of capsular contracture was significantly higher when an implant was placed in a subglandular pocket than in a submuscular pocket,^22,26 and that although texturing decreased capsular contracture compared with smooth implants, the effect was modest when a textured or smooth implant was placed in a submuscular location.²⁸ With regard to incision location, studies have reported that the incidence of capsular contracture is highest with transaxillary and periareolar incisions, and lowest with inframammary incisions.^20,21

The leading theory is that contamination of the implant (primarily from the mammary ducts) results in biofilm formation. Subclinical hematoma surrounding the implant may also provide key bacterial nutrients.²⁰

Textured implants induce a greater inflammatory response in the capsular tissue, resulting in a thicker capsule; however, contracture rates remain lower with textured than with smooth implants.^14,31 Interestingly, lower rates of capsular contracture have been observed with later-generation, cohesive-gel, form-stable implants than with those of earlier generations.¹²

Although more research is needed, silicone implants appear to confer a higher risk of capsular contracture than saline implants.^14,20

Irrigating the breast pocket intraoperatively with triple antibiotic solution (bacitracin, cefazolin, and gentamicin) before placing the implant may decrease the capsular contracture rate.^15,20

Treatments for capsular contracture include pocket modifications such as capsulotomy (making releasing, relaxing incisions in the scar capsule encasing the implant), capsulectomy (removing portions of or the entire capsule), and replacing the implant in the other pocket (ie, if the original implant was subglandular, the replacement is placed in the submuscular pocket). Patients who have contractures that fail to respond to these treatments may ultimately benefit from implant removal and autologous reconstruction (autoaugmentation) rather than implant replacement.^32,33

ADDITIONAL COMPLICATIONS

Other complications include infection, malposition, rippling, seroma, hematoma, and sensory alterations.

Irrigation during the implantation procedure with a triple antibiotic solution consisting of bacitracin, gentamycin, and cephalexin in normal saline decreases infection and seroma rates.^15,20,34

Some surgeons also choose to irrigate the pocket with a betadine solution, or to cleanse the skin with betadine and place sterile towels and redrape before inserting the implant. Additionally, many prefer using a sterile device much like a pastry funnel called a Keller funnel to insert the implant into the breast pocket.³⁵

Infection is less common with cosmetic augmentations than with implant-based breast reconstruction, likely because of healthier, well-vascularized tissue in patients undergoing cosmetic surgery than in those undergoing mastectomy.¹⁴

Seroma is thought to be a consequence of texturing, and more so with macro- vs microtexturing. Though poorly understood, an association between texturing and double capsules has also been reported.^12,20

After primary breast augmentation, 10-year follow-up rates of capsular contracture, seroma, rippling, and malposition vary across the 3 major silicone implant manufacturers.¹² Hematoma and infection occur in less than 1% of primary augmentation patients.¹⁵

Malposition of the implant over time is less frequent with textured implants because of the higher coefficient of friction compared with smooth implants^.6,8,15

Visible skin rippling may be a consequence of texturing and also of thin body habitus, eg, in patients with a body mass index less than 18.5 kg/m². If the soft-tissue layer of the breast is thin, the natural rippling of smooth saline implant shells are more likely to show when placed in the subglandular pocket. Form-stable implants, by contrast, resist rippling.^12,15

Large implants and extensive lateral dissection can cause alterations in nipple sensation and sensory loss within lower breast pole skin. Axillary incisions may traumatize or damage the intercostobrachial nerve, resulting in upper inner arm sensory aberrations.

Ultimately, the 10-year incidence of secondary surgery ranges from 0% to 36% and the 10-year incidence of capsular contracture ranges from 11% to 19%.¹⁵ Additional cosmetic complaints after augmentation with implants include enlargement of the areola and engorgement of breast veins.¹⁴

BREAST CANCER AND DETECTION

Patients with or without implants do not seem to differ with regard to breast cancer stage upon detection, tumor burden, recurrence, or survival. However, more patients with implants may present with palpable masses, invasive tumors, axillary metastasis, and falsely negative mammograms.

Breast implants may actually facilitate cancer detection on physical examination by providing a more dense or stable surface upon which to palpate the breast tissue. Although they do not necessarily impair mastectomy or breast reconstruction, they may result in an increased rate of revision surgery after breast conservation therapy.^24,36 Mammography remains the standard of care for radiologic diagnosis but can be further supported by MRI and ultrasonography if necessary in patients with implants.

 

 

AUTOIMMUNE DISEASES

Although concerns persist, multiple studies have demonstrated the safety of fourth- and fifth-generation silicone breast implants with regard to autoimmune disease.⁷

In various clinical studies in mastectomy patients who underwent breast reconstruction with either silicone implants or autologous tissue, no difference was found with regard to the incidence of autoimmune diseases.² Additionally, in meta-analyses of data from more than 87,000 women, no association was found between connective tissue disease and silicone breast implants.^2,11 One study^11,23 noted no increase in autoantibodies in patients with undamaged silicone implants vs patients who experienced rupture.

Studies have also demonstrated that in children born to mothers with breast implants, the risk of rheumatic disease, esophageal disorders, congenital malformations, and death during the perinatal period is comparable with that in controls.³⁷ Another study, examining breastfeeding in women with silicone breast implants, showed no significant difference in silicon levels (used as a proxy for silicone) in breast milk compared with controls without implants; silicon levels were found to be significantly higher in cow’s milk and store-bought formulas.³⁸

BREAST IMPLANT-ASSOCIATED ANAPLASTIC LARGE-CELL LYMPHOMA

Breast implant-associated anaplastic large-cell lymphoma (BIA-ALCL) is a subtype of T-cell lymphoma that develops in tissue adjacent to breast implants. It typically presents as breast swelling 2 to 38 years (mean of 8 years) after implant insertion.^39,40 The swelling may be secondary to periprosthetic seroma formation or, more rarely, palpable disease in the axilla. Patients occasionally complain of pain and, rarely, constitutional symptoms.²⁰ BIA-ALCL is not a disease of the surrounding breast tissue, but rather of the fibrous periprosthetic capsule.²¹

Of note, there is no documented case involving smooth implants,^41–43 but it may be related to fifth-generation textured implants.⁶ At present, it is not possible to definitively state which implant is associated with this condition; hence, more data are needed, and this association is currently under study.

The absolute risk of BIA-ALCL was reported in a Dutch study³⁹ as 1 in 35,000 by age 50, 1 in 12,000 by age 70, and 1 in 7,000 by age 75, with a number needed to harm of 6,920. Overall lifetime risk was estimated at 1 in 30,000 for women with textured implants in a 2015 US study.⁴⁰ In comparison, breast cancer risk is about 1 in 8 women. There is no apparent predilection for patients who underwent cosmetic augmentation vs reconstruction, or who received silicone vs saline implants.

The diagnosis is confirmed by ultrasonographically guided fine-needle aspiration of seroma fluid and subsequent immunohistochemical testing for CD30-positive and ALK-negative T lymphocytes. Other than positron-emission tomography for staging after diagnosis confirmation, imaging is ineffective. Expert opinion does not recommend routine screening unless the aforementioned symptoms arise.

Treatment involves implant removal and total capsulectomy, with samples sent for pathology study with cytokeratin staining.¹² Of note, in all cases of BIA-ALCL in which the disease was limited to the circumscribed scar tissue of the breast capsule, complete surgical excision has proved curative, whereas incomplete capsulectomy portends a greater risk of recurrence and decreased survival.⁴⁴

In cases of advanced or recurrent ALCL, diagnosed late or inappropriately, the National Comprehensive Cancer Network recommends a multidisciplinary approach involving adjuvant chemotherapy and radiation.⁴⁴ Anecdotally, at our institution, we have recently treated several cases of advanced ALCL presenting with invasive chest wall masses with extirpative surgery and subsequent reconstruction with the assistance of our thoracic surgery colleagues, as well as the aforementioned multidisciplinary approach using adjuvant therapy.

The mechanism of this malignancy is currently under investigation, but the current theory implicates an exaggerated lymphoproliferative response to bacterial contamination of the capsule superimposed upon genetic factors in susceptible patients.^42,43

National societies advise plastic surgeons to discuss the risk of BIA-ALCL with all patients at the time of breast augmentation consultation and to report all confirmed cases to the PROFILE registry (Patient Registry and Outcomes for Breast Implants and Anaplastic Large Cell Lymphoma Etiology and Epidemiology).⁴⁵

ARE PATIENTS HAPPIER AFTERWARD?

Studies have shown that after undergoing breast augmentation surgery, patients note improvement in body image, and satisfaction rates range from 85% to 95% with respect to self-confidence and body image.⁴⁶ An evaluation of patient responses on the validated BREAST-Q Augmentation Questionnaire showed the following satisfaction rates: breasts 83%, psychosocial well-being 88%, and sexual functioning 81%.¹⁵

Although epidemiologic studies have reported higher suicide rates in women with cosmetic breast implants, this likely stems from preoperative psychological factors and underscores the role of psychiatric referral in patients with a mental health history or in those whom the surgeon deems it necessary.⁴⁶

Several high-quality studies have demonstrated that quality of life and psychosocial functioning (including depression) markedly improve after breast augmentation surgery.⁴⁷ Among a cohort of Norwegian patients, breast implant surgery resulted in improved motivation to perform daily activities, as well as improved quality of life from both a psychosocial and aesthetic perspective.⁴⁸ Interestingly, a recent study reported that patients who underwent breast implant surgery alone reported greater satisfaction and psychosocial quality of life than patients who underwent combination breast augmentation and mastopexy (breast-lifting) surgery.⁴⁹

Additional data are needed to refine our understanding of the complex interplay between psychosocial factors before and after surgery in patients seeking and undergoing breast augmentation procedures.

At present, 300,000 US women undergo breast augmentation surgery each year,¹ making this the second most common aesthetic procedure in women (after liposuction),^2–4 and making it extremely likely that clinicians will encounter women who have breast implants. In addition, approximately 110,000 women undergo breast reconstructive surgery after mastectomy, of whom more than 88,000 (81%) receive implants (2016 data).⁵

This review discusses the evolution of breast implants, their complications, and key considerations with regard to aesthetic and reconstructive breast surgery, as the principles are similar.

EVOLUTION OF IMPLANTS

Reports of breast augmentation surgery, also known as augmentation mammoplasty, date back to 1895, when a fatty tumor (lipoma) was successfully transplanted from a patient’s back to a breast defect in a mastectomy patient.^2,3,6,7 In the 1930s, implantation of a glass ball into a patient’s breast marked the first implant-based breast augmentation.⁶ By 1954, attempts at breast augmentation using local dermal-fat flaps, adipose tissue, and even omentum were described.

Alloplastic materials gained popularity throughout the 1950s and 1960s and included polyurethane, polytetrafluoroethylene (Teflon), and other synthetics. Adverse reactions associated with alloplastic materials were plentiful: local tissue reactions, distortion of the breast mound, increased firmness, and discomfort all contributed to the eventual discontinuation of their use. The history of alloplastic breast augmentation also included epoxy resin, shellac, beeswax, paraffin, rubber, petroleum jelly, and liquefied silicone. Outcomes were not good, and many patients ultimately needed mastectomy.⁷

The first modern breast prosthesis was developed in 1961, and since then, implant composition and design have evolved significantly.⁸

From silicone to saline, and back again

The first silicone gel implants, introduced in the early 1960s,^8–19 had high complication rates—some centers reported an incidence of capsular contracture of up to 70%.^8,11 This is a foreign body reaction in which pathologic scar tissue encases the implant, causing it to distort, appear misshapen, harden, and even become painful.¹¹ Attempts to minimize this reaction led to later generations of silicone implants with polyurethane shells.¹²

Inflatable implants filled with sterile saline solution were originally developed in France in 1965. Unlike silicone implants, saline implants have undergone minimal changes since their inception, and grew in popularity during the 1970s in view of the high rates of capsular contracture with silicone implants.⁸ However, saline implants have their own problems, and as they became increasingly popular, deflation and the unnatural feel of saline sparked a renewed interest in silicone gel.

By the late 1980s, the thinner-shelled generation of silicone implants displayed its own frustrating complications including implant rupture, capsular contracture, infection, and possible systemic and disseminated granulomatous disease. From 1992 to 2006, the US Food and Drug Administration (FDA) placed a moratorium on silicone implants due to concerns about a possible link with autoimmune and connective tissue diseases and the possible carcinogenic nature of silicone.

While silicone implants were prohibited in the United States, development continued abroad, and eventually the moratorium was lifted after several meta-analyses failed to reveal any link regarding the aforementioned concerns.¹³

Today, silicone gel implants dominate the world market.¹⁴ In the United States, approximately 60% of implants contain silicone gel filler, and trends are similar in Europe.⁷

Table 1 summarizes the evolution of silicone breast implants over the last 50 years.^2,6,11,12Table 2 lists the advantages and disadvantages of silicone and saline breast implants.^2,6,8,15

CURRENT IMPLANT OPTIONS

Currently, 3 companies (Allergan, Mentor, Sientra) manufacture and distribute breast implants and implant-associated products such as tissue expanders and sizers in the US market.⁶

Another company, Motiva, makes an implant that is available in Europe, Asia, and Australia, and the device is currently undergoing a 10-year clinical trial in the United States that began recruiting patients in 16 centers in April 2018.¹⁶ Pending final approval, the Cleveland Clinic Department of Plastic Surgery may be among the centers involved in the clinical trial of the Motiva implant. Innovations in the Motiva implant include a high-performance shell that maintains consistent strength and includes a proprietary barrier layer, improved silicone gel filler, 3-D imprinted surface texturing, and an implant shape that adapts with vertical and horizontal movement. It also contains radio-frequency identification transponders that can transmit data about the implant wirelessly.^17–19

 

 

Surface (textured vs smooth)

Developed in the 1980s, texturing of the implant surface disrupts capsule formation around the prosthesis. Additionally, texturing stabilizes an anatomically shaped (teardrop) implant within the breast pocket, reducing malrotation.^20,21

The first textured implants were covered with polyurethane foam, but they were ultimately withdrawn from the US market because of concern for in vivo degradation to carcinogenic compounds. The focus subsequently turned to texturing implant shells by mechanically creating pores of different sizes. Smooth implants, by contrast, are manufactured by repeatedly dipping the implant shell into liquid silicone.²

The capsular contraction rate has been shown to be lower with textured silicone than with smooth silicone (number needed to treat = 7–9), and evidence suggests a lower risk of needing a secondary procedure.²¹

Form-stable vs fluid-form

Silicone is a polymer. The physical properties of polymers vary greatly and depend on the length of the individual chains and the degree to which those chains are cross-linked. Liquid silicone contains short chains and sparse cross-linking, resulting in an oily compound well suited for lubrication. Silicone gel contains longer chains and more cross-linking and is therefore more viscous.

In “form-stable” implants, the silicone interior has sufficient chain length and cross-linking to retain the designed shape even at rest,² but they require slightly larger incisions.⁷ “Fluid-form” refers to an implant with silicone filler with shorter chain length, less cross-linking, and more fluidity.⁶

Shell

As with silicone fillers, the properties of silicone implant shells also depend on chain length and cross-linking within the polymer. Silicone elastomer shells (Table 1) contain extensively cross-linked chains that impart a flexible yet rubbery character. Silicone elastomers can also be found in facial implants and tissue expanders.²

Implant shape (round vs anatomic)

The shape of an implant is determined by the gel distribution inside of it. To understand gel distribution and implant shape, one must understand the gel-shell ratio. This ratio increases as cohesivity of the filler increases, and it represents increased bonding of the gel filler to the shell and a preserved implant shape at rest.

The gel-shell ratio varies among manufacturers, and a less-viscous filler may be more prone to rippling or loss of upper pole fullness in some patients. For this reason, careful analysis, patient and implant selection, and discussion of complications remain paramount.²

No anatomically shaped implant is manufactured with a smooth shell, but rather with a textured shell that resists malrotation.^6,15 However, in the United States, 95% of patients receive round implants.¹⁶

PATIENT ASSESSMENT

Before breast augmentation surgery, the surgeon assesses a number of factors—physical and psychosocial—and helps the patient choose a type and size of implant. The surgeon and patient also plan where the implants will be placed—ie, above or beneath the chest wall muscle—and where the incisions will be made. Every decision is made in close consultation with the patient, taking into account the patient’s desires and expectations, as well as what the patient’s anatomy allows. An integral component of this shared decision-making process is a discussion of the possible complications, and often photographs to better illustrate what to expect postoperatively. 

Psychosocial factors

One must consider the patient’s psychology, motivations for surgery, and emotional stability. Here, we look for underlying body dysmorphic disorder; excessive or unusual encouragement to undergo the procedure by a spouse, friends, or others; a history of other aesthetic procedures; unrealistic expectations; and other factors influencing the desire to undergo this surgery.

Choosing an implant

Implant selection must take into account the patient’s height, weight,⁷ and overall body morphology: taller patients and those with wider hips or shoulders usually require larger implants. A reliable method for determining the appropriate implant must include the current breast shape, dimensions, volume, skin elasticity, soft-tissue thickness, and overall body habitus. Ultimately, the most important considerations include breast base diameter, implant volume,²⁰ and soft-tissue envelope.

Filler type, followed by shape (round or anatomically shaped), anterior-posterior profile, and shell type (smooth or textured) are subsequent considerations (Figure 1).

Preoperative sizing can involve placing sample implants within a brassiere so that the patient can preview possible outcomes. This method is particularly effective in minimizing dissatisfaction because it shares ownership of the decision-making process.¹⁵

A computerized implant selection program available in Europe suggests a “best-fit” implant based on a clinician’s measurements.⁷

 

 

Anatomic placement

Traditionally, plastic surgeons place breast implants either beneath the pectoralis major muscle (submuscular placement) or over the pectoralis⁸ but beneath the glandular breast parenchyma (subglandular placement) (Figure 2).⁷

Advantages of submuscular placement are a smoother transition of the upper breast pole from the chest wall and less rippling visible through the skin, due to the additional muscular coverage of the implant. Another advantage is that capsular contraction rates are lower with submuscular placement, likely due to possible contamination of implants by lactiferous ductal microbes when accessing the subglandular plane.^14,20 Disadvantages are pronounced discomfort after surgery and animation deformities with muscle contraction, particularly in young, highly active patients.

A popular modification of submuscular placement involves creating a surgical dissection plane between the subglandular tissue and the pectoralis major fascia. This “dual­­plane” approach allows the parenchyma to retract superiorly and reduce breast ptosis.⁷

Incisions

The incision is most commonly made along the inframammary fold (Figure 3), but it can also be done around the areola, in the axilla, or even through the umbilicus, although this approach is less commonly used.

Table 3 highlights important considerations with regard to incision location.^15,20,21

ANTIBIOTICS

Many surgeons give a single prophylactic dose of antibiotic before surgery, a practice that some studies have shown to be effective in reducing the risk of infection.¹⁵ However, the benefit of routine postoperative use of antibiotics remains unsubstantiated¹⁵: postoperative antibiotic use does not appear to protect against infection, capsular contracture, or overall complications in primary or secondary breast augmentation surgery.²⁰

PERIOPERATIVE PERIOD

At our institution, breast augmentation surgery is an ambulatory procedure—the patient goes home the same day unless circumstances such as pain control warrant admission. This is, however, according to surgeon preference, and differs on a case-by-case basis. General anesthesia is the standard of care.¹⁵

POSTOPERATIVE PERIOD

In the immediate postoperative period, patients are instructed to wear a surgical bra for up to 6 weeks to allow stable scarring. Early mobilization is encouraged.^7,15 Depending on the patient’s situation, recovery, and healing, she may be out of work for about 1 week, sometimes more, sometimes less. 

Additional instructions are surgeon-specific. However, the patient is instructed to avoid bathing, swimming, immersion in water, and wearing underwire brassieres that could impair healing of an inferior incision; instead, patients are often instructed to wear a surgical bra provided on the day of surgery until cleared in the clinic.

Showering is allowed the next day or the second day after surgery, and of course there is no driving while on narcotics. Additionally, patients are counseled extensively regarding hematoma formation and the signs and symptoms of infection.

Patients are typically seen in clinic 1 week after surgery.

The cost of surgery may be $5,000 to $6,000 but can vary significantly from center to center depending on who the patient sees and where, and whether the patient presents for breast reconstruction after cancer or repair of congenital anomalies, or in certain cases of transgender surgery. The patient is typically responsible for the fee, but again this depends on the patient, indications, and particular insurance concerns.

IMPLANT LONGEVITY AND RUPTURE

In the United States, implant rupture rates range from 1.1% to 17.7% at 6 to 10 years after primary augmentation, 2.9% to 14.7% after revision augmentation, 1.5% to 35.4% after primary breast reconstruction, and 0% to 19.6% after revision reconstruction.¹¹

Unfortunately, the existence of multiple implant manufacturers, numerous implant generations, and poorly standardized screening protocols and reporting systems make the true rate of implant rupture difficult to assess without definitive imaging or implant retrieval.¹¹

Damage from surgical instrumentation during implantation is the most common cause of silicone breast implant rupture (50% to 64% of cases).²² Other causes include underfilling and fold flaw from capsular contracture.

Leakage of silicone gel filler may be confined to the periprosthetic capsule (intracapsular rupture) or extend beyond and into the breast parenchyma (extracapsular rupture). One study reported that only 10% of intracapsular ruptures progressed extracapsularly, while 84% of patients with extracapsular involvement remained stable for up to 2 years,²³ indicating that intracapsular rupture may not portend worsening disease.¹¹

Implant rupture occurs silently in most cases, with no clinically detectable signs or symptoms. In other cases, patients may present with alterations in breast shape and size, sudden asymmetry, firmness, pronounced capsular contracture, contour irregularity, or pain.

Aside from physical examination, comprehensive diagnostic testing includes imaging—ultrasonography, mammography, computed tomography, and magnetic resonance imaging (MRI). Of these, MRI is the method of choice, with sensitivity and specificity exceeding 90% for detecting implant rupture.¹¹ Classic findings on MRI include the “linguine” sign from a deflating implant shell, or the teardrop sign from implant sagging. Classic findings on ultrasonography include the “snowstorm” sign of extracapsular rupture and the “stepladder” sign of intracapsular rupture.

