Applied Evidence

A 54-year-old white woman with a history of a cerebrovascular accident (CVA) a year earlier sought care at the local emergency department for numbness and weakness in her right foot. She reported no other neurologic symptoms. She had mild weakness in her right leg and a mildly unsteady gait. Her neurologic examination was otherwise normal.

Initial testing included a complete blood count (CBC), renal profile, and thyroid-stimulating hormone measurement. All results were normal. A noncontrast computed tomography (CT) scan of the head was normal. We admitted her for further evaluation of probable acute ischemic stroke.

By the following day, the patient’s leg weakness and unsteadiness had worsened. A magnetic resonance imaging (MRI) scan of her head showed a prior left pontine infarct, but no new findings. She developed right arm weakness, and an MRI scan of her spine (FIGURE 1) showed multiple intradural lesions. A lumbar puncture showed elevated protein and oligoclonal bands. CT scans of the chest, abdomen, and pelvis were unremarkable. Two lumbar punctures for cytology and culture evaluations yielded negative results. A full-body positron-emission tomography (PET) scan showed diffuse small inguinal adenopathy bilaterally, suggestive of metastatic disease or lymphoma.

FIGURE 1
MRI of the spine

What is your presumptive diagnosis?

Diagnosis: Sarcoidosis

Findings from the full-body PET scan (FIGURE 2) prompted a biopsy of a right inguinal node, which showed a noncaseating granuloma—a hallmark finding of sarcoidosis.

Sarcoidosis is a multisystem disease of unknown cause. The exact prevalence in the general population is estimated at 10 to 20 cases per 100,000.¹ A higher incidence occurs in blacks in the United States, with a 2.4% lifetime risk compared with 0.85% of whites.² Sarcoidosis usually appears in patients ages 20 to 40 years, and although this systemic disease usually affects the lungs, 5% to 10% of patients will have nervous system involvement.^3,4

FIGURE 2
Full-body PET scan

What you’ll see

The most common presenting symptoms of systemic sarcoidosis are chronic cough, shortness of breath, and chest pain. Fatigue, weight loss, and myalgias are also frequently part of the initial presentation.

Patients with sarcoidosis can present with neurologic symptoms suggestive of many diseases (TABLE 1), and in the absence of systemic symptoms the diagnosis of neurosarcoidosis is easily confused with CVA. Most patients with neurosarcoidosis have cranial nerve involvement (50%-75%).¹ Other common presentations include seizures, meningitis, psychiatric symptoms, mass lesions, or endocrine abnormalities.

TABLE 1
Differential diagnosis of an acute neurologic event

Useful studies in the clinical evaluation

Consider a diagnosis of sarcoidosis involving the nervous system when an initial work-up for CVA is negative. In addition to asking about systemic symptoms, perform a complete neurologic exam and skin exam, search for lymphadenopathy, and conduct an ophthalmologic evaluation. After the initial evaluation, a neurology consult will likely be needed to guide further testing.

Choice of serum studies will vary depending on presenting symptoms, but they usually include tests for infection (CBC, cultures, Lyme titers, rapid plasma reagin, tuberculin skin test), rheumatologic disorders (antinuclear antibodies, erythrocyte sedimentation rate, C-reactive protein), and neoplastic diseases (lactate dehydrogenase, peripheral smear).⁵ Serum angiotensin-converting enzyme (ACE) may be useful in the diagnosis of systemic sarcoidosis, with positive results seen in approximately 75% of cases.³

Examination of cerebrospinal fluid often reveals an elevated total protein with oligoclonal bands, normal to low glucose, and possibly mild pleocytosis of monocytic or lymphocytic predominance.³ Spinal fluid ACE is neither sensitive nor specific for neurosarcoidosis, as it may be elevated in infectious or malignant processes.³

Imaging studies should include contrast-enhanced brain MRI, which may reveal multiple white matter lesions.⁶ Although the specificity of PET for neurosarcoidosis is poor—with positive results being seen also in infectious and neoplastic processes—the scan may help in identifying extraneural sites for biopsy. Histology will generally show the classic noncaseating granuloma with surrounding lymphocytes, plasma cells, and mast cells.

Treat with high-dose steroids

The mainstay of treatment, based largely on expert opinion, is high-dose steroids that are gradually tapered over weeks (TABLE 2). Other agents may be added if the condition is poorly controlled with steroids alone, or may be given if symptoms recur while tapering the steroid dose. Recurrence of sarcoidosis is common after doses of <10 to 20 mg/d. Prophylactic measures to counteract the adverse effects of long-term steroid use include weight-bearing exercise programs; administration of calcium, vitamin D, and bisphosphonates; and resorting to a stress-dose steroid regimen in times of illness.

The prognosis with sarcoidosis can vary widely. Case studies show that two-thirds of patients may have a nonrecurring illness. Among the remaining one-third, the disease course may be relapsing-remitting or progressive. When confronted with an acute neurologic event, consider recurrent sarcoidosis and coordinate care between specialists. Also, take steps to prevent complications related to prolonged steroid use.

TABLE 2
Treatment of neurosarcoidosis³

Improvement for our patient
Based on cerebrospinal fluid study results, a positive peripheral lymph node biopsy, and the exclusion of other diagnoses, we regarded the diagnosis of sarcoidosis as highly probable and initiated high-dose intravenous corticosteroids. Over several weeks, our patient gradually improved with physical therapy and was walking unassisted at the time of discharge from a hospital-based rehabilitation unit. Repeat MRI scans showed a reduction in the size of her intradural lesions, and we slowly tapered her steroids.

CORRESPONDENCE
Hillary R. Mount, MD, 2123 Auburn Avenue,#340, Cincinnati, OH 45219; hillary.mount@thechristhospital.com

A 54-year-old white woman with a history of a cerebrovascular accident (CVA) a year earlier sought care at the local emergency department for numbness and weakness in her right foot. She reported no other neurologic symptoms. She had mild weakness in her right leg and a mildly unsteady gait. Her neurologic examination was otherwise normal.

Initial testing included a complete blood count (CBC), renal profile, and thyroid-stimulating hormone measurement. All results were normal. A noncontrast computed tomography (CT) scan of the head was normal. We admitted her for further evaluation of probable acute ischemic stroke.

By the following day, the patient’s leg weakness and unsteadiness had worsened. A magnetic resonance imaging (MRI) scan of her head showed a prior left pontine infarct, but no new findings. She developed right arm weakness, and an MRI scan of her spine (FIGURE 1) showed multiple intradural lesions. A lumbar puncture showed elevated protein and oligoclonal bands. CT scans of the chest, abdomen, and pelvis were unremarkable. Two lumbar punctures for cytology and culture evaluations yielded negative results. A full-body positron-emission tomography (PET) scan showed diffuse small inguinal adenopathy bilaterally, suggestive of metastatic disease or lymphoma.

FIGURE 1
MRI of the spine

What is your presumptive diagnosis?

Diagnosis: Sarcoidosis

Findings from the full-body PET scan (FIGURE 2) prompted a biopsy of a right inguinal node, which showed a noncaseating granuloma—a hallmark finding of sarcoidosis.

Sarcoidosis is a multisystem disease of unknown cause. The exact prevalence in the general population is estimated at 10 to 20 cases per 100,000.¹ A higher incidence occurs in blacks in the United States, with a 2.4% lifetime risk compared with 0.85% of whites.² Sarcoidosis usually appears in patients ages 20 to 40 years, and although this systemic disease usually affects the lungs, 5% to 10% of patients will have nervous system involvement.^3,4

FIGURE 2
Full-body PET scan

What you’ll see

The most common presenting symptoms of systemic sarcoidosis are chronic cough, shortness of breath, and chest pain. Fatigue, weight loss, and myalgias are also frequently part of the initial presentation.

Patients with sarcoidosis can present with neurologic symptoms suggestive of many diseases (TABLE 1), and in the absence of systemic symptoms the diagnosis of neurosarcoidosis is easily confused with CVA. Most patients with neurosarcoidosis have cranial nerve involvement (50%-75%).¹ Other common presentations include seizures, meningitis, psychiatric symptoms, mass lesions, or endocrine abnormalities.

TABLE 1
Differential diagnosis of an acute neurologic event

Useful studies in the clinical evaluation

Consider a diagnosis of sarcoidosis involving the nervous system when an initial work-up for CVA is negative. In addition to asking about systemic symptoms, perform a complete neurologic exam and skin exam, search for lymphadenopathy, and conduct an ophthalmologic evaluation. After the initial evaluation, a neurology consult will likely be needed to guide further testing.

Choice of serum studies will vary depending on presenting symptoms, but they usually include tests for infection (CBC, cultures, Lyme titers, rapid plasma reagin, tuberculin skin test), rheumatologic disorders (antinuclear antibodies, erythrocyte sedimentation rate, C-reactive protein), and neoplastic diseases (lactate dehydrogenase, peripheral smear).⁵ Serum angiotensin-converting enzyme (ACE) may be useful in the diagnosis of systemic sarcoidosis, with positive results seen in approximately 75% of cases.³

Examination of cerebrospinal fluid often reveals an elevated total protein with oligoclonal bands, normal to low glucose, and possibly mild pleocytosis of monocytic or lymphocytic predominance.³ Spinal fluid ACE is neither sensitive nor specific for neurosarcoidosis, as it may be elevated in infectious or malignant processes.³

Imaging studies should include contrast-enhanced brain MRI, which may reveal multiple white matter lesions.⁶ Although the specificity of PET for neurosarcoidosis is poor—with positive results being seen also in infectious and neoplastic processes—the scan may help in identifying extraneural sites for biopsy. Histology will generally show the classic noncaseating granuloma with surrounding lymphocytes, plasma cells, and mast cells.

Treat with high-dose steroids

The mainstay of treatment, based largely on expert opinion, is high-dose steroids that are gradually tapered over weeks (TABLE 2). Other agents may be added if the condition is poorly controlled with steroids alone, or may be given if symptoms recur while tapering the steroid dose. Recurrence of sarcoidosis is common after doses of <10 to 20 mg/d. Prophylactic measures to counteract the adverse effects of long-term steroid use include weight-bearing exercise programs; administration of calcium, vitamin D, and bisphosphonates; and resorting to a stress-dose steroid regimen in times of illness.

The prognosis with sarcoidosis can vary widely. Case studies show that two-thirds of patients may have a nonrecurring illness. Among the remaining one-third, the disease course may be relapsing-remitting or progressive. When confronted with an acute neurologic event, consider recurrent sarcoidosis and coordinate care between specialists. Also, take steps to prevent complications related to prolonged steroid use.

TABLE 2
Treatment of neurosarcoidosis³

Improvement for our patient
Based on cerebrospinal fluid study results, a positive peripheral lymph node biopsy, and the exclusion of other diagnoses, we regarded the diagnosis of sarcoidosis as highly probable and initiated high-dose intravenous corticosteroids. Over several weeks, our patient gradually improved with physical therapy and was walking unassisted at the time of discharge from a hospital-based rehabilitation unit. Repeat MRI scans showed a reduction in the size of her intradural lesions, and we slowly tapered her steroids.

CORRESPONDENCE
Hillary R. Mount, MD, 2123 Auburn Avenue,#340, Cincinnati, OH 45219; hillary.mount@thechristhospital.com

A 54-year-old white woman with a history of a cerebrovascular accident (CVA) a year earlier sought care at the local emergency department for numbness and weakness in her right foot. She reported no other neurologic symptoms. She had mild weakness in her right leg and a mildly unsteady gait. Her neurologic examination was otherwise normal.

Initial testing included a complete blood count (CBC), renal profile, and thyroid-stimulating hormone measurement. All results were normal. A noncontrast computed tomography (CT) scan of the head was normal. We admitted her for further evaluation of probable acute ischemic stroke.

By the following day, the patient’s leg weakness and unsteadiness had worsened. A magnetic resonance imaging (MRI) scan of her head showed a prior left pontine infarct, but no new findings. She developed right arm weakness, and an MRI scan of her spine (FIGURE 1) showed multiple intradural lesions. A lumbar puncture showed elevated protein and oligoclonal bands. CT scans of the chest, abdomen, and pelvis were unremarkable. Two lumbar punctures for cytology and culture evaluations yielded negative results. A full-body positron-emission tomography (PET) scan showed diffuse small inguinal adenopathy bilaterally, suggestive of metastatic disease or lymphoma.

FIGURE 1
MRI of the spine

What is your presumptive diagnosis?

