Cleveland Clinic Journal of Medicine

In Reply: We welcome the comments from Dr. Chongnarungsin on our article and the opportunity to further discuss our opinions.

In our paper, we discussed current recommendations for prophylaxis of stress ulcer-related bleeding in hospitalized patients and advocated against the blind administration of drugs without risk stratification.

The landmark trial that provides the most-cited definitions and the risk factors for clinically significant stress ulcer-related bleeding in critically ill patients was published in 1994 by Cook et al.¹ In their multicenter prospective cohort study of 2,252 patients, the authors reported that prolonged mechanical ventilation is an important risk factor for clinically significant stress ulcer-related bleeding.

Another major prospective cohort study observed an incidence rate of clinically significant stress ulcer-related bleeding of 3.5%.²

Dr. Chongnarungsin cites another prospective cohort study of 183 patients from the same era,³ wherein the authors defined stress ulcer-related bleeding as bleeding requiring transfusion of packed red blood cells, found on endoscopy or on postmortem evaluation. This was in contrast to the 1994 study of Cook et al,¹ who had a more rigorous and comprehensive definition for overt and clinically significant stress ulcer-related bleeding, applied by up to three independent adjudicators not involved in the patients’ care. Their definition not only entailed a more accurate transfusion-dependent bleeding criterion, but also included hemodynamic and laboratory criteria. As such, the “very low rate” of stress ulcer-related bleeding reported by Zandstra et al³ should be critically appraised. Of note, the authors in that study did not report the rates of patients who received early enteral feeding, and their patients received cefotaxime for digestive tract decontamination, an important confounder to the interpretation of the study results.

Indeed, the remarkable variation in estimates of the incidence of stress ulcer-related bleeding is probably related to the lack of a uniform definition. Even when rates of endoscopic and occult bleeding are set aside, agreement is lacking as to which category of bleeding is clinically significant.

Dr. Chongnarungsin also cites the study by Ellison et al⁴ of a cohort of 874 patients who had no previous gastrointestinal bleeding or peptic ulcer disease and who were enrolled in a multicenter randomized controlled trial of prophylactic intravenous immune globulin to prevent infections associated with an intensive care unit. In a secondary objective, the authors did not identify coagulopathy or prolonged mechanical ventilation as a principal risk factor for bleeding. The authors ascribed this discrepancy with previously published literature to their unique study population, which consisted predominantly of elderly men and rarely included trauma patients. In light of these unique peculiarities of their population, the lack of an association between prolonged mechanical ventilation and stress ulcer-related bleeding cannot be determined. Moreover, that study showed that prolonged nasogastric tube insertion was one of the risk factors for increased risk of gastrointestinal bleeding, and not the risk factor for development of stress ulcer as stated by Dr. Chongnarungsin.

The decrease in the incidence of stress ulcer-related bleeding in critically ill patients over the years could be attributed to an era effect, from advances in critical care medicine and prophylactic methods.⁵ We agree with Dr. Chongnarungsin that the increased introduction of early enteral feeding may have also contributed to the reduced incidence of stress ulcer-related bleeding.⁶ However, we think the conclusion that “mechanical ventilation for more than 48 hours does not seem to increase the risk of stress ulcer” is overelaborated, and we believe that strong evidence demonstrates this association.^1,2

Alternatively, we recognize the lack of mortality-benefit evidence for stress ulcer prophylaxis. This notwithstanding, according to recent Surviving Sepsis Campaign guidelines, the use of stress ulcer prophylaxis is listed as a 1B recommendation (strong recommendation) for severely septic patients who require prolonged mechanical ventilation. In addition, the updated 2014 guidelines of the American Society of Health-System Pharmacists⁷ continue to recommend stress ulcer prophylaxis in the context of mechanical ventilation, with H2 receptor antagonists being the preferred first-line agents.⁸

It is important to acknowledge that these recommendations were endorsed despite the lack of obvious mortality benefit, and it is our opinion that large randomized controlled studies are needed to evaluate the risks and mortality benefit of these prophylaxis methods.

In Reply: We welcome the comments from Dr. Chongnarungsin on our article and the opportunity to further discuss our opinions.

In our paper, we discussed current recommendations for prophylaxis of stress ulcer-related bleeding in hospitalized patients and advocated against the blind administration of drugs without risk stratification.

The landmark trial that provides the most-cited definitions and the risk factors for clinically significant stress ulcer-related bleeding in critically ill patients was published in 1994 by Cook et al.¹ In their multicenter prospective cohort study of 2,252 patients, the authors reported that prolonged mechanical ventilation is an important risk factor for clinically significant stress ulcer-related bleeding.

Another major prospective cohort study observed an incidence rate of clinically significant stress ulcer-related bleeding of 3.5%.²

Dr. Chongnarungsin cites another prospective cohort study of 183 patients from the same era,³ wherein the authors defined stress ulcer-related bleeding as bleeding requiring transfusion of packed red blood cells, found on endoscopy or on postmortem evaluation. This was in contrast to the 1994 study of Cook et al,¹ who had a more rigorous and comprehensive definition for overt and clinically significant stress ulcer-related bleeding, applied by up to three independent adjudicators not involved in the patients’ care. Their definition not only entailed a more accurate transfusion-dependent bleeding criterion, but also included hemodynamic and laboratory criteria. As such, the “very low rate” of stress ulcer-related bleeding reported by Zandstra et al³ should be critically appraised. Of note, the authors in that study did not report the rates of patients who received early enteral feeding, and their patients received cefotaxime for digestive tract decontamination, an important confounder to the interpretation of the study results.

Indeed, the remarkable variation in estimates of the incidence of stress ulcer-related bleeding is probably related to the lack of a uniform definition. Even when rates of endoscopic and occult bleeding are set aside, agreement is lacking as to which category of bleeding is clinically significant.

Dr. Chongnarungsin also cites the study by Ellison et al⁴ of a cohort of 874 patients who had no previous gastrointestinal bleeding or peptic ulcer disease and who were enrolled in a multicenter randomized controlled trial of prophylactic intravenous immune globulin to prevent infections associated with an intensive care unit. In a secondary objective, the authors did not identify coagulopathy or prolonged mechanical ventilation as a principal risk factor for bleeding. The authors ascribed this discrepancy with previously published literature to their unique study population, which consisted predominantly of elderly men and rarely included trauma patients. In light of these unique peculiarities of their population, the lack of an association between prolonged mechanical ventilation and stress ulcer-related bleeding cannot be determined. Moreover, that study showed that prolonged nasogastric tube insertion was one of the risk factors for increased risk of gastrointestinal bleeding, and not the risk factor for development of stress ulcer as stated by Dr. Chongnarungsin.

The decrease in the incidence of stress ulcer-related bleeding in critically ill patients over the years could be attributed to an era effect, from advances in critical care medicine and prophylactic methods.⁵ We agree with Dr. Chongnarungsin that the increased introduction of early enteral feeding may have also contributed to the reduced incidence of stress ulcer-related bleeding.⁶ However, we think the conclusion that “mechanical ventilation for more than 48 hours does not seem to increase the risk of stress ulcer” is overelaborated, and we believe that strong evidence demonstrates this association.^1,2

Alternatively, we recognize the lack of mortality-benefit evidence for stress ulcer prophylaxis. This notwithstanding, according to recent Surviving Sepsis Campaign guidelines, the use of stress ulcer prophylaxis is listed as a 1B recommendation (strong recommendation) for severely septic patients who require prolonged mechanical ventilation. In addition, the updated 2014 guidelines of the American Society of Health-System Pharmacists⁷ continue to recommend stress ulcer prophylaxis in the context of mechanical ventilation, with H2 receptor antagonists being the preferred first-line agents.⁸

It is important to acknowledge that these recommendations were endorsed despite the lack of obvious mortality benefit, and it is our opinion that large randomized controlled studies are needed to evaluate the risks and mortality benefit of these prophylaxis methods.

In Reply: We welcome the comments from Dr. Chongnarungsin on our article and the opportunity to further discuss our opinions.

In our paper, we discussed current recommendations for prophylaxis of stress ulcer-related bleeding in hospitalized patients and advocated against the blind administration of drugs without risk stratification.

The landmark trial that provides the most-cited definitions and the risk factors for clinically significant stress ulcer-related bleeding in critically ill patients was published in 1994 by Cook et al.¹ In their multicenter prospective cohort study of 2,252 patients, the authors reported that prolonged mechanical ventilation is an important risk factor for clinically significant stress ulcer-related bleeding.

Another major prospective cohort study observed an incidence rate of clinically significant stress ulcer-related bleeding of 3.5%.²

Dr. Chongnarungsin cites another prospective cohort study of 183 patients from the same era,³ wherein the authors defined stress ulcer-related bleeding as bleeding requiring transfusion of packed red blood cells, found on endoscopy or on postmortem evaluation. This was in contrast to the 1994 study of Cook et al,¹ who had a more rigorous and comprehensive definition for overt and clinically significant stress ulcer-related bleeding, applied by up to three independent adjudicators not involved in the patients’ care. Their definition not only entailed a more accurate transfusion-dependent bleeding criterion, but also included hemodynamic and laboratory criteria. As such, the “very low rate” of stress ulcer-related bleeding reported by Zandstra et al³ should be critically appraised. Of note, the authors in that study did not report the rates of patients who received early enteral feeding, and their patients received cefotaxime for digestive tract decontamination, an important confounder to the interpretation of the study results.

Indeed, the remarkable variation in estimates of the incidence of stress ulcer-related bleeding is probably related to the lack of a uniform definition. Even when rates of endoscopic and occult bleeding are set aside, agreement is lacking as to which category of bleeding is clinically significant.