Mammography effectively detects free silicone in breast tissue with extracapsular rupture (25% of ruptures according to some studies)²³; however, it cannot detect rupture within the implant capsule. As an aside, submuscular implant placement may interfere less with screening mammography than subglandular implants do.^14,24

Current FDA recommendations to detect implant rupture encourage women with silicone breast implants to undergo screening 3 years after implantation and then every 2 years thereafter; no long-term monitoring is suggested for saline implants.¹⁵ Many plastic surgeons evaluate silicone breast implant patients every 1 to 2 years for contracture and rupture.⁸ Of note, capsular contracture impairs the effectiveness of ultrasonography and may require MRI confirmation.¹¹

If implant rupture is confirmed, the current recommendation is to remove the implant and the capsule. Another implant may be placed depending on the patient’s preference. Rigorous washout remains a key feature of any surgical intervention for ruptured breast implants; however, in the event of extracapsular rupture, resection of silicone granulomas may also be required.¹¹

Reoperation rates for primary breast augmentation surgery approach 20% and are even higher for secondary augmentation over a patient’s lifetime—the highest rate of all aesthetic procedures.^7,14

 

 

CAPSULAR CONTRACTURE

Capsular contracture is the most common complication of breast augmentation,²⁵ typically presenting within the first postoperative year,^26,27 and the risk increases over time.²⁸ It occurs with both silicone and saline breast implants.

In some studies, the incidence exceeded 4% in the first 2 years after surgery,²⁹ and nearly 50% by 10 years.³⁰ Other studies found rates of 0% to 20% over 13 years.²⁰

The etiology is not well understood and is presumed to be multifactorial, with proposed mechanisms and factors that include bacterial contamination, surface texturing, the implant pocket selected, the incision type, drain placement, antibiotic use, and smoking.²⁵

A meta-analysis from 17,000 implants found that the risk of capsular contracture was significantly higher when an implant was placed in a subglandular pocket than in a submuscular pocket,^22,26 and that although texturing decreased capsular contracture compared with smooth implants, the effect was modest when a textured or smooth implant was placed in a submuscular location.²⁸ With regard to incision location, studies have reported that the incidence of capsular contracture is highest with transaxillary and periareolar incisions, and lowest with inframammary incisions.^20,21

The leading theory is that contamination of the implant (primarily from the mammary ducts) results in biofilm formation. Subclinical hematoma surrounding the implant may also provide key bacterial nutrients.²⁰

Textured implants induce a greater inflammatory response in the capsular tissue, resulting in a thicker capsule; however, contracture rates remain lower with textured than with smooth implants.^14,31 Interestingly, lower rates of capsular contracture have been observed with later-generation, cohesive-gel, form-stable implants than with those of earlier generations.¹²

Although more research is needed, silicone implants appear to confer a higher risk of capsular contracture than saline implants.^14,20

Irrigating the breast pocket intraoperatively with triple antibiotic solution (bacitracin, cefazolin, and gentamicin) before placing the implant may decrease the capsular contracture rate.^15,20

Treatments for capsular contracture include pocket modifications such as capsulotomy (making releasing, relaxing incisions in the scar capsule encasing the implant), capsulectomy (removing portions of or the entire capsule), and replacing the implant in the other pocket (ie, if the original implant was subglandular, the replacement is placed in the submuscular pocket). Patients who have contractures that fail to respond to these treatments may ultimately benefit from implant removal and autologous reconstruction (autoaugmentation) rather than implant replacement.^32,33

ADDITIONAL COMPLICATIONS

Other complications include infection, malposition, rippling, seroma, hematoma, and sensory alterations.

Irrigation during the implantation procedure with a triple antibiotic solution consisting of bacitracin, gentamycin, and cephalexin in normal saline decreases infection and seroma rates.^15,20,34

Some surgeons also choose to irrigate the pocket with a betadine solution, or to cleanse the skin with betadine and place sterile towels and redrape before inserting the implant. Additionally, many prefer using a sterile device much like a pastry funnel called a Keller funnel to insert the implant into the breast pocket.³⁵

Infection is less common with cosmetic augmentations than with implant-based breast reconstruction, likely because of healthier, well-vascularized tissue in patients undergoing cosmetic surgery than in those undergoing mastectomy.¹⁴

Seroma is thought to be a consequence of texturing, and more so with macro- vs microtexturing. Though poorly understood, an association between texturing and double capsules has also been reported.^12,20

After primary breast augmentation, 10-year follow-up rates of capsular contracture, seroma, rippling, and malposition vary across the 3 major silicone implant manufacturers.¹² Hematoma and infection occur in less than 1% of primary augmentation patients.¹⁵

Malposition of the implant over time is less frequent with textured implants because of the higher coefficient of friction compared with smooth implants^.6,8,15

Visible skin rippling may be a consequence of texturing and also of thin body habitus, eg, in patients with a body mass index less than 18.5 kg/m². If the soft-tissue layer of the breast is thin, the natural rippling of smooth saline implant shells are more likely to show when placed in the subglandular pocket. Form-stable implants, by contrast, resist rippling.^12,15

Large implants and extensive lateral dissection can cause alterations in nipple sensation and sensory loss within lower breast pole skin. Axillary incisions may traumatize or damage the intercostobrachial nerve, resulting in upper inner arm sensory aberrations.

Ultimately, the 10-year incidence of secondary surgery ranges from 0% to 36% and the 10-year incidence of capsular contracture ranges from 11% to 19%.¹⁵ Additional cosmetic complaints after augmentation with implants include enlargement of the areola and engorgement of breast veins.¹⁴

BREAST CANCER AND DETECTION

Patients with or without implants do not seem to differ with regard to breast cancer stage upon detection, tumor burden, recurrence, or survival. However, more patients with implants may present with palpable masses, invasive tumors, axillary metastasis, and falsely negative mammograms.

Breast implants may actually facilitate cancer detection on physical examination by providing a more dense or stable surface upon which to palpate the breast tissue. Although they do not necessarily impair mastectomy or breast reconstruction, they may result in an increased rate of revision surgery after breast conservation therapy.^24,36 Mammography remains the standard of care for radiologic diagnosis but can be further supported by MRI and ultrasonography if necessary in patients with implants.

 

 

AUTOIMMUNE DISEASES

Although concerns persist, multiple studies have demonstrated the safety of fourth- and fifth-generation silicone breast implants with regard to autoimmune disease.⁷

In various clinical studies in mastectomy patients who underwent breast reconstruction with either silicone implants or autologous tissue, no difference was found with regard to the incidence of autoimmune diseases.² Additionally, in meta-analyses of data from more than 87,000 women, no association was found between connective tissue disease and silicone breast implants.^2,11 One study^11,23 noted no increase in autoantibodies in patients with undamaged silicone implants vs patients who experienced rupture.

Studies have also demonstrated that in children born to mothers with breast implants, the risk of rheumatic disease, esophageal disorders, congenital malformations, and death during the perinatal period is comparable with that in controls.³⁷ Another study, examining breastfeeding in women with silicone breast implants, showed no significant difference in silicon levels (used as a proxy for silicone) in breast milk compared with controls without implants; silicon levels were found to be significantly higher in cow’s milk and store-bought formulas.³⁸

BREAST IMPLANT-ASSOCIATED ANAPLASTIC LARGE-CELL LYMPHOMA

Breast implant-associated anaplastic large-cell lymphoma (BIA-ALCL) is a subtype of T-cell lymphoma that develops in tissue adjacent to breast implants. It typically presents as breast swelling 2 to 38 years (mean of 8 years) after implant insertion.^39,40 The swelling may be secondary to periprosthetic seroma formation or, more rarely, palpable disease in the axilla. Patients occasionally complain of pain and, rarely, constitutional symptoms.²⁰ BIA-ALCL is not a disease of the surrounding breast tissue, but rather of the fibrous periprosthetic capsule.²¹

Of note, there is no documented case involving smooth implants,^41–43 but it may be related to fifth-generation textured implants.⁶ At present, it is not possible to definitively state which implant is associated with this condition; hence, more data are needed, and this association is currently under study.

The absolute risk of BIA-ALCL was reported in a Dutch study³⁹ as 1 in 35,000 by age 50, 1 in 12,000 by age 70, and 1 in 7,000 by age 75, with a number needed to harm of 6,920. Overall lifetime risk was estimated at 1 in 30,000 for women with textured implants in a 2015 US study.⁴⁰ In comparison, breast cancer risk is about 1 in 8 women. There is no apparent predilection for patients who underwent cosmetic augmentation vs reconstruction, or who received silicone vs saline implants.

The diagnosis is confirmed by ultrasonographically guided fine-needle aspiration of seroma fluid and subsequent immunohistochemical testing for CD30-positive and ALK-negative T lymphocytes. Other than positron-emission tomography for staging after diagnosis confirmation, imaging is ineffective. Expert opinion does not recommend routine screening unless the aforementioned symptoms arise.

Treatment involves implant removal and total capsulectomy, with samples sent for pathology study with cytokeratin staining.¹² Of note, in all cases of BIA-ALCL in which the disease was limited to the circumscribed scar tissue of the breast capsule, complete surgical excision has proved curative, whereas incomplete capsulectomy portends a greater risk of recurrence and decreased survival.⁴⁴

In cases of advanced or recurrent ALCL, diagnosed late or inappropriately, the National Comprehensive Cancer Network recommends a multidisciplinary approach involving adjuvant chemotherapy and radiation.⁴⁴ Anecdotally, at our institution, we have recently treated several cases of advanced ALCL presenting with invasive chest wall masses with extirpative surgery and subsequent reconstruction with the assistance of our thoracic surgery colleagues, as well as the aforementioned multidisciplinary approach using adjuvant therapy.

The mechanism of this malignancy is currently under investigation, but the current theory implicates an exaggerated lymphoproliferative response to bacterial contamination of the capsule superimposed upon genetic factors in susceptible patients.^42,43

National societies advise plastic surgeons to discuss the risk of BIA-ALCL with all patients at the time of breast augmentation consultation and to report all confirmed cases to the PROFILE registry (Patient Registry and Outcomes for Breast Implants and Anaplastic Large Cell Lymphoma Etiology and Epidemiology).⁴⁵

ARE PATIENTS HAPPIER AFTERWARD?

Studies have shown that after undergoing breast augmentation surgery, patients note improvement in body image, and satisfaction rates range from 85% to 95% with respect to self-confidence and body image.⁴⁶ An evaluation of patient responses on the validated BREAST-Q Augmentation Questionnaire showed the following satisfaction rates: breasts 83%, psychosocial well-being 88%, and sexual functioning 81%.¹⁵

Although epidemiologic studies have reported higher suicide rates in women with cosmetic breast implants, this likely stems from preoperative psychological factors and underscores the role of psychiatric referral in patients with a mental health history or in those whom the surgeon deems it necessary.⁴⁶

Several high-quality studies have demonstrated that quality of life and psychosocial functioning (including depression) markedly improve after breast augmentation surgery.⁴⁷ Among a cohort of Norwegian patients, breast implant surgery resulted in improved motivation to perform daily activities, as well as improved quality of life from both a psychosocial and aesthetic perspective.⁴⁸ Interestingly, a recent study reported that patients who underwent breast implant surgery alone reported greater satisfaction and psychosocial quality of life than patients who underwent combination breast augmentation and mastopexy (breast-lifting) surgery.⁴⁹

Additional data are needed to refine our understanding of the complex interplay between psychosocial factors before and after surgery in patients seeking and undergoing breast augmentation procedures.

At present, 300,000 US women undergo breast augmentation surgery each year,¹ making this the second most common aesthetic procedure in women (after liposuction),^2–4 and making it extremely likely that clinicians will encounter women who have breast implants. In addition, approximately 110,000 women undergo breast reconstructive surgery after mastectomy, of whom more than 88,000 (81%) receive implants (2016 data).⁵

This review discusses the evolution of breast implants, their complications, and key considerations with regard to aesthetic and reconstructive breast surgery, as the principles are similar.

EVOLUTION OF IMPLANTS

Reports of breast augmentation surgery, also known as augmentation mammoplasty, date back to 1895, when a fatty tumor (lipoma) was successfully transplanted from a patient’s back to a breast defect in a mastectomy patient.^2,3,6,7 In the 1930s, implantation of a glass ball into a patient’s breast marked the first implant-based breast augmentation.⁶ By 1954, attempts at breast augmentation using local dermal-fat flaps, adipose tissue, and even omentum were described.

Alloplastic materials gained popularity throughout the 1950s and 1960s and included polyurethane, polytetrafluoroethylene (Teflon), and other synthetics. Adverse reactions associated with alloplastic materials were plentiful: local tissue reactions, distortion of the breast mound, increased firmness, and discomfort all contributed to the eventual discontinuation of their use. The history of alloplastic breast augmentation also included epoxy resin, shellac, beeswax, paraffin, rubber, petroleum jelly, and liquefied silicone. Outcomes were not good, and many patients ultimately needed mastectomy.⁷

The first modern breast prosthesis was developed in 1961, and since then, implant composition and design have evolved significantly.⁸

From silicone to saline, and back again

The first silicone gel implants, introduced in the early 1960s,^8–19 had high complication rates—some centers reported an incidence of capsular contracture of up to 70%.^8,11 This is a foreign body reaction in which pathologic scar tissue encases the implant, causing it to distort, appear misshapen, harden, and even become painful.¹¹ Attempts to minimize this reaction led to later generations of silicone implants with polyurethane shells.¹²

Inflatable implants filled with sterile saline solution were originally developed in France in 1965. Unlike silicone implants, saline implants have undergone minimal changes since their inception, and grew in popularity during the 1970s in view of the high rates of capsular contracture with silicone implants.⁸ However, saline implants have their own problems, and as they became increasingly popular, deflation and the unnatural feel of saline sparked a renewed interest in silicone gel.

By the late 1980s, the thinner-shelled generation of silicone implants displayed its own frustrating complications including implant rupture, capsular contracture, infection, and possible systemic and disseminated granulomatous disease. From 1992 to 2006, the US Food and Drug Administration (FDA) placed a moratorium on silicone implants due to concerns about a possible link with autoimmune and connective tissue diseases and the possible carcinogenic nature of silicone.

While silicone implants were prohibited in the United States, development continued abroad, and eventually the moratorium was lifted after several meta-analyses failed to reveal any link regarding the aforementioned concerns.¹³

Today, silicone gel implants dominate the world market.¹⁴ In the United States, approximately 60% of implants contain silicone gel filler, and trends are similar in Europe.⁷

Table 1 summarizes the evolution of silicone breast implants over the last 50 years.^2,6,11,12Table 2 lists the advantages and disadvantages of silicone and saline breast implants.^2,6,8,15

CURRENT IMPLANT OPTIONS

Currently, 3 companies (Allergan, Mentor, Sientra) manufacture and distribute breast implants and implant-associated products such as tissue expanders and sizers in the US market.⁶

Another company, Motiva, makes an implant that is available in Europe, Asia, and Australia, and the device is currently undergoing a 10-year clinical trial in the United States that began recruiting patients in 16 centers in April 2018.¹⁶ Pending final approval, the Cleveland Clinic Department of Plastic Surgery may be among the centers involved in the clinical trial of the Motiva implant. Innovations in the Motiva implant include a high-performance shell that maintains consistent strength and includes a proprietary barrier layer, improved silicone gel filler, 3-D imprinted surface texturing, and an implant shape that adapts with vertical and horizontal movement. It also contains radio-frequency identification transponders that can transmit data about the implant wirelessly.^17–19

 

 

Surface (textured vs smooth)

Developed in the 1980s, texturing of the implant surface disrupts capsule formation around the prosthesis. Additionally, texturing stabilizes an anatomically shaped (teardrop) implant within the breast pocket, reducing malrotation.^20,21

The first textured implants were covered with polyurethane foam, but they were ultimately withdrawn from the US market because of concern for in vivo degradation to carcinogenic compounds. The focus subsequently turned to texturing implant shells by mechanically creating pores of different sizes. Smooth implants, by contrast, are manufactured by repeatedly dipping the implant shell into liquid silicone.²

The capsular contraction rate has been shown to be lower with textured silicone than with smooth silicone (number needed to treat = 7–9), and evidence suggests a lower risk of needing a secondary procedure.²¹

Form-stable vs fluid-form

Silicone is a polymer. The physical properties of polymers vary greatly and depend on the length of the individual chains and the degree to which those chains are cross-linked. Liquid silicone contains short chains and sparse cross-linking, resulting in an oily compound well suited for lubrication. Silicone gel contains longer chains and more cross-linking and is therefore more viscous.

In “form-stable” implants, the silicone interior has sufficient chain length and cross-linking to retain the designed shape even at rest,² but they require slightly larger incisions.⁷ “Fluid-form” refers to an implant with silicone filler with shorter chain length, less cross-linking, and more fluidity.⁶

Shell

As with silicone fillers, the properties of silicone implant shells also depend on chain length and cross-linking within the polymer. Silicone elastomer shells (Table 1) contain extensively cross-linked chains that impart a flexible yet rubbery character. Silicone elastomers can also be found in facial implants and tissue expanders.²

Implant shape (round vs anatomic)

The shape of an implant is determined by the gel distribution inside of it. To understand gel distribution and implant shape, one must understand the gel-shell ratio. This ratio increases as cohesivity of the filler increases, and it represents increased bonding of the gel filler to the shell and a preserved implant shape at rest.

The gel-shell ratio varies among manufacturers, and a less-viscous filler may be more prone to rippling or loss of upper pole fullness in some patients. For this reason, careful analysis, patient and implant selection, and discussion of complications remain paramount.²

No anatomically shaped implant is manufactured with a smooth shell, but rather with a textured shell that resists malrotation.^6,15 However, in the United States, 95% of patients receive round implants.¹⁶

PATIENT ASSESSMENT

Before breast augmentation surgery, the surgeon assesses a number of factors—physical and psychosocial—and helps the patient choose a type and size of implant. The surgeon and patient also plan where the implants will be placed—ie, above or beneath the chest wall muscle—and where the incisions will be made. Every decision is made in close consultation with the patient, taking into account the patient’s desires and expectations, as well as what the patient’s anatomy allows. An integral component of this shared decision-making process is a discussion of the possible complications, and often photographs to better illustrate what to expect postoperatively. 

Psychosocial factors

One must consider the patient’s psychology, motivations for surgery, and emotional stability. Here, we look for underlying body dysmorphic disorder; excessive or unusual encouragement to undergo the procedure by a spouse, friends, or others; a history of other aesthetic procedures; unrealistic expectations; and other factors influencing the desire to undergo this surgery.

Choosing an implant

Implant selection must take into account the patient’s height, weight,⁷ and overall body morphology: taller patients and those with wider hips or shoulders usually require larger implants. A reliable method for determining the appropriate implant must include the current breast shape, dimensions, volume, skin elasticity, soft-tissue thickness, and overall body habitus. Ultimately, the most important considerations include breast base diameter, implant volume,²⁰ and soft-tissue envelope.

Filler type, followed by shape (round or anatomically shaped), anterior-posterior profile, and shell type (smooth or textured) are subsequent considerations (Figure 1).

Preoperative sizing can involve placing sample implants within a brassiere so that the patient can preview possible outcomes. This method is particularly effective in minimizing dissatisfaction because it shares ownership of the decision-making process.¹⁵

A computerized implant selection program available in Europe suggests a “best-fit” implant based on a clinician’s measurements.⁷

 

 

Anatomic placement

Traditionally, plastic surgeons place breast implants either beneath the pectoralis major muscle (submuscular placement) or over the pectoralis⁸ but beneath the glandular breast parenchyma (subglandular placement) (Figure 2).⁷

Advantages of submuscular placement are a smoother transition of the upper breast pole from the chest wall and less rippling visible through the skin, due to the additional muscular coverage of the implant. Another advantage is that capsular contraction rates are lower with submuscular placement, likely due to possible contamination of implants by lactiferous ductal microbes when accessing the subglandular plane.^14,20 Disadvantages are pronounced discomfort after surgery and animation deformities with muscle contraction, particularly in young, highly active patients.

A popular modification of submuscular placement involves creating a surgical dissection plane between the subglandular tissue and the pectoralis major fascia. This “dual­­plane” approach allows the parenchyma to retract superiorly and reduce breast ptosis.⁷

Incisions

The incision is most commonly made along the inframammary fold (Figure 3), but it can also be done around the areola, in the axilla, or even through the umbilicus, although this approach is less commonly used.

Table 3 highlights important considerations with regard to incision location.^15,20,21

ANTIBIOTICS

Many surgeons give a single prophylactic dose of antibiotic before surgery, a practice that some studies have shown to be effective in reducing the risk of infection.¹⁵ However, the benefit of routine postoperative use of antibiotics remains unsubstantiated¹⁵: postoperative antibiotic use does not appear to protect against infection, capsular contracture, or overall complications in primary or secondary breast augmentation surgery.²⁰

PERIOPERATIVE PERIOD

At our institution, breast augmentation surgery is an ambulatory procedure—the patient goes home the same day unless circumstances such as pain control warrant admission. This is, however, according to surgeon preference, and differs on a case-by-case basis. General anesthesia is the standard of care.¹⁵

POSTOPERATIVE PERIOD

In the immediate postoperative period, patients are instructed to wear a surgical bra for up to 6 weeks to allow stable scarring. Early mobilization is encouraged.^7,15 Depending on the patient’s situation, recovery, and healing, she may be out of work for about 1 week, sometimes more, sometimes less. 

Additional instructions are surgeon-specific. However, the patient is instructed to avoid bathing, swimming, immersion in water, and wearing underwire brassieres that could impair healing of an inferior incision; instead, patients are often instructed to wear a surgical bra provided on the day of surgery until cleared in the clinic.

Showering is allowed the next day or the second day after surgery, and of course there is no driving while on narcotics. Additionally, patients are counseled extensively regarding hematoma formation and the signs and symptoms of infection.

Patients are typically seen in clinic 1 week after surgery.

The cost of surgery may be $5,000 to $6,000 but can vary significantly from center to center depending on who the patient sees and where, and whether the patient presents for breast reconstruction after cancer or repair of congenital anomalies, or in certain cases of transgender surgery. The patient is typically responsible for the fee, but again this depends on the patient, indications, and particular insurance concerns.

IMPLANT LONGEVITY AND RUPTURE

In the United States, implant rupture rates range from 1.1% to 17.7% at 6 to 10 years after primary augmentation, 2.9% to 14.7% after revision augmentation, 1.5% to 35.4% after primary breast reconstruction, and 0% to 19.6% after revision reconstruction.¹¹

Unfortunately, the existence of multiple implant manufacturers, numerous implant generations, and poorly standardized screening protocols and reporting systems make the true rate of implant rupture difficult to assess without definitive imaging or implant retrieval.¹¹

Damage from surgical instrumentation during implantation is the most common cause of silicone breast implant rupture (50% to 64% of cases).²² Other causes include underfilling and fold flaw from capsular contracture.

Leakage of silicone gel filler may be confined to the periprosthetic capsule (intracapsular rupture) or extend beyond and into the breast parenchyma (extracapsular rupture). One study reported that only 10% of intracapsular ruptures progressed extracapsularly, while 84% of patients with extracapsular involvement remained stable for up to 2 years,²³ indicating that intracapsular rupture may not portend worsening disease.¹¹

Implant rupture occurs silently in most cases, with no clinically detectable signs or symptoms. In other cases, patients may present with alterations in breast shape and size, sudden asymmetry, firmness, pronounced capsular contracture, contour irregularity, or pain.