Diagnosis: Sarcoidosis

Findings from the full-body PET scan (FIGURE 2) prompted a biopsy of a right inguinal node, which showed a noncaseating granuloma—a hallmark finding of sarcoidosis.

Sarcoidosis is a multisystem disease of unknown cause. The exact prevalence in the general population is estimated at 10 to 20 cases per 100,000.¹ A higher incidence occurs in blacks in the United States, with a 2.4% lifetime risk compared with 0.85% of whites.² Sarcoidosis usually appears in patients ages 20 to 40 years, and although this systemic disease usually affects the lungs, 5% to 10% of patients will have nervous system involvement.^3,4

FIGURE 2
Full-body PET scan

What you’ll see

The most common presenting symptoms of systemic sarcoidosis are chronic cough, shortness of breath, and chest pain. Fatigue, weight loss, and myalgias are also frequently part of the initial presentation.

Patients with sarcoidosis can present with neurologic symptoms suggestive of many diseases (TABLE 1), and in the absence of systemic symptoms the diagnosis of neurosarcoidosis is easily confused with CVA. Most patients with neurosarcoidosis have cranial nerve involvement (50%-75%).¹ Other common presentations include seizures, meningitis, psychiatric symptoms, mass lesions, or endocrine abnormalities.

TABLE 1
Differential diagnosis of an acute neurologic event

Useful studies in the clinical evaluation

Consider a diagnosis of sarcoidosis involving the nervous system when an initial work-up for CVA is negative. In addition to asking about systemic symptoms, perform a complete neurologic exam and skin exam, search for lymphadenopathy, and conduct an ophthalmologic evaluation. After the initial evaluation, a neurology consult will likely be needed to guide further testing.

Choice of serum studies will vary depending on presenting symptoms, but they usually include tests for infection (CBC, cultures, Lyme titers, rapid plasma reagin, tuberculin skin test), rheumatologic disorders (antinuclear antibodies, erythrocyte sedimentation rate, C-reactive protein), and neoplastic diseases (lactate dehydrogenase, peripheral smear).⁵ Serum angiotensin-converting enzyme (ACE) may be useful in the diagnosis of systemic sarcoidosis, with positive results seen in approximately 75% of cases.³

Examination of cerebrospinal fluid often reveals an elevated total protein with oligoclonal bands, normal to low glucose, and possibly mild pleocytosis of monocytic or lymphocytic predominance.³ Spinal fluid ACE is neither sensitive nor specific for neurosarcoidosis, as it may be elevated in infectious or malignant processes.³

Imaging studies should include contrast-enhanced brain MRI, which may reveal multiple white matter lesions.⁶ Although the specificity of PET for neurosarcoidosis is poor—with positive results being seen also in infectious and neoplastic processes—the scan may help in identifying extraneural sites for biopsy. Histology will generally show the classic noncaseating granuloma with surrounding lymphocytes, plasma cells, and mast cells.

Treat with high-dose steroids

The mainstay of treatment, based largely on expert opinion, is high-dose steroids that are gradually tapered over weeks (TABLE 2). Other agents may be added if the condition is poorly controlled with steroids alone, or may be given if symptoms recur while tapering the steroid dose. Recurrence of sarcoidosis is common after doses of <10 to 20 mg/d. Prophylactic measures to counteract the adverse effects of long-term steroid use include weight-bearing exercise programs; administration of calcium, vitamin D, and bisphosphonates; and resorting to a stress-dose steroid regimen in times of illness.

The prognosis with sarcoidosis can vary widely. Case studies show that two-thirds of patients may have a nonrecurring illness. Among the remaining one-third, the disease course may be relapsing-remitting or progressive. When confronted with an acute neurologic event, consider recurrent sarcoidosis and coordinate care between specialists. Also, take steps to prevent complications related to prolonged steroid use.

TABLE 2
Treatment of neurosarcoidosis³

Improvement for our patient
Based on cerebrospinal fluid study results, a positive peripheral lymph node biopsy, and the exclusion of other diagnoses, we regarded the diagnosis of sarcoidosis as highly probable and initiated high-dose intravenous corticosteroids. Over several weeks, our patient gradually improved with physical therapy and was walking unassisted at the time of discharge from a hospital-based rehabilitation unit. Repeat MRI scans showed a reduction in the size of her intradural lesions, and we slowly tapered her steroids.

CORRESPONDENCE
Hillary R. Mount, MD, 2123 Auburn Avenue,#340, Cincinnati, OH 45219; hillary.mount@thechristhospital.com

CASE Joe G, a 44-year-old man who has been your patient for years, comes to your office 48 hours after having a seizure. He has no history of seizures, had no warning signs or symptoms, and felt fine all day, but simply collapsed when the seizure occurred. He was transported to the emergency department (ED), and found to be postictal, with no further seizure activity. The ED work-up included a hemogram, comprehensive metabolic panel, and computed tomography brain scan, all of which were normal. An hour later, Joe had a normal neurological exam, then underwent electroencephalography (EEG) and magnetic resonance imaging (MRI) and was discharged home without medication.

How would you treat this patient?

About 10% of Americans will experience a seizure at some point in their lives,^1,2 and more than 3 million have epilepsy.³ The incidence ranges from 1% among 20-year-olds to more than 3% by the age of 75.^1,2

To adequately care for such patients—whether they have had multiple seizures or only one—you need to know whether they’re at risk for recurrences, when (or if) to prescribe an AED, and which agents provide optimal seizure control with the fewest adverse effects. You also need to know when a referral to an epilepsy specialist is indicated, when or whether it’s safe for patients to stop taking antiseizure medication, and how to address lifestyle issues that patients with epilepsy often need help with.

This review addresses these and other questions.

Is it epilepsy? How to respond to a single seizure

A seizure—a transient occurrence of signs or symptoms due to abnormal excessive or synchronous neural activity in the brain—can be either focal (partial) or generalized. In addition, seizures can be broadly divided into 2 categories, based on etiology:

Provoked seizures are caused by an acute structural, toxic, or metabolic insult to the brain, and, presumably, would not have occurred if the underlying medical condition did not exist. Treating the cause—eg, alcohol withdrawal, hyponatremia, or hypoglycemia—should prevent a recurrence.

Unprovoked seizures have no apparent underlying cause. Epilepsy is defined as a chronic condition characterized by ≥2 unprovoked seizures at least 24 hours apart, and epilepsy syndromes are classified as localization-related or generalized (TABLE 1).^1,4,5

Generally, epileptologists do not recommend symptomatic treatment of a first unprovoked seizure⁶—a consensus based on several randomized controlled trials that found immediate treatment with an AED reduced the risk of a subsequent seizure in the short term, but did not affect long-term outcomes or the development of epilepsy.⁷

Treatment should begin after a single seizure, however, if the seizure was prolonged or there is an increased risk of recurrence.⁶ Factors that increase this risk include an abnormal EEG, particularly if the abnormality is epileptiform; the presence of a brain lesion; a localized (focal) seizure; and an abnormal neurologic exam.⁸ A history of status epilepticus—a single, unremitting seizure lasting ≥5 to 10 minutes or frequent seizures without a return to neurologic baseline in between—or complex febrile seizures, and a family history of epilepsy are risk factors for recurrence, as well.⁷

When the patient is a child. Prescribing an AED for a child after a first unprovoked seizure is not indicated to prevent the development of epilepsy, but may be considered, as for adults, in circumstances where the benefit of reducing the risk of a second seizure outweighs the risk of pharmacologic and psychosocial adverse effects.⁹

CASE Joe’s ED records show that his MRI was normal, but his EEG revealed an epileptogenic focus on the right temporal region—a finding that indicates that he has an elevated risk of recurrence and is a candidate for an AED. Before selecting a particular agent, you review his chart.

Joe is taking a thiazide diuretic and a calcium channel blocker for hypertension. He was a heavy drinker until he had an episode of pancreatitis 10 years ago, and has been abstinent ever since. About 5 years ago, he suffered from depression and was treated with sertraline, but the depression resolved and the drug was discontinued 3 years ago. The patient’s mother and brother have type 2 diabetes and his father had a myocardial infarction before the age of 60. Joe was laid off from his sales job 18 months ago and is actively seeking employment. At this point, you consider a broad-spectrum AED that would not interact with his current medications or adversely affect his medical conditions, and would be relatively inexpensive.

TABLE 1
Identifying seizures and types of epilepsy:^1,4,5 International League Against Epilepsy classification

What to consider in a first-line drug

The number of AEDs on the market has increased sharply in the past few years, giving physicians many medications to choose from. Selecting the optimal drug is particularly important for the initial treatment, as many patients remain on the first AED for years. Second-generation AEDs have been found to be as effective as, and better tolerated than, first-generation antiseizure drugs. But all AEDs carry a warning of a potential increase in suicide risk and the need to monitor patients for behavior changes.¹⁰

Before selecting an AED for a particular patient, consider the following questions:

What type of seizure? AEDs are generally classified by spectrum of activity into “narrow-spectrum” and “broad-spectrum.” Narrow-spectrum drugs are more effective for controlling partial seizures, but have the potential to exacerbate generalized seizures; broad-spectrum AEDs can be used for both. (TABLE 2^11-18 lists indications for first- and second-generation AEDs based on type of epilepsy.) If there’s no definitive diagnosis of the type of epilepsy a patient has, use a broad-spectrum drug.

What other drugs is the patient taking? If the AED will be added to the patient’s current medication regimen, look closely at potential pharmacodynamic drug-drug interactions, and consider whether a dosage adjustment is needed. Determine, too, whether the patient has any comorbidities that could affect his or her response to the AED.

Side effects, such as weight gain or loss, urolithiasis, and hepatic enzyme induction, are key considerations. (TABLE W1,^19-24 which details dose, side effects, and costs of first- and second-generation AEDs, can be found at jfponline.com.)

Is the patient elderly? AED clearance is reduced in the elderly, so lower doses are needed. Reduction in serum albumin increases the free or active component of highly protein-bound drugs, increasing the likelihood of adverse effects.

Is the patient female? Some AEDs may have effects on women’s hormonal function, sexuality, bone health, and pregnancy.²⁵ Hepatic enzyme inducers increase the clearance of oral contraceptives, reducing their efficacy. Vitamin D and calcium metabolism can also be affected, which can lead to osteomalacia. Valproate treatment in women is associated with higher levels of insulin, testosterone, and triglycerides.²⁶ Cytochrome P-450-activating AEDs in general are associated with higher testosterone levels and reduced libido.²⁷

Potential pregnancy is another consideration. Women with epilepsy are able to bear healthy children. What’s more, patients whose seizures are controlled with AEDs should be maintained on medication throughout pregnancy, as the risk of fetal harm from seizures generally outweighs the teratogenicity of the drug.²⁸

Although large studies are limited, a study of 1532 infants exposed to AEDs in the first trimester did not find an increase in major birth defects compared with infants without such exposure.²⁹ More recently, a large observational cohort study conducted in more than 40 countries found that the possibility of harm to a developing fetus is not only drug-specific but also dose-related.³⁰ (To learn more, see “Pregnancy and epilepsy—when you’re managing both,” in the December 2010 issue of The Journal of Family Practice.)

Is cost a factor? Finally, consider the cost of the AED you would like to prescribe, and whether the patient has a prescription drug plan or the means to pay for his prescription.

CASE After a discussion of potential side effects, including the potential for suicidal ideation associated with AEDs, you prescribe carbamazepine for Joe as seizure prophylaxis, because it is the least expensive of the broad-spectrum AEDs and is unlikely to exacerbate his previous pancreatitis or interact with his current medications.

TABLE 2
Choosing an AED: What to consider^11-18

TABLE W1
A closer look at antiepileptic drugs^19-24

When to add a second AED

Monotherapy is the preferred method of epilepsy treatment, and controls seizures for 70% to 90% of patients.^31,32 If seizures continue and potential adverse effects prevent you from increasing the dosage, switching to a different AED, then tapering off the first agent, is recommended.^33,34

If the new AED fails to provide adequate seizure control, consider combination therapy. An additional 10% to 15% of patients with epilepsy achieve control with dual therapy.^33,34

Many second-generation agents are approved for adjunctive therapy. However, the use of 2 AEDs increases the risk of toxicities and drug interactions, and requires complex dosage adjustments, which should be done slowly and cautiously. Combination therapy also increases costs and may cause a decrease in compliance.^33,34

Noncompliance is the single most common reason for treatment failure in patients with epilepsy, occurring at an estimated rate of up to 60%.^35,36 The complexity of the drug regimen is the major cause, regardless of patient age, sex, psychomotor development, seizure type, or seizure frequency.^35,36

Because of the lack of good clinical trials of combination antiepilepsy therapy, no evidence is available to indicate which AEDs are safe and effective when taken together. There is, however, evidence that certain combinations should be avoided due to the risk of increased adverse effects. These include phenobarbital/valproate, phenytoin/carbamazepine, and carbamazepine/lamotrigine.²⁵

Managing the patient who is seizure-free

After a patient has been seizure-free for 2 to 5 years, consider a reduction in, or a discontinuation of, his or her AED. The relapse rate varies from 10% to 70%, with meta-analyses showing a rate of 25% in the first year and 29% in the second year.^19,37 The American Academy of Neurology (AAN) has published an evidence-based guideline for discontinuing AEDs in seizure-free patients, available at www.aan.com/professionals/practice/pdfs/gl0007.pdf.