Dr. Chongnarungsin also cites the study by Ellison et al⁴ of a cohort of 874 patients who had no previous gastrointestinal bleeding or peptic ulcer disease and who were enrolled in a multicenter randomized controlled trial of prophylactic intravenous immune globulin to prevent infections associated with an intensive care unit. In a secondary objective, the authors did not identify coagulopathy or prolonged mechanical ventilation as a principal risk factor for bleeding. The authors ascribed this discrepancy with previously published literature to their unique study population, which consisted predominantly of elderly men and rarely included trauma patients. In light of these unique peculiarities of their population, the lack of an association between prolonged mechanical ventilation and stress ulcer-related bleeding cannot be determined. Moreover, that study showed that prolonged nasogastric tube insertion was one of the risk factors for increased risk of gastrointestinal bleeding, and not the risk factor for development of stress ulcer as stated by Dr. Chongnarungsin.

The decrease in the incidence of stress ulcer-related bleeding in critically ill patients over the years could be attributed to an era effect, from advances in critical care medicine and prophylactic methods.⁵ We agree with Dr. Chongnarungsin that the increased introduction of early enteral feeding may have also contributed to the reduced incidence of stress ulcer-related bleeding.⁶ However, we think the conclusion that “mechanical ventilation for more than 48 hours does not seem to increase the risk of stress ulcer” is overelaborated, and we believe that strong evidence demonstrates this association.^1,2

Alternatively, we recognize the lack of mortality-benefit evidence for stress ulcer prophylaxis. This notwithstanding, according to recent Surviving Sepsis Campaign guidelines, the use of stress ulcer prophylaxis is listed as a 1B recommendation (strong recommendation) for severely septic patients who require prolonged mechanical ventilation. In addition, the updated 2014 guidelines of the American Society of Health-System Pharmacists⁷ continue to recommend stress ulcer prophylaxis in the context of mechanical ventilation, with H2 receptor antagonists being the preferred first-line agents.⁸

It is important to acknowledge that these recommendations were endorsed despite the lack of obvious mortality benefit, and it is our opinion that large randomized controlled studies are needed to evaluate the risks and mortality benefit of these prophylaxis methods.

To the Editor: Stephen et al¹ have written a nice review of the implications of albuminuria. However, they are clearly incorrect when they state, “Most of the protein in the urine is albumin filtered from the plasma.”¹ First, as they later point out in the article, the normal upper limit of protein excretion is about 150 mg/day, and only about 20 mg/day is normally albumin. Therefore, most of the protein in normally found in urine is not albumin, but instead is mostly a variety of globulins. Tamm-Horsfall mucoprotein or uromodulin is usually the protein found in highest concentration in normal urine.

To the Editor: Stephen et al¹ have written a nice review of the implications of albuminuria. However, they are clearly incorrect when they state, “Most of the protein in the urine is albumin filtered from the plasma.”¹ First, as they later point out in the article, the normal upper limit of protein excretion is about 150 mg/day, and only about 20 mg/day is normally albumin. Therefore, most of the protein in normally found in urine is not albumin, but instead is mostly a variety of globulins. Tamm-Horsfall mucoprotein or uromodulin is usually the protein found in highest concentration in normal urine.

To the Editor: Stephen et al¹ have written a nice review of the implications of albuminuria. However, they are clearly incorrect when they state, “Most of the protein in the urine is albumin filtered from the plasma.”¹ First, as they later point out in the article, the normal upper limit of protein excretion is about 150 mg/day, and only about 20 mg/day is normally albumin. Therefore, most of the protein in normally found in urine is not albumin, but instead is mostly a variety of globulins. Tamm-Horsfall mucoprotein or uromodulin is usually the protein found in highest concentration in normal urine.

Primum non nocere: First, do no harm—a principle taught across the world to all medical students. It reminds the health care provider to consider the possible harm that any intervention might produce. Never is it more relevant in the mind of a clinician than when prescribing a medication for a pregnant woman. We are, after all, brought up in a society averse to medical risk.

When managing a pregnant patient, should the baby be the highest priority, whatever the mother may face? Or to take the extreme opposite position, should the mother be treated with the best possible options and the baby ignored?

And what about the views of the patient? There is a widespread cultural belief about the vulnerability of the mother and fetus during pregnancy. Therefore, when faced with the decision of whether to use a medication or not, what is the best recourse for the pregnant patient? Should she be the “good mother” and avoid all risk to the baby, or should she be the “responsible mother” who follows medical advice and takes treatment as recommended?

In truth, the path to safe management of a pregnant patient is rarely so dichotomous. In most cases, what is best for the mother is also best for the baby. However, caring for a pregnant or lactating woman can be challenging for clinicians facing insufficient information regarding medication safety, overestimation of the risk of medication by both the patient and the care provider, and increasing litigation costs.

This article provides key principles to guide clinicians caring for pregnant patients, as we find ourselves increasingly dependent on pharmacotherapy. It also includes sources of information clinicians can turn to when they need additional pregnancy safety data about a certain drug and when they want advice about conditions commonly seen in pregnancy and medications that can be justifiably used in those circumstances.

KEY CONCEPTS FOR PRESCRIBING IN PREGNANCY

The following concepts are key to prescribing for a pregnant patient:

No protective barrier exists between the maternal and fetal environments

The placenta contains a semipermeable membrane that selectively allows some substances to pass from the maternal to the fetal blood and excludes others. However, it is not really a “protective mechanism” when it comes to medications. Assume that the fetus will have exposure, at least to some degree.

In general, drugs that are lipophilic, of a low molecular weight, or not ionized at physiologic pH cross the placenta more efficiently than others. Heparin and insulin are notable exceptions to the rule that most drugs cross the placenta. They do not.

The gestational stage may determine the effect of a medication on the fetus

In animals and in humans, exposure of the embryo or fetus to a teratogen may produce a permanent abnormality of structure or function.

First-trimester exposures are most worrisome for structural malformations. However, fetal neurologic and behavioral development, fetal survival, and function of specific organs can be affected even after the first trimester. For example, while first-trimester exposure to angiotensin-converting enzyme inhibitors has been linked to a slight increase in congenital heart defects, exposure in the second or third trimester can result in fetal oligohydramnios, neonatal anuria, pulmonary hypoplasia, intrauterine growth restriction, and fetal death.

Physiologic changes of pregnancy affect the pharmacokinetics of medications

Pregnancy is associated with increased plasma volume, increased glomerular filtration rate, and dilutional hypoalbuminemia, which can all affect the bioavailability of medications. Absorption of oral agents also may be affected by slowed gastric motility in pregnancy.

Although these physiologic alterations do not routinely warrant a change in drug dosage, they may be important considerations when choosing an appropriate agent. For example, medications taken in multiple doses per day are more likely to have a sustained effect than once-daily medications, which would be rapidly cleared in a pregnant patient.

Sole reliance on the FDA pregnancy safety category may be inadequate

To help clinicians prescribe medications for pregnant women, the US Food and Drug Administration (FDA) assigns medications to one of five categories of risk (A, B, C, D, or X) (Table 1). Unfortunately, this classification system has several shortcomings:

• The categories are often seen as a grading system in which the risk increases from the lowest in category A to highest in category X, and the safety information in the accompanying narrative is not always appreciated by prescribers.
• Clinicians incorrectly assume that drugs in a particular category carry a similar risk. However, 65% to 70% of all medications are in category C. This category includes medications with adverse animal data or no animal data at all. In addition, adverse animal data may vary in severity from decreased fetal weight to major structural malformation and fetal loss, indicating a difference in expected risk.
• Most of the data on medication safety in pregnancy comes from animal studies, case reports, case series, case-control studies, or pregnancy registries, and each of these sources has significant limitations.
• The categories do not distinguish between supporting data from animal studies and human studies. For instance, a category-B drug may have animal studies that show no risk but no adequate human studies, or may have animal studies showing risk but human studies that do not.

Looking at the pregnancy risk classifications used in the United States (ie, the FDA system), Australia, and Sweden, researchers compared the classification of 236 drugs between the three systems and found that only one in four drugs was similarly classified into the same risk category. This discrepancy further brings into question the usefulness and reliability of these classifications.¹

Finally, none of the classification systems tells us the potential harm from withholding a medication in pregnancy.

RESOURCES TO ASSESS MEDICATION SAFETY IN PREGNANCY

The FDA has proposed changes in the labeling of medications related to pregnancy and lactation.² The proposed changes would eliminate the current categories and instead require a summary of the risks, the effects of the drug on the fetus, and clinical considerations for use during pregnancy. In addition, labeling would include a description of the medication’s effects on milk production, the amount of drug present in milk, and possible effects on the infant.

Until such changes are in place, what other resources can a busy clinician turn to for support?

The official drug labeling (or the package insert), also published in the Physicians’ Desk Reference, is one source of information, but it rarely provides up-to-date information about teratogenic risks in human pregnancies.

Several online databases review, summarize, and periodically update information from the peer-reviewed medical literature.^3–7 The REPRORISK system^4–7 maintained by Micromedex (Greenwood Village, CO) provides access to several databases that contain information about a wide range of individual medications: REPROTEXT, REPROTOX,⁵ Shepard’s Catalog of Teratogenic Agents,⁷ and the Teratogen Information System (TERIS).⁴ Online access and a smartphone “app” for these databases are available for a subscription fee. Summaries for individual medications can be ordered directly from TERIS, also for a fee. Several other resources are available in textbook format.^8–10

In addition, health care providers can obtain information from or can refer pregnant and breastfeeding patients to a teratology information service for information and counseling about medication exposures. MotherToBaby,¹¹ a service of the nonprofit Organization of Teratology Information Specialists, provides fact sheets, free phone consultation, risk assessment, and counseling by trained teratogen information specialists about environmental exposures, including prescription and over-the-counter medications and dietary and herbal supplements. Counselors from these services gather and synthesize information about exposures from the databases mentioned above, from the peer-reviewed medical literature, from drug manufacturers, and from other sources.