Aside from physical examination, comprehensive diagnostic testing includes imaging—ultrasonography, mammography, computed tomography, and magnetic resonance imaging (MRI). Of these, MRI is the method of choice, with sensitivity and specificity exceeding 90% for detecting implant rupture.¹¹ Classic findings on MRI include the “linguine” sign from a deflating implant shell, or the teardrop sign from implant sagging. Classic findings on ultrasonography include the “snowstorm” sign of extracapsular rupture and the “stepladder” sign of intracapsular rupture.

Mammography effectively detects free silicone in breast tissue with extracapsular rupture (25% of ruptures according to some studies)²³; however, it cannot detect rupture within the implant capsule. As an aside, submuscular implant placement may interfere less with screening mammography than subglandular implants do.^14,24

Current FDA recommendations to detect implant rupture encourage women with silicone breast implants to undergo screening 3 years after implantation and then every 2 years thereafter; no long-term monitoring is suggested for saline implants.¹⁵ Many plastic surgeons evaluate silicone breast implant patients every 1 to 2 years for contracture and rupture.⁸ Of note, capsular contracture impairs the effectiveness of ultrasonography and may require MRI confirmation.¹¹

If implant rupture is confirmed, the current recommendation is to remove the implant and the capsule. Another implant may be placed depending on the patient’s preference. Rigorous washout remains a key feature of any surgical intervention for ruptured breast implants; however, in the event of extracapsular rupture, resection of silicone granulomas may also be required.¹¹

Reoperation rates for primary breast augmentation surgery approach 20% and are even higher for secondary augmentation over a patient’s lifetime—the highest rate of all aesthetic procedures.^7,14

 

 

CAPSULAR CONTRACTURE

Capsular contracture is the most common complication of breast augmentation,²⁵ typically presenting within the first postoperative year,^26,27 and the risk increases over time.²⁸ It occurs with both silicone and saline breast implants.

In some studies, the incidence exceeded 4% in the first 2 years after surgery,²⁹ and nearly 50% by 10 years.³⁰ Other studies found rates of 0% to 20% over 13 years.²⁰

The etiology is not well understood and is presumed to be multifactorial, with proposed mechanisms and factors that include bacterial contamination, surface texturing, the implant pocket selected, the incision type, drain placement, antibiotic use, and smoking.²⁵

A meta-analysis from 17,000 implants found that the risk of capsular contracture was significantly higher when an implant was placed in a subglandular pocket than in a submuscular pocket,^22,26 and that although texturing decreased capsular contracture compared with smooth implants, the effect was modest when a textured or smooth implant was placed in a submuscular location.²⁸ With regard to incision location, studies have reported that the incidence of capsular contracture is highest with transaxillary and periareolar incisions, and lowest with inframammary incisions.^20,21

The leading theory is that contamination of the implant (primarily from the mammary ducts) results in biofilm formation. Subclinical hematoma surrounding the implant may also provide key bacterial nutrients.²⁰

Textured implants induce a greater inflammatory response in the capsular tissue, resulting in a thicker capsule; however, contracture rates remain lower with textured than with smooth implants.^14,31 Interestingly, lower rates of capsular contracture have been observed with later-generation, cohesive-gel, form-stable implants than with those of earlier generations.¹²

Although more research is needed, silicone implants appear to confer a higher risk of capsular contracture than saline implants.^14,20

Irrigating the breast pocket intraoperatively with triple antibiotic solution (bacitracin, cefazolin, and gentamicin) before placing the implant may decrease the capsular contracture rate.^15,20

Treatments for capsular contracture include pocket modifications such as capsulotomy (making releasing, relaxing incisions in the scar capsule encasing the implant), capsulectomy (removing portions of or the entire capsule), and replacing the implant in the other pocket (ie, if the original implant was subglandular, the replacement is placed in the submuscular pocket). Patients who have contractures that fail to respond to these treatments may ultimately benefit from implant removal and autologous reconstruction (autoaugmentation) rather than implant replacement.^32,33

ADDITIONAL COMPLICATIONS

Other complications include infection, malposition, rippling, seroma, hematoma, and sensory alterations.

Irrigation during the implantation procedure with a triple antibiotic solution consisting of bacitracin, gentamycin, and cephalexin in normal saline decreases infection and seroma rates.^15,20,34

Some surgeons also choose to irrigate the pocket with a betadine solution, or to cleanse the skin with betadine and place sterile towels and redrape before inserting the implant. Additionally, many prefer using a sterile device much like a pastry funnel called a Keller funnel to insert the implant into the breast pocket.³⁵

Infection is less common with cosmetic augmentations than with implant-based breast reconstruction, likely because of healthier, well-vascularized tissue in patients undergoing cosmetic surgery than in those undergoing mastectomy.¹⁴

Seroma is thought to be a consequence of texturing, and more so with macro- vs microtexturing. Though poorly understood, an association between texturing and double capsules has also been reported.^12,20

After primary breast augmentation, 10-year follow-up rates of capsular contracture, seroma, rippling, and malposition vary across the 3 major silicone implant manufacturers.¹² Hematoma and infection occur in less than 1% of primary augmentation patients.¹⁵

Malposition of the implant over time is less frequent with textured implants because of the higher coefficient of friction compared with smooth implants^.6,8,15

Visible skin rippling may be a consequence of texturing and also of thin body habitus, eg, in patients with a body mass index less than 18.5 kg/m². If the soft-tissue layer of the breast is thin, the natural rippling of smooth saline implant shells are more likely to show when placed in the subglandular pocket. Form-stable implants, by contrast, resist rippling.^12,15

Large implants and extensive lateral dissection can cause alterations in nipple sensation and sensory loss within lower breast pole skin. Axillary incisions may traumatize or damage the intercostobrachial nerve, resulting in upper inner arm sensory aberrations.

Ultimately, the 10-year incidence of secondary surgery ranges from 0% to 36% and the 10-year incidence of capsular contracture ranges from 11% to 19%.¹⁵ Additional cosmetic complaints after augmentation with implants include enlargement of the areola and engorgement of breast veins.¹⁴

BREAST CANCER AND DETECTION

Patients with or without implants do not seem to differ with regard to breast cancer stage upon detection, tumor burden, recurrence, or survival. However, more patients with implants may present with palpable masses, invasive tumors, axillary metastasis, and falsely negative mammograms.

Breast implants may actually facilitate cancer detection on physical examination by providing a more dense or stable surface upon which to palpate the breast tissue. Although they do not necessarily impair mastectomy or breast reconstruction, they may result in an increased rate of revision surgery after breast conservation therapy.^24,36 Mammography remains the standard of care for radiologic diagnosis but can be further supported by MRI and ultrasonography if necessary in patients with implants.

 

 

AUTOIMMUNE DISEASES

Although concerns persist, multiple studies have demonstrated the safety of fourth- and fifth-generation silicone breast implants with regard to autoimmune disease.⁷

In various clinical studies in mastectomy patients who underwent breast reconstruction with either silicone implants or autologous tissue, no difference was found with regard to the incidence of autoimmune diseases.² Additionally, in meta-analyses of data from more than 87,000 women, no association was found between connective tissue disease and silicone breast implants.^2,11 One study^11,23 noted no increase in autoantibodies in patients with undamaged silicone implants vs patients who experienced rupture.

Studies have also demonstrated that in children born to mothers with breast implants, the risk of rheumatic disease, esophageal disorders, congenital malformations, and death during the perinatal period is comparable with that in controls.³⁷ Another study, examining breastfeeding in women with silicone breast implants, showed no significant difference in silicon levels (used as a proxy for silicone) in breast milk compared with controls without implants; silicon levels were found to be significantly higher in cow’s milk and store-bought formulas.³⁸

BREAST IMPLANT-ASSOCIATED ANAPLASTIC LARGE-CELL LYMPHOMA

Breast implant-associated anaplastic large-cell lymphoma (BIA-ALCL) is a subtype of T-cell lymphoma that develops in tissue adjacent to breast implants. It typically presents as breast swelling 2 to 38 years (mean of 8 years) after implant insertion.^39,40 The swelling may be secondary to periprosthetic seroma formation or, more rarely, palpable disease in the axilla. Patients occasionally complain of pain and, rarely, constitutional symptoms.²⁰ BIA-ALCL is not a disease of the surrounding breast tissue, but rather of the fibrous periprosthetic capsule.²¹

Of note, there is no documented case involving smooth implants,^41–43 but it may be related to fifth-generation textured implants.⁶ At present, it is not possible to definitively state which implant is associated with this condition; hence, more data are needed, and this association is currently under study.

The absolute risk of BIA-ALCL was reported in a Dutch study³⁹ as 1 in 35,000 by age 50, 1 in 12,000 by age 70, and 1 in 7,000 by age 75, with a number needed to harm of 6,920. Overall lifetime risk was estimated at 1 in 30,000 for women with textured implants in a 2015 US study.⁴⁰ In comparison, breast cancer risk is about 1 in 8 women. There is no apparent predilection for patients who underwent cosmetic augmentation vs reconstruction, or who received silicone vs saline implants.

The diagnosis is confirmed by ultrasonographically guided fine-needle aspiration of seroma fluid and subsequent immunohistochemical testing for CD30-positive and ALK-negative T lymphocytes. Other than positron-emission tomography for staging after diagnosis confirmation, imaging is ineffective. Expert opinion does not recommend routine screening unless the aforementioned symptoms arise.

Treatment involves implant removal and total capsulectomy, with samples sent for pathology study with cytokeratin staining.¹² Of note, in all cases of BIA-ALCL in which the disease was limited to the circumscribed scar tissue of the breast capsule, complete surgical excision has proved curative, whereas incomplete capsulectomy portends a greater risk of recurrence and decreased survival.⁴⁴

In cases of advanced or recurrent ALCL, diagnosed late or inappropriately, the National Comprehensive Cancer Network recommends a multidisciplinary approach involving adjuvant chemotherapy and radiation.⁴⁴ Anecdotally, at our institution, we have recently treated several cases of advanced ALCL presenting with invasive chest wall masses with extirpative surgery and subsequent reconstruction with the assistance of our thoracic surgery colleagues, as well as the aforementioned multidisciplinary approach using adjuvant therapy.

The mechanism of this malignancy is currently under investigation, but the current theory implicates an exaggerated lymphoproliferative response to bacterial contamination of the capsule superimposed upon genetic factors in susceptible patients.^42,43

National societies advise plastic surgeons to discuss the risk of BIA-ALCL with all patients at the time of breast augmentation consultation and to report all confirmed cases to the PROFILE registry (Patient Registry and Outcomes for Breast Implants and Anaplastic Large Cell Lymphoma Etiology and Epidemiology).⁴⁵

ARE PATIENTS HAPPIER AFTERWARD?

Studies have shown that after undergoing breast augmentation surgery, patients note improvement in body image, and satisfaction rates range from 85% to 95% with respect to self-confidence and body image.⁴⁶ An evaluation of patient responses on the validated BREAST-Q Augmentation Questionnaire showed the following satisfaction rates: breasts 83%, psychosocial well-being 88%, and sexual functioning 81%.¹⁵

Although epidemiologic studies have reported higher suicide rates in women with cosmetic breast implants, this likely stems from preoperative psychological factors and underscores the role of psychiatric referral in patients with a mental health history or in those whom the surgeon deems it necessary.⁴⁶

Several high-quality studies have demonstrated that quality of life and psychosocial functioning (including depression) markedly improve after breast augmentation surgery.⁴⁷ Among a cohort of Norwegian patients, breast implant surgery resulted in improved motivation to perform daily activities, as well as improved quality of life from both a psychosocial and aesthetic perspective.⁴⁸ Interestingly, a recent study reported that patients who underwent breast implant surgery alone reported greater satisfaction and psychosocial quality of life than patients who underwent combination breast augmentation and mastopexy (breast-lifting) surgery.⁴⁹

Additional data are needed to refine our understanding of the complex interplay between psychosocial factors before and after surgery in patients seeking and undergoing breast augmentation procedures.

In 2017, the American College of Cardiology (ACC), American Heart Association (AHA), and Heart Failure Society of America (HFSA) jointly released a focused update¹ of the 2013 ACC/AHA guideline for managing heart failure.² This is the second focused update of the 2013 guidelines; the first update,³ in 2016, covered 2 new drugs (sacubitril-valsartan and ivabradine) for chronic stage C heart failure with reduced ejection fraction (HFrEF).

Rather than focus on new medication classes, this second update provides recommendations regarding:

• Preventing the progression to left ventricular dysfunction or heart failure in patients at high risk (stage A) through screening with B-type natriuretic peptide (BNP) and aiming for more aggressive blood pressure control
• Inpatient biomarker use
• Medications in heart failure with preserved ejection fraction (HFpEF, or diastolic heart failure)
• Blood pressure targets in stage C heart failure
• Managing important comorbidities such as iron deficiency and sleep-disordered breathing to decrease morbidity, improve functional capacity, and enhance quality of life.

These guidelines and the data that underlie them are explored below. We also discuss potential applications to the management of hospitalization for acute decompensated heart failure (ADHF).

COMMON, COSTLY, AND DEBILITATING

Heart failure—defined by the ACC/AHA as the complex clinical syndrome that results from any structural or functional impairment of ventricular filling or ejection of blood—remains one of the most common, costly, and debilitating diseases in the United States.² Based on National Health and Nutrition Examination Survey data from 2011 to 2014, an estimated 6.5 million US adults have it, with projections of more than 8 million by 2030.^4,5 More than 960,000 new cases are thought to occur annually, with a lifetime risk of developing it of roughly 20% to 45%.⁶

Despite ever-growing familiarity and some significant strides in management, the death rate in this syndrome is substantial. After admissions for heart failure (which number 1 million per year), the mortality rate is roughly 10% at 1 year and 40% at 5 years.⁶ Also staggering are the associated costs, with $30.7 billion attributed to heart failure in 2012 and a projected $69.7 billion annually by 2030.⁵ Thus, we must direct efforts not only to treatment, but also to prevention.

Preventive efforts would target patients  with ACC/AHA stage A heart failure—those at high risk for developing but currently without evidence of structural heart disease or heart failure symptoms (Table 1).⁷ This group may represent up to one-third of the US adult population, or 75 million people, when including the well-recognized risk factors of coronary artery disease, hypertension, diabetes mellitus, and chronic kidney disease in those without left ventricular dysfunction or heart failure.⁸

BIOMARKERS FOR PREVENTION

Past ACC/AHA heart failure guidelines² have included recommendations on the use of biomarkers to aid in diagnosis and prognosis and, to a lesser degree, to guide treatment of heart failure. Largely based on 2 trials (see below), the 2017 guidelines go further, issuing a recommendation on the use of natriuretic peptide biomarkers in a screening strategy to prompt early intervention and prevent the progression to clinical heart failure in high-risk patients (stage A heart failure).

The PONTIAC trial

The NT-proBNP Selected Prevention of Cardiac Events in a Population of Diabetic Patients Without a History of Cardiac Disease (PONTIAC) trial⁹ randomized 300 outpatients with type 2 diabetes mellitus and an elevated N-terminal proBNP (NT-proBNP) level (> 125 pg/mL) to standard medical care vs standard care plus intensive up-titration of renin-angiotensin system antagonists and beta-blockers in a cardiac clinic over 2 years.

Earlier studies¹⁰ had shown NT-proBNP levels to have predictive value for cardiac events in diabetic patients, while the neurohormonal treatments were thought to have an established record of preventing primary and secondary cardiovascular events. In PONTIAC, a significant reduction was seen in the primary end point of hospitalization or death due to cardiac disease (hazard ratio [HR] 0.351, P = .044), as well as in the secondary end point of hospitalization due to heart failure (P < .05), in the aggressive-intervention group. These results laid the foundation for the larger St. Vincent’s Screening to Prevent Heart Failure (STOP-HF) trial.¹¹

 

 

The STOP-HF trial

The STOP-HF trial randomized 1,235 outpatients who were at high risk but without left ventricular dysfunction or heart failure symptoms (stage A) to annual screening alone vs annual screening plus BNP testing, in which a BNP level higher than 50 pg/mL triggered echocardiography and evaluation by a cardiologist who would then assist with medications.¹¹

Eligible patients were over age 40 and had 1 or more of the following risk factors:

• Diabetes mellitus
• Hypertension
• Hypercholesterolemia
• Obesity (body mass index > 30 kg/m²)
• Vascular disease (coronary, cerebral, or peripheral arterial disease)
• Arrhythmia requiring treatment
• Moderate to severe valvular disease.

After a mean follow-up of 4.3 years, the primary end point, ie, asymptomatic left ventricular dysfunction with or without newly diagnosed heart failure, was found in 9.7% of the control group and in only 5.9% of the intervention group with BNP screening, a 42% relative risk reduction (P = .013).

Similarly, the incidence of secondary end points of emergency hospitalization for a cardiovascular event (arrhythmia, transient ischemic attack, stroke, myocardial infarction, peripheral or pulmonary thrombosis or embolization, or heart failure) was also lower at 45.2 vs 24.4 per 1,000 patient-years, a 46% relative risk reduction.

An important difference in medications between the 2 groups was an increase in subsequently prescribed renin-angiotensin-aldosterone system therapy, mainly consisting of angiotensin II receptor blockers (ARBs), in those with elevated BNP in the intervention group. Notably, blood pressure was about the same in the 2 groups.¹¹

Although these findings are encouraging, larger studies are needed, as the lack of blinding, low event rates, and small absolute risk reduction make the results difficult to generalize.

New or modified recommendations for screening


The 2017 update¹ provided a class IIa (moderate) recommendation for natriuretic peptide biomarker-based screening with subsequent guideline-based treatment directed by a cardiovascular specialist in patients at high risk of heart failure but without structural heart disease or heart failure symptoms (stage A) (Table 2).

Employing this novel prevention strategy in the extremely large number of patients with stage A heart failure, thought to be up to one-third of the US adult population, may serve as a way to best direct and utilize limited medical resources.⁸

BIOMARKERS FOR PROGNOSIS OR ADDED RISK STRATIFICATION

The 2013 guidelines² recognized that a significant body of work had accumulated showing that natriuretic peptide levels can predict outcomes in both chronic and acute heart failure. Thus, in both conditions, the guidelines contained separate class Ia recommendations to obtain a natriuretic peptide level, troponin level, or both to establish prognosis or disease severity.

The 2017 update¹ underscores the importance of timing in measuring natriuretic peptide levels during admission for ADHF, with emphasis on obtaining them at admission and at discharge for acute and postdischarge prognosis. The completely new class IIa recommendation to obtain a predischarge natriuretic peptide level for postdischarge prognosis was based on a number of observational studies, some of which we explore below.

The ELAN-HF meta-analysis

The European Collaboration on Acute Decompensated Heart Failure (ELAN-HF)¹² performed a meta-analysis to develop a discharge prognostication score for ADHF that included both absolute level and percent change in natriuretic peptide levels at the time of discharge.

Using data from 7 prospective cohorts totaling 1,301 patients, the authors found that incorporation of these values into a subsequently validated risk model led to significant improvements in the ability to predict the end points of all-cause mortality and the combined end point of all-cause mortality or first readmission for a cardiovascular reason within 180 days.

The OPTIMIZE-HF retrospective analysis

Data from the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF) were retrospectively analyzed¹³ to determine whether postdischarge outcomes were best predicted by natriuretic peptide levels at admission or discharge or by the relative change in natriuretic peptide level. More than 7,000 patients age 65 or older, in 220 hospitals, were included, and Cox prediction models were compared using clinical variables alone or in combination with the natriuretic peptide levels.

The model that included the discharge natriuretic peptide level was found to be the most predictive, with a c-index of 0.693 for predicting mortality and a c-index of 0.606 for mortality or rehospitalization at 1 year.

New or modified recommendations on biomarkers for prognosis

The 2017 update¹ modified the earlier recommendation to obtain a natriuretic peptide or troponin level or both at admission for ADHF to establish prognosis. This now has a class Ia recommendation, emphasizing that such levels be obtained on admission. In addition, a new class IIa recommendation is made to obtain a predischarge natriuretic peptide level for postdischarge prognosis. The former class Ia recommendation to obtain a natriuretic peptide level in chronic heart failure to establish prognosis or disease severity remains unchanged.

Also worth noting is what the 2017 update does not recommend in regard to obtaining biomarker levels. It emphasizes that many patients, particularly those with advanced (stage D) heart failure, have a poor prognosis that is well established with or without biomarker levels. Additionally, there are many cardiac and noncardiac causes of natriuretic peptide elevation; thus, clinical judgment remains paramount.

The 2017 update¹ also cautions against setting targets of percent change in or absolute levels of natriuretic peptide at discharge despite observational and retrospective studies demonstrating better outcomes when levels are reduced, as treating for any specific target has never been studied in a large prospective study. Thus, doing so may result in unintended harm. Rather, clinical judgment and optimization of guideline-directed management and therapy are encouraged (Table 2).

 

 

PHARMACOLOGIC TREATMENT FOR STAGE C HFpEF

Although the 2013 guidelines² contain many class I recommendations for various medications in chronic HFrEF, not a single such recommendation is found for chronic HFpEF. A review by Okwuosa et al⁷ covered HFrEF, including the most recent additions on which the 2016 update was based, sacubitril-valsartan and ivabradine. The 2016 update was similarly devoid of recommendations regarding specific medications in HFpEF, leaving only the 2013 class IIb recommendation to consider using an ARB to decrease hospitalizations in HFpEF.

Evidence behind this recommendation came from the Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity program’s randomized controlled trial in 3,025 patients with New York Heart Association (NYHA) class II to IV heart failure and left ventricular ejection fraction over 40%, who were treated with candesartan or placebo.¹⁴ Over a median follow-up of 36.6 months, there was no significant difference in the primary composite outcome of cardiovascular death or admission for heart failure, but significantly fewer patients in the candesartan arm were admitted (230 vs 270, P = .017). Thus the recommendation.

Although this finding was encouraging, it was clear that no blockbuster drug for HFpEF had been identified. Considering that roughly half of all heart failure patients have preserved ejection fraction, the discovery of such a drug for HFpEF would be met with much excitement.¹⁵ Subsequently, other medication classes have been evaluated in the hope of benefit, allowing the 2017 update to provide specific recommendations for aldosterone antagonists, nitrates, and phosphodiesterase-5 inhibitors in HFpEF.

ALDOSTERONE ANTAGONISTS FOR HFpEF

Mineralocorticoid receptor antagonists had previously been shown to significantly reduce morbidity and mortality rates in patients with HFrEF.¹⁶ In addition to aldosterone’s effects on sodium retention and many other pathophysiologic mechanisms relating to heart failure, this hormone is also known to play a role in promoting myocardial fibrosis.¹⁷ Accordingly, some have wondered whether aldosterone antagonists could improve diastolic dysfunction, and perhaps outcomes, in HFpEF.