Withdrawal should be gradual and, for patients on combination therapy, carried out one drug at a time to prevent a recurrence of seizures or status epilepticus. The AAN recommends a 2- to- 3-month withdrawal period for AEDs (and longer for benzodiazepines), although relapse rates have been found to be lower when the medication is withdrawn more slowly, over about 6 months.^19,34 If seizures recur after withdrawal, restart the AEDs at previous dosages.^19,34,38

Should the patient drive?

For patients with epilepsy, loss of independence related to driving restrictions is a major source of stress. A 10-year follow-up study of Danish patients with epilepsy found a 7-fold increase in motor vehicle accidents (MVAs) in patients with seizure disorders.³⁹ Other studies have shown that the seizure-free interval is the best predictor of involvement in an MVA.⁴⁰

The risk of driving accidents decreases as the seizure-free interval increases. Unfortunately, however, a decline in patient compliance is also associated with longer seizure-free intervals—creating the potential for recurrence and driving risk. Because of this discrepancy, a consensus statement from the AAN, American Epilepsy Society, and Epilepsy Foundation of America recommends a minimum 3-month seizure-free interval before patients are allowed to drive.⁴¹

Use clinical judgment in deciding whether to extend the seizure-free period. State laws vary widely regarding the need to report patients with seizure disorders, limitations on professional drivers, and seizure-free intervals required, so it is important to be familiar with the laws in your state. The Epilepsy Foundation has a helpful online resource with a database detailing individual state statutes (http://www.epilepsyfoundation.org/living/wellness/transportation/driverlicensing.cfm).

The danger of uncontrolled seizures

Overall, AEDs effectively control 70% of 80% of cases; the remaining 20% to 30% are considered medically refractory.³⁸ What’s more, after 2 AED failures, a patient’s chances of achieving full seizure control with additional drugs are no better than 10% to 20%.⁴² And, as more drugs are tried, the likelihood of full control declines even further.⁴³

Patients with uncontrolled seizures have a cumulative risk of sudden unexpected death in epilepsy (SUDEP) of 0.5% per year.⁴⁴ Cognitive decline is associated with uncontrolled epilepsy, as well. In children, frequent seizures may significantly alter neuronal networks, affecting cognitive and motor development.

Is your patient a candidate for surgery?

Patients with disabling complex partial seizures that remain uncontrolled after 2 or more AED trials (either as monotherapy or in combination) should be referred to an epilepsy specialty center for evaluation for surgery.⁴⁵ This should be considered as early as possible to afford the patient the best chance of achieving seizure control.

Successful epilepsy surgery—in which the portion of the brain causing the misfiring that causes the seizures is removed—often results in a better quality of life; it is also cost effective.⁴⁶ Not everyone with refractory epilepsy is a candidate for surgery, of course. Among those who are, however, 50% to 70% of patients can expect to have improved seizure control.⁴⁷

Status epilepticus is a medical emergency

A patient who develops status epilepticus is at high risk and requires immediate, and simultaneous, evaluation and treatment. Status epilepticus carries nearly a 20% mortality from the first episode,⁴⁸ and the 10-year mortality rate after an episode of status epilepticus is as high as 40%.⁴⁹

Although most of the deaths associated with status epilepticus are due to the underlying pathology, early treatment can prevent or ameliorate complications from rhabdomyolysis and irreversible anoxic neuronal damage.⁵⁰

A benzodiazepine (typically, a 10-mg IV bolus of diazepam) is the initial treatment for status epilepticus, followed by or concurrent with fosphenytoin (15-18 mg/kg). If status epilepticus remains refractory to first-line drugs (lasting >30 minutes), intubation and transfer to an intensive care setting may be required, and a neurological consult should be obtained.

Pharmacologic treatment of status epilepticus falls into 3 main classes: benzodiazepines, standard AEDs, and general anesthetics such as propofol. Benzodiazepines act very rapidly to control most prolonged seizures, and are the first-line treatment choice. Diazepam has long been the mainstay of treatment, and is usually readily available. However, in both a large systematic review and a head-to-head trial, lorazepam was found to be superior to diazepam in ending seizure activity and maintaining seizure control without the use of other medications^51,52—and is now the drug of choice for initial treatment of status epilepticus.

CASE You continue to see Joe every 3 to 4 months to monitor his basic blood work and mood. A year after his seizure, he remains seizure-free and is tolerating the AED without any adverse effects.

CORRESPONDENCE
William J. Geiger, MD, FAAFP, Medical College of Wisconsin, Columbia St. Mary’s Family Medicine Residency, 1121 East North Avenue, Milwaukee, WI 53212; bgeiger@mcw.edu

CASE Joe G, a 44-year-old man who has been your patient for years, comes to your office 48 hours after having a seizure. He has no history of seizures, had no warning signs or symptoms, and felt fine all day, but simply collapsed when the seizure occurred. He was transported to the emergency department (ED), and found to be postictal, with no further seizure activity. The ED work-up included a hemogram, comprehensive metabolic panel, and computed tomography brain scan, all of which were normal. An hour later, Joe had a normal neurological exam, then underwent electroencephalography (EEG) and magnetic resonance imaging (MRI) and was discharged home without medication.

How would you treat this patient?

About 10% of Americans will experience a seizure at some point in their lives,^1,2 and more than 3 million have epilepsy.³ The incidence ranges from 1% among 20-year-olds to more than 3% by the age of 75.^1,2

To adequately care for such patients—whether they have had multiple seizures or only one—you need to know whether they’re at risk for recurrences, when (or if) to prescribe an AED, and which agents provide optimal seizure control with the fewest adverse effects. You also need to know when a referral to an epilepsy specialist is indicated, when or whether it’s safe for patients to stop taking antiseizure medication, and how to address lifestyle issues that patients with epilepsy often need help with.

This review addresses these and other questions.

Is it epilepsy? How to respond to a single seizure

A seizure—a transient occurrence of signs or symptoms due to abnormal excessive or synchronous neural activity in the brain—can be either focal (partial) or generalized. In addition, seizures can be broadly divided into 2 categories, based on etiology:

Provoked seizures are caused by an acute structural, toxic, or metabolic insult to the brain, and, presumably, would not have occurred if the underlying medical condition did not exist. Treating the cause—eg, alcohol withdrawal, hyponatremia, or hypoglycemia—should prevent a recurrence.

Unprovoked seizures have no apparent underlying cause. Epilepsy is defined as a chronic condition characterized by ≥2 unprovoked seizures at least 24 hours apart, and epilepsy syndromes are classified as localization-related or generalized (TABLE 1).^1,4,5

Generally, epileptologists do not recommend symptomatic treatment of a first unprovoked seizure⁶—a consensus based on several randomized controlled trials that found immediate treatment with an AED reduced the risk of a subsequent seizure in the short term, but did not affect long-term outcomes or the development of epilepsy.⁷

Treatment should begin after a single seizure, however, if the seizure was prolonged or there is an increased risk of recurrence.⁶ Factors that increase this risk include an abnormal EEG, particularly if the abnormality is epileptiform; the presence of a brain lesion; a localized (focal) seizure; and an abnormal neurologic exam.⁸ A history of status epilepticus—a single, unremitting seizure lasting ≥5 to 10 minutes or frequent seizures without a return to neurologic baseline in between—or complex febrile seizures, and a family history of epilepsy are risk factors for recurrence, as well.⁷

When the patient is a child. Prescribing an AED for a child after a first unprovoked seizure is not indicated to prevent the development of epilepsy, but may be considered, as for adults, in circumstances where the benefit of reducing the risk of a second seizure outweighs the risk of pharmacologic and psychosocial adverse effects.⁹

CASE Joe’s ED records show that his MRI was normal, but his EEG revealed an epileptogenic focus on the right temporal region—a finding that indicates that he has an elevated risk of recurrence and is a candidate for an AED. Before selecting a particular agent, you review his chart.

Joe is taking a thiazide diuretic and a calcium channel blocker for hypertension. He was a heavy drinker until he had an episode of pancreatitis 10 years ago, and has been abstinent ever since. About 5 years ago, he suffered from depression and was treated with sertraline, but the depression resolved and the drug was discontinued 3 years ago. The patient’s mother and brother have type 2 diabetes and his father had a myocardial infarction before the age of 60. Joe was laid off from his sales job 18 months ago and is actively seeking employment. At this point, you consider a broad-spectrum AED that would not interact with his current medications or adversely affect his medical conditions, and would be relatively inexpensive.

TABLE 1
Identifying seizures and types of epilepsy:^1,4,5 International League Against Epilepsy classification

What to consider in a first-line drug

The number of AEDs on the market has increased sharply in the past few years, giving physicians many medications to choose from. Selecting the optimal drug is particularly important for the initial treatment, as many patients remain on the first AED for years. Second-generation AEDs have been found to be as effective as, and better tolerated than, first-generation antiseizure drugs. But all AEDs carry a warning of a potential increase in suicide risk and the need to monitor patients for behavior changes.¹⁰

Before selecting an AED for a particular patient, consider the following questions:

What type of seizure? AEDs are generally classified by spectrum of activity into “narrow-spectrum” and “broad-spectrum.” Narrow-spectrum drugs are more effective for controlling partial seizures, but have the potential to exacerbate generalized seizures; broad-spectrum AEDs can be used for both. (TABLE 2^11-18 lists indications for first- and second-generation AEDs based on type of epilepsy.) If there’s no definitive diagnosis of the type of epilepsy a patient has, use a broad-spectrum drug.

What other drugs is the patient taking? If the AED will be added to the patient’s current medication regimen, look closely at potential pharmacodynamic drug-drug interactions, and consider whether a dosage adjustment is needed. Determine, too, whether the patient has any comorbidities that could affect his or her response to the AED.

Side effects, such as weight gain or loss, urolithiasis, and hepatic enzyme induction, are key considerations. (TABLE W1,^19-24 which details dose, side effects, and costs of first- and second-generation AEDs, can be found at jfponline.com.)

Is the patient elderly? AED clearance is reduced in the elderly, so lower doses are needed. Reduction in serum albumin increases the free or active component of highly protein-bound drugs, increasing the likelihood of adverse effects.

Is the patient female? Some AEDs may have effects on women’s hormonal function, sexuality, bone health, and pregnancy.²⁵ Hepatic enzyme inducers increase the clearance of oral contraceptives, reducing their efficacy. Vitamin D and calcium metabolism can also be affected, which can lead to osteomalacia. Valproate treatment in women is associated with higher levels of insulin, testosterone, and triglycerides.²⁶ Cytochrome P-450-activating AEDs in general are associated with higher testosterone levels and reduced libido.²⁷

Potential pregnancy is another consideration. Women with epilepsy are able to bear healthy children. What’s more, patients whose seizures are controlled with AEDs should be maintained on medication throughout pregnancy, as the risk of fetal harm from seizures generally outweighs the teratogenicity of the drug.²⁸

Although large studies are limited, a study of 1532 infants exposed to AEDs in the first trimester did not find an increase in major birth defects compared with infants without such exposure.²⁹ More recently, a large observational cohort study conducted in more than 40 countries found that the possibility of harm to a developing fetus is not only drug-specific but also dose-related.³⁰ (To learn more, see “Pregnancy and epilepsy—when you’re managing both,” in the December 2010 issue of The Journal of Family Practice.)

Is cost a factor? Finally, consider the cost of the AED you would like to prescribe, and whether the patient has a prescription drug plan or the means to pay for his prescription.

CASE After a discussion of potential side effects, including the potential for suicidal ideation associated with AEDs, you prescribe carbamazepine for Joe as seizure prophylaxis, because it is the least expensive of the broad-spectrum AEDs and is unlikely to exacerbate his previous pancreatitis or interact with his current medications.