With the advent of electronic medical records and computerized provider order entry, clinical decision support systems hold promise as an additional resource for safe prescribing in pregnancy.

Fortunately, the list of teratogenic medications that are absolutely contraindicated in pregnancy remains small (Table 2).^12,13

THE FOUR-QUESTION APPROACH TO CARING FOR THE PREGNANT PATIENT

Is the symptom self-limited or amenable to nonpharmacologic management?

It has been said that we live in a culture where every symptom warrants a pill. If this is true, there can be no better time for reevaluating this practice than during pregnancy.

Many of the medications most commonly used in pregnancy are for upper-respiratory-tract infections, headache, or psychological distress. Pregnancy is the ideal time to educate patients about the limited effectiveness of most cough-and-cold remedies and the inappropriateness of antibiotics for colds and viral bronchitis. It is also an ideal time for a trial of lifestyle modifications, relaxation, and biofeedback for a chronic headache problem. For cases of mild to moderate depression, it may be worth considering treatment with psychotherapy rather than medications.

Offering patients the option of no treatment or nonpharmacologic treatment for self-limited symptoms is an option worth considering.

How do the patient’s (and your) values and understanding affect the decision?

Is the patient willing to take medication? What are her beliefs with regard to her problem and how it should be managed in pregnancy?

Women and clinicians bring many worries and prejudices to the use of medications in pregnancy. The experiences of the patient and her family and friends may present huge obstacles to needed medication use in pregnancy. Misinformation in the media and from family members, friends, and other health care providers are other obstacles. The only way to deal with this issue is to ask your patient directly about her fears and concerns regarding each prescription written.

Clinicians also need to address fears or prejudices they themselves may have about medication safety in pregnancy. These may arise from a single bad experience in caring for a pregnant woman, discomfort with uncertainty, or a belief that pregnant women should avoid any and all risks of exposures, even when the mother’s condition warrants pharmacologic treatment.

Being informed, both scientifically and about one’s own biases or tendencies, is an essential foundation for rational prescribing in pregnancy.

Is the problem affected by pregnancy, and how?

Pregnancy can affect many medical conditions, and in different ways. Conditions such as asthma, migraine headache, and cardiac arrhythmia are exacerbated in pregnancy, placing the mother and fetus at increased risk of morbidity. Conditions such as Graves disease and hypertension may improve as pregnancy progresses, and medications often can be withdrawn as the patient progresses further along in gestation.

Understanding the effect of pregnancy on a particular problem may help the clinician to make an informed decision about medication use in pregnancy.

How does the problem affect pregnancy?

Considering the risk of untreated disease to the pregnancy may help in decision-making.

Many medical conditions can negatively affect the development of the fetus. A glaring example is diabetes mellitus, with poor glycemic control being linked to congenital malformations, spontaneous abortion, and fetal demise. Chronic conditions with periodic exacerbations such as asthma or epilepsy place the fetus at increased risk during a flare-up.

Therefore, for chronic conditions, continuing maintenance therapy is best. Preconception counseling in such cases is crucial, so that a drug with adequate safety data can be substituted before pregnancy. In this way, any risk to the mother or the embryo from exacerbation of disease as such adjustments are made is avoided.

For conditions arising de novo in pregnancy, the underlying principle remains the same. Is the risk of pharmacotherapy more than the risk of untreated disease? Invariably, the answer to this question supports medication use, and an educated provider will be able to choose a treatment that is justifiable in most circumstances.

CHOOSING A MEDICATION

Fetal well-being depends on maternal well-being. It therefore helps to think of medication use in pregnancy as “justified or not” rather than “safe or not.” Table 3 lists some conditions commonly seen in pregnancy, selected drugs of choice that can be safely used for treating those conditions, and alternates that may be justified in some circumstances.^5,6,14–18

GOOD PRACTICES WHEN PRESCRIBING IN PREGNANCY

Prescribing in pregnancy will be most successful when both the patient and the prescribing physician consider the fetal benefit gained from optimizing maternal health. Good prescribing practices to ensure optimum therapeutic benefit when caring for a pregnant patient are to:

• Involve the patient in decision-making. Recognize her concerns, worries, and preferences regarding her illness and its treatment.
• Inform the patient of the risk of an untreated medical condition, weighed against the risk of medication.
• Choose medications with the most available safety data. Let the patient know what resources you have referred to in choosing the medication.
• It is advisable to perform a search each time a prescription is written for a pregnant or lactating woman.
• When possible, have the discussion in the preconception period.
• Consider the dynamic physiology of gestation. Choose the right drug for the right trimester.
• Discuss the plan with the patient and other providers.
• Define clear criteria for when to discontinue the treatment.

Primum non nocere: First, do no harm—a principle taught across the world to all medical students. It reminds the health care provider to consider the possible harm that any intervention might produce. Never is it more relevant in the mind of a clinician than when prescribing a medication for a pregnant woman. We are, after all, brought up in a society averse to medical risk.

When managing a pregnant patient, should the baby be the highest priority, whatever the mother may face? Or to take the extreme opposite position, should the mother be treated with the best possible options and the baby ignored?

And what about the views of the patient? There is a widespread cultural belief about the vulnerability of the mother and fetus during pregnancy. Therefore, when faced with the decision of whether to use a medication or not, what is the best recourse for the pregnant patient? Should she be the “good mother” and avoid all risk to the baby, or should she be the “responsible mother” who follows medical advice and takes treatment as recommended?

In truth, the path to safe management of a pregnant patient is rarely so dichotomous. In most cases, what is best for the mother is also best for the baby. However, caring for a pregnant or lactating woman can be challenging for clinicians facing insufficient information regarding medication safety, overestimation of the risk of medication by both the patient and the care provider, and increasing litigation costs.

This article provides key principles to guide clinicians caring for pregnant patients, as we find ourselves increasingly dependent on pharmacotherapy. It also includes sources of information clinicians can turn to when they need additional pregnancy safety data about a certain drug and when they want advice about conditions commonly seen in pregnancy and medications that can be justifiably used in those circumstances.

KEY CONCEPTS FOR PRESCRIBING IN PREGNANCY

The following concepts are key to prescribing for a pregnant patient:

No protective barrier exists between the maternal and fetal environments

The placenta contains a semipermeable membrane that selectively allows some substances to pass from the maternal to the fetal blood and excludes others. However, it is not really a “protective mechanism” when it comes to medications. Assume that the fetus will have exposure, at least to some degree.

In general, drugs that are lipophilic, of a low molecular weight, or not ionized at physiologic pH cross the placenta more efficiently than others. Heparin and insulin are notable exceptions to the rule that most drugs cross the placenta. They do not.

The gestational stage may determine the effect of a medication on the fetus

In animals and in humans, exposure of the embryo or fetus to a teratogen may produce a permanent abnormality of structure or function.

First-trimester exposures are most worrisome for structural malformations. However, fetal neurologic and behavioral development, fetal survival, and function of specific organs can be affected even after the first trimester. For example, while first-trimester exposure to angiotensin-converting enzyme inhibitors has been linked to a slight increase in congenital heart defects, exposure in the second or third trimester can result in fetal oligohydramnios, neonatal anuria, pulmonary hypoplasia, intrauterine growth restriction, and fetal death.

Physiologic changes of pregnancy affect the pharmacokinetics of medications

Pregnancy is associated with increased plasma volume, increased glomerular filtration rate, and dilutional hypoalbuminemia, which can all affect the bioavailability of medications. Absorption of oral agents also may be affected by slowed gastric motility in pregnancy.

Although these physiologic alterations do not routinely warrant a change in drug dosage, they may be important considerations when choosing an appropriate agent. For example, medications taken in multiple doses per day are more likely to have a sustained effect than once-daily medications, which would be rapidly cleared in a pregnant patient.

Sole reliance on the FDA pregnancy safety category may be inadequate

To help clinicians prescribe medications for pregnant women, the US Food and Drug Administration (FDA) assigns medications to one of five categories of risk (A, B, C, D, or X) (Table 1). Unfortunately, this classification system has several shortcomings:

• The categories are often seen as a grading system in which the risk increases from the lowest in category A to highest in category X, and the safety information in the accompanying narrative is not always appreciated by prescribers.
• Clinicians incorrectly assume that drugs in a particular category carry a similar risk. However, 65% to 70% of all medications are in category C. This category includes medications with adverse animal data or no animal data at all. In addition, adverse animal data may vary in severity from decreased fetal weight to major structural malformation and fetal loss, indicating a difference in expected risk.
• Most of the data on medication safety in pregnancy comes from animal studies, case reports, case series, case-control studies, or pregnancy registries, and each of these sources has significant limitations.
• The categories do not distinguish between supporting data from animal studies and human studies. For instance, a category-B drug may have animal studies that show no risk but no adequate human studies, or may have animal studies showing risk but human studies that do not.

Looking at the pregnancy risk classifications used in the United States (ie, the FDA system), Australia, and Sweden, researchers compared the classification of 236 drugs between the three systems and found that only one in four drugs was similarly classified into the same risk category. This discrepancy further brings into question the usefulness and reliability of these classifications.¹

Finally, none of the classification systems tells us the potential harm from withholding a medication in pregnancy.

RESOURCES TO ASSESS MEDICATION SAFETY IN PREGNANCY

The FDA has proposed changes in the labeling of medications related to pregnancy and lactation.² The proposed changes would eliminate the current categories and instead require a summary of the risks, the effects of the drug on the fetus, and clinical considerations for use during pregnancy. In addition, labeling would include a description of the medication’s effects on milk production, the amount of drug present in milk, and possible effects on the infant.