The Aldo-DHF trial

The Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial investigated whether the aldosterone antagonist spironolactone would improve diastolic function or maximal exercise capacity in chronic HFpEF.¹⁸ It randomized 422 ambulatory patients with NYHA stage II or III heart failure, preserved left ventricular ejection fraction (≥ 50%), and echocardiographic evidence of diastolic dysfunction to receive spironolactone 25 mg daily or placebo.

Although no significant difference was seen in maximal exercise capacity, follow-up over 1 year nevertheless showed significant improvement in echocardiographic diastolic dysfunction (E/e') and perhaps reverse remodeling (decreased left ventricular mass index). These improvements spurred larger trials powered to detect whether clinical outcomes could also be improved.

The TOPCAT trial

The Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial¹⁹ was a large, multicenter, international, double-blind, placebo-controlled trial that investigated whether spironolactone could improve clinical outcomes in HFpEF. It randomized 3,445 patients with symptomatic heart failure and left ventricular ejection fraction of 45% or more to spironolactone 15 to 45 mg daily or placebo.

The effect on a composite primary outcome of death from cardiovascular cause, aborted cardiac arrest, or hospitalization for heart failure was evaluated over a mean follow-up of 3.3 years, with only a small (HR 0.89), nonclinically significant reduction evident. Those in the spironolactone group did have a significantly lower incidence of hospitalization for heart failure (12.0% vs 14.2%, P = .04).

Although the results were disappointing in this essentially negative trial, significant regional variations evident on post hoc analysis prompted further investigation and much controversy since the trial’s publication in 2014.

Participants came in roughly equal proportions from the Americas (United States, Canada, Brazil, and Argentina—51%) and from Russia and Georgia (49%), but outcomes between the two groups were markedly different. Concern was first raised when immediate review discovered a 4-fold lower rate of the primary outcome in the placebo groups from Russia and Georgia (8.4%), a rate in fact similar to that in patients without heart failure.¹⁹ This led to further exploration that identified other red flags that called into question the data integrity from the non-American sites.²⁰

Not only did patients receiving spironolactone in Russia and Georgia not experience the reduction in clinical outcomes seen in their American counterparts, they also did not manifest the expected elevations in potassium and creatinine, and spironolactone metabolites were undetectable in almost one-third of patients.²¹

These findings prompted a post hoc analysis that included only the 51% (1,767 patients) of the study population coming from the Americas; in this subgroup, treatment with spironolactone was associated with a statistically significant 18% relative risk reduction in the primary composite outcome, a 26% reduction in cardiovascular mortality, and an 18% reduction in hospitalization for heart failure.²⁰

New or modified recommendations on aldosterone receptor antagonists

Recognizing both the encouraging data above and the limitations of post hoc analyses, the 2017 focused update provides a class IIb (weak) recommendation stating that aldosterone receptor antagonists might be considered to decrease hospitalizations in appropriately selected patients with HFpEF (Table 3).¹

Nitrates and phosphodiesterase-5 inhibitors

Earlier studies indicated that long-acting nitrates are prescribed in 15% to 50% of patients with HFpEF, perhaps based on extrapolation from studies in HFrEF suggesting that they might improve exercise intolerance.²² Some have speculated that the hemodynamic effects of nitrates, such as decreasing pulmonary congestion, might improve exercise intolerance in those with the stiff ventricles of HFpEF as well, prompting further study.

 

 

The NEAT-HFpEF trial

The Nitrate’s Effect on Activity Tolerance in Heart Failure With Preserved Ejection Fraction (NEAT-HFpEF) trial²² investigated whether extended-release isosorbide mononitrate would increase daily activity levels in patients with HFpEF. This double-blind, crossover study randomized 110 patients with HFpEF (ejection fraction ≥ 50%) and persistent dyspnea to escalating doses of isosorbide mononitrate or placebo over 6 weeks, then to the other arm for another 6 weeks. Daily activity levels during the 120-mg phase were measured with a continuously worn accelerometer.

No beneficial effect of nitrates was evident, with a nonsignificant trend towards decreased activity levels, a significant decrease in hours of activity per day (–0.30 hours, P = .02), and no change in the other secondary end points such as quality-of-life score, 6-minute walk distance, or natriuretic peptide level.

Suggested explanations for these negative findings include the possibility of rapid dose escalation leading to increased subtle side effects (headache, dizziness, fatigue) that, in turn, decreased activity. Additionally, given the imprecise diagnostic criteria for HFpEF, difficulties with patient selection may have led to inclusion of a large number of patients without elevated left-sided filling pressures.²³

The RELAX trial

The Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure With Preserved Ejection Fraction (RELAX) trial²⁴ investigated whether the phosphodiesterase-5 inhibitor sildenafil would improve exercise capacity in HFpEF. Improvements in both exercise capacity and clinical outcomes had already been seen in earlier trials in patients with pulmonary hypertension, as well as in those with HFrEF.²⁵ A smaller study in HFpEF patients with pulmonary hypertension was also encouraging.²⁶

Thus, it was disappointing that, after randomizing 216 outpatients with HFpEF to sildenafil or placebo for 24 weeks, no benefit was seen in the primary end point of change in peak oxygen consumption or in secondary end points of change in 6-minute walk distance or composite clinical score. Unlike in NEAT-HFpEF, patients here were required to have elevated natriuretic peptide levels or elevated invasively measured filling pressures.

The study authors speculated that pulmonary arterial hypertension and right ventricular systolic failure might need to be significant for patients with HFpEF to benefit from phosphodiesterase-5 inhibitors, with their known effects of dilation of pulmonary vasculature and increasing contractility of the right ventricle.²⁴

New or modified recommendations on nitrates or phosphodiesterase-5 drugs

Given these disappointing results, the 2017 update provides a class III (no benefit) recommendation against the routine use of nitrates or phosphodiesterase-5 inhibitors to improve exercise tolerance or quality of life in HFpEF, citing them as ineffective (Table 3).¹

IRON DEFICIENCY IN HEART FAILURE

Not only is iron deficiency present in roughly 50% of patients with symptomatic heart failure (stage C and D HFrEF),²⁷ it is also associated with increased heart failure symptoms such as fatigue and exercise intolerance,²⁸ reduced functional capacity, decreased quality of life, and increased mortality.

Notably, this association exists regardless of the hemoglobin level.²⁹ In fact, even in those without heart failure or anemia, iron deficiency alone results in worsened aerobic performance, exercise intolerance, and increased fatigue.³⁰ Conversely, improvement in symptoms, exercise tolerance, and cognition have been shown with repletion of iron stores in such patients.³¹

At the time of the 2013 guidelines, only a single large trial of intravenous iron in HFrEF and iron deficiency had been carried out (see below), and although the results were promising, it was felt that the evidence base on which to make recommendations was inadequate. Thus, recommendations were deferred until more data could be obtained.

Of note, in all the trials discussed below, iron deficiency was diagnosed in the setting of heart failure as ferritin less than 100 mg/mL (absolute iron deficiency) or as ferritin 100 to 300 mg/mL with transferrin saturation less than 20% (relative deficiency).³²

The CONFIRM-HF trial

As in the Ferinject Assessment in Patients With Iron Deficiency and Chronic Heart Failure (FAIR-HF) trial,³³ the subsequent Ferric Carboxymaltose Evaluation on Performance in Patients With Iron Deficiency in Combination With Chronic Heart Failure (CONFIRM-HF) trial³⁴ involved the intravenous infusion of iron (ferric carboxymaltose) in outpatients with symptomatic HFrEF and iron deficiency. It showed that benefits remained evident with a more objective primary end point (change in 6-minute walk test distance at 24 weeks), and that such benefits were sustained, as seen in numerous secondary end points related to functional capacity at 52 weeks. Benefits in CONFIRM-HF were evident independently from anemia, specifically whether hemoglobin was under or over 12 g/dL.

Although these results were promising, it remained unclear whether such improvements could be obtained with a much easier to administer, more readily available, and less expensive oral iron formulation.

The IRONOUT-HF trial

The Iron Repletion Effects on Oxygen Uptake in Heart Failure (IRONOUT-HF) trial³⁵ investigated whether oral, rather than intravenous, iron supplementation could improve peak exercise capacity in patients with HFrEF and iron deficiency. This double-blind, placebo-controlled trial randomized 225 patients with NYHA class II to IV HFrEF and iron deficiency to treatment with oral iron polysaccharide (150 mg twice daily) or placebo for 16 weeks.

Contrary to the supportive findings above, no significant change was seen in the primary end point of change in peak oxygen uptake or in any of the secondary end points (change in 6-minute walk, quality of life). Also, despite a 15-fold increase in the amount of iron administered in oral form compared with intravenously, little change was evident in the indices of iron stores over the course of the study, with only a 3% increase in transferrin saturation and an 11 ng/mL increase in ferritin. The intravenous trials resulted in a 4-fold greater increase in transferrin saturation and a 20-fold greater increase in ferritin.³⁶

What keeps heart failure patients from absorbing oral iron? It is unclear why oral iron administration in HFrEF, such as in IRONOUT-HF, seems to be so ineffective, but hepcidin—a protein hormone made by the liver that shuts down intestinal iron absorption and iron release from macrophages—may play a central role.³⁷ When iron stores are adequate, hepcidin is upregulated to prevent iron overload. However, hepcidin is also increased in inflammatory states, and chronic heart failure is often associated with inflammation.

With this in mind, the IRONOUT-HF investigators measured baseline hepcidin levels at the beginning and at the end of the 16 weeks and found that high baseline hepcidin levels predicted poorer response to oral iron. Other inflammatory mediators, such as interleukin 6, may also play a role.^38,39 Unlike oral iron formulations such as iron polysaccharide, intravenous iron (ferric carboxymaltose) bypasses these regulatory mechanisms, which may partly explain its much more significant effect on the indices of iron stores and outcomes.

 

 

New or modified recommendations on iron

The 2017 update¹ makes recommendations regarding iron deficiency and anemia in heart failure for the first time.

A class IIb recommendation states that it might be reasonable to treat NYHA class II and III heart failure patients with iron deficiency with intravenous iron to improve functional status and quality of life. A strong recommendation has been deferred until more is known about morbidity and mortality effects from adequately powered trials, some of which are under way and explored further below.

The 2017 update also withholds any recommendations regarding oral iron supplementation in heart failure, citing an uncertain evidence base. Certainly, the subsequent IRONOUT-HF trial does not lend enthusiasm for this approach.

Lastly, given the lack of benefit coupled with the increased risk of thromboembolic events evident in a trial of darbepoetin alfa vs placebo in non-iron deficiency-related anemia in HFrEF,^40,41 the 2017 update provides a class III (no benefit) recommendation against using erythropoietin-stimulating agents in heart failure and anemia.

HYPERTENSION IN HEART FAILURE

The 2013 guidelines for the management of heart failure simply provided a class I recommendation to control hypertension and lipid disorders in accordance with contemporary guidelines to lower the risk of heart failure.¹

SPRINT

The Systolic Blood Pressure Intervention Trial (SPRINT)⁴² sought to determine whether a lower systolic blood pressure target (120 vs 140 mm Hg) would reduce clinical events in patients at high risk for cardiovascular events but without diabetes mellitus. Patients at high risk were defined as over age 75, or with known vascular disease, chronic kidney disease, or a Framingham Risk Score higher than 15%. This multicenter, open-label controlled trial randomized 9,361 patients to intensive treatment (goal systolic blood pressure < 120 mm Hg) or standard treatment (goal systolic blood pressure < 140 mm Hg).

SPRINT was stopped early at a median follow-up of 3.26 years when a 25% relative risk reduction in the primary composite outcome of myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes became evident in the intensive-treatment group (1.65% vs 2.19% per year, HR 0.75, P < .0001).

All-cause mortality was also lower in the intensive-treatment group (HR 0.73, P = .003), while the incidence of serious adverse events (hypotension, syncope, electrolyte abnormalities, acute kidney injury, and noninjurious falls) was only slightly higher (38.3% vs 37.1%, P = .25). Most pertinent, a significant 38% relative risk reduction in heart failure and a 43% relative risk reduction in cardiovascular events were also evident.

Of note, blood pressure measurements were taken as the average of 3 measurements obtained by an automated cuff taken after the patient had been sitting quietly alone in a room for 5 minutes.

New or modified recommendations on hypertension in heart failure

Given the impressive 25% relative risk reduction in myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes in SPRINT,⁴² the 2017 update¹ incorporated the intensive targets of SPRINT into its recommendations. However, to compensate for what are expected to be higher blood pressures obtained in real-world clinical practice as opposed to the near-perfect conditions used in SPRINT, a slightly higher blood pressure goal of less than 130/80 mm Hg was set.

Specific blood pressure guidelines have not been given for stage A heart failure in the past. However, as for other new approaches to prevent heart failure in this update and given the 38% relative risk reduction in heart failure seen in SPRINT, a class I recommendation is given to target a blood pressure goal of less than 130/80 mm Hg in stage A heart failure with hypertension (Table 4).

Although not specifically included in SPRINT, given the lack of trial data on specific blood pressure targets in HFrEF and the decreased cardiovascular events noted above, a class I (level of evidence C, expert opinion) recommendation to target a goal systolic blood pressure less than 130 mm Hg in stage C HFrEF with hypertension is also given. Standard guideline-directed medications in the treatment of HFrEF are to be used (Table 4).

Similarly, a new class I (level of evidence C, expert opinion) recommendation is given for hypertension in HFpEF to target a systolic blood pressure of less than 130 mm Hg, with special mention to first manage any element of volume overload with diuretics. Other than avoiding nitrates (unless used for angina) and phosphodiesterase inhibitors, it is noted that few data exist to guide the choice of antihypertensive further, although perhaps renin-angiotensin-aldosterone system inhibition, especially aldosterone antagonists, may be considered. These recommendations are fully in line with the 2017 ACC/AHA high blood pressure clinical practice guidelines,⁴³ ie, that renin-angiotensin-aldosterone system inhibition with an angiotensin-converting enzyme (ACE) inhibitor or ARB and especially mineralocorticoid receptor antagonists would be the preferred choice (Table 4).

SLEEP-DISORDERED BREATHING IN HEART FAILURE

Sleep-disordered breathing, either obstructive sleep apnea (OSA) or central sleep apnea, is quite commonly associated with symptomatic HFrEF.⁴⁴ Whereas OSA is found in roughly 18% and central sleep apnea in 1% of the general population, sleep-disordered breathing is found in nearly 60% of patients with HFrEF, with some studies showing a nearly equal proportion of OSA and central sleep apnea.⁴⁵ A similar prevalence is seen in HFpEF, although with a much higher proportion of OSA.⁴⁶ Central sleep apnea tends to be a marker of more severe heart failure, as it is strongly associated with severe cardiac systolic dysfunction and worse functional capacity.⁴⁷

Not surprisingly, the underlying mechanism of central sleep apnea is quite different from that of OSA. Whereas OSA predominantly occurs because of repeated obstruction of the pharynx due to nocturnal pharyngeal muscle relaxation, no such airway patency issues or strained breathing patterns exist in central sleep apnea. Central sleep apnea, which can manifest as Cheyne-Stokes respirations, is thought to occur due to an abnormal ventilatory control system with complex pathophysiology such as altered sensitivity of central chemoreceptors to carbon dioxide, interplay of pulmonary congestion, subsequent hyperventilation, and prolonged circulation times due to reduced cardiac output.⁴⁸

What the two types of sleep-disordered breathing have in common is an association with negative health outcomes. Both appear to induce inflammation and sympathetic nervous system activity via oxidative stress from intermittent nocturnal hypoxemia and hypercapnea.⁴⁹ OSA was already known to be associated with significant morbidity and mortality rates in the general population,⁵⁰ and central sleep apnea had been identified as an independent predictor of mortality in HFrEF.⁵¹

At the time of the 2013 guidelines, only small or observational studies with limited results had been done evaluating treatment effects of continuous positive airway pressure therapy (CPAP) on OSA and central sleep apnea. Given the relative paucity of data, only a single class IIa recommendation stating that CPAP could be beneficial to increase left ventricular ejection fraction and functional status in concomitant sleep apnea and heart failure was given in 2013. However, many larger trials were under way,^52–59 some with surprising results such as a significant increase in cardiovascular and all-cause mortality (Table 5).⁵⁴

 

 

New or modified recommendations on sleep-disordered breathing

Stemming from several trials,^54,56 3 new recommendations on sleep-disordered breathing were made in the 2017 update (Table 6).

Given the common association with heart failure (60%)⁴⁵ and the marked variation in response to treatment, including potential for harm with adaptive servo-ventilation and central sleep apnea, a class IIa recommendation is made stating that it is reasonable to obtain a formal sleep study in any patient with symptomatic (NYHA class II–IV) heart failure.¹

Due to the potential for harm with adaptive servo-ventilation in patients with central sleep apnea and NYHA class II to IV HFrEF, a class III (harm) recommendation is made against its use.

Largely based on the results of the Sleep Apnea Cardiovascular Endpoints (SAVE) trial,⁵⁶ a class IIb, level of evidence B-R (moderate, based on randomized trials) recommendation is given, stating that the use of CPAP in those with OSA and known cardiovascular disease may be reasonable to improve sleep quality and reduce daytime sleepiness.

POTENTIAL APPLICATIONS IN ACUTE DECOMPENSATED HEART FAILURE

Although the 2017 update¹ is directed mostly toward managing chronic heart failure, it is worth considering how it might apply to the management of ADHF.

SHOULD WE USE BIOMARFER TARGETS TO GUIDE THERAPY IN ADHF?

The 2017 update¹ does offer direct recommendations regarding the use of biomarker levels during admissions for ADHF. Mainly, they emphasize that the admission biomarker levels provide valuable information regarding acute prognosis and risk stratification (class I recommendation), while natriuretic peptide levels just before discharge provide the same for the postdischarge timeframe (class IIa recommendation).

The update also explicitly cautions against using a natriuretic peptide level-guided treatment strategy, such as setting targets for predischarge absolute level or percent change in level of natriuretic peptides during admissions for ADHF. Although observational and retrospective studies have shown better outcomes when levels are reduced at discharge, treating for any specific inpatient target has never been tested in any large, prospective study; thus, doing so could result in unintended harm.

So what do we know?

McQuade et al systematic review

McQuade et al⁵⁷ performed a systematic review of more than 40 ADHF trials, which showed that, indeed, patients who achieved a target absolute natriuretic peptide level (BNP ≤ 250 pg/mL) or percent reduction (≥ 30%) at time of discharge had significantly improved outcomes such as reduced postdischarge all-cause mortality and rehospitalization rates. However, these were mostly prospective cohort studies that did not use any type of natriuretic peptide level-guided treatment protocol, leaving it unclear whether such a strategy could positively influence outcomes.

For this reason, both McQuade et al⁵⁷ and, in an accompanying editorial, Felker et al⁵⁸ called for properly designed, randomized controlled trials to investigate such a strategy. Felker noted that only 2 such phase II trials in ADHF have been completed,^59,60 with unconvincing results.

PRIMA II

The Multicenter, Randomized Clinical Trial to Study the Impact of In-hospital Guidance for Acute Decompensated Heart Failure Treatment by a Predefined NT-ProBNP Target on the Reduction of Readmission and Mortality Rates (PRIMA II)⁶⁰ randomized patients to natriuretic peptide level-guided treatment or standard care during admission for ADHF.

Many participants (60%) reached the predetermined target of 30% reduction in natriuretic peptide levels at the time of clinical stabilization and randomization; 405 patients were randomized. Patients in the natriuretic peptide level-guided treatment group underwent a prespecified treatment algorithm, with repeat natriuretic peptide levels measured again after the protocol.

Natriuretic peptide-guided therapy failed to show any significant benefit in any clinical outcomes, including the primary composite end point of mortality or heart failure readmissions at 180 days (36% vs 38%, HR 0.99, 95% confidence interval 0.72–1.36). Consistent with the review by McQuade et al,⁵⁷ achieving the 30% reduction in natriuretic peptide at discharge, in either arm, was associated with a better prognosis, with significantly lower mortality and readmission rates at 180 days (HR 0.39 for rehospitalization or death, 95% confidence interval 0.27–0.55).

As in the observational studies, those who achieved the target natriuretic peptide level at the time of discharge had a better prognosis than those who did not, but neither study showed an improvement in clinical outcomes using a natriuretic peptide level-targeting treatment strategy.

No larger randomized controlled trial results are available for guided therapy in ADHF. However, additional insight may be gained from a subsequent trial⁶¹ that evaluated biomarker-guided titration of guideline-directed medical therapy in outpatients with chronic HFrEF.

The GUIDE-IT trial

That trial, the Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure (GUIDE-IT)⁶¹ trial, was a large multicenter attempt to determine whether a natriuretic peptide-guided treatment strategy was more effective than standard care in the management of 894 high-risk outpatients with chronic HFrEF. Earlier, promising results had been obtained in a meta-analysis⁶² of more than 11 similar trials in 2,000 outpatients, with a decreased mortality rate (HR 0.62) seen in the biomarker-guided arm. However, the results had not been definitive due to being underpowered.⁶²

Unfortunately, the results of GUIDE-IT were disappointing, with no significant difference in either the combined primary end point of mortality or hospitalization for heart failure, or the secondary end points evident at 15 months, prompting early termination for futility.⁶¹ Among other factors, the study authors postulated that this may have partly resulted from a patient population with more severe heart failure and resultant azotemia, limiting the ability to titrate neurohormonal medications to the desired dosage.

The question of whether patients who cannot achieve such biomarker targets need more intensive therapy or whether their heart failure is too severe to respond adequately echoes the question often raised in discussions of inpatient biomarker-guided therapy.⁵⁸ Thus, only limited insight is gained, and it remains unclear whether a natriuretic peptide-guided treatment strategy can improve outpatient or inpatient outcomes. Until this is clarified, clinical judgment and optimization of guideline-directed management and therapy should remain the bedrock of treatment.

 

 

SHOULD ALDOSTERONE ANTAGONISTS BE USED IN ACUTE HFpEF?

Given the encouraging results in chronic HFpEF from post hoc analyses of TOPCAT, are there any additional recent data suggesting a role for aldosterone antagonists such as spironolactone in acute HFpEF?

The ATHENA-HF trial

The Aldosterone Targeted Neurohormonal Combined With Natriuresis Therapy in Heart Failure (ATHENA-HF) trial⁶³ compared treatment with high-dose spironolactone (100 mg) for 96 hours vs usual care in 360 patients with ADHF. The patient population included those with HFrEF and HFpEF, and usual care included low-dose spironolactone (12.5–25 mg) in roughly 15% of patients. High-dose mineralocorticoid receptor antagonists have been shown to overcome diuretic resistance, improve pulmonary vascular congestion, and partially combat the adverse neurohormonal activation seen in ADHF.