TABLE 2
Choosing an AED: What to consider^11-18

TABLE W1
A closer look at antiepileptic drugs^19-24

When to add a second AED

Monotherapy is the preferred method of epilepsy treatment, and controls seizures for 70% to 90% of patients.^31,32 If seizures continue and potential adverse effects prevent you from increasing the dosage, switching to a different AED, then tapering off the first agent, is recommended.^33,34

If the new AED fails to provide adequate seizure control, consider combination therapy. An additional 10% to 15% of patients with epilepsy achieve control with dual therapy.^33,34

Many second-generation agents are approved for adjunctive therapy. However, the use of 2 AEDs increases the risk of toxicities and drug interactions, and requires complex dosage adjustments, which should be done slowly and cautiously. Combination therapy also increases costs and may cause a decrease in compliance.^33,34

Noncompliance is the single most common reason for treatment failure in patients with epilepsy, occurring at an estimated rate of up to 60%.^35,36 The complexity of the drug regimen is the major cause, regardless of patient age, sex, psychomotor development, seizure type, or seizure frequency.^35,36

Because of the lack of good clinical trials of combination antiepilepsy therapy, no evidence is available to indicate which AEDs are safe and effective when taken together. There is, however, evidence that certain combinations should be avoided due to the risk of increased adverse effects. These include phenobarbital/valproate, phenytoin/carbamazepine, and carbamazepine/lamotrigine.²⁵

Managing the patient who is seizure-free

After a patient has been seizure-free for 2 to 5 years, consider a reduction in, or a discontinuation of, his or her AED. The relapse rate varies from 10% to 70%, with meta-analyses showing a rate of 25% in the first year and 29% in the second year.^19,37 The American Academy of Neurology (AAN) has published an evidence-based guideline for discontinuing AEDs in seizure-free patients, available at www.aan.com/professionals/practice/pdfs/gl0007.pdf.

Withdrawal should be gradual and, for patients on combination therapy, carried out one drug at a time to prevent a recurrence of seizures or status epilepticus. The AAN recommends a 2- to- 3-month withdrawal period for AEDs (and longer for benzodiazepines), although relapse rates have been found to be lower when the medication is withdrawn more slowly, over about 6 months.^19,34 If seizures recur after withdrawal, restart the AEDs at previous dosages.^19,34,38

Should the patient drive?

For patients with epilepsy, loss of independence related to driving restrictions is a major source of stress. A 10-year follow-up study of Danish patients with epilepsy found a 7-fold increase in motor vehicle accidents (MVAs) in patients with seizure disorders.³⁹ Other studies have shown that the seizure-free interval is the best predictor of involvement in an MVA.⁴⁰

The risk of driving accidents decreases as the seizure-free interval increases. Unfortunately, however, a decline in patient compliance is also associated with longer seizure-free intervals—creating the potential for recurrence and driving risk. Because of this discrepancy, a consensus statement from the AAN, American Epilepsy Society, and Epilepsy Foundation of America recommends a minimum 3-month seizure-free interval before patients are allowed to drive.⁴¹

Use clinical judgment in deciding whether to extend the seizure-free period. State laws vary widely regarding the need to report patients with seizure disorders, limitations on professional drivers, and seizure-free intervals required, so it is important to be familiar with the laws in your state. The Epilepsy Foundation has a helpful online resource with a database detailing individual state statutes (http://www.epilepsyfoundation.org/living/wellness/transportation/driverlicensing.cfm).

The danger of uncontrolled seizures

Overall, AEDs effectively control 70% of 80% of cases; the remaining 20% to 30% are considered medically refractory.³⁸ What’s more, after 2 AED failures, a patient’s chances of achieving full seizure control with additional drugs are no better than 10% to 20%.⁴² And, as more drugs are tried, the likelihood of full control declines even further.⁴³

Patients with uncontrolled seizures have a cumulative risk of sudden unexpected death in epilepsy (SUDEP) of 0.5% per year.⁴⁴ Cognitive decline is associated with uncontrolled epilepsy, as well. In children, frequent seizures may significantly alter neuronal networks, affecting cognitive and motor development.

Is your patient a candidate for surgery?

Patients with disabling complex partial seizures that remain uncontrolled after 2 or more AED trials (either as monotherapy or in combination) should be referred to an epilepsy specialty center for evaluation for surgery.⁴⁵ This should be considered as early as possible to afford the patient the best chance of achieving seizure control.

Successful epilepsy surgery—in which the portion of the brain causing the misfiring that causes the seizures is removed—often results in a better quality of life; it is also cost effective.⁴⁶ Not everyone with refractory epilepsy is a candidate for surgery, of course. Among those who are, however, 50% to 70% of patients can expect to have improved seizure control.⁴⁷

Status epilepticus is a medical emergency

A patient who develops status epilepticus is at high risk and requires immediate, and simultaneous, evaluation and treatment. Status epilepticus carries nearly a 20% mortality from the first episode,⁴⁸ and the 10-year mortality rate after an episode of status epilepticus is as high as 40%.⁴⁹

Although most of the deaths associated with status epilepticus are due to the underlying pathology, early treatment can prevent or ameliorate complications from rhabdomyolysis and irreversible anoxic neuronal damage.⁵⁰

A benzodiazepine (typically, a 10-mg IV bolus of diazepam) is the initial treatment for status epilepticus, followed by or concurrent with fosphenytoin (15-18 mg/kg). If status epilepticus remains refractory to first-line drugs (lasting >30 minutes), intubation and transfer to an intensive care setting may be required, and a neurological consult should be obtained.

Pharmacologic treatment of status epilepticus falls into 3 main classes: benzodiazepines, standard AEDs, and general anesthetics such as propofol. Benzodiazepines act very rapidly to control most prolonged seizures, and are the first-line treatment choice. Diazepam has long been the mainstay of treatment, and is usually readily available. However, in both a large systematic review and a head-to-head trial, lorazepam was found to be superior to diazepam in ending seizure activity and maintaining seizure control without the use of other medications^51,52—and is now the drug of choice for initial treatment of status epilepticus.

CASE You continue to see Joe every 3 to 4 months to monitor his basic blood work and mood. A year after his seizure, he remains seizure-free and is tolerating the AED without any adverse effects.

CORRESPONDENCE
William J. Geiger, MD, FAAFP, Medical College of Wisconsin, Columbia St. Mary’s Family Medicine Residency, 1121 East North Avenue, Milwaukee, WI 53212; bgeiger@mcw.edu

CASE Joe G, a 44-year-old man who has been your patient for years, comes to your office 48 hours after having a seizure. He has no history of seizures, had no warning signs or symptoms, and felt fine all day, but simply collapsed when the seizure occurred. He was transported to the emergency department (ED), and found to be postictal, with no further seizure activity. The ED work-up included a hemogram, comprehensive metabolic panel, and computed tomography brain scan, all of which were normal. An hour later, Joe had a normal neurological exam, then underwent electroencephalography (EEG) and magnetic resonance imaging (MRI) and was discharged home without medication.

How would you treat this patient?

About 10% of Americans will experience a seizure at some point in their lives,^1,2 and more than 3 million have epilepsy.³ The incidence ranges from 1% among 20-year-olds to more than 3% by the age of 75.^1,2

To adequately care for such patients—whether they have had multiple seizures or only one—you need to know whether they’re at risk for recurrences, when (or if) to prescribe an AED, and which agents provide optimal seizure control with the fewest adverse effects. You also need to know when a referral to an epilepsy specialist is indicated, when or whether it’s safe for patients to stop taking antiseizure medication, and how to address lifestyle issues that patients with epilepsy often need help with.

This review addresses these and other questions.

Is it epilepsy? How to respond to a single seizure

A seizure—a transient occurrence of signs or symptoms due to abnormal excessive or synchronous neural activity in the brain—can be either focal (partial) or generalized. In addition, seizures can be broadly divided into 2 categories, based on etiology:

Provoked seizures are caused by an acute structural, toxic, or metabolic insult to the brain, and, presumably, would not have occurred if the underlying medical condition did not exist. Treating the cause—eg, alcohol withdrawal, hyponatremia, or hypoglycemia—should prevent a recurrence.

Unprovoked seizures have no apparent underlying cause. Epilepsy is defined as a chronic condition characterized by ≥2 unprovoked seizures at least 24 hours apart, and epilepsy syndromes are classified as localization-related or generalized (TABLE 1).^1,4,5

Generally, epileptologists do not recommend symptomatic treatment of a first unprovoked seizure⁶—a consensus based on several randomized controlled trials that found immediate treatment with an AED reduced the risk of a subsequent seizure in the short term, but did not affect long-term outcomes or the development of epilepsy.⁷

Treatment should begin after a single seizure, however, if the seizure was prolonged or there is an increased risk of recurrence.⁶ Factors that increase this risk include an abnormal EEG, particularly if the abnormality is epileptiform; the presence of a brain lesion; a localized (focal) seizure; and an abnormal neurologic exam.⁸ A history of status epilepticus—a single, unremitting seizure lasting ≥5 to 10 minutes or frequent seizures without a return to neurologic baseline in between—or complex febrile seizures, and a family history of epilepsy are risk factors for recurrence, as well.⁷

When the patient is a child. Prescribing an AED for a child after a first unprovoked seizure is not indicated to prevent the development of epilepsy, but may be considered, as for adults, in circumstances where the benefit of reducing the risk of a second seizure outweighs the risk of pharmacologic and psychosocial adverse effects.⁹

CASE Joe’s ED records show that his MRI was normal, but his EEG revealed an epileptogenic focus on the right temporal region—a finding that indicates that he has an elevated risk of recurrence and is a candidate for an AED. Before selecting a particular agent, you review his chart.

Joe is taking a thiazide diuretic and a calcium channel blocker for hypertension. He was a heavy drinker until he had an episode of pancreatitis 10 years ago, and has been abstinent ever since. About 5 years ago, he suffered from depression and was treated with sertraline, but the depression resolved and the drug was discontinued 3 years ago. The patient’s mother and brother have type 2 diabetes and his father had a myocardial infarction before the age of 60. Joe was laid off from his sales job 18 months ago and is actively seeking employment. At this point, you consider a broad-spectrum AED that would not interact with his current medications or adversely affect his medical conditions, and would be relatively inexpensive.

TABLE 1
Identifying seizures and types of epilepsy:^1,4,5 International League Against Epilepsy classification

What to consider in a first-line drug

The number of AEDs on the market has increased sharply in the past few years, giving physicians many medications to choose from. Selecting the optimal drug is particularly important for the initial treatment, as many patients remain on the first AED for years. Second-generation AEDs have been found to be as effective as, and better tolerated than, first-generation antiseizure drugs. But all AEDs carry a warning of a potential increase in suicide risk and the need to monitor patients for behavior changes.¹⁰

Before selecting an AED for a particular patient, consider the following questions:

What type of seizure? AEDs are generally classified by spectrum of activity into “narrow-spectrum” and “broad-spectrum.” Narrow-spectrum drugs are more effective for controlling partial seizures, but have the potential to exacerbate generalized seizures; broad-spectrum AEDs can be used for both. (TABLE 2^11-18 lists indications for first- and second-generation AEDs based on type of epilepsy.) If there’s no definitive diagnosis of the type of epilepsy a patient has, use a broad-spectrum drug.

What other drugs is the patient taking? If the AED will be added to the patient’s current medication regimen, look closely at potential pharmacodynamic drug-drug interactions, and consider whether a dosage adjustment is needed. Determine, too, whether the patient has any comorbidities that could affect his or her response to the AED.

Side effects, such as weight gain or loss, urolithiasis, and hepatic enzyme induction, are key considerations. (TABLE W1,^19-24 which details dose, side effects, and costs of first- and second-generation AEDs, can be found at jfponline.com.)

Is the patient elderly? AED clearance is reduced in the elderly, so lower doses are needed. Reduction in serum albumin increases the free or active component of highly protein-bound drugs, increasing the likelihood of adverse effects.