Until such changes are in place, what other resources can a busy clinician turn to for support?

The official drug labeling (or the package insert), also published in the Physicians’ Desk Reference, is one source of information, but it rarely provides up-to-date information about teratogenic risks in human pregnancies.

Several online databases review, summarize, and periodically update information from the peer-reviewed medical literature.^3–7 The REPRORISK system^4–7 maintained by Micromedex (Greenwood Village, CO) provides access to several databases that contain information about a wide range of individual medications: REPROTEXT, REPROTOX,⁵ Shepard’s Catalog of Teratogenic Agents,⁷ and the Teratogen Information System (TERIS).⁴ Online access and a smartphone “app” for these databases are available for a subscription fee. Summaries for individual medications can be ordered directly from TERIS, also for a fee. Several other resources are available in textbook format.^8–10

In addition, health care providers can obtain information from or can refer pregnant and breastfeeding patients to a teratology information service for information and counseling about medication exposures. MotherToBaby,¹¹ a service of the nonprofit Organization of Teratology Information Specialists, provides fact sheets, free phone consultation, risk assessment, and counseling by trained teratogen information specialists about environmental exposures, including prescription and over-the-counter medications and dietary and herbal supplements. Counselors from these services gather and synthesize information about exposures from the databases mentioned above, from the peer-reviewed medical literature, from drug manufacturers, and from other sources.

With the advent of electronic medical records and computerized provider order entry, clinical decision support systems hold promise as an additional resource for safe prescribing in pregnancy.

Fortunately, the list of teratogenic medications that are absolutely contraindicated in pregnancy remains small (Table 2).^12,13

THE FOUR-QUESTION APPROACH TO CARING FOR THE PREGNANT PATIENT

Is the symptom self-limited or amenable to nonpharmacologic management?

It has been said that we live in a culture where every symptom warrants a pill. If this is true, there can be no better time for reevaluating this practice than during pregnancy.

Many of the medications most commonly used in pregnancy are for upper-respiratory-tract infections, headache, or psychological distress. Pregnancy is the ideal time to educate patients about the limited effectiveness of most cough-and-cold remedies and the inappropriateness of antibiotics for colds and viral bronchitis. It is also an ideal time for a trial of lifestyle modifications, relaxation, and biofeedback for a chronic headache problem. For cases of mild to moderate depression, it may be worth considering treatment with psychotherapy rather than medications.

Offering patients the option of no treatment or nonpharmacologic treatment for self-limited symptoms is an option worth considering.

How do the patient’s (and your) values and understanding affect the decision?

Is the patient willing to take medication? What are her beliefs with regard to her problem and how it should be managed in pregnancy?

Women and clinicians bring many worries and prejudices to the use of medications in pregnancy. The experiences of the patient and her family and friends may present huge obstacles to needed medication use in pregnancy. Misinformation in the media and from family members, friends, and other health care providers are other obstacles. The only way to deal with this issue is to ask your patient directly about her fears and concerns regarding each prescription written.

Clinicians also need to address fears or prejudices they themselves may have about medication safety in pregnancy. These may arise from a single bad experience in caring for a pregnant woman, discomfort with uncertainty, or a belief that pregnant women should avoid any and all risks of exposures, even when the mother’s condition warrants pharmacologic treatment.

Being informed, both scientifically and about one’s own biases or tendencies, is an essential foundation for rational prescribing in pregnancy.

Is the problem affected by pregnancy, and how?

Pregnancy can affect many medical conditions, and in different ways. Conditions such as asthma, migraine headache, and cardiac arrhythmia are exacerbated in pregnancy, placing the mother and fetus at increased risk of morbidity. Conditions such as Graves disease and hypertension may improve as pregnancy progresses, and medications often can be withdrawn as the patient progresses further along in gestation.

Understanding the effect of pregnancy on a particular problem may help the clinician to make an informed decision about medication use in pregnancy.

How does the problem affect pregnancy?

Considering the risk of untreated disease to the pregnancy may help in decision-making.

Many medical conditions can negatively affect the development of the fetus. A glaring example is diabetes mellitus, with poor glycemic control being linked to congenital malformations, spontaneous abortion, and fetal demise. Chronic conditions with periodic exacerbations such as asthma or epilepsy place the fetus at increased risk during a flare-up.

Therefore, for chronic conditions, continuing maintenance therapy is best. Preconception counseling in such cases is crucial, so that a drug with adequate safety data can be substituted before pregnancy. In this way, any risk to the mother or the embryo from exacerbation of disease as such adjustments are made is avoided.

For conditions arising de novo in pregnancy, the underlying principle remains the same. Is the risk of pharmacotherapy more than the risk of untreated disease? Invariably, the answer to this question supports medication use, and an educated provider will be able to choose a treatment that is justifiable in most circumstances.

CHOOSING A MEDICATION

Fetal well-being depends on maternal well-being. It therefore helps to think of medication use in pregnancy as “justified or not” rather than “safe or not.” Table 3 lists some conditions commonly seen in pregnancy, selected drugs of choice that can be safely used for treating those conditions, and alternates that may be justified in some circumstances.^5,6,14–18

GOOD PRACTICES WHEN PRESCRIBING IN PREGNANCY

Prescribing in pregnancy will be most successful when both the patient and the prescribing physician consider the fetal benefit gained from optimizing maternal health. Good prescribing practices to ensure optimum therapeutic benefit when caring for a pregnant patient are to:

• Involve the patient in decision-making. Recognize her concerns, worries, and preferences regarding her illness and its treatment.
• Inform the patient of the risk of an untreated medical condition, weighed against the risk of medication.
• Choose medications with the most available safety data. Let the patient know what resources you have referred to in choosing the medication.
• It is advisable to perform a search each time a prescription is written for a pregnant or lactating woman.
• When possible, have the discussion in the preconception period.
• Consider the dynamic physiology of gestation. Choose the right drug for the right trimester.
• Discuss the plan with the patient and other providers.
• Define clear criteria for when to discontinue the treatment.

Primum non nocere: First, do no harm—a principle taught across the world to all medical students. It reminds the health care provider to consider the possible harm that any intervention might produce. Never is it more relevant in the mind of a clinician than when prescribing a medication for a pregnant woman. We are, after all, brought up in a society averse to medical risk.

When managing a pregnant patient, should the baby be the highest priority, whatever the mother may face? Or to take the extreme opposite position, should the mother be treated with the best possible options and the baby ignored?

And what about the views of the patient? There is a widespread cultural belief about the vulnerability of the mother and fetus during pregnancy. Therefore, when faced with the decision of whether to use a medication or not, what is the best recourse for the pregnant patient? Should she be the “good mother” and avoid all risk to the baby, or should she be the “responsible mother” who follows medical advice and takes treatment as recommended?

In truth, the path to safe management of a pregnant patient is rarely so dichotomous. In most cases, what is best for the mother is also best for the baby. However, caring for a pregnant or lactating woman can be challenging for clinicians facing insufficient information regarding medication safety, overestimation of the risk of medication by both the patient and the care provider, and increasing litigation costs.

This article provides key principles to guide clinicians caring for pregnant patients, as we find ourselves increasingly dependent on pharmacotherapy. It also includes sources of information clinicians can turn to when they need additional pregnancy safety data about a certain drug and when they want advice about conditions commonly seen in pregnancy and medications that can be justifiably used in those circumstances.

KEY CONCEPTS FOR PRESCRIBING IN PREGNANCY

The following concepts are key to prescribing for a pregnant patient:

No protective barrier exists between the maternal and fetal environments

The placenta contains a semipermeable membrane that selectively allows some substances to pass from the maternal to the fetal blood and excludes others. However, it is not really a “protective mechanism” when it comes to medications. Assume that the fetus will have exposure, at least to some degree.

In general, drugs that are lipophilic, of a low molecular weight, or not ionized at physiologic pH cross the placenta more efficiently than others. Heparin and insulin are notable exceptions to the rule that most drugs cross the placenta. They do not.

The gestational stage may determine the effect of a medication on the fetus

In animals and in humans, exposure of the embryo or fetus to a teratogen may produce a permanent abnormality of structure or function.

First-trimester exposures are most worrisome for structural malformations. However, fetal neurologic and behavioral development, fetal survival, and function of specific organs can be affected even after the first trimester. For example, while first-trimester exposure to angiotensin-converting enzyme inhibitors has been linked to a slight increase in congenital heart defects, exposure in the second or third trimester can result in fetal oligohydramnios, neonatal anuria, pulmonary hypoplasia, intrauterine growth restriction, and fetal death.

Physiologic changes of pregnancy affect the pharmacokinetics of medications

Pregnancy is associated with increased plasma volume, increased glomerular filtration rate, and dilutional hypoalbuminemia, which can all affect the bioavailability of medications. Absorption of oral agents also may be affected by slowed gastric motility in pregnancy.

Although these physiologic alterations do not routinely warrant a change in drug dosage, they may be important considerations when choosing an appropriate agent. For example, medications taken in multiple doses per day are more likely to have a sustained effect than once-daily medications, which would be rapidly cleared in a pregnant patient.

Sole reliance on the FDA pregnancy safety category may be inadequate

To help clinicians prescribe medications for pregnant women, the US Food and Drug Administration (FDA) assigns medications to one of five categories of risk (A, B, C, D, or X) (Table 1). Unfortunately, this classification system has several shortcomings:

• The categories are often seen as a grading system in which the risk increases from the lowest in category A to highest in category X, and the safety information in the accompanying narrative is not always appreciated by prescribers.
• Clinicians incorrectly assume that drugs in a particular category carry a similar risk. However, 65% to 70% of all medications are in category C. This category includes medications with adverse animal data or no animal data at all. In addition, adverse animal data may vary in severity from decreased fetal weight to major structural malformation and fetal loss, indicating a difference in expected risk.
• Most of the data on medication safety in pregnancy comes from animal studies, case reports, case series, case-control studies, or pregnancy registries, and each of these sources has significant limitations.
• The categories do not distinguish between supporting data from animal studies and human studies. For instance, a category-B drug may have animal studies that show no risk but no adequate human studies, or may have animal studies showing risk but human studies that do not.