Unfortunately, the trial was completely neutral in regard to the primary end point of reduction in natriuretic peptide levels as well as to the secondary end points of 30-day mortality rate, heart failure readmission, clinical congestion scores, urine output, and change in weight. No suggestion of additional benefit was seen in subgroup analysis of patients with acute HFpEF (ejection fraction > 45%), which yielded similar results.⁶³

Given these lackluster findings, routine use of high-dose spironolactone in ADHF is not recommended.⁶⁴ However, the treatment was well tolerated, without significant adverse effects of hyperkalemia or kidney injury, leaving the door open as to whether it may have utility in selected patients with diuretic resistance.

Should ARNIs and ivabradine be started during ADHF admissions?

The first half of the focused update³ of the 2013 guidelines,² reviewed by Okwuosa et al,⁷ provided recommendations for the use of sacubitril-valsartan, an angiotensin-neprilysin inhibitor (ARNI), and ivabradine, a selective sinoatrial node If channel inhibitor, in chronic HFrEF.

Sacubitril-valsartan was given a class I recommendation for use in patients with NYHA class II or III chronic HFrEF who tolerate an ACE inhibitor or an ARB. This recommendation was given largely based on the benefits in mortality and heart failure hospitalizations seen in PARADIGM-HF (the Prospective Comparison of ARNI With ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure)⁶⁵ compared with enalapril (HR 0.80, 95% CI 0.73–0.87, P < .001).

There is currently no recommendation on initiation or use of ARNIs during admissions for ADHF, but a recent trial may lend some insight.⁶⁶

THE PIONEER-HF trial

The Comparison of Sacubitril/Valsartan vs Enalapril on Effect on NT-proBNP in Patients Stabilized From an Acute Heart Failure Episode (PIONEER-HF) trial⁶⁶ randomized patients admitted for acute HFrEF, once stabilized, to sacubitril-valsartan or enalapril. Encouragingly, the percentage change of natriuretic peptide levels from the time of inpatient initiation to 4 and 8 weeks thereafter, the primary efficacy end point, was 46.7% with sacubitril-valsartan versus 25.3% with enalapril alone (ratio of change 0.71, 95% CI 0.63–0.81, P < .001). Although not powered for such, a prespecified analysis of a composite of clinical outcomes was also favorable for sacubitril-valsartan, largely driven by a 44% decreased rate of rehospitalization. More definitive, and quite reassuring, was that no significant difference was seen in the key safety outcomes of worsening renal function, hyperkalemia, symptomatic hypotension, and angioedema. These results were also applicable to the one-third of study participants who had no former diagnosis of heart failure, the one-third identifying as African American, and the one-third who had not been taking an ACE inhibitor or ARB. These results, taken together with the notion that at study completion the patients become similar to those included in PARADIGM-HF, have led some to assert that PIONEER-HF has the potential to change clinical practice.

Ivabradine was given a class IIa recommendation for use in patients with NYHA class II or III chronic HFrEF with a resting heart rate of at least 70 bpm, in sinus rhythm, despite being on optimal medical therapy including a beta-blocker at a maximum tolerated dose.

This recommendation was largely based on SHIFT (Systolic Heart Failure Treatment With the If Inhibitor Ivabradine Trial), which randomized patients to ivabradine or placebo to evaluate the effects of isolated lowering of the heart rate on the composite primary outcome of cardiovascular death or hospitalization. A significant reduction was seen in the ivabradine arm (HR 0.82, 95% CI 0.75–0.90, P < .0001), mainly driven by decreased hospitalizations.⁶⁷

Subsequently, a small unblinded single-center study was undertaken to evaluate the efficacy and safety of initiating ivabradine during admissions for ADHF.⁶⁸

THE ETHIC-AHF trial

The Effect of Early Treatment With Ivabradine Combined With Beta-Blockers vs Beta-Blockers Alone in Patients Hospitalized With Heart Failure and Reduced Left Ventricular Ejection Fraction (ETHIC-AHF) trial⁶⁸ sought to determine the safety and effectiveness of early coadministration of ivabradine with beta-blockers in patients with acute HFrEF.

This single-center, unblinded study randomized 71 patients to ivabradine and beta-blockade or beta-blockade alone upon clinical stabilization (24–48 hours) after admission for acute decompensated HFrEF.

The primary end point was heart rate at 28 days, with the ivabradine group showing a statistically significant decrease (64 vs 70 bpm, P = .01), which persisted at 4 months. There was no significant difference in the secondary end points of adverse drug effects or the composite of clinical event outcomes (all-cause mortality, admission for heart failure or cardiovascular cause), but a number of surrogate end points including left ventricular ejection fraction, BNP level, and NYHA functional class at 4 months showed mild improvement.

Although this study provided evidence that the coadministration of ivabradine and a beta-blocker is safe and was positive in regard to clinical outcomes, the significant limitations due to its size and study design (single-center, unblinded, 4-month follow-up) simply serve to support the pursuit of larger studies with more stringent design and longer follow-up in order to determine the clinical efficacy.

 

 

The PRIME-HF trial

The Predischarge Initiation of Ivabradine in the Management of Heart Failure (PRIME-HF) trial⁶⁹ is a randomized, open-label, multicenter trial comparing standard care vs the initiation of ivabradine before discharge, but after clinical stabilization, during admissions for ADHF in patients with chronic HFrEF (left ventricular ejection fraction ≤ 35%). At subsequent outpatient visits, the dosage can be modified in the ivabradine group, or ivabradine can be initiated at the provider’s discretion in the usual-care group.

PRIME-HF is attempting to determine whether initiating ivabradine before discharge will result in more patients taking ivabradine at 180 days, its primary end point, as well as in changes in secondary end points including heart rate and patient-centered outcomes. The study is active, with reporting expected in 2019.

As these trials all come to completion, it will not be long before we have further guidance regarding the inpatient initiation of these new and exciting therapeutic agents.

SHOULD INTRAVENOUS IRON BE GIVEN DURING ADHF ADMISSIONS?

Given the high prevalence of iron deficiency in symptomatic HFrEF, its independent association with mortality, improvements in quality of life and functional capacity suggested by repleting with intravenous iron (in FAIR-HF and CONFIRM-HF), the seeming inefficacy of oral iron in IRONOUT, and the logistical challenges of intravenous administration during standard clinic visits, could giving intravenous iron soon be incorporated into admissions for ADHF?

Caution has been advised for several reasons. As discussed above, larger randomized controlled trials powered to detect more definitive clinical end points such as death and the rate of hospitalization are still needed before a stronger recommendation can be made for intravenous iron in HFrEF. Also, without such data, it seems unwise to add the considerable economic burden of routinely assessing for iron deficiency and providing intravenous iron during ADHF admissions to the already staggering costs of heart failure.

Thus far, only a single meta-analysis is available, including 893 patients⁷⁰ largely from the FAIR-HF and CONFIRM-HF trials. While it does suggest benefit in both cardiovascular mortality and recurrent hospitalizations for heart failure (rate ratio 0.59, 95% CI 0.40–0.88; P = .009), more definitive guidance will be provided by the results from 4 large randomized placebo-controlled studies  currently under way or recruiting. All 4 seek to examine the effects of intravenous iron on morbidity and mortality in patients with HFrEF and iron deficiency, using a variety of end points ranging from exercise tolerance, to hospitalizations, to mortality (Table 7).^71–74

The effects seen on morbidity and mortality that become evident in these trials over the next 5 years will help determine future guidelines and whether intravenous iron is routinely administered in bridge clinics, during inpatient admissions for ADHF, or not at all in patients with HFrEF and iron deficiency.

INTERNISTS ARE KEY

Heart failure remains one of the most common, morbid, complex, and costly diseases in the United States, and its prevalence is expected only to increase.^4,5 The 2017 update1 of the 2013 guideline² for the management of heart failure provides recommendations aimed not only at management of heart failure, but also at its comorbidities and, for the first time ever, at its prevention.

Internists provide care for the majority of heart failure patients, as well as for their comorbidities, and are most often the first to come into contact with patients at high risk of developing heart failure. Thus, a thorough understanding of these guidelines and how to apply them to the management of acute decompensated heart failure is of critical importance.

In 2017, the American College of Cardiology (ACC), American Heart Association (AHA), and Heart Failure Society of America (HFSA) jointly released a focused update¹ of the 2013 ACC/AHA guideline for managing heart failure.² This is the second focused update of the 2013 guidelines; the first update,³ in 2016, covered 2 new drugs (sacubitril-valsartan and ivabradine) for chronic stage C heart failure with reduced ejection fraction (HFrEF).

Rather than focus on new medication classes, this second update provides recommendations regarding:

• Preventing the progression to left ventricular dysfunction or heart failure in patients at high risk (stage A) through screening with B-type natriuretic peptide (BNP) and aiming for more aggressive blood pressure control
• Inpatient biomarker use
• Medications in heart failure with preserved ejection fraction (HFpEF, or diastolic heart failure)
• Blood pressure targets in stage C heart failure
• Managing important comorbidities such as iron deficiency and sleep-disordered breathing to decrease morbidity, improve functional capacity, and enhance quality of life.

These guidelines and the data that underlie them are explored below. We also discuss potential applications to the management of hospitalization for acute decompensated heart failure (ADHF).

COMMON, COSTLY, AND DEBILITATING

Heart failure—defined by the ACC/AHA as the complex clinical syndrome that results from any structural or functional impairment of ventricular filling or ejection of blood—remains one of the most common, costly, and debilitating diseases in the United States.² Based on National Health and Nutrition Examination Survey data from 2011 to 2014, an estimated 6.5 million US adults have it, with projections of more than 8 million by 2030.^4,5 More than 960,000 new cases are thought to occur annually, with a lifetime risk of developing it of roughly 20% to 45%.⁶

Despite ever-growing familiarity and some significant strides in management, the death rate in this syndrome is substantial. After admissions for heart failure (which number 1 million per year), the mortality rate is roughly 10% at 1 year and 40% at 5 years.⁶ Also staggering are the associated costs, with $30.7 billion attributed to heart failure in 2012 and a projected $69.7 billion annually by 2030.⁵ Thus, we must direct efforts not only to treatment, but also to prevention.

Preventive efforts would target patients  with ACC/AHA stage A heart failure—those at high risk for developing but currently without evidence of structural heart disease or heart failure symptoms (Table 1).⁷ This group may represent up to one-third of the US adult population, or 75 million people, when including the well-recognized risk factors of coronary artery disease, hypertension, diabetes mellitus, and chronic kidney disease in those without left ventricular dysfunction or heart failure.⁸

BIOMARKERS FOR PREVENTION

Past ACC/AHA heart failure guidelines² have included recommendations on the use of biomarkers to aid in diagnosis and prognosis and, to a lesser degree, to guide treatment of heart failure. Largely based on 2 trials (see below), the 2017 guidelines go further, issuing a recommendation on the use of natriuretic peptide biomarkers in a screening strategy to prompt early intervention and prevent the progression to clinical heart failure in high-risk patients (stage A heart failure).

The PONTIAC trial

The NT-proBNP Selected Prevention of Cardiac Events in a Population of Diabetic Patients Without a History of Cardiac Disease (PONTIAC) trial⁹ randomized 300 outpatients with type 2 diabetes mellitus and an elevated N-terminal proBNP (NT-proBNP) level (> 125 pg/mL) to standard medical care vs standard care plus intensive up-titration of renin-angiotensin system antagonists and beta-blockers in a cardiac clinic over 2 years.

Earlier studies¹⁰ had shown NT-proBNP levels to have predictive value for cardiac events in diabetic patients, while the neurohormonal treatments were thought to have an established record of preventing primary and secondary cardiovascular events. In PONTIAC, a significant reduction was seen in the primary end point of hospitalization or death due to cardiac disease (hazard ratio [HR] 0.351, P = .044), as well as in the secondary end point of hospitalization due to heart failure (P < .05), in the aggressive-intervention group. These results laid the foundation for the larger St. Vincent’s Screening to Prevent Heart Failure (STOP-HF) trial.¹¹

 

 

The STOP-HF trial

The STOP-HF trial randomized 1,235 outpatients who were at high risk but without left ventricular dysfunction or heart failure symptoms (stage A) to annual screening alone vs annual screening plus BNP testing, in which a BNP level higher than 50 pg/mL triggered echocardiography and evaluation by a cardiologist who would then assist with medications.¹¹

Eligible patients were over age 40 and had 1 or more of the following risk factors:

• Diabetes mellitus
• Hypertension
• Hypercholesterolemia
• Obesity (body mass index > 30 kg/m²)
• Vascular disease (coronary, cerebral, or peripheral arterial disease)
• Arrhythmia requiring treatment
• Moderate to severe valvular disease.

After a mean follow-up of 4.3 years, the primary end point, ie, asymptomatic left ventricular dysfunction with or without newly diagnosed heart failure, was found in 9.7% of the control group and in only 5.9% of the intervention group with BNP screening, a 42% relative risk reduction (P = .013).

Similarly, the incidence of secondary end points of emergency hospitalization for a cardiovascular event (arrhythmia, transient ischemic attack, stroke, myocardial infarction, peripheral or pulmonary thrombosis or embolization, or heart failure) was also lower at 45.2 vs 24.4 per 1,000 patient-years, a 46% relative risk reduction.

An important difference in medications between the 2 groups was an increase in subsequently prescribed renin-angiotensin-aldosterone system therapy, mainly consisting of angiotensin II receptor blockers (ARBs), in those with elevated BNP in the intervention group. Notably, blood pressure was about the same in the 2 groups.¹¹

Although these findings are encouraging, larger studies are needed, as the lack of blinding, low event rates, and small absolute risk reduction make the results difficult to generalize.

New or modified recommendations for screening


The 2017 update¹ provided a class IIa (moderate) recommendation for natriuretic peptide biomarker-based screening with subsequent guideline-based treatment directed by a cardiovascular specialist in patients at high risk of heart failure but without structural heart disease or heart failure symptoms (stage A) (Table 2).

Employing this novel prevention strategy in the extremely large number of patients with stage A heart failure, thought to be up to one-third of the US adult population, may serve as a way to best direct and utilize limited medical resources.⁸

BIOMARKERS FOR PROGNOSIS OR ADDED RISK STRATIFICATION

The 2013 guidelines² recognized that a significant body of work had accumulated showing that natriuretic peptide levels can predict outcomes in both chronic and acute heart failure. Thus, in both conditions, the guidelines contained separate class Ia recommendations to obtain a natriuretic peptide level, troponin level, or both to establish prognosis or disease severity.

The 2017 update¹ underscores the importance of timing in measuring natriuretic peptide levels during admission for ADHF, with emphasis on obtaining them at admission and at discharge for acute and postdischarge prognosis. The completely new class IIa recommendation to obtain a predischarge natriuretic peptide level for postdischarge prognosis was based on a number of observational studies, some of which we explore below.

The ELAN-HF meta-analysis

The European Collaboration on Acute Decompensated Heart Failure (ELAN-HF)¹² performed a meta-analysis to develop a discharge prognostication score for ADHF that included both absolute level and percent change in natriuretic peptide levels at the time of discharge.

Using data from 7 prospective cohorts totaling 1,301 patients, the authors found that incorporation of these values into a subsequently validated risk model led to significant improvements in the ability to predict the end points of all-cause mortality and the combined end point of all-cause mortality or first readmission for a cardiovascular reason within 180 days.

The OPTIMIZE-HF retrospective analysis

Data from the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF) were retrospectively analyzed¹³ to determine whether postdischarge outcomes were best predicted by natriuretic peptide levels at admission or discharge or by the relative change in natriuretic peptide level. More than 7,000 patients age 65 or older, in 220 hospitals, were included, and Cox prediction models were compared using clinical variables alone or in combination with the natriuretic peptide levels.

The model that included the discharge natriuretic peptide level was found to be the most predictive, with a c-index of 0.693 for predicting mortality and a c-index of 0.606 for mortality or rehospitalization at 1 year.

New or modified recommendations on biomarkers for prognosis

The 2017 update¹ modified the earlier recommendation to obtain a natriuretic peptide or troponin level or both at admission for ADHF to establish prognosis. This now has a class Ia recommendation, emphasizing that such levels be obtained on admission. In addition, a new class IIa recommendation is made to obtain a predischarge natriuretic peptide level for postdischarge prognosis. The former class Ia recommendation to obtain a natriuretic peptide level in chronic heart failure to establish prognosis or disease severity remains unchanged.

Also worth noting is what the 2017 update does not recommend in regard to obtaining biomarker levels. It emphasizes that many patients, particularly those with advanced (stage D) heart failure, have a poor prognosis that is well established with or without biomarker levels. Additionally, there are many cardiac and noncardiac causes of natriuretic peptide elevation; thus, clinical judgment remains paramount.

The 2017 update¹ also cautions against setting targets of percent change in or absolute levels of natriuretic peptide at discharge despite observational and retrospective studies demonstrating better outcomes when levels are reduced, as treating for any specific target has never been studied in a large prospective study. Thus, doing so may result in unintended harm. Rather, clinical judgment and optimization of guideline-directed management and therapy are encouraged (Table 2).

 

 

PHARMACOLOGIC TREATMENT FOR STAGE C HFpEF

Although the 2013 guidelines² contain many class I recommendations for various medications in chronic HFrEF, not a single such recommendation is found for chronic HFpEF. A review by Okwuosa et al⁷ covered HFrEF, including the most recent additions on which the 2016 update was based, sacubitril-valsartan and ivabradine. The 2016 update was similarly devoid of recommendations regarding specific medications in HFpEF, leaving only the 2013 class IIb recommendation to consider using an ARB to decrease hospitalizations in HFpEF.

Evidence behind this recommendation came from the Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity program’s randomized controlled trial in 3,025 patients with New York Heart Association (NYHA) class II to IV heart failure and left ventricular ejection fraction over 40%, who were treated with candesartan or placebo.¹⁴ Over a median follow-up of 36.6 months, there was no significant difference in the primary composite outcome of cardiovascular death or admission for heart failure, but significantly fewer patients in the candesartan arm were admitted (230 vs 270, P = .017). Thus the recommendation.

Although this finding was encouraging, it was clear that no blockbuster drug for HFpEF had been identified. Considering that roughly half of all heart failure patients have preserved ejection fraction, the discovery of such a drug for HFpEF would be met with much excitement.¹⁵ Subsequently, other medication classes have been evaluated in the hope of benefit, allowing the 2017 update to provide specific recommendations for aldosterone antagonists, nitrates, and phosphodiesterase-5 inhibitors in HFpEF.

ALDOSTERONE ANTAGONISTS FOR HFpEF

Mineralocorticoid receptor antagonists had previously been shown to significantly reduce morbidity and mortality rates in patients with HFrEF.¹⁶ In addition to aldosterone’s effects on sodium retention and many other pathophysiologic mechanisms relating to heart failure, this hormone is also known to play a role in promoting myocardial fibrosis.¹⁷ Accordingly, some have wondered whether aldosterone antagonists could improve diastolic dysfunction, and perhaps outcomes, in HFpEF.

The Aldo-DHF trial

The Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial investigated whether the aldosterone antagonist spironolactone would improve diastolic function or maximal exercise capacity in chronic HFpEF.¹⁸ It randomized 422 ambulatory patients with NYHA stage II or III heart failure, preserved left ventricular ejection fraction (≥ 50%), and echocardiographic evidence of diastolic dysfunction to receive spironolactone 25 mg daily or placebo.

Although no significant difference was seen in maximal exercise capacity, follow-up over 1 year nevertheless showed significant improvement in echocardiographic diastolic dysfunction (E/e') and perhaps reverse remodeling (decreased left ventricular mass index). These improvements spurred larger trials powered to detect whether clinical outcomes could also be improved.

The TOPCAT trial

The Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial¹⁹ was a large, multicenter, international, double-blind, placebo-controlled trial that investigated whether spironolactone could improve clinical outcomes in HFpEF. It randomized 3,445 patients with symptomatic heart failure and left ventricular ejection fraction of 45% or more to spironolactone 15 to 45 mg daily or placebo.

The effect on a composite primary outcome of death from cardiovascular cause, aborted cardiac arrest, or hospitalization for heart failure was evaluated over a mean follow-up of 3.3 years, with only a small (HR 0.89), nonclinically significant reduction evident. Those in the spironolactone group did have a significantly lower incidence of hospitalization for heart failure (12.0% vs 14.2%, P = .04).

Although the results were disappointing in this essentially negative trial, significant regional variations evident on post hoc analysis prompted further investigation and much controversy since the trial’s publication in 2014.

Participants came in roughly equal proportions from the Americas (United States, Canada, Brazil, and Argentina—51%) and from Russia and Georgia (49%), but outcomes between the two groups were markedly different. Concern was first raised when immediate review discovered a 4-fold lower rate of the primary outcome in the placebo groups from Russia and Georgia (8.4%), a rate in fact similar to that in patients without heart failure.¹⁹ This led to further exploration that identified other red flags that called into question the data integrity from the non-American sites.²⁰

Not only did patients receiving spironolactone in Russia and Georgia not experience the reduction in clinical outcomes seen in their American counterparts, they also did not manifest the expected elevations in potassium and creatinine, and spironolactone metabolites were undetectable in almost one-third of patients.²¹

These findings prompted a post hoc analysis that included only the 51% (1,767 patients) of the study population coming from the Americas; in this subgroup, treatment with spironolactone was associated with a statistically significant 18% relative risk reduction in the primary composite outcome, a 26% reduction in cardiovascular mortality, and an 18% reduction in hospitalization for heart failure.²⁰

New or modified recommendations on aldosterone receptor antagonists

Recognizing both the encouraging data above and the limitations of post hoc analyses, the 2017 focused update provides a class IIb (weak) recommendation stating that aldosterone receptor antagonists might be considered to decrease hospitalizations in appropriately selected patients with HFpEF (Table 3).¹

Nitrates and phosphodiesterase-5 inhibitors

Earlier studies indicated that long-acting nitrates are prescribed in 15% to 50% of patients with HFpEF, perhaps based on extrapolation from studies in HFrEF suggesting that they might improve exercise intolerance.²² Some have speculated that the hemodynamic effects of nitrates, such as decreasing pulmonary congestion, might improve exercise intolerance in those with the stiff ventricles of HFpEF as well, prompting further study.

 

 

The NEAT-HFpEF trial

The Nitrate’s Effect on Activity Tolerance in Heart Failure With Preserved Ejection Fraction (NEAT-HFpEF) trial²² investigated whether extended-release isosorbide mononitrate would increase daily activity levels in patients with HFpEF. This double-blind, crossover study randomized 110 patients with HFpEF (ejection fraction ≥ 50%) and persistent dyspnea to escalating doses of isosorbide mononitrate or placebo over 6 weeks, then to the other arm for another 6 weeks. Daily activity levels during the 120-mg phase were measured with a continuously worn accelerometer.

No beneficial effect of nitrates was evident, with a nonsignificant trend towards decreased activity levels, a significant decrease in hours of activity per day (–0.30 hours, P = .02), and no change in the other secondary end points such as quality-of-life score, 6-minute walk distance, or natriuretic peptide level.