Is the patient female? Some AEDs may have effects on women’s hormonal function, sexuality, bone health, and pregnancy.²⁵ Hepatic enzyme inducers increase the clearance of oral contraceptives, reducing their efficacy. Vitamin D and calcium metabolism can also be affected, which can lead to osteomalacia. Valproate treatment in women is associated with higher levels of insulin, testosterone, and triglycerides.²⁶ Cytochrome P-450-activating AEDs in general are associated with higher testosterone levels and reduced libido.²⁷

Potential pregnancy is another consideration. Women with epilepsy are able to bear healthy children. What’s more, patients whose seizures are controlled with AEDs should be maintained on medication throughout pregnancy, as the risk of fetal harm from seizures generally outweighs the teratogenicity of the drug.²⁸

Although large studies are limited, a study of 1532 infants exposed to AEDs in the first trimester did not find an increase in major birth defects compared with infants without such exposure.²⁹ More recently, a large observational cohort study conducted in more than 40 countries found that the possibility of harm to a developing fetus is not only drug-specific but also dose-related.³⁰ (To learn more, see “Pregnancy and epilepsy—when you’re managing both,” in the December 2010 issue of The Journal of Family Practice.)

Is cost a factor? Finally, consider the cost of the AED you would like to prescribe, and whether the patient has a prescription drug plan or the means to pay for his prescription.

CASE After a discussion of potential side effects, including the potential for suicidal ideation associated with AEDs, you prescribe carbamazepine for Joe as seizure prophylaxis, because it is the least expensive of the broad-spectrum AEDs and is unlikely to exacerbate his previous pancreatitis or interact with his current medications.

TABLE 2
Choosing an AED: What to consider^11-18

TABLE W1
A closer look at antiepileptic drugs^19-24

When to add a second AED

Monotherapy is the preferred method of epilepsy treatment, and controls seizures for 70% to 90% of patients.^31,32 If seizures continue and potential adverse effects prevent you from increasing the dosage, switching to a different AED, then tapering off the first agent, is recommended.^33,34

If the new AED fails to provide adequate seizure control, consider combination therapy. An additional 10% to 15% of patients with epilepsy achieve control with dual therapy.^33,34

Many second-generation agents are approved for adjunctive therapy. However, the use of 2 AEDs increases the risk of toxicities and drug interactions, and requires complex dosage adjustments, which should be done slowly and cautiously. Combination therapy also increases costs and may cause a decrease in compliance.^33,34

Noncompliance is the single most common reason for treatment failure in patients with epilepsy, occurring at an estimated rate of up to 60%.^35,36 The complexity of the drug regimen is the major cause, regardless of patient age, sex, psychomotor development, seizure type, or seizure frequency.^35,36

Because of the lack of good clinical trials of combination antiepilepsy therapy, no evidence is available to indicate which AEDs are safe and effective when taken together. There is, however, evidence that certain combinations should be avoided due to the risk of increased adverse effects. These include phenobarbital/valproate, phenytoin/carbamazepine, and carbamazepine/lamotrigine.²⁵

Managing the patient who is seizure-free

After a patient has been seizure-free for 2 to 5 years, consider a reduction in, or a discontinuation of, his or her AED. The relapse rate varies from 10% to 70%, with meta-analyses showing a rate of 25% in the first year and 29% in the second year.^19,37 The American Academy of Neurology (AAN) has published an evidence-based guideline for discontinuing AEDs in seizure-free patients, available at www.aan.com/professionals/practice/pdfs/gl0007.pdf.

Withdrawal should be gradual and, for patients on combination therapy, carried out one drug at a time to prevent a recurrence of seizures or status epilepticus. The AAN recommends a 2- to- 3-month withdrawal period for AEDs (and longer for benzodiazepines), although relapse rates have been found to be lower when the medication is withdrawn more slowly, over about 6 months.^19,34 If seizures recur after withdrawal, restart the AEDs at previous dosages.^19,34,38

Should the patient drive?

For patients with epilepsy, loss of independence related to driving restrictions is a major source of stress. A 10-year follow-up study of Danish patients with epilepsy found a 7-fold increase in motor vehicle accidents (MVAs) in patients with seizure disorders.³⁹ Other studies have shown that the seizure-free interval is the best predictor of involvement in an MVA.⁴⁰

The risk of driving accidents decreases as the seizure-free interval increases. Unfortunately, however, a decline in patient compliance is also associated with longer seizure-free intervals—creating the potential for recurrence and driving risk. Because of this discrepancy, a consensus statement from the AAN, American Epilepsy Society, and Epilepsy Foundation of America recommends a minimum 3-month seizure-free interval before patients are allowed to drive.⁴¹

Use clinical judgment in deciding whether to extend the seizure-free period. State laws vary widely regarding the need to report patients with seizure disorders, limitations on professional drivers, and seizure-free intervals required, so it is important to be familiar with the laws in your state. The Epilepsy Foundation has a helpful online resource with a database detailing individual state statutes (http://www.epilepsyfoundation.org/living/wellness/transportation/driverlicensing.cfm).

The danger of uncontrolled seizures

Overall, AEDs effectively control 70% of 80% of cases; the remaining 20% to 30% are considered medically refractory.³⁸ What’s more, after 2 AED failures, a patient’s chances of achieving full seizure control with additional drugs are no better than 10% to 20%.⁴² And, as more drugs are tried, the likelihood of full control declines even further.⁴³

Patients with uncontrolled seizures have a cumulative risk of sudden unexpected death in epilepsy (SUDEP) of 0.5% per year.⁴⁴ Cognitive decline is associated with uncontrolled epilepsy, as well. In children, frequent seizures may significantly alter neuronal networks, affecting cognitive and motor development.

Is your patient a candidate for surgery?

Patients with disabling complex partial seizures that remain uncontrolled after 2 or more AED trials (either as monotherapy or in combination) should be referred to an epilepsy specialty center for evaluation for surgery.⁴⁵ This should be considered as early as possible to afford the patient the best chance of achieving seizure control.

Successful epilepsy surgery—in which the portion of the brain causing the misfiring that causes the seizures is removed—often results in a better quality of life; it is also cost effective.⁴⁶ Not everyone with refractory epilepsy is a candidate for surgery, of course. Among those who are, however, 50% to 70% of patients can expect to have improved seizure control.⁴⁷

Status epilepticus is a medical emergency

A patient who develops status epilepticus is at high risk and requires immediate, and simultaneous, evaluation and treatment. Status epilepticus carries nearly a 20% mortality from the first episode,⁴⁸ and the 10-year mortality rate after an episode of status epilepticus is as high as 40%.⁴⁹

Although most of the deaths associated with status epilepticus are due to the underlying pathology, early treatment can prevent or ameliorate complications from rhabdomyolysis and irreversible anoxic neuronal damage.⁵⁰

A benzodiazepine (typically, a 10-mg IV bolus of diazepam) is the initial treatment for status epilepticus, followed by or concurrent with fosphenytoin (15-18 mg/kg). If status epilepticus remains refractory to first-line drugs (lasting >30 minutes), intubation and transfer to an intensive care setting may be required, and a neurological consult should be obtained.

Pharmacologic treatment of status epilepticus falls into 3 main classes: benzodiazepines, standard AEDs, and general anesthetics such as propofol. Benzodiazepines act very rapidly to control most prolonged seizures, and are the first-line treatment choice. Diazepam has long been the mainstay of treatment, and is usually readily available. However, in both a large systematic review and a head-to-head trial, lorazepam was found to be superior to diazepam in ending seizure activity and maintaining seizure control without the use of other medications^51,52—and is now the drug of choice for initial treatment of status epilepticus.

CASE You continue to see Joe every 3 to 4 months to monitor his basic blood work and mood. A year after his seizure, he remains seizure-free and is tolerating the AED without any adverse effects.

CORRESPONDENCE
William J. Geiger, MD, FAAFP, Medical College of Wisconsin, Columbia St. Mary’s Family Medicine Residency, 1121 East North Avenue, Milwaukee, WI 53212; bgeiger@mcw.edu

CASE When Joan C, a 35-year-old patient whom you’ve known for years, comes in for a physical, you notice that she’s coughing frequently. Upon questioning, Joan says she first noticed the cough several months ago; she also reports that she’s frequently hoarse, but has no other symptoms. Joan is a former smoker, and quit 4 years ago.

If Joan were your patient, would you suspect that she had an upper respiratory infection and prescribe an antibiotic such as azithromycin? Would you include laryngopharyngeal reflux disease in the differential diagnosis?

Laryngopharyngeal reflux disease (LPRD) is a common condition that most primary care physicians encounter frequently. It is also frequently misdiagnosed by clinicians who are unfamiliar with the differences between LPRD and gastroesophageal reflux disease (GERD).

The American Academy of Otolaryngology–Head and Neck Surgery defines laryngopharyngeal reflux as the retrograde movement of gastric contents into the laryngopharynx.¹ Common symptoms include hoarseness/dysphonia, chronic throat clearing, dysphagia, globus pharyngeus, and chronic cough, as well as postnasal drip, paroxysmal laryngospasm, odynophagia, excessive throat mucus, and a strange taste in the mouth.²

The diversity and vagueness of these symptoms, as well as the lack of a gold standard diagnostic test for LPRD, make it difficult to estimate its prevalence. In addition, signs of gastroesophageal reflux can be found in the laryngopharynx of up to 86% of healthy individuals, further complicating the clinical picture.³ To avoid missing this often overlooked reflux disease, you need to know how it develops, what signs and symptoms to look for, and which distinguishing features to keep in mind.

Pathophysiology and distinguishing features

The precise way in which LPRD develops is not known, but there are 2 proposed means of laryngeal injury—direct and indirect. In the first case, chemical irritants in the gastric refluxate enter the laryngopharynx and cause local mucosal injury. In the second, gastric reflux irritates the esophageal tissue enough to evoke laryngeal reflexes without ever reaching the larynx—a vagally mediated response associated with symptoms such as chronic cough, throat-clearing sensations, and bronchoconstriction.⁴

Unlike the esophageal lining, laryngeal epithelium is not protected against chemical injury from gastric acid, as it lacks both the stripping motion of esophageal peristalsis and the neutralizing bicarbonate in saliva.⁴ Thus, while far smaller amounts of gastric reflux make it into the laryngopharynx, the acid remains there longer and may cause greater injury.⁵ In some cases, this occurs as often as 50 times a day, although as few as 3 episodes per week have been known to cause LPRD.⁵

Heartburn is not the rule
Heartburn is a primary complaint of patients with GERD. It is reported by little more than a third (35%) of those with LPRD,^5,6 however, (which is why it is sometimes called the “silent” reflux disease). This is because heartburn is caused by esophagitis due to esophageal dysmotility and lower esophageal sphincter dysfunction,³ while most patients with LPRD have normal esophageal motor function and upper esophageal sphincter dysfunction. The fact that only a minimal amount of reflux enters the laryngopharynx may be part of the reason heartburn is less likely in patients with LPRD.

Onset of symptoms. When reflux occurs is another thing that distinguishes LPRD and GERD. Symptoms of GERD typically worsen when the individual is supine, while laryngopharyngeal reflux usually occurs when he or she is upright.⁷ The frequency with which these 2 conditions overlap is debatable, as there are few studies differentiating LPRD and GERD based on standardized signs and symptoms.⁷

Making sense of signs and symptoms

Most patients with LPRD seek treatment from their primary care physician, typically reporting symptoms that they don’t associate with gastric reflux, such as hoarseness, a chronic cough or sore throat, or the sensation of a lump in the throat (TABLE 1). Less common manifestations include “water brash”—excessive mucus in the mouth caused by a release of salivary bicarbonate to help neutralize acidity⁸—otitis media, sinus disease, and dental caries.⁵

Laryngeal endoscopy may reveal many changes from diffuse irritation. Diffuse erythema, edema, and interarytenoid hypertrophy/cobblestoning are the most useful findings for an LPRD diagnosis.^9,10 But in most cases, only a few nonspecific signs with a number of possible causes (infection, environmental irritants, allergies, temperature/climate change, among others) are seen on endoscopic examination, with little correlation with symptom severity. In fact, 74% of otolaryngologists responding to a recent survey said they relied more on patient symptoms than on laryngeal signs for an LPRD diagnosis.¹⁰

The Reflux Finding Score (RFS), available at http://www.nature.com/gimo/contents/pt1/fig_tab/gimo46_T3.html, is a clinical tool developed to quantify laryngeal inflammation and standardize objective endoscopic findings. The RFS incorporates the following endolaryngeal signs:

• subglottic edema
• ventricular obliteration
• erythema/hyperemia
• vocal cord edema
• diffuse laryngeal edema
• posterior commissure hypertrophy
• granuloma/granulation tissue
• thick endolaryngeal mucus.