Looking at the pregnancy risk classifications used in the United States (ie, the FDA system), Australia, and Sweden, researchers compared the classification of 236 drugs between the three systems and found that only one in four drugs was similarly classified into the same risk category. This discrepancy further brings into question the usefulness and reliability of these classifications.¹

Finally, none of the classification systems tells us the potential harm from withholding a medication in pregnancy.

RESOURCES TO ASSESS MEDICATION SAFETY IN PREGNANCY

The FDA has proposed changes in the labeling of medications related to pregnancy and lactation.² The proposed changes would eliminate the current categories and instead require a summary of the risks, the effects of the drug on the fetus, and clinical considerations for use during pregnancy. In addition, labeling would include a description of the medication’s effects on milk production, the amount of drug present in milk, and possible effects on the infant.

Until such changes are in place, what other resources can a busy clinician turn to for support?

The official drug labeling (or the package insert), also published in the Physicians’ Desk Reference, is one source of information, but it rarely provides up-to-date information about teratogenic risks in human pregnancies.

Several online databases review, summarize, and periodically update information from the peer-reviewed medical literature.^3–7 The REPRORISK system^4–7 maintained by Micromedex (Greenwood Village, CO) provides access to several databases that contain information about a wide range of individual medications: REPROTEXT, REPROTOX,⁵ Shepard’s Catalog of Teratogenic Agents,⁷ and the Teratogen Information System (TERIS).⁴ Online access and a smartphone “app” for these databases are available for a subscription fee. Summaries for individual medications can be ordered directly from TERIS, also for a fee. Several other resources are available in textbook format.^8–10

In addition, health care providers can obtain information from or can refer pregnant and breastfeeding patients to a teratology information service for information and counseling about medication exposures. MotherToBaby,¹¹ a service of the nonprofit Organization of Teratology Information Specialists, provides fact sheets, free phone consultation, risk assessment, and counseling by trained teratogen information specialists about environmental exposures, including prescription and over-the-counter medications and dietary and herbal supplements. Counselors from these services gather and synthesize information about exposures from the databases mentioned above, from the peer-reviewed medical literature, from drug manufacturers, and from other sources.

With the advent of electronic medical records and computerized provider order entry, clinical decision support systems hold promise as an additional resource for safe prescribing in pregnancy.

Fortunately, the list of teratogenic medications that are absolutely contraindicated in pregnancy remains small (Table 2).^12,13

THE FOUR-QUESTION APPROACH TO CARING FOR THE PREGNANT PATIENT

Is the symptom self-limited or amenable to nonpharmacologic management?

It has been said that we live in a culture where every symptom warrants a pill. If this is true, there can be no better time for reevaluating this practice than during pregnancy.

Many of the medications most commonly used in pregnancy are for upper-respiratory-tract infections, headache, or psychological distress. Pregnancy is the ideal time to educate patients about the limited effectiveness of most cough-and-cold remedies and the inappropriateness of antibiotics for colds and viral bronchitis. It is also an ideal time for a trial of lifestyle modifications, relaxation, and biofeedback for a chronic headache problem. For cases of mild to moderate depression, it may be worth considering treatment with psychotherapy rather than medications.

Offering patients the option of no treatment or nonpharmacologic treatment for self-limited symptoms is an option worth considering.

How do the patient’s (and your) values and understanding affect the decision?

Is the patient willing to take medication? What are her beliefs with regard to her problem and how it should be managed in pregnancy?

Women and clinicians bring many worries and prejudices to the use of medications in pregnancy. The experiences of the patient and her family and friends may present huge obstacles to needed medication use in pregnancy. Misinformation in the media and from family members, friends, and other health care providers are other obstacles. The only way to deal with this issue is to ask your patient directly about her fears and concerns regarding each prescription written.

Clinicians also need to address fears or prejudices they themselves may have about medication safety in pregnancy. These may arise from a single bad experience in caring for a pregnant woman, discomfort with uncertainty, or a belief that pregnant women should avoid any and all risks of exposures, even when the mother’s condition warrants pharmacologic treatment.

Being informed, both scientifically and about one’s own biases or tendencies, is an essential foundation for rational prescribing in pregnancy.

Is the problem affected by pregnancy, and how?

Pregnancy can affect many medical conditions, and in different ways. Conditions such as asthma, migraine headache, and cardiac arrhythmia are exacerbated in pregnancy, placing the mother and fetus at increased risk of morbidity. Conditions such as Graves disease and hypertension may improve as pregnancy progresses, and medications often can be withdrawn as the patient progresses further along in gestation.

Understanding the effect of pregnancy on a particular problem may help the clinician to make an informed decision about medication use in pregnancy.

How does the problem affect pregnancy?

Considering the risk of untreated disease to the pregnancy may help in decision-making.

Many medical conditions can negatively affect the development of the fetus. A glaring example is diabetes mellitus, with poor glycemic control being linked to congenital malformations, spontaneous abortion, and fetal demise. Chronic conditions with periodic exacerbations such as asthma or epilepsy place the fetus at increased risk during a flare-up.

Therefore, for chronic conditions, continuing maintenance therapy is best. Preconception counseling in such cases is crucial, so that a drug with adequate safety data can be substituted before pregnancy. In this way, any risk to the mother or the embryo from exacerbation of disease as such adjustments are made is avoided.

For conditions arising de novo in pregnancy, the underlying principle remains the same. Is the risk of pharmacotherapy more than the risk of untreated disease? Invariably, the answer to this question supports medication use, and an educated provider will be able to choose a treatment that is justifiable in most circumstances.

CHOOSING A MEDICATION

Fetal well-being depends on maternal well-being. It therefore helps to think of medication use in pregnancy as “justified or not” rather than “safe or not.” Table 3 lists some conditions commonly seen in pregnancy, selected drugs of choice that can be safely used for treating those conditions, and alternates that may be justified in some circumstances.^5,6,14–18

GOOD PRACTICES WHEN PRESCRIBING IN PREGNANCY

Prescribing in pregnancy will be most successful when both the patient and the prescribing physician consider the fetal benefit gained from optimizing maternal health. Good prescribing practices to ensure optimum therapeutic benefit when caring for a pregnant patient are to:

• Involve the patient in decision-making. Recognize her concerns, worries, and preferences regarding her illness and its treatment.
• Inform the patient of the risk of an untreated medical condition, weighed against the risk of medication.
• Choose medications with the most available safety data. Let the patient know what resources you have referred to in choosing the medication.
• It is advisable to perform a search each time a prescription is written for a pregnant or lactating woman.
• When possible, have the discussion in the preconception period.
• Consider the dynamic physiology of gestation. Choose the right drug for the right trimester.
• Discuss the plan with the patient and other providers.
• Define clear criteria for when to discontinue the treatment.

A 32-year-old woman presents with a history of pelvic pain, dysmenorrhea, dyspareunia, dyschezia, and dysuria, with exacerbation of the symptoms during her menstrual cycles. Her menarche occurred at the age of 13 and her menses are regular. She has never undergone surgery and has no relevant pathologic processes. She also reports that for the past 18 months she has been unsuccessfully trying to conceive.

Two months ago, she went to the emergency department because of an acute episode of severe pelvic pain associated with abdominal cramps, vomiting, and dyschezia, occurring at the beginning of her menstrual cycle. At that time, her vital signs were within normal limits, but deep palpation of the right iliac fossa was painful. On that occasion, acute abdomen and bowel obstruction were excluded.

Now, vaginal examination reveals a bluish, painful, bulky induration in the posterior fornix. Digital rectal examination reveals a circular infiltrated area in the anterior rectal wall. Her cancer antigen 125 (CA 125) level is 230 U/mL (normal range 0–35 U/mL).

MENSES-RELATED SYMPTOMS AND THE DIAGNOSIS OF ENDOMETRIOSIS

The diagnosis of endometriosis should be considered in the patient described above. Many of her signs and symptoms can be associated with several diseases. However, the diagnostic hypothesis points strongly toward endometriosis, since her symptoms recur at the beginning of every menstrual cycle.¹

Endometriosis is the presence of endometrial tissue outside the uterine cavity. The affected organs usually include the ovaries, fallopian tubes,² peritoneal surface, vagina, cervix, abdominal wall,³ scar tissue, pouch of Douglas, urinary tract, and bowel. However, any organ can be involved.

So-called deeply infiltrating endometriosis is an endometriotic lesion penetrating into the retroperitoneal space (most often affecting the uterosacral ligaments and the rectovaginal septum) or the pelvic-organ wall to a depth of at least 5 mm and involving structures such as the rectum, vagina, ureters, and bladder.⁴ Its clinical presentation is highly variable, ranging from no symptoms to severe pain and dysfunction of pelvic organs.

Endometriosis can be diagnosed with certainty only when the endometriotic lesions are observed by laparoscopy or laparotomy and after the histologic examination of surgically resected lesions (Figure 1).¹ However, a presumptive diagnosis can be made on the basis of imaging findings, which can be useful in the differential diagnostic process (Table 1).

EXAMINATION AND BLOOD MARKERS PROVIDE LIMITED INFORMATION

Knowing the history of the patient, along with a physical examination that includes speculum and bimanual vaginal and rectal examination, can be helpful in the diagnostic process even if nothing abnormal is found.

Pelvic examination has a poor predictive value, as demonstrated in a study conducted by Nezhat et al⁵ in 91 patients with surgically confirmed endometriosis, 47% of whom had a normal bimanual examination.