Suggested explanations for these negative findings include the possibility of rapid dose escalation leading to increased subtle side effects (headache, dizziness, fatigue) that, in turn, decreased activity. Additionally, given the imprecise diagnostic criteria for HFpEF, difficulties with patient selection may have led to inclusion of a large number of patients without elevated left-sided filling pressures.²³

The RELAX trial

The Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure With Preserved Ejection Fraction (RELAX) trial²⁴ investigated whether the phosphodiesterase-5 inhibitor sildenafil would improve exercise capacity in HFpEF. Improvements in both exercise capacity and clinical outcomes had already been seen in earlier trials in patients with pulmonary hypertension, as well as in those with HFrEF.²⁵ A smaller study in HFpEF patients with pulmonary hypertension was also encouraging.²⁶

Thus, it was disappointing that, after randomizing 216 outpatients with HFpEF to sildenafil or placebo for 24 weeks, no benefit was seen in the primary end point of change in peak oxygen consumption or in secondary end points of change in 6-minute walk distance or composite clinical score. Unlike in NEAT-HFpEF, patients here were required to have elevated natriuretic peptide levels or elevated invasively measured filling pressures.

The study authors speculated that pulmonary arterial hypertension and right ventricular systolic failure might need to be significant for patients with HFpEF to benefit from phosphodiesterase-5 inhibitors, with their known effects of dilation of pulmonary vasculature and increasing contractility of the right ventricle.²⁴

New or modified recommendations on nitrates or phosphodiesterase-5 drugs

Given these disappointing results, the 2017 update provides a class III (no benefit) recommendation against the routine use of nitrates or phosphodiesterase-5 inhibitors to improve exercise tolerance or quality of life in HFpEF, citing them as ineffective (Table 3).¹

IRON DEFICIENCY IN HEART FAILURE

Not only is iron deficiency present in roughly 50% of patients with symptomatic heart failure (stage C and D HFrEF),²⁷ it is also associated with increased heart failure symptoms such as fatigue and exercise intolerance,²⁸ reduced functional capacity, decreased quality of life, and increased mortality.

Notably, this association exists regardless of the hemoglobin level.²⁹ In fact, even in those without heart failure or anemia, iron deficiency alone results in worsened aerobic performance, exercise intolerance, and increased fatigue.³⁰ Conversely, improvement in symptoms, exercise tolerance, and cognition have been shown with repletion of iron stores in such patients.³¹

At the time of the 2013 guidelines, only a single large trial of intravenous iron in HFrEF and iron deficiency had been carried out (see below), and although the results were promising, it was felt that the evidence base on which to make recommendations was inadequate. Thus, recommendations were deferred until more data could be obtained.

Of note, in all the trials discussed below, iron deficiency was diagnosed in the setting of heart failure as ferritin less than 100 mg/mL (absolute iron deficiency) or as ferritin 100 to 300 mg/mL with transferrin saturation less than 20% (relative deficiency).³²

The CONFIRM-HF trial

As in the Ferinject Assessment in Patients With Iron Deficiency and Chronic Heart Failure (FAIR-HF) trial,³³ the subsequent Ferric Carboxymaltose Evaluation on Performance in Patients With Iron Deficiency in Combination With Chronic Heart Failure (CONFIRM-HF) trial³⁴ involved the intravenous infusion of iron (ferric carboxymaltose) in outpatients with symptomatic HFrEF and iron deficiency. It showed that benefits remained evident with a more objective primary end point (change in 6-minute walk test distance at 24 weeks), and that such benefits were sustained, as seen in numerous secondary end points related to functional capacity at 52 weeks. Benefits in CONFIRM-HF were evident independently from anemia, specifically whether hemoglobin was under or over 12 g/dL.

Although these results were promising, it remained unclear whether such improvements could be obtained with a much easier to administer, more readily available, and less expensive oral iron formulation.

The IRONOUT-HF trial

The Iron Repletion Effects on Oxygen Uptake in Heart Failure (IRONOUT-HF) trial³⁵ investigated whether oral, rather than intravenous, iron supplementation could improve peak exercise capacity in patients with HFrEF and iron deficiency. This double-blind, placebo-controlled trial randomized 225 patients with NYHA class II to IV HFrEF and iron deficiency to treatment with oral iron polysaccharide (150 mg twice daily) or placebo for 16 weeks.

Contrary to the supportive findings above, no significant change was seen in the primary end point of change in peak oxygen uptake or in any of the secondary end points (change in 6-minute walk, quality of life). Also, despite a 15-fold increase in the amount of iron administered in oral form compared with intravenously, little change was evident in the indices of iron stores over the course of the study, with only a 3% increase in transferrin saturation and an 11 ng/mL increase in ferritin. The intravenous trials resulted in a 4-fold greater increase in transferrin saturation and a 20-fold greater increase in ferritin.³⁶

What keeps heart failure patients from absorbing oral iron? It is unclear why oral iron administration in HFrEF, such as in IRONOUT-HF, seems to be so ineffective, but hepcidin—a protein hormone made by the liver that shuts down intestinal iron absorption and iron release from macrophages—may play a central role.³⁷ When iron stores are adequate, hepcidin is upregulated to prevent iron overload. However, hepcidin is also increased in inflammatory states, and chronic heart failure is often associated with inflammation.

With this in mind, the IRONOUT-HF investigators measured baseline hepcidin levels at the beginning and at the end of the 16 weeks and found that high baseline hepcidin levels predicted poorer response to oral iron. Other inflammatory mediators, such as interleukin 6, may also play a role.^38,39 Unlike oral iron formulations such as iron polysaccharide, intravenous iron (ferric carboxymaltose) bypasses these regulatory mechanisms, which may partly explain its much more significant effect on the indices of iron stores and outcomes.

 

 

New or modified recommendations on iron

The 2017 update¹ makes recommendations regarding iron deficiency and anemia in heart failure for the first time.

A class IIb recommendation states that it might be reasonable to treat NYHA class II and III heart failure patients with iron deficiency with intravenous iron to improve functional status and quality of life. A strong recommendation has been deferred until more is known about morbidity and mortality effects from adequately powered trials, some of which are under way and explored further below.

The 2017 update also withholds any recommendations regarding oral iron supplementation in heart failure, citing an uncertain evidence base. Certainly, the subsequent IRONOUT-HF trial does not lend enthusiasm for this approach.

Lastly, given the lack of benefit coupled with the increased risk of thromboembolic events evident in a trial of darbepoetin alfa vs placebo in non-iron deficiency-related anemia in HFrEF,^40,41 the 2017 update provides a class III (no benefit) recommendation against using erythropoietin-stimulating agents in heart failure and anemia.

HYPERTENSION IN HEART FAILURE

The 2013 guidelines for the management of heart failure simply provided a class I recommendation to control hypertension and lipid disorders in accordance with contemporary guidelines to lower the risk of heart failure.¹

SPRINT

The Systolic Blood Pressure Intervention Trial (SPRINT)⁴² sought to determine whether a lower systolic blood pressure target (120 vs 140 mm Hg) would reduce clinical events in patients at high risk for cardiovascular events but without diabetes mellitus. Patients at high risk were defined as over age 75, or with known vascular disease, chronic kidney disease, or a Framingham Risk Score higher than 15%. This multicenter, open-label controlled trial randomized 9,361 patients to intensive treatment (goal systolic blood pressure < 120 mm Hg) or standard treatment (goal systolic blood pressure < 140 mm Hg).

SPRINT was stopped early at a median follow-up of 3.26 years when a 25% relative risk reduction in the primary composite outcome of myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes became evident in the intensive-treatment group (1.65% vs 2.19% per year, HR 0.75, P < .0001).

All-cause mortality was also lower in the intensive-treatment group (HR 0.73, P = .003), while the incidence of serious adverse events (hypotension, syncope, electrolyte abnormalities, acute kidney injury, and noninjurious falls) was only slightly higher (38.3% vs 37.1%, P = .25). Most pertinent, a significant 38% relative risk reduction in heart failure and a 43% relative risk reduction in cardiovascular events were also evident.

Of note, blood pressure measurements were taken as the average of 3 measurements obtained by an automated cuff taken after the patient had been sitting quietly alone in a room for 5 minutes.

New or modified recommendations on hypertension in heart failure

Given the impressive 25% relative risk reduction in myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes in SPRINT,⁴² the 2017 update¹ incorporated the intensive targets of SPRINT into its recommendations. However, to compensate for what are expected to be higher blood pressures obtained in real-world clinical practice as opposed to the near-perfect conditions used in SPRINT, a slightly higher blood pressure goal of less than 130/80 mm Hg was set.

Specific blood pressure guidelines have not been given for stage A heart failure in the past. However, as for other new approaches to prevent heart failure in this update and given the 38% relative risk reduction in heart failure seen in SPRINT, a class I recommendation is given to target a blood pressure goal of less than 130/80 mm Hg in stage A heart failure with hypertension (Table 4).

Although not specifically included in SPRINT, given the lack of trial data on specific blood pressure targets in HFrEF and the decreased cardiovascular events noted above, a class I (level of evidence C, expert opinion) recommendation to target a goal systolic blood pressure less than 130 mm Hg in stage C HFrEF with hypertension is also given. Standard guideline-directed medications in the treatment of HFrEF are to be used (Table 4).

Similarly, a new class I (level of evidence C, expert opinion) recommendation is given for hypertension in HFpEF to target a systolic blood pressure of less than 130 mm Hg, with special mention to first manage any element of volume overload with diuretics. Other than avoiding nitrates (unless used for angina) and phosphodiesterase inhibitors, it is noted that few data exist to guide the choice of antihypertensive further, although perhaps renin-angiotensin-aldosterone system inhibition, especially aldosterone antagonists, may be considered. These recommendations are fully in line with the 2017 ACC/AHA high blood pressure clinical practice guidelines,⁴³ ie, that renin-angiotensin-aldosterone system inhibition with an angiotensin-converting enzyme (ACE) inhibitor or ARB and especially mineralocorticoid receptor antagonists would be the preferred choice (Table 4).

SLEEP-DISORDERED BREATHING IN HEART FAILURE

Sleep-disordered breathing, either obstructive sleep apnea (OSA) or central sleep apnea, is quite commonly associated with symptomatic HFrEF.⁴⁴ Whereas OSA is found in roughly 18% and central sleep apnea in 1% of the general population, sleep-disordered breathing is found in nearly 60% of patients with HFrEF, with some studies showing a nearly equal proportion of OSA and central sleep apnea.⁴⁵ A similar prevalence is seen in HFpEF, although with a much higher proportion of OSA.⁴⁶ Central sleep apnea tends to be a marker of more severe heart failure, as it is strongly associated with severe cardiac systolic dysfunction and worse functional capacity.⁴⁷

Not surprisingly, the underlying mechanism of central sleep apnea is quite different from that of OSA. Whereas OSA predominantly occurs because of repeated obstruction of the pharynx due to nocturnal pharyngeal muscle relaxation, no such airway patency issues or strained breathing patterns exist in central sleep apnea. Central sleep apnea, which can manifest as Cheyne-Stokes respirations, is thought to occur due to an abnormal ventilatory control system with complex pathophysiology such as altered sensitivity of central chemoreceptors to carbon dioxide, interplay of pulmonary congestion, subsequent hyperventilation, and prolonged circulation times due to reduced cardiac output.⁴⁸

What the two types of sleep-disordered breathing have in common is an association with negative health outcomes. Both appear to induce inflammation and sympathetic nervous system activity via oxidative stress from intermittent nocturnal hypoxemia and hypercapnea.⁴⁹ OSA was already known to be associated with significant morbidity and mortality rates in the general population,⁵⁰ and central sleep apnea had been identified as an independent predictor of mortality in HFrEF.⁵¹

At the time of the 2013 guidelines, only small or observational studies with limited results had been done evaluating treatment effects of continuous positive airway pressure therapy (CPAP) on OSA and central sleep apnea. Given the relative paucity of data, only a single class IIa recommendation stating that CPAP could be beneficial to increase left ventricular ejection fraction and functional status in concomitant sleep apnea and heart failure was given in 2013. However, many larger trials were under way,^52–59 some with surprising results such as a significant increase in cardiovascular and all-cause mortality (Table 5).⁵⁴

 

 

New or modified recommendations on sleep-disordered breathing

Stemming from several trials,^54,56 3 new recommendations on sleep-disordered breathing were made in the 2017 update (Table 6).

Given the common association with heart failure (60%)⁴⁵ and the marked variation in response to treatment, including potential for harm with adaptive servo-ventilation and central sleep apnea, a class IIa recommendation is made stating that it is reasonable to obtain a formal sleep study in any patient with symptomatic (NYHA class II–IV) heart failure.¹

Due to the potential for harm with adaptive servo-ventilation in patients with central sleep apnea and NYHA class II to IV HFrEF, a class III (harm) recommendation is made against its use.

Largely based on the results of the Sleep Apnea Cardiovascular Endpoints (SAVE) trial,⁵⁶ a class IIb, level of evidence B-R (moderate, based on randomized trials) recommendation is given, stating that the use of CPAP in those with OSA and known cardiovascular disease may be reasonable to improve sleep quality and reduce daytime sleepiness.

POTENTIAL APPLICATIONS IN ACUTE DECOMPENSATED HEART FAILURE

Although the 2017 update¹ is directed mostly toward managing chronic heart failure, it is worth considering how it might apply to the management of ADHF.

SHOULD WE USE BIOMARFER TARGETS TO GUIDE THERAPY IN ADHF?

The 2017 update¹ does offer direct recommendations regarding the use of biomarker levels during admissions for ADHF. Mainly, they emphasize that the admission biomarker levels provide valuable information regarding acute prognosis and risk stratification (class I recommendation), while natriuretic peptide levels just before discharge provide the same for the postdischarge timeframe (class IIa recommendation).

The update also explicitly cautions against using a natriuretic peptide level-guided treatment strategy, such as setting targets for predischarge absolute level or percent change in level of natriuretic peptides during admissions for ADHF. Although observational and retrospective studies have shown better outcomes when levels are reduced at discharge, treating for any specific inpatient target has never been tested in any large, prospective study; thus, doing so could result in unintended harm.

So what do we know?

McQuade et al systematic review

McQuade et al⁵⁷ performed a systematic review of more than 40 ADHF trials, which showed that, indeed, patients who achieved a target absolute natriuretic peptide level (BNP ≤ 250 pg/mL) or percent reduction (≥ 30%) at time of discharge had significantly improved outcomes such as reduced postdischarge all-cause mortality and rehospitalization rates. However, these were mostly prospective cohort studies that did not use any type of natriuretic peptide level-guided treatment protocol, leaving it unclear whether such a strategy could positively influence outcomes.

For this reason, both McQuade et al⁵⁷ and, in an accompanying editorial, Felker et al⁵⁸ called for properly designed, randomized controlled trials to investigate such a strategy. Felker noted that only 2 such phase II trials in ADHF have been completed,^59,60 with unconvincing results.

PRIMA II

The Multicenter, Randomized Clinical Trial to Study the Impact of In-hospital Guidance for Acute Decompensated Heart Failure Treatment by a Predefined NT-ProBNP Target on the Reduction of Readmission and Mortality Rates (PRIMA II)⁶⁰ randomized patients to natriuretic peptide level-guided treatment or standard care during admission for ADHF.

Many participants (60%) reached the predetermined target of 30% reduction in natriuretic peptide levels at the time of clinical stabilization and randomization; 405 patients were randomized. Patients in the natriuretic peptide level-guided treatment group underwent a prespecified treatment algorithm, with repeat natriuretic peptide levels measured again after the protocol.

Natriuretic peptide-guided therapy failed to show any significant benefit in any clinical outcomes, including the primary composite end point of mortality or heart failure readmissions at 180 days (36% vs 38%, HR 0.99, 95% confidence interval 0.72–1.36). Consistent with the review by McQuade et al,⁵⁷ achieving the 30% reduction in natriuretic peptide at discharge, in either arm, was associated with a better prognosis, with significantly lower mortality and readmission rates at 180 days (HR 0.39 for rehospitalization or death, 95% confidence interval 0.27–0.55).

As in the observational studies, those who achieved the target natriuretic peptide level at the time of discharge had a better prognosis than those who did not, but neither study showed an improvement in clinical outcomes using a natriuretic peptide level-targeting treatment strategy.

No larger randomized controlled trial results are available for guided therapy in ADHF. However, additional insight may be gained from a subsequent trial⁶¹ that evaluated biomarker-guided titration of guideline-directed medical therapy in outpatients with chronic HFrEF.

The GUIDE-IT trial

That trial, the Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure (GUIDE-IT)⁶¹ trial, was a large multicenter attempt to determine whether a natriuretic peptide-guided treatment strategy was more effective than standard care in the management of 894 high-risk outpatients with chronic HFrEF. Earlier, promising results had been obtained in a meta-analysis⁶² of more than 11 similar trials in 2,000 outpatients, with a decreased mortality rate (HR 0.62) seen in the biomarker-guided arm. However, the results had not been definitive due to being underpowered.⁶²

Unfortunately, the results of GUIDE-IT were disappointing, with no significant difference in either the combined primary end point of mortality or hospitalization for heart failure, or the secondary end points evident at 15 months, prompting early termination for futility.⁶¹ Among other factors, the study authors postulated that this may have partly resulted from a patient population with more severe heart failure and resultant azotemia, limiting the ability to titrate neurohormonal medications to the desired dosage.

The question of whether patients who cannot achieve such biomarker targets need more intensive therapy or whether their heart failure is too severe to respond adequately echoes the question often raised in discussions of inpatient biomarker-guided therapy.⁵⁸ Thus, only limited insight is gained, and it remains unclear whether a natriuretic peptide-guided treatment strategy can improve outpatient or inpatient outcomes. Until this is clarified, clinical judgment and optimization of guideline-directed management and therapy should remain the bedrock of treatment.

 

 

SHOULD ALDOSTERONE ANTAGONISTS BE USED IN ACUTE HFpEF?

Given the encouraging results in chronic HFpEF from post hoc analyses of TOPCAT, are there any additional recent data suggesting a role for aldosterone antagonists such as spironolactone in acute HFpEF?

The ATHENA-HF trial

The Aldosterone Targeted Neurohormonal Combined With Natriuresis Therapy in Heart Failure (ATHENA-HF) trial⁶³ compared treatment with high-dose spironolactone (100 mg) for 96 hours vs usual care in 360 patients with ADHF. The patient population included those with HFrEF and HFpEF, and usual care included low-dose spironolactone (12.5–25 mg) in roughly 15% of patients. High-dose mineralocorticoid receptor antagonists have been shown to overcome diuretic resistance, improve pulmonary vascular congestion, and partially combat the adverse neurohormonal activation seen in ADHF.

Unfortunately, the trial was completely neutral in regard to the primary end point of reduction in natriuretic peptide levels as well as to the secondary end points of 30-day mortality rate, heart failure readmission, clinical congestion scores, urine output, and change in weight. No suggestion of additional benefit was seen in subgroup analysis of patients with acute HFpEF (ejection fraction > 45%), which yielded similar results.⁶³

Given these lackluster findings, routine use of high-dose spironolactone in ADHF is not recommended.⁶⁴ However, the treatment was well tolerated, without significant adverse effects of hyperkalemia or kidney injury, leaving the door open as to whether it may have utility in selected patients with diuretic resistance.

Should ARNIs and ivabradine be started during ADHF admissions?

The first half of the focused update³ of the 2013 guidelines,² reviewed by Okwuosa et al,⁷ provided recommendations for the use of sacubitril-valsartan, an angiotensin-neprilysin inhibitor (ARNI), and ivabradine, a selective sinoatrial node If channel inhibitor, in chronic HFrEF.

Sacubitril-valsartan was given a class I recommendation for use in patients with NYHA class II or III chronic HFrEF who tolerate an ACE inhibitor or an ARB. This recommendation was given largely based on the benefits in mortality and heart failure hospitalizations seen in PARADIGM-HF (the Prospective Comparison of ARNI With ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure)⁶⁵ compared with enalapril (HR 0.80, 95% CI 0.73–0.87, P < .001).

There is currently no recommendation on initiation or use of ARNIs during admissions for ADHF, but a recent trial may lend some insight.⁶⁶

THE PIONEER-HF trial

The Comparison of Sacubitril/Valsartan vs Enalapril on Effect on NT-proBNP in Patients Stabilized From an Acute Heart Failure Episode (PIONEER-HF) trial⁶⁶ randomized patients admitted for acute HFrEF, once stabilized, to sacubitril-valsartan or enalapril. Encouragingly, the percentage change of natriuretic peptide levels from the time of inpatient initiation to 4 and 8 weeks thereafter, the primary efficacy end point, was 46.7% with sacubitril-valsartan versus 25.3% with enalapril alone (ratio of change 0.71, 95% CI 0.63–0.81, P < .001). Although not powered for such, a prespecified analysis of a composite of clinical outcomes was also favorable for sacubitril-valsartan, largely driven by a 44% decreased rate of rehospitalization. More definitive, and quite reassuring, was that no significant difference was seen in the key safety outcomes of worsening renal function, hyperkalemia, symptomatic hypotension, and angioedema. These results were also applicable to the one-third of study participants who had no former diagnosis of heart failure, the one-third identifying as African American, and the one-third who had not been taking an ACE inhibitor or ARB. These results, taken together with the notion that at study completion the patients become similar to those included in PARADIGM-HF, have led some to assert that PIONEER-HF has the potential to change clinical practice.

Ivabradine was given a class IIa recommendation for use in patients with NYHA class II or III chronic HFrEF with a resting heart rate of at least 70 bpm, in sinus rhythm, despite being on optimal medical therapy including a beta-blocker at a maximum tolerated dose.

This recommendation was largely based on SHIFT (Systolic Heart Failure Treatment With the If Inhibitor Ivabradine Trial), which randomized patients to ivabradine or placebo to evaluate the effects of isolated lowering of the heart rate on the composite primary outcome of cardiovascular death or hospitalization. A significant reduction was seen in the ivabradine arm (HR 0.82, 95% CI 0.75–0.90, P < .0001), mainly driven by decreased hospitalizations.⁶⁷

Subsequently, a small unblinded single-center study was undertaken to evaluate the efficacy and safety of initiating ivabradine during admissions for ADHF.⁶⁸

THE ETHIC-AHF trial

The Effect of Early Treatment With Ivabradine Combined With Beta-Blockers vs Beta-Blockers Alone in Patients Hospitalized With Heart Failure and Reduced Left Ventricular Ejection Fraction (ETHIC-AHF) trial⁶⁸ sought to determine the safety and effectiveness of early coadministration of ivabradine with beta-blockers in patients with acute HFrEF.

This single-center, unblinded study randomized 71 patients to ivabradine and beta-blockade or beta-blockade alone upon clinical stabilization (24–48 hours) after admission for acute decompensated HFrEF.