A numeric value is assigned to each, based on whether it is present or absent; partial or complete; local or diffuse; or mild or severe. However, the RFS, too, is an imperfect tool. Clinicians who have used the RFS report that a score higher than 7 identifies LPRD with 95% sensitivity.¹¹ But laryngeal findings may be due to other causes, such as infection, autoimmune reaction, or even allergies, and studies have found the RFS to have poor specificity and inter-rater reliability.^12-14

Ambulatory dual probe pH monitoring was considered to be the gold standard test for LPRD at one time, but newer studies have raised questions about its validity and usefulness, especially in patients taking proton-pump inhibitors (PPIs).^1,5,7 Newer advanced probes featuring less invasive data collection and greater sensitivity are under development. Ambulatory 24-hour multichannel intraluminal impedance with pH monitoring is the most promising new diagnostic tool, as it can monitor both acidic and nonacidic reflux and distinguish between gas and liquid.¹⁵

TABLE 1
When to suspect laryngopharyngeal reflux disease^1,5,24

Treatment, like diagnosis, is not clear-cut

LPRD is often called a diagnosis of exclusion, because of the nonspecific nature of its signs and symptoms and the importance of considering a range of other etiologies. The differential diagnosis includes excessive voice use, postnasal drip, upper respiratory infection, habitual throat clearing, allergic rhinitis, environmental irritants, temperature/climate change, chronic or episodic use of alcohol and/or tobacco, and psychological problems related to tics, such as habitual throat clearing or coughing.⁵

Diagnosis is often based on an empiric trial of high-dose PPIs, with confirmation dependent on symptom relief. Because there have been few placebo-controlled trials with PPIs and those that have been completed had conflicting results, diagnosis based on a combination of medical history and endoscopic laryngeal examination may be a better approach.^16,17

Acid suppression therapy with either PPIs or histamine-2 (H2) receptor blockers such as ranitidine or famotidine is the mainstay of treatment for LPRD. But medical societies offer conflicting advice. The American Gastroenterological Association cautions clinicians not to prescribe acid-suppression therapy for patients with LPRD unless they also have GERD.⁶ The American Academy of Otolaryngology–Head and Neck Surgery recommends twice-daily PPI use for ≥6 months.^1,13 The general consensus, based on clinical experience alone, is that patients should be treated with high doses of PPIs (eg, 40 mg omeprazole twice a day) for ≥6 months, with the addition of an H2 receptor blocker to help reduce overnight acid production.^1,18 Prophylactic antacid use is also recommended in anticipation of reflux, such as before exercising and right after a meal.

Symptoms should start to improve within 6 to 8 weeks, and patients should be reassessed in about 3 months. To avoid a rebound effect from the abrupt cessation of medications, we suggest a gradual taper over 16 weeks. For the first 8 weeks, the H2 blocker should be discontinued and the PPI decreased from twice a day to once. If symptoms are still controlled, the PPI dose can be reduced to once every other day for another 8 weeks, then stopped if symptoms do not recur.¹⁸

Lifestyle and dietary changes (TABLE 2), such as smoking cessation, weight loss, and avoidance of alcohol, are an important part of LPRD treatment, and may be used as a first-line therapy before prescribing medication.¹⁹ In fact, some studies have found PPI therapy to be inferior to behavioral/lifestyle modifications.¹⁷

Fundoplication surgery, a procedure in which the gastric fundus of the stomach is wrapped around the lower end of the esophagus and stitched in place to prevent reflux, may be an option for patients who do not respond to, or cannot tolerate, aggressive medical treatment for LPRD. A 2006 prospective controlled study found that surgical fundoplication did not consistently relieve laryngeal symptoms.²⁰ But other studies have found that a carefully selected population with medically unresponsive laryngopharyngeal symptoms can benefit from this procedure.^21,22 One study showed a significant improvement within one month of fundoplication, with continued improvement observed during a 3-year follow-up.²¹ In another prospective study, researchers showed that while LPRD-related laryngeal symptoms such as coughing and throat-clearing improved with both medical therapy and laparoscopic fundoplication, voice quality and endoscopic laryngeal/pharyngeal findings improved significantly only with the surgical procedure.²³

TABLE 2
Recommend these lifestyle modifications¹⁹

CORRESPONDENCE
Shoib Sana, DO, Detroit Medical Center, Otolaryngology-Head and Neck Surgery, 6533 East Jefferson Avenue, Apartment 316, Detroit, MI 48207; ssana@dmc.org

CASE When Joan C, a 35-year-old patient whom you’ve known for years, comes in for a physical, you notice that she’s coughing frequently. Upon questioning, Joan says she first noticed the cough several months ago; she also reports that she’s frequently hoarse, but has no other symptoms. Joan is a former smoker, and quit 4 years ago.

If Joan were your patient, would you suspect that she had an upper respiratory infection and prescribe an antibiotic such as azithromycin? Would you include laryngopharyngeal reflux disease in the differential diagnosis?

Laryngopharyngeal reflux disease (LPRD) is a common condition that most primary care physicians encounter frequently. It is also frequently misdiagnosed by clinicians who are unfamiliar with the differences between LPRD and gastroesophageal reflux disease (GERD).

The American Academy of Otolaryngology–Head and Neck Surgery defines laryngopharyngeal reflux as the retrograde movement of gastric contents into the laryngopharynx.¹ Common symptoms include hoarseness/dysphonia, chronic throat clearing, dysphagia, globus pharyngeus, and chronic cough, as well as postnasal drip, paroxysmal laryngospasm, odynophagia, excessive throat mucus, and a strange taste in the mouth.²

The diversity and vagueness of these symptoms, as well as the lack of a gold standard diagnostic test for LPRD, make it difficult to estimate its prevalence. In addition, signs of gastroesophageal reflux can be found in the laryngopharynx of up to 86% of healthy individuals, further complicating the clinical picture.³ To avoid missing this often overlooked reflux disease, you need to know how it develops, what signs and symptoms to look for, and which distinguishing features to keep in mind.

Pathophysiology and distinguishing features

The precise way in which LPRD develops is not known, but there are 2 proposed means of laryngeal injury—direct and indirect. In the first case, chemical irritants in the gastric refluxate enter the laryngopharynx and cause local mucosal injury. In the second, gastric reflux irritates the esophageal tissue enough to evoke laryngeal reflexes without ever reaching the larynx—a vagally mediated response associated with symptoms such as chronic cough, throat-clearing sensations, and bronchoconstriction.⁴

Unlike the esophageal lining, laryngeal epithelium is not protected against chemical injury from gastric acid, as it lacks both the stripping motion of esophageal peristalsis and the neutralizing bicarbonate in saliva.⁴ Thus, while far smaller amounts of gastric reflux make it into the laryngopharynx, the acid remains there longer and may cause greater injury.⁵ In some cases, this occurs as often as 50 times a day, although as few as 3 episodes per week have been known to cause LPRD.⁵

Heartburn is not the rule
Heartburn is a primary complaint of patients with GERD. It is reported by little more than a third (35%) of those with LPRD,^5,6 however, (which is why it is sometimes called the “silent” reflux disease). This is because heartburn is caused by esophagitis due to esophageal dysmotility and lower esophageal sphincter dysfunction,³ while most patients with LPRD have normal esophageal motor function and upper esophageal sphincter dysfunction. The fact that only a minimal amount of reflux enters the laryngopharynx may be part of the reason heartburn is less likely in patients with LPRD.

Onset of symptoms. When reflux occurs is another thing that distinguishes LPRD and GERD. Symptoms of GERD typically worsen when the individual is supine, while laryngopharyngeal reflux usually occurs when he or she is upright.⁷ The frequency with which these 2 conditions overlap is debatable, as there are few studies differentiating LPRD and GERD based on standardized signs and symptoms.⁷

Making sense of signs and symptoms

Most patients with LPRD seek treatment from their primary care physician, typically reporting symptoms that they don’t associate with gastric reflux, such as hoarseness, a chronic cough or sore throat, or the sensation of a lump in the throat (TABLE 1). Less common manifestations include “water brash”—excessive mucus in the mouth caused by a release of salivary bicarbonate to help neutralize acidity⁸—otitis media, sinus disease, and dental caries.⁵

Laryngeal endoscopy may reveal many changes from diffuse irritation. Diffuse erythema, edema, and interarytenoid hypertrophy/cobblestoning are the most useful findings for an LPRD diagnosis.^9,10 But in most cases, only a few nonspecific signs with a number of possible causes (infection, environmental irritants, allergies, temperature/climate change, among others) are seen on endoscopic examination, with little correlation with symptom severity. In fact, 74% of otolaryngologists responding to a recent survey said they relied more on patient symptoms than on laryngeal signs for an LPRD diagnosis.¹⁰

The Reflux Finding Score (RFS), available at http://www.nature.com/gimo/contents/pt1/fig_tab/gimo46_T3.html, is a clinical tool developed to quantify laryngeal inflammation and standardize objective endoscopic findings. The RFS incorporates the following endolaryngeal signs:

• subglottic edema
• ventricular obliteration
• erythema/hyperemia
• vocal cord edema
• diffuse laryngeal edema
• posterior commissure hypertrophy
• granuloma/granulation tissue
• thick endolaryngeal mucus.

A numeric value is assigned to each, based on whether it is present or absent; partial or complete; local or diffuse; or mild or severe. However, the RFS, too, is an imperfect tool. Clinicians who have used the RFS report that a score higher than 7 identifies LPRD with 95% sensitivity.¹¹ But laryngeal findings may be due to other causes, such as infection, autoimmune reaction, or even allergies, and studies have found the RFS to have poor specificity and inter-rater reliability.^12-14

Ambulatory dual probe pH monitoring was considered to be the gold standard test for LPRD at one time, but newer studies have raised questions about its validity and usefulness, especially in patients taking proton-pump inhibitors (PPIs).^1,5,7 Newer advanced probes featuring less invasive data collection and greater sensitivity are under development. Ambulatory 24-hour multichannel intraluminal impedance with pH monitoring is the most promising new diagnostic tool, as it can monitor both acidic and nonacidic reflux and distinguish between gas and liquid.¹⁵

TABLE 1
When to suspect laryngopharyngeal reflux disease^1,5,24

Treatment, like diagnosis, is not clear-cut

LPRD is often called a diagnosis of exclusion, because of the nonspecific nature of its signs and symptoms and the importance of considering a range of other etiologies. The differential diagnosis includes excessive voice use, postnasal drip, upper respiratory infection, habitual throat clearing, allergic rhinitis, environmental irritants, temperature/climate change, chronic or episodic use of alcohol and/or tobacco, and psychological problems related to tics, such as habitual throat clearing or coughing.⁵

Diagnosis is often based on an empiric trial of high-dose PPIs, with confirmation dependent on symptom relief. Because there have been few placebo-controlled trials with PPIs and those that have been completed had conflicting results, diagnosis based on a combination of medical history and endoscopic laryngeal examination may be a better approach.^16,17

Acid suppression therapy with either PPIs or histamine-2 (H2) receptor blockers such as ranitidine or famotidine is the mainstay of treatment for LPRD. But medical societies offer conflicting advice. The American Gastroenterological Association cautions clinicians not to prescribe acid-suppression therapy for patients with LPRD unless they also have GERD.⁶ The American Academy of Otolaryngology–Head and Neck Surgery recommends twice-daily PPI use for ≥6 months.^1,13 The general consensus, based on clinical experience alone, is that patients should be treated with high doses of PPIs (eg, 40 mg omeprazole twice a day) for ≥6 months, with the addition of an H2 receptor blocker to help reduce overnight acid production.^1,18 Prophylactic antacid use is also recommended in anticipation of reflux, such as before exercising and right after a meal.

Symptoms should start to improve within 6 to 8 weeks, and patients should be reassessed in about 3 months. To avoid a rebound effect from the abrupt cessation of medications, we suggest a gradual taper over 16 weeks. For the first 8 weeks, the H2 blocker should be discontinued and the PPI decreased from twice a day to once. If symptoms are still controlled, the PPI dose can be reduced to once every other day for another 8 weeks, then stopped if symptoms do not recur.¹⁸

Lifestyle and dietary changes (TABLE 2), such as smoking cessation, weight loss, and avoidance of alcohol, are an important part of LPRD treatment, and may be used as a first-line therapy before prescribing medication.¹⁹ In fact, some studies have found PPI therapy to be inferior to behavioral/lifestyle modifications.¹⁷

Fundoplication surgery, a procedure in which the gastric fundus of the stomach is wrapped around the lower end of the esophagus and stitched in place to prevent reflux, may be an option for patients who do not respond to, or cannot tolerate, aggressive medical treatment for LPRD. A 2006 prospective controlled study found that surgical fundoplication did not consistently relieve laryngeal symptoms.²⁰ But other studies have found that a carefully selected population with medically unresponsive laryngopharyngeal symptoms can benefit from this procedure.^21,22 One study showed a significant improvement within one month of fundoplication, with continued improvement observed during a 3-year follow-up.²¹ In another prospective study, researchers showed that while LPRD-related laryngeal symptoms such as coughing and throat-clearing improved with both medical therapy and laparoscopic fundoplication, voice quality and endoscopic laryngeal/pharyngeal findings improved significantly only with the surgical procedure.²³

TABLE 2
Recommend these lifestyle modifications¹⁹

CORRESPONDENCE
Shoib Sana, DO, Detroit Medical Center, Otolaryngology-Head and Neck Surgery, 6533 East Jefferson Avenue, Apartment 316, Detroit, MI 48207; ssana@dmc.org

CASE When Joan C, a 35-year-old patient whom you’ve known for years, comes in for a physical, you notice that she’s coughing frequently. Upon questioning, Joan says she first noticed the cough several months ago; she also reports that she’s frequently hoarse, but has no other symptoms. Joan is a former smoker, and quit 4 years ago.