CA 125 is the serologic marker most often used for diagnosing endometriosis. Levels are usually high in the sera of patients with endometriosis, especially in the advanced stages.⁶ However, levels increase both in the physiologic menstrual cycle and in epithelial ovarian cancers.⁷ Thus, the diagnostic value of CA 125 is limited in terms of both sensitivity and specificity.

INCLUDE IMAGING IN THE DIAGNOSTIC WORKUP

Surgical treatment is frequently offered to patients who have severe pelvic pain that does not respond to medical treatment, or in cases of infertility. Imaging investigations are mandatory both to ascertain the diagnosis and to assess involvement of internal organs before surgery. Moreover, imaging helps minimize the surgical risks.

The primary aim of the radiologic examination is to describe the precise location, the depth, and the number of pelvic endometriotic lesions. Furthermore, imaging is useful to check for endometriotic foci in pelvic organs such as the bowel, ureters, and bladder, which are often involved in the pathologic process.

Transvaginal ultrasonography and magnetic resonance imaging (MRI) can accurately delineate deeply infiltrating lesions of endometriosis that are not easily accessible laparoscopically.

Transvaginal ultrasonography

Transvaginal ultrasonography is the first-line imaging study when endometriosis is suspected: it is powerful, simple, widely available, and cost-effective. In particular, it is recommended for diagnosing endometriotic ovarian cysts (endometriomas)^8,9 and endometriosis of the bladder.¹⁰ However, its value for the assessment of superficial peritoneal lesions, ovarian foci, and deeply infiltrating endometriosis is questionable.

Although uncomfortable for the patient, transvaginal ultrasonography should be performed during menses, or when the pain reaches its highest level. In fact, during menstrual bleeding the endometrial implants grow and become easier to detect.

Mais et al⁸ reported that transvaginal ultrasonography has a sensitivity of 88% in differentiating endometriomas from other ovarian masses, and a specificity of 90% (Figure 2). Furthermore, its specificity is as high as that of MRI.^8,9

Endometriotic nodules detected in the uterosacral ligaments, rectovaginal septum, vagina, vesicouterine pouch, bladder (Figure 3), and ureters can be signs of deeply infiltrating endometriosis. Pelvic adhesions can be suspected when pelvic organs appear fixed to each other, when hyperechogenic plaques are found between the serosal surfaces of the different organs, and when the pouch of Douglas is partially or completely obliterated.

The accuracy of transvaginal ultrasonography strongly depends on the operator’s skill. Furthermore, lesions of the sigmoid colon are impossible to visualize by transvaginal ultrasonography; hence, further diagnostic procedures are required. Transvaginal ultrasonography is the most accurate technique in detecting endometriotic nodules of the bladder wall in patients with urinary symptoms.

Transvaginal ultrasonography combined with color Doppler can also demonstrate the flow of urine through the ureters to the bladder, thereby ascertaining the patency of the ureters and clarifying the anatomic relationship between the ureters and any endometriotic lesions in the detrusors.¹⁰ Hydronephrosis can arise from ureteral restriction caused by endometriotic nodules. Thus, transabdominal ultrasonography of the kidneys is always recommended when deeply infiltrating endometriosis is suspected.

Some centers use a bowel-preparation protocol consisting of a laxative taken 24 hours before the procedure, combined with a low-residue diet and an enema 1 hour before the examination to cleanse the rectosigmoid colon of fecal content and gas, which can interfere with the visual examination of the pelvic structures.¹¹

Transabdominal ultrasonography

Transabdominal ultrasonography can be used instead of transvaginal ultrasonography, eg, in young girls and women who have never been sexually active. When transabdominal ultrasonography is selected, the patient should have a full bladder to maximize the visualization of the pelvic structures. However, transvaginal ultrasonography is generally more sensitive than transabdominal in detecting adnexal masses and pelvic nodules.¹²

Magnetic resonance imaging

MRI has been recently introduced in the diagnosis of endometriosis. MRI is less operator-dependent than transvaginal ultrasonography and is more sensitive for detecting foci of deeply infiltrating endometriosis, because of its ability to completely survey the anterior and posterior compartments of the pelvis. However, its diagnostic value in cases of bladder endometriosis, superficial peritoneal lesions, and ovarian foci is still controversial.^13–16

On MRI, lesions of deeply infiltrating endometriosis mainly appear as areas or nodules with regular, irregular, indistinct, or stellate margins. A distortion of the normal pelvic anatomy or the detection of a loculated fluid collection can indirectly signal the presence of adhesions.

MRI has high specificity for the diagnosis of endometriomas as a result of its ability to detect aged hemorrhagic content (Figure 4).¹⁷ Despite the many studies that point to the limits of MRI in detecting small endometriotic lesions, recent studies demonstrated that MRI also has good sensitivity for small peritoneal implants and adhesions.^18,19 The injection of gadolinium contrast is still a debatable measure, because contrast-enhanced imaging cannot differentiate infiltrating lesions from other normal fibromuscular pelvic anatomic structures.^15,20

Bowel preparation can be done with an oral laxative the day before imaging, complemented by a low-residue diet. A single dose of a ready-to-use enema is given 30 minutes before the examination to cleanse the terminal section of the intestinal tract. To avoid motion artifacts caused by bowel peristalsis, images are obtained after intramuscular injections of a myorelaxant are given, if there is no contraindication. Bowel preparation is useful to eliminate fecal residue and gas, thereby allowing proper visualization of lesions of deeply infiltrating endometriosis, but it is not routinely prescribed in all centers.¹¹

In most cases, endometriotic lesions have an MRI signal intensity that comes very close to that of the surrounding fibromuscular structures. In this regard, vaginal and rectal distention and opacification using ultrasonographic gel clearly help to delineate the cervix, vaginal fornices, and vaginal wall, as well as the rectum and wall of the rectosigmoid junction (Figure  5).²⁰

PRESURGICAL IMAGING

Rectal endoscopic ultrasonography

Even though it should not be included in the routine diagnostic workup, rectal endoscopic ultrasonography, using a flexible echoendoscope, is suitable in certain presurgical cases. The aim of this imaging technique is to assess the depth of bowel wall infiltration thanks to the visualization of the different layers.²¹

Double-contrast barium enema and multislice computed tomography

Double-contrast barium enema is extensively used for the diagnosis of bowel endometriosis, once the decision to perform surgery has been made. It allows evaluation of the degree and length of the bowel occlusion at the level of the sigmoid or high rectosigmoid tract, but it does not permit differentiation of bowel endometriosis from other pathologies.

Multislice computed tomography offers the opportunity to evaluate the depth of the lesions with excellent precision. ²²

The most relevant disadvantage of both procedures is the exposure of women of reproductive age to ionizing radiation. In addition, multislice computed tomography requires the administration of an intravenous iodinated contrast medium and a retrograde colonic distention with about 2 L of water.

A 32-year-old woman presents with a history of pelvic pain, dysmenorrhea, dyspareunia, dyschezia, and dysuria, with exacerbation of the symptoms during her menstrual cycles. Her menarche occurred at the age of 13 and her menses are regular. She has never undergone surgery and has no relevant pathologic processes. She also reports that for the past 18 months she has been unsuccessfully trying to conceive.

Two months ago, she went to the emergency department because of an acute episode of severe pelvic pain associated with abdominal cramps, vomiting, and dyschezia, occurring at the beginning of her menstrual cycle. At that time, her vital signs were within normal limits, but deep palpation of the right iliac fossa was painful. On that occasion, acute abdomen and bowel obstruction were excluded.

Now, vaginal examination reveals a bluish, painful, bulky induration in the posterior fornix. Digital rectal examination reveals a circular infiltrated area in the anterior rectal wall. Her cancer antigen 125 (CA 125) level is 230 U/mL (normal range 0–35 U/mL).

MENSES-RELATED SYMPTOMS AND THE DIAGNOSIS OF ENDOMETRIOSIS

The diagnosis of endometriosis should be considered in the patient described above. Many of her signs and symptoms can be associated with several diseases. However, the diagnostic hypothesis points strongly toward endometriosis, since her symptoms recur at the beginning of every menstrual cycle.¹

Endometriosis is the presence of endometrial tissue outside the uterine cavity. The affected organs usually include the ovaries, fallopian tubes,² peritoneal surface, vagina, cervix, abdominal wall,³ scar tissue, pouch of Douglas, urinary tract, and bowel. However, any organ can be involved.

So-called deeply infiltrating endometriosis is an endometriotic lesion penetrating into the retroperitoneal space (most often affecting the uterosacral ligaments and the rectovaginal septum) or the pelvic-organ wall to a depth of at least 5 mm and involving structures such as the rectum, vagina, ureters, and bladder.⁴ Its clinical presentation is highly variable, ranging from no symptoms to severe pain and dysfunction of pelvic organs.

Endometriosis can be diagnosed with certainty only when the endometriotic lesions are observed by laparoscopy or laparotomy and after the histologic examination of surgically resected lesions (Figure 1).¹ However, a presumptive diagnosis can be made on the basis of imaging findings, which can be useful in the differential diagnostic process (Table 1).

EXAMINATION AND BLOOD MARKERS PROVIDE LIMITED INFORMATION

Knowing the history of the patient, along with a physical examination that includes speculum and bimanual vaginal and rectal examination, can be helpful in the diagnostic process even if nothing abnormal is found.

Pelvic examination has a poor predictive value, as demonstrated in a study conducted by Nezhat et al⁵ in 91 patients with surgically confirmed endometriosis, 47% of whom had a normal bimanual examination.