The primary end point was heart rate at 28 days, with the ivabradine group showing a statistically significant decrease (64 vs 70 bpm, P = .01), which persisted at 4 months. There was no significant difference in the secondary end points of adverse drug effects or the composite of clinical event outcomes (all-cause mortality, admission for heart failure or cardiovascular cause), but a number of surrogate end points including left ventricular ejection fraction, BNP level, and NYHA functional class at 4 months showed mild improvement.

Although this study provided evidence that the coadministration of ivabradine and a beta-blocker is safe and was positive in regard to clinical outcomes, the significant limitations due to its size and study design (single-center, unblinded, 4-month follow-up) simply serve to support the pursuit of larger studies with more stringent design and longer follow-up in order to determine the clinical efficacy.

 

 

The PRIME-HF trial

The Predischarge Initiation of Ivabradine in the Management of Heart Failure (PRIME-HF) trial⁶⁹ is a randomized, open-label, multicenter trial comparing standard care vs the initiation of ivabradine before discharge, but after clinical stabilization, during admissions for ADHF in patients with chronic HFrEF (left ventricular ejection fraction ≤ 35%). At subsequent outpatient visits, the dosage can be modified in the ivabradine group, or ivabradine can be initiated at the provider’s discretion in the usual-care group.

PRIME-HF is attempting to determine whether initiating ivabradine before discharge will result in more patients taking ivabradine at 180 days, its primary end point, as well as in changes in secondary end points including heart rate and patient-centered outcomes. The study is active, with reporting expected in 2019.

As these trials all come to completion, it will not be long before we have further guidance regarding the inpatient initiation of these new and exciting therapeutic agents.

SHOULD INTRAVENOUS IRON BE GIVEN DURING ADHF ADMISSIONS?

Given the high prevalence of iron deficiency in symptomatic HFrEF, its independent association with mortality, improvements in quality of life and functional capacity suggested by repleting with intravenous iron (in FAIR-HF and CONFIRM-HF), the seeming inefficacy of oral iron in IRONOUT, and the logistical challenges of intravenous administration during standard clinic visits, could giving intravenous iron soon be incorporated into admissions for ADHF?

Caution has been advised for several reasons. As discussed above, larger randomized controlled trials powered to detect more definitive clinical end points such as death and the rate of hospitalization are still needed before a stronger recommendation can be made for intravenous iron in HFrEF. Also, without such data, it seems unwise to add the considerable economic burden of routinely assessing for iron deficiency and providing intravenous iron during ADHF admissions to the already staggering costs of heart failure.

Thus far, only a single meta-analysis is available, including 893 patients⁷⁰ largely from the FAIR-HF and CONFIRM-HF trials. While it does suggest benefit in both cardiovascular mortality and recurrent hospitalizations for heart failure (rate ratio 0.59, 95% CI 0.40–0.88; P = .009), more definitive guidance will be provided by the results from 4 large randomized placebo-controlled studies  currently under way or recruiting. All 4 seek to examine the effects of intravenous iron on morbidity and mortality in patients with HFrEF and iron deficiency, using a variety of end points ranging from exercise tolerance, to hospitalizations, to mortality (Table 7).^71–74

The effects seen on morbidity and mortality that become evident in these trials over the next 5 years will help determine future guidelines and whether intravenous iron is routinely administered in bridge clinics, during inpatient admissions for ADHF, or not at all in patients with HFrEF and iron deficiency.

INTERNISTS ARE KEY

Heart failure remains one of the most common, morbid, complex, and costly diseases in the United States, and its prevalence is expected only to increase.^4,5 The 2017 update1 of the 2013 guideline² for the management of heart failure provides recommendations aimed not only at management of heart failure, but also at its comorbidities and, for the first time ever, at its prevention.

Internists provide care for the majority of heart failure patients, as well as for their comorbidities, and are most often the first to come into contact with patients at high risk of developing heart failure. Thus, a thorough understanding of these guidelines and how to apply them to the management of acute decompensated heart failure is of critical importance.

In 2017, the American College of Cardiology (ACC), American Heart Association (AHA), and Heart Failure Society of America (HFSA) jointly released a focused update¹ of the 2013 ACC/AHA guideline for managing heart failure.² This is the second focused update of the 2013 guidelines; the first update,³ in 2016, covered 2 new drugs (sacubitril-valsartan and ivabradine) for chronic stage C heart failure with reduced ejection fraction (HFrEF).

Rather than focus on new medication classes, this second update provides recommendations regarding:

• Preventing the progression to left ventricular dysfunction or heart failure in patients at high risk (stage A) through screening with B-type natriuretic peptide (BNP) and aiming for more aggressive blood pressure control
• Inpatient biomarker use
• Medications in heart failure with preserved ejection fraction (HFpEF, or diastolic heart failure)
• Blood pressure targets in stage C heart failure
• Managing important comorbidities such as iron deficiency and sleep-disordered breathing to decrease morbidity, improve functional capacity, and enhance quality of life.

These guidelines and the data that underlie them are explored below. We also discuss potential applications to the management of hospitalization for acute decompensated heart failure (ADHF).

COMMON, COSTLY, AND DEBILITATING

Heart failure—defined by the ACC/AHA as the complex clinical syndrome that results from any structural or functional impairment of ventricular filling or ejection of blood—remains one of the most common, costly, and debilitating diseases in the United States.² Based on National Health and Nutrition Examination Survey data from 2011 to 2014, an estimated 6.5 million US adults have it, with projections of more than 8 million by 2030.^4,5 More than 960,000 new cases are thought to occur annually, with a lifetime risk of developing it of roughly 20% to 45%.⁶

Despite ever-growing familiarity and some significant strides in management, the death rate in this syndrome is substantial. After admissions for heart failure (which number 1 million per year), the mortality rate is roughly 10% at 1 year and 40% at 5 years.⁶ Also staggering are the associated costs, with $30.7 billion attributed to heart failure in 2012 and a projected $69.7 billion annually by 2030.⁵ Thus, we must direct efforts not only to treatment, but also to prevention.

Preventive efforts would target patients  with ACC/AHA stage A heart failure—those at high risk for developing but currently without evidence of structural heart disease or heart failure symptoms (Table 1).⁷ This group may represent up to one-third of the US adult population, or 75 million people, when including the well-recognized risk factors of coronary artery disease, hypertension, diabetes mellitus, and chronic kidney disease in those without left ventricular dysfunction or heart failure.⁸

BIOMARKERS FOR PREVENTION

Past ACC/AHA heart failure guidelines² have included recommendations on the use of biomarkers to aid in diagnosis and prognosis and, to a lesser degree, to guide treatment of heart failure. Largely based on 2 trials (see below), the 2017 guidelines go further, issuing a recommendation on the use of natriuretic peptide biomarkers in a screening strategy to prompt early intervention and prevent the progression to clinical heart failure in high-risk patients (stage A heart failure).

The PONTIAC trial

The NT-proBNP Selected Prevention of Cardiac Events in a Population of Diabetic Patients Without a History of Cardiac Disease (PONTIAC) trial⁹ randomized 300 outpatients with type 2 diabetes mellitus and an elevated N-terminal proBNP (NT-proBNP) level (> 125 pg/mL) to standard medical care vs standard care plus intensive up-titration of renin-angiotensin system antagonists and beta-blockers in a cardiac clinic over 2 years.

Earlier studies¹⁰ had shown NT-proBNP levels to have predictive value for cardiac events in diabetic patients, while the neurohormonal treatments were thought to have an established record of preventing primary and secondary cardiovascular events. In PONTIAC, a significant reduction was seen in the primary end point of hospitalization or death due to cardiac disease (hazard ratio [HR] 0.351, P = .044), as well as in the secondary end point of hospitalization due to heart failure (P < .05), in the aggressive-intervention group. These results laid the foundation for the larger St. Vincent’s Screening to Prevent Heart Failure (STOP-HF) trial.¹¹

 

 

The STOP-HF trial

The STOP-HF trial randomized 1,235 outpatients who were at high risk but without left ventricular dysfunction or heart failure symptoms (stage A) to annual screening alone vs annual screening plus BNP testing, in which a BNP level higher than 50 pg/mL triggered echocardiography and evaluation by a cardiologist who would then assist with medications.¹¹

Eligible patients were over age 40 and had 1 or more of the following risk factors:

• Diabetes mellitus
• Hypertension
• Hypercholesterolemia
• Obesity (body mass index > 30 kg/m²)
• Vascular disease (coronary, cerebral, or peripheral arterial disease)
• Arrhythmia requiring treatment
• Moderate to severe valvular disease.

After a mean follow-up of 4.3 years, the primary end point, ie, asymptomatic left ventricular dysfunction with or without newly diagnosed heart failure, was found in 9.7% of the control group and in only 5.9% of the intervention group with BNP screening, a 42% relative risk reduction (P = .013).

Similarly, the incidence of secondary end points of emergency hospitalization for a cardiovascular event (arrhythmia, transient ischemic attack, stroke, myocardial infarction, peripheral or pulmonary thrombosis or embolization, or heart failure) was also lower at 45.2 vs 24.4 per 1,000 patient-years, a 46% relative risk reduction.

An important difference in medications between the 2 groups was an increase in subsequently prescribed renin-angiotensin-aldosterone system therapy, mainly consisting of angiotensin II receptor blockers (ARBs), in those with elevated BNP in the intervention group. Notably, blood pressure was about the same in the 2 groups.¹¹

Although these findings are encouraging, larger studies are needed, as the lack of blinding, low event rates, and small absolute risk reduction make the results difficult to generalize.

New or modified recommendations for screening


The 2017 update¹ provided a class IIa (moderate) recommendation for natriuretic peptide biomarker-based screening with subsequent guideline-based treatment directed by a cardiovascular specialist in patients at high risk of heart failure but without structural heart disease or heart failure symptoms (stage A) (Table 2).

Employing this novel prevention strategy in the extremely large number of patients with stage A heart failure, thought to be up to one-third of the US adult population, may serve as a way to best direct and utilize limited medical resources.⁸

BIOMARKERS FOR PROGNOSIS OR ADDED RISK STRATIFICATION

The 2013 guidelines² recognized that a significant body of work had accumulated showing that natriuretic peptide levels can predict outcomes in both chronic and acute heart failure. Thus, in both conditions, the guidelines contained separate class Ia recommendations to obtain a natriuretic peptide level, troponin level, or both to establish prognosis or disease severity.

The 2017 update¹ underscores the importance of timing in measuring natriuretic peptide levels during admission for ADHF, with emphasis on obtaining them at admission and at discharge for acute and postdischarge prognosis. The completely new class IIa recommendation to obtain a predischarge natriuretic peptide level for postdischarge prognosis was based on a number of observational studies, some of which we explore below.

The ELAN-HF meta-analysis

The European Collaboration on Acute Decompensated Heart Failure (ELAN-HF)¹² performed a meta-analysis to develop a discharge prognostication score for ADHF that included both absolute level and percent change in natriuretic peptide levels at the time of discharge.

Using data from 7 prospective cohorts totaling 1,301 patients, the authors found that incorporation of these values into a subsequently validated risk model led to significant improvements in the ability to predict the end points of all-cause mortality and the combined end point of all-cause mortality or first readmission for a cardiovascular reason within 180 days.

The OPTIMIZE-HF retrospective analysis

Data from the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF) were retrospectively analyzed¹³ to determine whether postdischarge outcomes were best predicted by natriuretic peptide levels at admission or discharge or by the relative change in natriuretic peptide level. More than 7,000 patients age 65 or older, in 220 hospitals, were included, and Cox prediction models were compared using clinical variables alone or in combination with the natriuretic peptide levels.

The model that included the discharge natriuretic peptide level was found to be the most predictive, with a c-index of 0.693 for predicting mortality and a c-index of 0.606 for mortality or rehospitalization at 1 year.

New or modified recommendations on biomarkers for prognosis

The 2017 update¹ modified the earlier recommendation to obtain a natriuretic peptide or troponin level or both at admission for ADHF to establish prognosis. This now has a class Ia recommendation, emphasizing that such levels be obtained on admission. In addition, a new class IIa recommendation is made to obtain a predischarge natriuretic peptide level for postdischarge prognosis. The former class Ia recommendation to obtain a natriuretic peptide level in chronic heart failure to establish prognosis or disease severity remains unchanged.

Also worth noting is what the 2017 update does not recommend in regard to obtaining biomarker levels. It emphasizes that many patients, particularly those with advanced (stage D) heart failure, have a poor prognosis that is well established with or without biomarker levels. Additionally, there are many cardiac and noncardiac causes of natriuretic peptide elevation; thus, clinical judgment remains paramount.

The 2017 update¹ also cautions against setting targets of percent change in or absolute levels of natriuretic peptide at discharge despite observational and retrospective studies demonstrating better outcomes when levels are reduced, as treating for any specific target has never been studied in a large prospective study. Thus, doing so may result in unintended harm. Rather, clinical judgment and optimization of guideline-directed management and therapy are encouraged (Table 2).

 

 

PHARMACOLOGIC TREATMENT FOR STAGE C HFpEF

Although the 2013 guidelines² contain many class I recommendations for various medications in chronic HFrEF, not a single such recommendation is found for chronic HFpEF. A review by Okwuosa et al⁷ covered HFrEF, including the most recent additions on which the 2016 update was based, sacubitril-valsartan and ivabradine. The 2016 update was similarly devoid of recommendations regarding specific medications in HFpEF, leaving only the 2013 class IIb recommendation to consider using an ARB to decrease hospitalizations in HFpEF.

Evidence behind this recommendation came from the Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity program’s randomized controlled trial in 3,025 patients with New York Heart Association (NYHA) class II to IV heart failure and left ventricular ejection fraction over 40%, who were treated with candesartan or placebo.¹⁴ Over a median follow-up of 36.6 months, there was no significant difference in the primary composite outcome of cardiovascular death or admission for heart failure, but significantly fewer patients in the candesartan arm were admitted (230 vs 270, P = .017). Thus the recommendation.

Although this finding was encouraging, it was clear that no blockbuster drug for HFpEF had been identified. Considering that roughly half of all heart failure patients have preserved ejection fraction, the discovery of such a drug for HFpEF would be met with much excitement.¹⁵ Subsequently, other medication classes have been evaluated in the hope of benefit, allowing the 2017 update to provide specific recommendations for aldosterone antagonists, nitrates, and phosphodiesterase-5 inhibitors in HFpEF.

ALDOSTERONE ANTAGONISTS FOR HFpEF

Mineralocorticoid receptor antagonists had previously been shown to significantly reduce morbidity and mortality rates in patients with HFrEF.¹⁶ In addition to aldosterone’s effects on sodium retention and many other pathophysiologic mechanisms relating to heart failure, this hormone is also known to play a role in promoting myocardial fibrosis.¹⁷ Accordingly, some have wondered whether aldosterone antagonists could improve diastolic dysfunction, and perhaps outcomes, in HFpEF.

The Aldo-DHF trial

The Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial investigated whether the aldosterone antagonist spironolactone would improve diastolic function or maximal exercise capacity in chronic HFpEF.¹⁸ It randomized 422 ambulatory patients with NYHA stage II or III heart failure, preserved left ventricular ejection fraction (≥ 50%), and echocardiographic evidence of diastolic dysfunction to receive spironolactone 25 mg daily or placebo.

Although no significant difference was seen in maximal exercise capacity, follow-up over 1 year nevertheless showed significant improvement in echocardiographic diastolic dysfunction (E/e') and perhaps reverse remodeling (decreased left ventricular mass index). These improvements spurred larger trials powered to detect whether clinical outcomes could also be improved.

The TOPCAT trial

The Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial¹⁹ was a large, multicenter, international, double-blind, placebo-controlled trial that investigated whether spironolactone could improve clinical outcomes in HFpEF. It randomized 3,445 patients with symptomatic heart failure and left ventricular ejection fraction of 45% or more to spironolactone 15 to 45 mg daily or placebo.

The effect on a composite primary outcome of death from cardiovascular cause, aborted cardiac arrest, or hospitalization for heart failure was evaluated over a mean follow-up of 3.3 years, with only a small (HR 0.89), nonclinically significant reduction evident. Those in the spironolactone group did have a significantly lower incidence of hospitalization for heart failure (12.0% vs 14.2%, P = .04).

Although the results were disappointing in this essentially negative trial, significant regional variations evident on post hoc analysis prompted further investigation and much controversy since the trial’s publication in 2014.

Participants came in roughly equal proportions from the Americas (United States, Canada, Brazil, and Argentina—51%) and from Russia and Georgia (49%), but outcomes between the two groups were markedly different. Concern was first raised when immediate review discovered a 4-fold lower rate of the primary outcome in the placebo groups from Russia and Georgia (8.4%), a rate in fact similar to that in patients without heart failure.¹⁹ This led to further exploration that identified other red flags that called into question the data integrity from the non-American sites.²⁰

Not only did patients receiving spironolactone in Russia and Georgia not experience the reduction in clinical outcomes seen in their American counterparts, they also did not manifest the expected elevations in potassium and creatinine, and spironolactone metabolites were undetectable in almost one-third of patients.²¹

These findings prompted a post hoc analysis that included only the 51% (1,767 patients) of the study population coming from the Americas; in this subgroup, treatment with spironolactone was associated with a statistically significant 18% relative risk reduction in the primary composite outcome, a 26% reduction in cardiovascular mortality, and an 18% reduction in hospitalization for heart failure.²⁰

New or modified recommendations on aldosterone receptor antagonists

Recognizing both the encouraging data above and the limitations of post hoc analyses, the 2017 focused update provides a class IIb (weak) recommendation stating that aldosterone receptor antagonists might be considered to decrease hospitalizations in appropriately selected patients with HFpEF (Table 3).¹

Nitrates and phosphodiesterase-5 inhibitors

Earlier studies indicated that long-acting nitrates are prescribed in 15% to 50% of patients with HFpEF, perhaps based on extrapolation from studies in HFrEF suggesting that they might improve exercise intolerance.²² Some have speculated that the hemodynamic effects of nitrates, such as decreasing pulmonary congestion, might improve exercise intolerance in those with the stiff ventricles of HFpEF as well, prompting further study.

 

 

The NEAT-HFpEF trial

The Nitrate’s Effect on Activity Tolerance in Heart Failure With Preserved Ejection Fraction (NEAT-HFpEF) trial²² investigated whether extended-release isosorbide mononitrate would increase daily activity levels in patients with HFpEF. This double-blind, crossover study randomized 110 patients with HFpEF (ejection fraction ≥ 50%) and persistent dyspnea to escalating doses of isosorbide mononitrate or placebo over 6 weeks, then to the other arm for another 6 weeks. Daily activity levels during the 120-mg phase were measured with a continuously worn accelerometer.

No beneficial effect of nitrates was evident, with a nonsignificant trend towards decreased activity levels, a significant decrease in hours of activity per day (–0.30 hours, P = .02), and no change in the other secondary end points such as quality-of-life score, 6-minute walk distance, or natriuretic peptide level.

Suggested explanations for these negative findings include the possibility of rapid dose escalation leading to increased subtle side effects (headache, dizziness, fatigue) that, in turn, decreased activity. Additionally, given the imprecise diagnostic criteria for HFpEF, difficulties with patient selection may have led to inclusion of a large number of patients without elevated left-sided filling pressures.²³

The RELAX trial

The Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure With Preserved Ejection Fraction (RELAX) trial²⁴ investigated whether the phosphodiesterase-5 inhibitor sildenafil would improve exercise capacity in HFpEF. Improvements in both exercise capacity and clinical outcomes had already been seen in earlier trials in patients with pulmonary hypertension, as well as in those with HFrEF.²⁵ A smaller study in HFpEF patients with pulmonary hypertension was also encouraging.²⁶

Thus, it was disappointing that, after randomizing 216 outpatients with HFpEF to sildenafil or placebo for 24 weeks, no benefit was seen in the primary end point of change in peak oxygen consumption or in secondary end points of change in 6-minute walk distance or composite clinical score. Unlike in NEAT-HFpEF, patients here were required to have elevated natriuretic peptide levels or elevated invasively measured filling pressures.

The study authors speculated that pulmonary arterial hypertension and right ventricular systolic failure might need to be significant for patients with HFpEF to benefit from phosphodiesterase-5 inhibitors, with their known effects of dilation of pulmonary vasculature and increasing contractility of the right ventricle.²⁴

New or modified recommendations on nitrates or phosphodiesterase-5 drugs

Given these disappointing results, the 2017 update provides a class III (no benefit) recommendation against the routine use of nitrates or phosphodiesterase-5 inhibitors to improve exercise tolerance or quality of life in HFpEF, citing them as ineffective (Table 3).¹

IRON DEFICIENCY IN HEART FAILURE

Not only is iron deficiency present in roughly 50% of patients with symptomatic heart failure (stage C and D HFrEF),²⁷ it is also associated with increased heart failure symptoms such as fatigue and exercise intolerance,²⁸ reduced functional capacity, decreased quality of life, and increased mortality.

Notably, this association exists regardless of the hemoglobin level.²⁹ In fact, even in those without heart failure or anemia, iron deficiency alone results in worsened aerobic performance, exercise intolerance, and increased fatigue.³⁰ Conversely, improvement in symptoms, exercise tolerance, and cognition have been shown with repletion of iron stores in such patients.³¹

At the time of the 2013 guidelines, only a single large trial of intravenous iron in HFrEF and iron deficiency had been carried out (see below), and although the results were promising, it was felt that the evidence base on which to make recommendations was inadequate. Thus, recommendations were deferred until more data could be obtained.

Of note, in all the trials discussed below, iron deficiency was diagnosed in the setting of heart failure as ferritin less than 100 mg/mL (absolute iron deficiency) or as ferritin 100 to 300 mg/mL with transferrin saturation less than 20% (relative deficiency).³²

The CONFIRM-HF trial

As in the Ferinject Assessment in Patients With Iron Deficiency and Chronic Heart Failure (FAIR-HF) trial,³³ the subsequent Ferric Carboxymaltose Evaluation on Performance in Patients With Iron Deficiency in Combination With Chronic Heart Failure (CONFIRM-HF) trial³⁴ involved the intravenous infusion of iron (ferric carboxymaltose) in outpatients with symptomatic HFrEF and iron deficiency. It showed that benefits remained evident with a more objective primary end point (change in 6-minute walk test distance at 24 weeks), and that such benefits were sustained, as seen in numerous secondary end points related to functional capacity at 52 weeks. Benefits in CONFIRM-HF were evident independently from anemia, specifically whether hemoglobin was under or over 12 g/dL.

Although these results were promising, it remained unclear whether such improvements could be obtained with a much easier to administer, more readily available, and less expensive oral iron formulation.