If Joan were your patient, would you suspect that she had an upper respiratory infection and prescribe an antibiotic such as azithromycin? Would you include laryngopharyngeal reflux disease in the differential diagnosis?

Laryngopharyngeal reflux disease (LPRD) is a common condition that most primary care physicians encounter frequently. It is also frequently misdiagnosed by clinicians who are unfamiliar with the differences between LPRD and gastroesophageal reflux disease (GERD).

The American Academy of Otolaryngology–Head and Neck Surgery defines laryngopharyngeal reflux as the retrograde movement of gastric contents into the laryngopharynx.¹ Common symptoms include hoarseness/dysphonia, chronic throat clearing, dysphagia, globus pharyngeus, and chronic cough, as well as postnasal drip, paroxysmal laryngospasm, odynophagia, excessive throat mucus, and a strange taste in the mouth.²

The diversity and vagueness of these symptoms, as well as the lack of a gold standard diagnostic test for LPRD, make it difficult to estimate its prevalence. In addition, signs of gastroesophageal reflux can be found in the laryngopharynx of up to 86% of healthy individuals, further complicating the clinical picture.³ To avoid missing this often overlooked reflux disease, you need to know how it develops, what signs and symptoms to look for, and which distinguishing features to keep in mind.

Pathophysiology and distinguishing features

The precise way in which LPRD develops is not known, but there are 2 proposed means of laryngeal injury—direct and indirect. In the first case, chemical irritants in the gastric refluxate enter the laryngopharynx and cause local mucosal injury. In the second, gastric reflux irritates the esophageal tissue enough to evoke laryngeal reflexes without ever reaching the larynx—a vagally mediated response associated with symptoms such as chronic cough, throat-clearing sensations, and bronchoconstriction.⁴

Unlike the esophageal lining, laryngeal epithelium is not protected against chemical injury from gastric acid, as it lacks both the stripping motion of esophageal peristalsis and the neutralizing bicarbonate in saliva.⁴ Thus, while far smaller amounts of gastric reflux make it into the laryngopharynx, the acid remains there longer and may cause greater injury.⁵ In some cases, this occurs as often as 50 times a day, although as few as 3 episodes per week have been known to cause LPRD.⁵

Heartburn is not the rule
Heartburn is a primary complaint of patients with GERD. It is reported by little more than a third (35%) of those with LPRD,^5,6 however, (which is why it is sometimes called the “silent” reflux disease). This is because heartburn is caused by esophagitis due to esophageal dysmotility and lower esophageal sphincter dysfunction,³ while most patients with LPRD have normal esophageal motor function and upper esophageal sphincter dysfunction. The fact that only a minimal amount of reflux enters the laryngopharynx may be part of the reason heartburn is less likely in patients with LPRD.

Onset of symptoms. When reflux occurs is another thing that distinguishes LPRD and GERD. Symptoms of GERD typically worsen when the individual is supine, while laryngopharyngeal reflux usually occurs when he or she is upright.⁷ The frequency with which these 2 conditions overlap is debatable, as there are few studies differentiating LPRD and GERD based on standardized signs and symptoms.⁷

Making sense of signs and symptoms

Most patients with LPRD seek treatment from their primary care physician, typically reporting symptoms that they don’t associate with gastric reflux, such as hoarseness, a chronic cough or sore throat, or the sensation of a lump in the throat (TABLE 1). Less common manifestations include “water brash”—excessive mucus in the mouth caused by a release of salivary bicarbonate to help neutralize acidity⁸—otitis media, sinus disease, and dental caries.⁵

Laryngeal endoscopy may reveal many changes from diffuse irritation. Diffuse erythema, edema, and interarytenoid hypertrophy/cobblestoning are the most useful findings for an LPRD diagnosis.^9,10 But in most cases, only a few nonspecific signs with a number of possible causes (infection, environmental irritants, allergies, temperature/climate change, among others) are seen on endoscopic examination, with little correlation with symptom severity. In fact, 74% of otolaryngologists responding to a recent survey said they relied more on patient symptoms than on laryngeal signs for an LPRD diagnosis.¹⁰

The Reflux Finding Score (RFS), available at http://www.nature.com/gimo/contents/pt1/fig_tab/gimo46_T3.html, is a clinical tool developed to quantify laryngeal inflammation and standardize objective endoscopic findings. The RFS incorporates the following endolaryngeal signs:

• subglottic edema
• ventricular obliteration
• erythema/hyperemia
• vocal cord edema
• diffuse laryngeal edema
• posterior commissure hypertrophy
• granuloma/granulation tissue
• thick endolaryngeal mucus.

A numeric value is assigned to each, based on whether it is present or absent; partial or complete; local or diffuse; or mild or severe. However, the RFS, too, is an imperfect tool. Clinicians who have used the RFS report that a score higher than 7 identifies LPRD with 95% sensitivity.¹¹ But laryngeal findings may be due to other causes, such as infection, autoimmune reaction, or even allergies, and studies have found the RFS to have poor specificity and inter-rater reliability.^12-14

Ambulatory dual probe pH monitoring was considered to be the gold standard test for LPRD at one time, but newer studies have raised questions about its validity and usefulness, especially in patients taking proton-pump inhibitors (PPIs).^1,5,7 Newer advanced probes featuring less invasive data collection and greater sensitivity are under development. Ambulatory 24-hour multichannel intraluminal impedance with pH monitoring is the most promising new diagnostic tool, as it can monitor both acidic and nonacidic reflux and distinguish between gas and liquid.¹⁵

TABLE 1
When to suspect laryngopharyngeal reflux disease^1,5,24

Treatment, like diagnosis, is not clear-cut

LPRD is often called a diagnosis of exclusion, because of the nonspecific nature of its signs and symptoms and the importance of considering a range of other etiologies. The differential diagnosis includes excessive voice use, postnasal drip, upper respiratory infection, habitual throat clearing, allergic rhinitis, environmental irritants, temperature/climate change, chronic or episodic use of alcohol and/or tobacco, and psychological problems related to tics, such as habitual throat clearing or coughing.⁵

Diagnosis is often based on an empiric trial of high-dose PPIs, with confirmation dependent on symptom relief. Because there have been few placebo-controlled trials with PPIs and those that have been completed had conflicting results, diagnosis based on a combination of medical history and endoscopic laryngeal examination may be a better approach.^16,17

Acid suppression therapy with either PPIs or histamine-2 (H2) receptor blockers such as ranitidine or famotidine is the mainstay of treatment for LPRD. But medical societies offer conflicting advice. The American Gastroenterological Association cautions clinicians not to prescribe acid-suppression therapy for patients with LPRD unless they also have GERD.⁶ The American Academy of Otolaryngology–Head and Neck Surgery recommends twice-daily PPI use for ≥6 months.^1,13 The general consensus, based on clinical experience alone, is that patients should be treated with high doses of PPIs (eg, 40 mg omeprazole twice a day) for ≥6 months, with the addition of an H2 receptor blocker to help reduce overnight acid production.^1,18 Prophylactic antacid use is also recommended in anticipation of reflux, such as before exercising and right after a meal.

Symptoms should start to improve within 6 to 8 weeks, and patients should be reassessed in about 3 months. To avoid a rebound effect from the abrupt cessation of medications, we suggest a gradual taper over 16 weeks. For the first 8 weeks, the H2 blocker should be discontinued and the PPI decreased from twice a day to once. If symptoms are still controlled, the PPI dose can be reduced to once every other day for another 8 weeks, then stopped if symptoms do not recur.¹⁸

Lifestyle and dietary changes (TABLE 2), such as smoking cessation, weight loss, and avoidance of alcohol, are an important part of LPRD treatment, and may be used as a first-line therapy before prescribing medication.¹⁹ In fact, some studies have found PPI therapy to be inferior to behavioral/lifestyle modifications.¹⁷

Fundoplication surgery, a procedure in which the gastric fundus of the stomach is wrapped around the lower end of the esophagus and stitched in place to prevent reflux, may be an option for patients who do not respond to, or cannot tolerate, aggressive medical treatment for LPRD. A 2006 prospective controlled study found that surgical fundoplication did not consistently relieve laryngeal symptoms.²⁰ But other studies have found that a carefully selected population with medically unresponsive laryngopharyngeal symptoms can benefit from this procedure.^21,22 One study showed a significant improvement within one month of fundoplication, with continued improvement observed during a 3-year follow-up.²¹ In another prospective study, researchers showed that while LPRD-related laryngeal symptoms such as coughing and throat-clearing improved with both medical therapy and laparoscopic fundoplication, voice quality and endoscopic laryngeal/pharyngeal findings improved significantly only with the surgical procedure.²³

TABLE 2
Recommend these lifestyle modifications¹⁹

CORRESPONDENCE
Shoib Sana, DO, Detroit Medical Center, Otolaryngology-Head and Neck Surgery, 6533 East Jefferson Avenue, Apartment 316, Detroit, MI 48207; ssana@dmc.org

CASE 1 A 31-year-old African American woman sought treatment at her local emergency department (ED) for nausea, vomiting, and diarrhea. She reported passing more than 6 yellowish-brown, watery, nonbloody stools during the previous 2 days. She felt weak, feverish, and light-headed and showed signs of dehydration.

The patient had Crohn’s disease and had undergone a partial colectomy 5 years earlier. She told the ED physician that 2 days before visiting the ED she had gone to a “cleansing center” for a colonic cleansing, but was unable to complete the process because she developed cramps 15 minutes into the procedure. Less than an hour later, she developed diarrhea, nausea, and vomiting.

In the ED, her serum potassium was 2.9 mEq/L, blood urea nitrogen was 26 mg/dL, and creatinine was 1.9 mg/dL. She was afebrile, with a blood pressure of 135/75 mm Hg and a heart rate of 113 beats per minute. After receiving 2 liters of normal saline and 90 mEq of potassium chloride replacement, the patient felt better and was later discharged from the ED.

Three days later, the patient came to our residency clinic. She described her stools as being loose, but not watery or bloody, and passed in small amounts, about 4 times daily. She still had some abdominal cramping just before passing stool, but bowel movements relieved that. Her vital signs were within normal limits, and her physical exam was benign. The patient was instructed to follow her normal diet, as tolerated, and drink plenty of fluids to maintain good hydration. Her symptoms resolved by the following week.

CASE 2 A 49-year-old African American man came to our community hospital because of vomiting, diarrhea, and abdominal pain he had been experiencing for 4 days. He linked the symptoms to eating a large fast-food breakfast, followed by a big lunch the day before. He described having multiple episodes of nonbloody, nonbilious vomiting, nonbloody watery diarrhea, and “twisting” abdominal pain that was constant but temporarily relieved with a warm compress or positional maneuvers. He had never had a similar episode and had not taken any antibiotics recently.

Upon further questioning, the patient revealed that he had used a colon cleanser a few days earlier. A review showed that he had lost 24 pounds in 10 days. Vitals were within normal limits. Serum potassium was 2.9 mEq/L, and creatinine was 2.1 mg/dL. A computed tomography scan of the abdomen revealed moderate to moderately severe dilatation of multiple small bowel loops with multiple air fluid levels, suggesting an early or partial small bowel obstruction. We obtained a surgical consultation, but surgery was not required. He was discharged after 2 days.

The patient returned to the hospital 3 days later with similar symptoms and severe weakness associated with dizziness. At that time his serum potassium was 2.4 mEq/L and creatinine was 4.0 mg/dL. Aspartate aminotransferase was 29 U/L, alanine aminotransferase was 80 U/L, lipase was 418 U/L, and amylase was 94 U/L.