CA 125 is the serologic marker most often used for diagnosing endometriosis. Levels are usually high in the sera of patients with endometriosis, especially in the advanced stages.⁶ However, levels increase both in the physiologic menstrual cycle and in epithelial ovarian cancers.⁷ Thus, the diagnostic value of CA 125 is limited in terms of both sensitivity and specificity.

INCLUDE IMAGING IN THE DIAGNOSTIC WORKUP

Surgical treatment is frequently offered to patients who have severe pelvic pain that does not respond to medical treatment, or in cases of infertility. Imaging investigations are mandatory both to ascertain the diagnosis and to assess involvement of internal organs before surgery. Moreover, imaging helps minimize the surgical risks.

The primary aim of the radiologic examination is to describe the precise location, the depth, and the number of pelvic endometriotic lesions. Furthermore, imaging is useful to check for endometriotic foci in pelvic organs such as the bowel, ureters, and bladder, which are often involved in the pathologic process.

Transvaginal ultrasonography and magnetic resonance imaging (MRI) can accurately delineate deeply infiltrating lesions of endometriosis that are not easily accessible laparoscopically.

Transvaginal ultrasonography

Transvaginal ultrasonography is the first-line imaging study when endometriosis is suspected: it is powerful, simple, widely available, and cost-effective. In particular, it is recommended for diagnosing endometriotic ovarian cysts (endometriomas)^8,9 and endometriosis of the bladder.¹⁰ However, its value for the assessment of superficial peritoneal lesions, ovarian foci, and deeply infiltrating endometriosis is questionable.

Although uncomfortable for the patient, transvaginal ultrasonography should be performed during menses, or when the pain reaches its highest level. In fact, during menstrual bleeding the endometrial implants grow and become easier to detect.

Mais et al⁸ reported that transvaginal ultrasonography has a sensitivity of 88% in differentiating endometriomas from other ovarian masses, and a specificity of 90% (Figure 2). Furthermore, its specificity is as high as that of MRI.^8,9

Endometriotic nodules detected in the uterosacral ligaments, rectovaginal septum, vagina, vesicouterine pouch, bladder (Figure 3), and ureters can be signs of deeply infiltrating endometriosis. Pelvic adhesions can be suspected when pelvic organs appear fixed to each other, when hyperechogenic plaques are found between the serosal surfaces of the different organs, and when the pouch of Douglas is partially or completely obliterated.

The accuracy of transvaginal ultrasonography strongly depends on the operator’s skill. Furthermore, lesions of the sigmoid colon are impossible to visualize by transvaginal ultrasonography; hence, further diagnostic procedures are required. Transvaginal ultrasonography is the most accurate technique in detecting endometriotic nodules of the bladder wall in patients with urinary symptoms.

Transvaginal ultrasonography combined with color Doppler can also demonstrate the flow of urine through the ureters to the bladder, thereby ascertaining the patency of the ureters and clarifying the anatomic relationship between the ureters and any endometriotic lesions in the detrusors.¹⁰ Hydronephrosis can arise from ureteral restriction caused by endometriotic nodules. Thus, transabdominal ultrasonography of the kidneys is always recommended when deeply infiltrating endometriosis is suspected.

Some centers use a bowel-preparation protocol consisting of a laxative taken 24 hours before the procedure, combined with a low-residue diet and an enema 1 hour before the examination to cleanse the rectosigmoid colon of fecal content and gas, which can interfere with the visual examination of the pelvic structures.¹¹

Transabdominal ultrasonography

Transabdominal ultrasonography can be used instead of transvaginal ultrasonography, eg, in young girls and women who have never been sexually active. When transabdominal ultrasonography is selected, the patient should have a full bladder to maximize the visualization of the pelvic structures. However, transvaginal ultrasonography is generally more sensitive than transabdominal in detecting adnexal masses and pelvic nodules.¹²

Magnetic resonance imaging

MRI has been recently introduced in the diagnosis of endometriosis. MRI is less operator-dependent than transvaginal ultrasonography and is more sensitive for detecting foci of deeply infiltrating endometriosis, because of its ability to completely survey the anterior and posterior compartments of the pelvis. However, its diagnostic value in cases of bladder endometriosis, superficial peritoneal lesions, and ovarian foci is still controversial.^13–16

On MRI, lesions of deeply infiltrating endometriosis mainly appear as areas or nodules with regular, irregular, indistinct, or stellate margins. A distortion of the normal pelvic anatomy or the detection of a loculated fluid collection can indirectly signal the presence of adhesions.

MRI has high specificity for the diagnosis of endometriomas as a result of its ability to detect aged hemorrhagic content (Figure 4).¹⁷ Despite the many studies that point to the limits of MRI in detecting small endometriotic lesions, recent studies demonstrated that MRI also has good sensitivity for small peritoneal implants and adhesions.^18,19 The injection of gadolinium contrast is still a debatable measure, because contrast-enhanced imaging cannot differentiate infiltrating lesions from other normal fibromuscular pelvic anatomic structures.^15,20

Bowel preparation can be done with an oral laxative the day before imaging, complemented by a low-residue diet. A single dose of a ready-to-use enema is given 30 minutes before the examination to cleanse the terminal section of the intestinal tract. To avoid motion artifacts caused by bowel peristalsis, images are obtained after intramuscular injections of a myorelaxant are given, if there is no contraindication. Bowel preparation is useful to eliminate fecal residue and gas, thereby allowing proper visualization of lesions of deeply infiltrating endometriosis, but it is not routinely prescribed in all centers.¹¹

In most cases, endometriotic lesions have an MRI signal intensity that comes very close to that of the surrounding fibromuscular structures. In this regard, vaginal and rectal distention and opacification using ultrasonographic gel clearly help to delineate the cervix, vaginal fornices, and vaginal wall, as well as the rectum and wall of the rectosigmoid junction (Figure  5).²⁰

PRESURGICAL IMAGING

Rectal endoscopic ultrasonography

Even though it should not be included in the routine diagnostic workup, rectal endoscopic ultrasonography, using a flexible echoendoscope, is suitable in certain presurgical cases. The aim of this imaging technique is to assess the depth of bowel wall infiltration thanks to the visualization of the different layers.²¹

Double-contrast barium enema and multislice computed tomography

Double-contrast barium enema is extensively used for the diagnosis of bowel endometriosis, once the decision to perform surgery has been made. It allows evaluation of the degree and length of the bowel occlusion at the level of the sigmoid or high rectosigmoid tract, but it does not permit differentiation of bowel endometriosis from other pathologies.

Multislice computed tomography offers the opportunity to evaluate the depth of the lesions with excellent precision. ²²

The most relevant disadvantage of both procedures is the exposure of women of reproductive age to ionizing radiation. In addition, multislice computed tomography requires the administration of an intravenous iodinated contrast medium and a retrograde colonic distention with about 2 L of water.

A 32-year-old woman presents with a history of pelvic pain, dysmenorrhea, dyspareunia, dyschezia, and dysuria, with exacerbation of the symptoms during her menstrual cycles. Her menarche occurred at the age of 13 and her menses are regular. She has never undergone surgery and has no relevant pathologic processes. She also reports that for the past 18 months she has been unsuccessfully trying to conceive.

Two months ago, she went to the emergency department because of an acute episode of severe pelvic pain associated with abdominal cramps, vomiting, and dyschezia, occurring at the beginning of her menstrual cycle. At that time, her vital signs were within normal limits, but deep palpation of the right iliac fossa was painful. On that occasion, acute abdomen and bowel obstruction were excluded.

Now, vaginal examination reveals a bluish, painful, bulky induration in the posterior fornix. Digital rectal examination reveals a circular infiltrated area in the anterior rectal wall. Her cancer antigen 125 (CA 125) level is 230 U/mL (normal range 0–35 U/mL).

MENSES-RELATED SYMPTOMS AND THE DIAGNOSIS OF ENDOMETRIOSIS

The diagnosis of endometriosis should be considered in the patient described above. Many of her signs and symptoms can be associated with several diseases. However, the diagnostic hypothesis points strongly toward endometriosis, since her symptoms recur at the beginning of every menstrual cycle.¹

Endometriosis is the presence of endometrial tissue outside the uterine cavity. The affected organs usually include the ovaries, fallopian tubes,² peritoneal surface, vagina, cervix, abdominal wall,³ scar tissue, pouch of Douglas, urinary tract, and bowel. However, any organ can be involved.

So-called deeply infiltrating endometriosis is an endometriotic lesion penetrating into the retroperitoneal space (most often affecting the uterosacral ligaments and the rectovaginal septum) or the pelvic-organ wall to a depth of at least 5 mm and involving structures such as the rectum, vagina, ureters, and bladder.⁴ Its clinical presentation is highly variable, ranging from no symptoms to severe pain and dysfunction of pelvic organs.

Endometriosis can be diagnosed with certainty only when the endometriotic lesions are observed by laparoscopy or laparotomy and after the histologic examination of surgically resected lesions (Figure 1).¹ However, a presumptive diagnosis can be made on the basis of imaging findings, which can be useful in the differential diagnostic process (Table 1).

EXAMINATION AND BLOOD MARKERS PROVIDE LIMITED INFORMATION

Knowing the history of the patient, along with a physical examination that includes speculum and bimanual vaginal and rectal examination, can be helpful in the diagnostic process even if nothing abnormal is found.

Pelvic examination has a poor predictive value, as demonstrated in a study conducted by Nezhat et al⁵ in 91 patients with surgically confirmed endometriosis, 47% of whom had a normal bimanual examination.

CA 125 is the serologic marker most often used for diagnosing endometriosis. Levels are usually high in the sera of patients with endometriosis, especially in the advanced stages.⁶ However, levels increase both in the physiologic menstrual cycle and in epithelial ovarian cancers.⁷ Thus, the diagnostic value of CA 125 is limited in terms of both sensitivity and specificity.