The IRONOUT-HF trial

The Iron Repletion Effects on Oxygen Uptake in Heart Failure (IRONOUT-HF) trial³⁵ investigated whether oral, rather than intravenous, iron supplementation could improve peak exercise capacity in patients with HFrEF and iron deficiency. This double-blind, placebo-controlled trial randomized 225 patients with NYHA class II to IV HFrEF and iron deficiency to treatment with oral iron polysaccharide (150 mg twice daily) or placebo for 16 weeks.

Contrary to the supportive findings above, no significant change was seen in the primary end point of change in peak oxygen uptake or in any of the secondary end points (change in 6-minute walk, quality of life). Also, despite a 15-fold increase in the amount of iron administered in oral form compared with intravenously, little change was evident in the indices of iron stores over the course of the study, with only a 3% increase in transferrin saturation and an 11 ng/mL increase in ferritin. The intravenous trials resulted in a 4-fold greater increase in transferrin saturation and a 20-fold greater increase in ferritin.³⁶

What keeps heart failure patients from absorbing oral iron? It is unclear why oral iron administration in HFrEF, such as in IRONOUT-HF, seems to be so ineffective, but hepcidin—a protein hormone made by the liver that shuts down intestinal iron absorption and iron release from macrophages—may play a central role.³⁷ When iron stores are adequate, hepcidin is upregulated to prevent iron overload. However, hepcidin is also increased in inflammatory states, and chronic heart failure is often associated with inflammation.

With this in mind, the IRONOUT-HF investigators measured baseline hepcidin levels at the beginning and at the end of the 16 weeks and found that high baseline hepcidin levels predicted poorer response to oral iron. Other inflammatory mediators, such as interleukin 6, may also play a role.^38,39 Unlike oral iron formulations such as iron polysaccharide, intravenous iron (ferric carboxymaltose) bypasses these regulatory mechanisms, which may partly explain its much more significant effect on the indices of iron stores and outcomes.

 

 

New or modified recommendations on iron

The 2017 update¹ makes recommendations regarding iron deficiency and anemia in heart failure for the first time.

A class IIb recommendation states that it might be reasonable to treat NYHA class II and III heart failure patients with iron deficiency with intravenous iron to improve functional status and quality of life. A strong recommendation has been deferred until more is known about morbidity and mortality effects from adequately powered trials, some of which are under way and explored further below.

The 2017 update also withholds any recommendations regarding oral iron supplementation in heart failure, citing an uncertain evidence base. Certainly, the subsequent IRONOUT-HF trial does not lend enthusiasm for this approach.

Lastly, given the lack of benefit coupled with the increased risk of thromboembolic events evident in a trial of darbepoetin alfa vs placebo in non-iron deficiency-related anemia in HFrEF,^40,41 the 2017 update provides a class III (no benefit) recommendation against using erythropoietin-stimulating agents in heart failure and anemia.

HYPERTENSION IN HEART FAILURE

The 2013 guidelines for the management of heart failure simply provided a class I recommendation to control hypertension and lipid disorders in accordance with contemporary guidelines to lower the risk of heart failure.¹

SPRINT

The Systolic Blood Pressure Intervention Trial (SPRINT)⁴² sought to determine whether a lower systolic blood pressure target (120 vs 140 mm Hg) would reduce clinical events in patients at high risk for cardiovascular events but without diabetes mellitus. Patients at high risk were defined as over age 75, or with known vascular disease, chronic kidney disease, or a Framingham Risk Score higher than 15%. This multicenter, open-label controlled trial randomized 9,361 patients to intensive treatment (goal systolic blood pressure < 120 mm Hg) or standard treatment (goal systolic blood pressure < 140 mm Hg).

SPRINT was stopped early at a median follow-up of 3.26 years when a 25% relative risk reduction in the primary composite outcome of myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes became evident in the intensive-treatment group (1.65% vs 2.19% per year, HR 0.75, P < .0001).

All-cause mortality was also lower in the intensive-treatment group (HR 0.73, P = .003), while the incidence of serious adverse events (hypotension, syncope, electrolyte abnormalities, acute kidney injury, and noninjurious falls) was only slightly higher (38.3% vs 37.1%, P = .25). Most pertinent, a significant 38% relative risk reduction in heart failure and a 43% relative risk reduction in cardiovascular events were also evident.

Of note, blood pressure measurements were taken as the average of 3 measurements obtained by an automated cuff taken after the patient had been sitting quietly alone in a room for 5 minutes.

New or modified recommendations on hypertension in heart failure

Given the impressive 25% relative risk reduction in myocardial infarction, other acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes in SPRINT,⁴² the 2017 update¹ incorporated the intensive targets of SPRINT into its recommendations. However, to compensate for what are expected to be higher blood pressures obtained in real-world clinical practice as opposed to the near-perfect conditions used in SPRINT, a slightly higher blood pressure goal of less than 130/80 mm Hg was set.

Specific blood pressure guidelines have not been given for stage A heart failure in the past. However, as for other new approaches to prevent heart failure in this update and given the 38% relative risk reduction in heart failure seen in SPRINT, a class I recommendation is given to target a blood pressure goal of less than 130/80 mm Hg in stage A heart failure with hypertension (Table 4).

Although not specifically included in SPRINT, given the lack of trial data on specific blood pressure targets in HFrEF and the decreased cardiovascular events noted above, a class I (level of evidence C, expert opinion) recommendation to target a goal systolic blood pressure less than 130 mm Hg in stage C HFrEF with hypertension is also given. Standard guideline-directed medications in the treatment of HFrEF are to be used (Table 4).

Similarly, a new class I (level of evidence C, expert opinion) recommendation is given for hypertension in HFpEF to target a systolic blood pressure of less than 130 mm Hg, with special mention to first manage any element of volume overload with diuretics. Other than avoiding nitrates (unless used for angina) and phosphodiesterase inhibitors, it is noted that few data exist to guide the choice of antihypertensive further, although perhaps renin-angiotensin-aldosterone system inhibition, especially aldosterone antagonists, may be considered. These recommendations are fully in line with the 2017 ACC/AHA high blood pressure clinical practice guidelines,⁴³ ie, that renin-angiotensin-aldosterone system inhibition with an angiotensin-converting enzyme (ACE) inhibitor or ARB and especially mineralocorticoid receptor antagonists would be the preferred choice (Table 4).

SLEEP-DISORDERED BREATHING IN HEART FAILURE

Sleep-disordered breathing, either obstructive sleep apnea (OSA) or central sleep apnea, is quite commonly associated with symptomatic HFrEF.⁴⁴ Whereas OSA is found in roughly 18% and central sleep apnea in 1% of the general population, sleep-disordered breathing is found in nearly 60% of patients with HFrEF, with some studies showing a nearly equal proportion of OSA and central sleep apnea.⁴⁵ A similar prevalence is seen in HFpEF, although with a much higher proportion of OSA.⁴⁶ Central sleep apnea tends to be a marker of more severe heart failure, as it is strongly associated with severe cardiac systolic dysfunction and worse functional capacity.⁴⁷

Not surprisingly, the underlying mechanism of central sleep apnea is quite different from that of OSA. Whereas OSA predominantly occurs because of repeated obstruction of the pharynx due to nocturnal pharyngeal muscle relaxation, no such airway patency issues or strained breathing patterns exist in central sleep apnea. Central sleep apnea, which can manifest as Cheyne-Stokes respirations, is thought to occur due to an abnormal ventilatory control system with complex pathophysiology such as altered sensitivity of central chemoreceptors to carbon dioxide, interplay of pulmonary congestion, subsequent hyperventilation, and prolonged circulation times due to reduced cardiac output.⁴⁸

What the two types of sleep-disordered breathing have in common is an association with negative health outcomes. Both appear to induce inflammation and sympathetic nervous system activity via oxidative stress from intermittent nocturnal hypoxemia and hypercapnea.⁴⁹ OSA was already known to be associated with significant morbidity and mortality rates in the general population,⁵⁰ and central sleep apnea had been identified as an independent predictor of mortality in HFrEF.⁵¹

At the time of the 2013 guidelines, only small or observational studies with limited results had been done evaluating treatment effects of continuous positive airway pressure therapy (CPAP) on OSA and central sleep apnea. Given the relative paucity of data, only a single class IIa recommendation stating that CPAP could be beneficial to increase left ventricular ejection fraction and functional status in concomitant sleep apnea and heart failure was given in 2013. However, many larger trials were under way,^52–59 some with surprising results such as a significant increase in cardiovascular and all-cause mortality (Table 5).⁵⁴

 

 

New or modified recommendations on sleep-disordered breathing

Stemming from several trials,^54,56 3 new recommendations on sleep-disordered breathing were made in the 2017 update (Table 6).

Given the common association with heart failure (60%)⁴⁵ and the marked variation in response to treatment, including potential for harm with adaptive servo-ventilation and central sleep apnea, a class IIa recommendation is made stating that it is reasonable to obtain a formal sleep study in any patient with symptomatic (NYHA class II–IV) heart failure.¹

Due to the potential for harm with adaptive servo-ventilation in patients with central sleep apnea and NYHA class II to IV HFrEF, a class III (harm) recommendation is made against its use.

Largely based on the results of the Sleep Apnea Cardiovascular Endpoints (SAVE) trial,⁵⁶ a class IIb, level of evidence B-R (moderate, based on randomized trials) recommendation is given, stating that the use of CPAP in those with OSA and known cardiovascular disease may be reasonable to improve sleep quality and reduce daytime sleepiness.

POTENTIAL APPLICATIONS IN ACUTE DECOMPENSATED HEART FAILURE

Although the 2017 update¹ is directed mostly toward managing chronic heart failure, it is worth considering how it might apply to the management of ADHF.

SHOULD WE USE BIOMARFER TARGETS TO GUIDE THERAPY IN ADHF?

The 2017 update¹ does offer direct recommendations regarding the use of biomarker levels during admissions for ADHF. Mainly, they emphasize that the admission biomarker levels provide valuable information regarding acute prognosis and risk stratification (class I recommendation), while natriuretic peptide levels just before discharge provide the same for the postdischarge timeframe (class IIa recommendation).

The update also explicitly cautions against using a natriuretic peptide level-guided treatment strategy, such as setting targets for predischarge absolute level or percent change in level of natriuretic peptides during admissions for ADHF. Although observational and retrospective studies have shown better outcomes when levels are reduced at discharge, treating for any specific inpatient target has never been tested in any large, prospective study; thus, doing so could result in unintended harm.

So what do we know?

McQuade et al systematic review

McQuade et al⁵⁷ performed a systematic review of more than 40 ADHF trials, which showed that, indeed, patients who achieved a target absolute natriuretic peptide level (BNP ≤ 250 pg/mL) or percent reduction (≥ 30%) at time of discharge had significantly improved outcomes such as reduced postdischarge all-cause mortality and rehospitalization rates. However, these were mostly prospective cohort studies that did not use any type of natriuretic peptide level-guided treatment protocol, leaving it unclear whether such a strategy could positively influence outcomes.

For this reason, both McQuade et al⁵⁷ and, in an accompanying editorial, Felker et al⁵⁸ called for properly designed, randomized controlled trials to investigate such a strategy. Felker noted that only 2 such phase II trials in ADHF have been completed,^59,60 with unconvincing results.

PRIMA II

The Multicenter, Randomized Clinical Trial to Study the Impact of In-hospital Guidance for Acute Decompensated Heart Failure Treatment by a Predefined NT-ProBNP Target on the Reduction of Readmission and Mortality Rates (PRIMA II)⁶⁰ randomized patients to natriuretic peptide level-guided treatment or standard care during admission for ADHF.

Many participants (60%) reached the predetermined target of 30% reduction in natriuretic peptide levels at the time of clinical stabilization and randomization; 405 patients were randomized. Patients in the natriuretic peptide level-guided treatment group underwent a prespecified treatment algorithm, with repeat natriuretic peptide levels measured again after the protocol.

Natriuretic peptide-guided therapy failed to show any significant benefit in any clinical outcomes, including the primary composite end point of mortality or heart failure readmissions at 180 days (36% vs 38%, HR 0.99, 95% confidence interval 0.72–1.36). Consistent with the review by McQuade et al,⁵⁷ achieving the 30% reduction in natriuretic peptide at discharge, in either arm, was associated with a better prognosis, with significantly lower mortality and readmission rates at 180 days (HR 0.39 for rehospitalization or death, 95% confidence interval 0.27–0.55).

As in the observational studies, those who achieved the target natriuretic peptide level at the time of discharge had a better prognosis than those who did not, but neither study showed an improvement in clinical outcomes using a natriuretic peptide level-targeting treatment strategy.

No larger randomized controlled trial results are available for guided therapy in ADHF. However, additional insight may be gained from a subsequent trial⁶¹ that evaluated biomarker-guided titration of guideline-directed medical therapy in outpatients with chronic HFrEF.

The GUIDE-IT trial

That trial, the Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure (GUIDE-IT)⁶¹ trial, was a large multicenter attempt to determine whether a natriuretic peptide-guided treatment strategy was more effective than standard care in the management of 894 high-risk outpatients with chronic HFrEF. Earlier, promising results had been obtained in a meta-analysis⁶² of more than 11 similar trials in 2,000 outpatients, with a decreased mortality rate (HR 0.62) seen in the biomarker-guided arm. However, the results had not been definitive due to being underpowered.⁶²

Unfortunately, the results of GUIDE-IT were disappointing, with no significant difference in either the combined primary end point of mortality or hospitalization for heart failure, or the secondary end points evident at 15 months, prompting early termination for futility.⁶¹ Among other factors, the study authors postulated that this may have partly resulted from a patient population with more severe heart failure and resultant azotemia, limiting the ability to titrate neurohormonal medications to the desired dosage.

The question of whether patients who cannot achieve such biomarker targets need more intensive therapy or whether their heart failure is too severe to respond adequately echoes the question often raised in discussions of inpatient biomarker-guided therapy.⁵⁸ Thus, only limited insight is gained, and it remains unclear whether a natriuretic peptide-guided treatment strategy can improve outpatient or inpatient outcomes. Until this is clarified, clinical judgment and optimization of guideline-directed management and therapy should remain the bedrock of treatment.

 

 

SHOULD ALDOSTERONE ANTAGONISTS BE USED IN ACUTE HFpEF?

Given the encouraging results in chronic HFpEF from post hoc analyses of TOPCAT, are there any additional recent data suggesting a role for aldosterone antagonists such as spironolactone in acute HFpEF?

The ATHENA-HF trial

The Aldosterone Targeted Neurohormonal Combined With Natriuresis Therapy in Heart Failure (ATHENA-HF) trial⁶³ compared treatment with high-dose spironolactone (100 mg) for 96 hours vs usual care in 360 patients with ADHF. The patient population included those with HFrEF and HFpEF, and usual care included low-dose spironolactone (12.5–25 mg) in roughly 15% of patients. High-dose mineralocorticoid receptor antagonists have been shown to overcome diuretic resistance, improve pulmonary vascular congestion, and partially combat the adverse neurohormonal activation seen in ADHF.

Unfortunately, the trial was completely neutral in regard to the primary end point of reduction in natriuretic peptide levels as well as to the secondary end points of 30-day mortality rate, heart failure readmission, clinical congestion scores, urine output, and change in weight. No suggestion of additional benefit was seen in subgroup analysis of patients with acute HFpEF (ejection fraction > 45%), which yielded similar results.⁶³

Given these lackluster findings, routine use of high-dose spironolactone in ADHF is not recommended.⁶⁴ However, the treatment was well tolerated, without significant adverse effects of hyperkalemia or kidney injury, leaving the door open as to whether it may have utility in selected patients with diuretic resistance.

Should ARNIs and ivabradine be started during ADHF admissions?

The first half of the focused update³ of the 2013 guidelines,² reviewed by Okwuosa et al,⁷ provided recommendations for the use of sacubitril-valsartan, an angiotensin-neprilysin inhibitor (ARNI), and ivabradine, a selective sinoatrial node If channel inhibitor, in chronic HFrEF.

Sacubitril-valsartan was given a class I recommendation for use in patients with NYHA class II or III chronic HFrEF who tolerate an ACE inhibitor or an ARB. This recommendation was given largely based on the benefits in mortality and heart failure hospitalizations seen in PARADIGM-HF (the Prospective Comparison of ARNI With ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure)⁶⁵ compared with enalapril (HR 0.80, 95% CI 0.73–0.87, P < .001).

There is currently no recommendation on initiation or use of ARNIs during admissions for ADHF, but a recent trial may lend some insight.⁶⁶

THE PIONEER-HF trial

The Comparison of Sacubitril/Valsartan vs Enalapril on Effect on NT-proBNP in Patients Stabilized From an Acute Heart Failure Episode (PIONEER-HF) trial⁶⁶ randomized patients admitted for acute HFrEF, once stabilized, to sacubitril-valsartan or enalapril. Encouragingly, the percentage change of natriuretic peptide levels from the time of inpatient initiation to 4 and 8 weeks thereafter, the primary efficacy end point, was 46.7% with sacubitril-valsartan versus 25.3% with enalapril alone (ratio of change 0.71, 95% CI 0.63–0.81, P < .001). Although not powered for such, a prespecified analysis of a composite of clinical outcomes was also favorable for sacubitril-valsartan, largely driven by a 44% decreased rate of rehospitalization. More definitive, and quite reassuring, was that no significant difference was seen in the key safety outcomes of worsening renal function, hyperkalemia, symptomatic hypotension, and angioedema. These results were also applicable to the one-third of study participants who had no former diagnosis of heart failure, the one-third identifying as African American, and the one-third who had not been taking an ACE inhibitor or ARB. These results, taken together with the notion that at study completion the patients become similar to those included in PARADIGM-HF, have led some to assert that PIONEER-HF has the potential to change clinical practice.

Ivabradine was given a class IIa recommendation for use in patients with NYHA class II or III chronic HFrEF with a resting heart rate of at least 70 bpm, in sinus rhythm, despite being on optimal medical therapy including a beta-blocker at a maximum tolerated dose.

This recommendation was largely based on SHIFT (Systolic Heart Failure Treatment With the If Inhibitor Ivabradine Trial), which randomized patients to ivabradine or placebo to evaluate the effects of isolated lowering of the heart rate on the composite primary outcome of cardiovascular death or hospitalization. A significant reduction was seen in the ivabradine arm (HR 0.82, 95% CI 0.75–0.90, P < .0001), mainly driven by decreased hospitalizations.⁶⁷

Subsequently, a small unblinded single-center study was undertaken to evaluate the efficacy and safety of initiating ivabradine during admissions for ADHF.⁶⁸

THE ETHIC-AHF trial

The Effect of Early Treatment With Ivabradine Combined With Beta-Blockers vs Beta-Blockers Alone in Patients Hospitalized With Heart Failure and Reduced Left Ventricular Ejection Fraction (ETHIC-AHF) trial⁶⁸ sought to determine the safety and effectiveness of early coadministration of ivabradine with beta-blockers in patients with acute HFrEF.

This single-center, unblinded study randomized 71 patients to ivabradine and beta-blockade or beta-blockade alone upon clinical stabilization (24–48 hours) after admission for acute decompensated HFrEF.

The primary end point was heart rate at 28 days, with the ivabradine group showing a statistically significant decrease (64 vs 70 bpm, P = .01), which persisted at 4 months. There was no significant difference in the secondary end points of adverse drug effects or the composite of clinical event outcomes (all-cause mortality, admission for heart failure or cardiovascular cause), but a number of surrogate end points including left ventricular ejection fraction, BNP level, and NYHA functional class at 4 months showed mild improvement.

Although this study provided evidence that the coadministration of ivabradine and a beta-blocker is safe and was positive in regard to clinical outcomes, the significant limitations due to its size and study design (single-center, unblinded, 4-month follow-up) simply serve to support the pursuit of larger studies with more stringent design and longer follow-up in order to determine the clinical efficacy.

 

 

The PRIME-HF trial

The Predischarge Initiation of Ivabradine in the Management of Heart Failure (PRIME-HF) trial⁶⁹ is a randomized, open-label, multicenter trial comparing standard care vs the initiation of ivabradine before discharge, but after clinical stabilization, during admissions for ADHF in patients with chronic HFrEF (left ventricular ejection fraction ≤ 35%). At subsequent outpatient visits, the dosage can be modified in the ivabradine group, or ivabradine can be initiated at the provider’s discretion in the usual-care group.

PRIME-HF is attempting to determine whether initiating ivabradine before discharge will result in more patients taking ivabradine at 180 days, its primary end point, as well as in changes in secondary end points including heart rate and patient-centered outcomes. The study is active, with reporting expected in 2019.

As these trials all come to completion, it will not be long before we have further guidance regarding the inpatient initiation of these new and exciting therapeutic agents.

SHOULD INTRAVENOUS IRON BE GIVEN DURING ADHF ADMISSIONS?

Given the high prevalence of iron deficiency in symptomatic HFrEF, its independent association with mortality, improvements in quality of life and functional capacity suggested by repleting with intravenous iron (in FAIR-HF and CONFIRM-HF), the seeming inefficacy of oral iron in IRONOUT, and the logistical challenges of intravenous administration during standard clinic visits, could giving intravenous iron soon be incorporated into admissions for ADHF?

Caution has been advised for several reasons. As discussed above, larger randomized controlled trials powered to detect more definitive clinical end points such as death and the rate of hospitalization are still needed before a stronger recommendation can be made for intravenous iron in HFrEF. Also, without such data, it seems unwise to add the considerable economic burden of routinely assessing for iron deficiency and providing intravenous iron during ADHF admissions to the already staggering costs of heart failure.

Thus far, only a single meta-analysis is available, including 893 patients⁷⁰ largely from the FAIR-HF and CONFIRM-HF trials. While it does suggest benefit in both cardiovascular mortality and recurrent hospitalizations for heart failure (rate ratio 0.59, 95% CI 0.40–0.88; P = .009), more definitive guidance will be provided by the results from 4 large randomized placebo-controlled studies  currently under way or recruiting. All 4 seek to examine the effects of intravenous iron on morbidity and mortality in patients with HFrEF and iron deficiency, using a variety of end points ranging from exercise tolerance, to hospitalizations, to mortality (Table 7).^71–74

The effects seen on morbidity and mortality that become evident in these trials over the next 5 years will help determine future guidelines and whether intravenous iron is routinely administered in bridge clinics, during inpatient admissions for ADHF, or not at all in patients with HFrEF and iron deficiency.

INTERNISTS ARE KEY

Heart failure remains one of the most common, morbid, complex, and costly diseases in the United States, and its prevalence is expected only to increase.^4,5 The 2017 update1 of the 2013 guideline² for the management of heart failure provides recommendations aimed not only at management of heart failure, but also at its comorbidities and, for the first time ever, at its prevention.

Internists provide care for the majority of heart failure patients, as well as for their comorbidities, and are most often the first to come into contact with patients at high risk of developing heart failure. Thus, a thorough understanding of these guidelines and how to apply them to the management of acute decompensated heart failure is of critical importance.