The patient was readmitted for dehydration, hypokalemia, and pancreatitis and, following a colonoscopy and biopsy that revealed chronic and acute inflammation, a gastroenterologist made a diagnosis of “herbal intoxication.” The patient was hydrated, his electrolytes were replaced, and his diet was slowly returned to normal. He was discharged after 5 days.

An old practice rediscovered

Colon cleansing has been around since ancient times, when its purported benefits were based on the belief that intestinal waste can poison the body (“autointoxication”).¹ The procedure became popular in the early 1900s, but in a 1919 paper, the American Medical Association discounted the autointoxication theory and condemned the practice.¹ The procedure then fell out of favor, albeit temporarily.² Colon cleansing has staged a comeback in recent years.

Colon cleansing basics
Colon cleansing, also called colonic irrigation or colonic hydrotherapy, is performed by colonic hygienists or colon therapists, or can be self-administered. The procedure works like an enema. The patient generally lies on a table and water (with or without additional herbs or compounds) is pumped through the rectum via a tube.

Unlike enemas, for which a small amount of fluid is used, however, colon cleansing calls for a large volume of fluid—up to 60 liters—to be introduced into the rectum.^3,4 Fluids and waste are expelled through another tube. The procedure may be repeated several times.

Products go by many names
Most colon cleansing products come in the forms of laxatives, teas, powders, and capsules. They can be taken by mouth or inserted into the rectum. They often contain sodium phosphate, coffee, probiotics, enzymes, or any of a variety of herbs.⁵ Some products contain fiber preparations, including psyllium, flaxseed, and laxatives such as cascara, magnesium oxide, cat’s claw, artichoke leaves, burdock root, licorice, and milk thistle.²

With names such as Nature’s Bounty Colon Cleanser Natural Detox Formula, Health Plus Inc. Colon Cleanse, and 7-Day Miracle Cleanse, as well as endorsements by movie stars, these colon cleansing products are actively promoted as a natural way to enhance one’s well-being. Advertisements promising that colon cleansing will alleviate fatigue, headache, weight gain, and low energy are ubiquitous on the Internet and in newspapers and magazines. The ads tout the safety of “herbal” and “natural” preparations. These materials also provide anecdotal support for claims that colon cleansing improves the immune and circulatory systems, enhances cognitive abilities, and aids weight loss through “detoxification.”⁶

Individuals who want to cleanse their own colons can choose among home kits, some of which include disposable tubing, while others have components that can be reused if they are sterilized after each use.^5,7 But many people turn to a “hydrotherapist” for colon irrigation. The services are also increasingly being offered by practitioners who describe themselves as “colon hygienists.”

These individuals sometimes belong to organizations such as the National Board for Colon Hydrotherapy (NBCH) or the International Association for Colon Hydrotherapy (I-ACT).^8,9 These practitioners are not licensed, but they are required to have a high school or equivalent degree plus 3 semesters of postsecondary education and to be certified in cardiopulmonary resuscitation. They also take various seminars and continuing education courses from the NBCH and I-ACT.

How many individuals have used colon cleansing is unclear, although one study suggested that in the United Kingdom, registered practitioners carry out an estimated 5600 procedures every month.¹⁰

Where’s the evidence?

Despite colon cleansing’s long history and current popularity, the literature does not support its purported benefits. Historically, colon cleansing was thought to prevent autointoxication from toxins originating in the colon, but the evidence for this claim is limited.¹¹ A search of the literature using the terms “colon cleansing,” “herbal colon cleanse,” “colon detoxification,” and “colon irrigation,” yielded no scientifically robust studies in support of this practice. One study suggested that lymphocytes might migrate from the gut into the circulation after the procedure, which may “improve colon and immune system function.”¹²

Even though colon cleansing is touted as a commonly used form of holistic, complementary and alternative medicine, the Natural Standard Professional Database concluded in a monograph that there is “limited clinical evidence validating colon therapy as a health promotion practice” and noted a “lack of sufficient evidence” for most of its prescribed uses.¹³

Adverse effects: From cramping to renal failure
Most reports in the literature note a variety of adverse effects of colon cleansing that range from mild (eg, cramping, abdominal pain, fullness, bloating, nausea, vomiting, perianal irritation, and soreness) to severe (eg, electrolyte imbalance and renal failure).^11,14-17 Some herbal preparations have also been associated with aplastic anemia and liver toxicity.¹⁸

Case reports also have noted back and pelvic abscesses after colonic hydrotherapy, fatal aeroportia (gas accumulation in the mesenteric veins) with air emboli, rectal perforations, perineal gangrene, acute water intoxication, coffee enema-associated colitis and septicemia, and deaths due to amebiasis.^2,3,19-21

The FDA has issued many warning letters
The preparations used for colon cleansing are considered dietary supplements, and the US Food and Drug Administration (FDA) requires that they be labeled as such; the FDA does not preapprove these substances, however. The FDA also requires that colonic hydrotherapy and irrigation system devices meet certain requirements, but the agency has never approved any system for general nonmedical purposes, such as colon cleansing.

The devices have an FDA Class III designation, indicating that if a device is used for purposes beyond what is medically indicated (preparation for radiologic and endoscopic procedures), the manufacturer must obtain premarket approval from the FDA, which is based on evaluation of the safety and effectiveness of the device as shown by available scientific evidence and current regulations.²² During the past decade the FDA has issued numerous warning letters to manufacturers for unapproved use of the devices for colon cleansing.^23-26

Raise the issue with patients
Given the current popularity of colon cleansing, it’s important to recognize that some of your patients may engage in, or be thinking about, the practice. (See “4 things to tell patients about colon cleansing”.) Be sure to tell patients about the potential consequences of colon cleansing and to emphasize that there is a lack of evidence to back up supporters’ claims.

CORRESPONDENCE
Ranit Mishori, MD, MHS, Georgetown University School of Medicine, 3900 Reservoir Road, NW, Washington, DC 20007; mishorir@georgetown.edu

CASE 1 A 31-year-old African American woman sought treatment at her local emergency department (ED) for nausea, vomiting, and diarrhea. She reported passing more than 6 yellowish-brown, watery, nonbloody stools during the previous 2 days. She felt weak, feverish, and light-headed and showed signs of dehydration.

The patient had Crohn’s disease and had undergone a partial colectomy 5 years earlier. She told the ED physician that 2 days before visiting the ED she had gone to a “cleansing center” for a colonic cleansing, but was unable to complete the process because she developed cramps 15 minutes into the procedure. Less than an hour later, she developed diarrhea, nausea, and vomiting.

In the ED, her serum potassium was 2.9 mEq/L, blood urea nitrogen was 26 mg/dL, and creatinine was 1.9 mg/dL. She was afebrile, with a blood pressure of 135/75 mm Hg and a heart rate of 113 beats per minute. After receiving 2 liters of normal saline and 90 mEq of potassium chloride replacement, the patient felt better and was later discharged from the ED.

Three days later, the patient came to our residency clinic. She described her stools as being loose, but not watery or bloody, and passed in small amounts, about 4 times daily. She still had some abdominal cramping just before passing stool, but bowel movements relieved that. Her vital signs were within normal limits, and her physical exam was benign. The patient was instructed to follow her normal diet, as tolerated, and drink plenty of fluids to maintain good hydration. Her symptoms resolved by the following week.

CASE 2 A 49-year-old African American man came to our community hospital because of vomiting, diarrhea, and abdominal pain he had been experiencing for 4 days. He linked the symptoms to eating a large fast-food breakfast, followed by a big lunch the day before. He described having multiple episodes of nonbloody, nonbilious vomiting, nonbloody watery diarrhea, and “twisting” abdominal pain that was constant but temporarily relieved with a warm compress or positional maneuvers. He had never had a similar episode and had not taken any antibiotics recently.

Upon further questioning, the patient revealed that he had used a colon cleanser a few days earlier. A review showed that he had lost 24 pounds in 10 days. Vitals were within normal limits. Serum potassium was 2.9 mEq/L, and creatinine was 2.1 mg/dL. A computed tomography scan of the abdomen revealed moderate to moderately severe dilatation of multiple small bowel loops with multiple air fluid levels, suggesting an early or partial small bowel obstruction. We obtained a surgical consultation, but surgery was not required. He was discharged after 2 days.

The patient returned to the hospital 3 days later with similar symptoms and severe weakness associated with dizziness. At that time his serum potassium was 2.4 mEq/L and creatinine was 4.0 mg/dL. Aspartate aminotransferase was 29 U/L, alanine aminotransferase was 80 U/L, lipase was 418 U/L, and amylase was 94 U/L.

The patient was readmitted for dehydration, hypokalemia, and pancreatitis and, following a colonoscopy and biopsy that revealed chronic and acute inflammation, a gastroenterologist made a diagnosis of “herbal intoxication.” The patient was hydrated, his electrolytes were replaced, and his diet was slowly returned to normal. He was discharged after 5 days.

An old practice rediscovered

Colon cleansing has been around since ancient times, when its purported benefits were based on the belief that intestinal waste can poison the body (“autointoxication”).¹ The procedure became popular in the early 1900s, but in a 1919 paper, the American Medical Association discounted the autointoxication theory and condemned the practice.¹ The procedure then fell out of favor, albeit temporarily.² Colon cleansing has staged a comeback in recent years.

Colon cleansing basics
Colon cleansing, also called colonic irrigation or colonic hydrotherapy, is performed by colonic hygienists or colon therapists, or can be self-administered. The procedure works like an enema. The patient generally lies on a table and water (with or without additional herbs or compounds) is pumped through the rectum via a tube.

Unlike enemas, for which a small amount of fluid is used, however, colon cleansing calls for a large volume of fluid—up to 60 liters—to be introduced into the rectum.^3,4 Fluids and waste are expelled through another tube. The procedure may be repeated several times.

Products go by many names
Most colon cleansing products come in the forms of laxatives, teas, powders, and capsules. They can be taken by mouth or inserted into the rectum. They often contain sodium phosphate, coffee, probiotics, enzymes, or any of a variety of herbs.⁵ Some products contain fiber preparations, including psyllium, flaxseed, and laxatives such as cascara, magnesium oxide, cat’s claw, artichoke leaves, burdock root, licorice, and milk thistle.²

With names such as Nature’s Bounty Colon Cleanser Natural Detox Formula, Health Plus Inc. Colon Cleanse, and 7-Day Miracle Cleanse, as well as endorsements by movie stars, these colon cleansing products are actively promoted as a natural way to enhance one’s well-being. Advertisements promising that colon cleansing will alleviate fatigue, headache, weight gain, and low energy are ubiquitous on the Internet and in newspapers and magazines. The ads tout the safety of “herbal” and “natural” preparations. These materials also provide anecdotal support for claims that colon cleansing improves the immune and circulatory systems, enhances cognitive abilities, and aids weight loss through “detoxification.”⁶

Individuals who want to cleanse their own colons can choose among home kits, some of which include disposable tubing, while others have components that can be reused if they are sterilized after each use.^5,7 But many people turn to a “hydrotherapist” for colon irrigation. The services are also increasingly being offered by practitioners who describe themselves as “colon hygienists.”

These individuals sometimes belong to organizations such as the National Board for Colon Hydrotherapy (NBCH) or the International Association for Colon Hydrotherapy (I-ACT).^8,9 These practitioners are not licensed, but they are required to have a high school or equivalent degree plus 3 semesters of postsecondary education and to be certified in cardiopulmonary resuscitation. They also take various seminars and continuing education courses from the NBCH and I-ACT.

How many individuals have used colon cleansing is unclear, although one study suggested that in the United Kingdom, registered practitioners carry out an estimated 5600 procedures every month.¹⁰

Where’s the evidence?

Despite colon cleansing’s long history and current popularity, the literature does not support its purported benefits. Historically, colon cleansing was thought to prevent autointoxication from toxins originating in the colon, but the evidence for this claim is limited.¹¹ A search of the literature using the terms “colon cleansing,” “herbal colon cleanse,” “colon detoxification,” and “colon irrigation,” yielded no scientifically robust studies in support of this practice. One study suggested that lymphocytes might migrate from the gut into the circulation after the procedure, which may “improve colon and immune system function.”¹²

Even though colon cleansing is touted as a commonly used form of holistic, complementary and alternative medicine, the Natural Standard Professional Database concluded in a monograph that there is “limited clinical evidence validating colon therapy as a health promotion practice” and noted a “lack of sufficient evidence” for most of its prescribed uses.¹³