INCLUDE IMAGING IN THE DIAGNOSTIC WORKUP

Surgical treatment is frequently offered to patients who have severe pelvic pain that does not respond to medical treatment, or in cases of infertility. Imaging investigations are mandatory both to ascertain the diagnosis and to assess involvement of internal organs before surgery. Moreover, imaging helps minimize the surgical risks.

The primary aim of the radiologic examination is to describe the precise location, the depth, and the number of pelvic endometriotic lesions. Furthermore, imaging is useful to check for endometriotic foci in pelvic organs such as the bowel, ureters, and bladder, which are often involved in the pathologic process.

Transvaginal ultrasonography and magnetic resonance imaging (MRI) can accurately delineate deeply infiltrating lesions of endometriosis that are not easily accessible laparoscopically.

Transvaginal ultrasonography

Transvaginal ultrasonography is the first-line imaging study when endometriosis is suspected: it is powerful, simple, widely available, and cost-effective. In particular, it is recommended for diagnosing endometriotic ovarian cysts (endometriomas)^8,9 and endometriosis of the bladder.¹⁰ However, its value for the assessment of superficial peritoneal lesions, ovarian foci, and deeply infiltrating endometriosis is questionable.

Although uncomfortable for the patient, transvaginal ultrasonography should be performed during menses, or when the pain reaches its highest level. In fact, during menstrual bleeding the endometrial implants grow and become easier to detect.

Mais et al⁸ reported that transvaginal ultrasonography has a sensitivity of 88% in differentiating endometriomas from other ovarian masses, and a specificity of 90% (Figure 2). Furthermore, its specificity is as high as that of MRI.^8,9

Endometriotic nodules detected in the uterosacral ligaments, rectovaginal septum, vagina, vesicouterine pouch, bladder (Figure 3), and ureters can be signs of deeply infiltrating endometriosis. Pelvic adhesions can be suspected when pelvic organs appear fixed to each other, when hyperechogenic plaques are found between the serosal surfaces of the different organs, and when the pouch of Douglas is partially or completely obliterated.

The accuracy of transvaginal ultrasonography strongly depends on the operator’s skill. Furthermore, lesions of the sigmoid colon are impossible to visualize by transvaginal ultrasonography; hence, further diagnostic procedures are required. Transvaginal ultrasonography is the most accurate technique in detecting endometriotic nodules of the bladder wall in patients with urinary symptoms.

Transvaginal ultrasonography combined with color Doppler can also demonstrate the flow of urine through the ureters to the bladder, thereby ascertaining the patency of the ureters and clarifying the anatomic relationship between the ureters and any endometriotic lesions in the detrusors.¹⁰ Hydronephrosis can arise from ureteral restriction caused by endometriotic nodules. Thus, transabdominal ultrasonography of the kidneys is always recommended when deeply infiltrating endometriosis is suspected.

Some centers use a bowel-preparation protocol consisting of a laxative taken 24 hours before the procedure, combined with a low-residue diet and an enema 1 hour before the examination to cleanse the rectosigmoid colon of fecal content and gas, which can interfere with the visual examination of the pelvic structures.¹¹

Transabdominal ultrasonography

Transabdominal ultrasonography can be used instead of transvaginal ultrasonography, eg, in young girls and women who have never been sexually active. When transabdominal ultrasonography is selected, the patient should have a full bladder to maximize the visualization of the pelvic structures. However, transvaginal ultrasonography is generally more sensitive than transabdominal in detecting adnexal masses and pelvic nodules.¹²

Magnetic resonance imaging

MRI has been recently introduced in the diagnosis of endometriosis. MRI is less operator-dependent than transvaginal ultrasonography and is more sensitive for detecting foci of deeply infiltrating endometriosis, because of its ability to completely survey the anterior and posterior compartments of the pelvis. However, its diagnostic value in cases of bladder endometriosis, superficial peritoneal lesions, and ovarian foci is still controversial.^13–16

On MRI, lesions of deeply infiltrating endometriosis mainly appear as areas or nodules with regular, irregular, indistinct, or stellate margins. A distortion of the normal pelvic anatomy or the detection of a loculated fluid collection can indirectly signal the presence of adhesions.

MRI has high specificity for the diagnosis of endometriomas as a result of its ability to detect aged hemorrhagic content (Figure 4).¹⁷ Despite the many studies that point to the limits of MRI in detecting small endometriotic lesions, recent studies demonstrated that MRI also has good sensitivity for small peritoneal implants and adhesions.^18,19 The injection of gadolinium contrast is still a debatable measure, because contrast-enhanced imaging cannot differentiate infiltrating lesions from other normal fibromuscular pelvic anatomic structures.^15,20

Bowel preparation can be done with an oral laxative the day before imaging, complemented by a low-residue diet. A single dose of a ready-to-use enema is given 30 minutes before the examination to cleanse the terminal section of the intestinal tract. To avoid motion artifacts caused by bowel peristalsis, images are obtained after intramuscular injections of a myorelaxant are given, if there is no contraindication. Bowel preparation is useful to eliminate fecal residue and gas, thereby allowing proper visualization of lesions of deeply infiltrating endometriosis, but it is not routinely prescribed in all centers.¹¹

In most cases, endometriotic lesions have an MRI signal intensity that comes very close to that of the surrounding fibromuscular structures. In this regard, vaginal and rectal distention and opacification using ultrasonographic gel clearly help to delineate the cervix, vaginal fornices, and vaginal wall, as well as the rectum and wall of the rectosigmoid junction (Figure  5).²⁰

PRESURGICAL IMAGING

Rectal endoscopic ultrasonography

Even though it should not be included in the routine diagnostic workup, rectal endoscopic ultrasonography, using a flexible echoendoscope, is suitable in certain presurgical cases. The aim of this imaging technique is to assess the depth of bowel wall infiltration thanks to the visualization of the different layers.²¹

Double-contrast barium enema and multislice computed tomography

Double-contrast barium enema is extensively used for the diagnosis of bowel endometriosis, once the decision to perform surgery has been made. It allows evaluation of the degree and length of the bowel occlusion at the level of the sigmoid or high rectosigmoid tract, but it does not permit differentiation of bowel endometriosis from other pathologies.

Multislice computed tomography offers the opportunity to evaluate the depth of the lesions with excellent precision. ²²

The most relevant disadvantage of both procedures is the exposure of women of reproductive age to ionizing radiation. In addition, multislice computed tomography requires the administration of an intravenous iodinated contrast medium and a retrograde colonic distention with about 2 L of water.

Managing a patient with chronic alcohol abuse who is beginning to withdraw is a situation in which to expect the unexpected.

A 61-year-old man with a 40-year history of alcohol abuse was admitted to the hospital after presenting with a 6-month history of weakness and increasing frequency of falls, as well as a 2-day history of myalgias and fatigue. He had anorexia and chronic diarrhea, and he reported an unintentional 100-lb weight loss over the past year. He consumed at least three to five alcoholic drinks daily, including a daily “eye-opener,” and his diet was essentially devoid of meat, bread, fruits, and vegetables. His last drink had been on the morning of admission.

On physical examination, his vital signs were notable for marked orthostatic hypotension. He had angular cheilitis, an erythematous, scaly facial rash, and poorly healing severe sunburn on the dorsal surface of the left forearm, acquired 1 month earlier after minimal sun exposure while driving his car (Figure 1).

The neurologic examination noted decreased sensation in a stocking distribution, reduced proprioception in both great toes, and an intention tremor. His upper extremities were slightly hyperreflexic, he had down-going toes, he did not have clonus, and his gait was wide-based.

Laboratory results were notable for mild anemia, with an elevated mean corpuscular volume of 109.2 fL and a normal thyrotropin level. Urinalysis and urine culture were normal. Stool testing for Clostridium difficile toxin was negative. Computed tomography of the head was unremarkable.

He was placed on the Clinical Institute Withdrawal Assessment for Alcohol revised protocol,1 and he was started on careful fluid, electrolyte, and nutritional management, including intravenous supplementation with thiamine 100 mg daily, folic acid 1 mg twice daily, and an oral multivitamin. Despite these interventions, he became delirious and developed horizontal nystagmus, visual hallucinations, dizziness, and increasing weakness. Benzodiazepine toxicity, which may cause similar findings, was thought unlikely, as he had received only a maximum of 4 mg of lorazepam in any single 24-hour period. His ongoing confusion, orthostatic hypotension, ataxia, and nystagmus raised concern for Wernicke encephalopathy. Thiamine was increased to 500 mg intravenously three times daily for 2 days, followed by 500 mg daily for 2 days, as recommended for acute Wernicke encephalopathy by the Royal College of Physicians.2

Although his ocular findings resolved and his blood pressure improved with high-dose thiamine, he remained confused. Other causes of delirium were considered, including other micronutrient deficiencies. He was evaluated for magnesium deficiency, as magnesium is a necessary cofactor in thiamine metabolism,³ but his level was within normal limits. The constellation of symptoms, including persistent delirium, diarrhea, and skin findings, appeared to be consistent with pellagra, and niacin deficiency became a concern.

PELLAGRA AND ALCOHOL ABUSE

Niacin is a water-soluble vitamin converted to two important coenzymes involved in carbohydrate metabolism and fatty-acid synthesis.⁴ Body stores are minimal, although the liver can convert tryptophan to niacin.⁴

In the United States, pellagra has become rare since niacin enrichment of bread and flour began in the 1940s. However, the risk is increased in those with chronic alcohol dependency, metabolic disease such as carcinoid syndrome, and malabsorptive states such as Crohn disease.⁵ Alcohol abuse is commonly associated with micronutrient deficiencies. Even in alcoholics with sufficient food intake, their ability to absorb niacin may be impaired both directly and indirectly because of the toxic effects of alcohol on the gastrointestinal tract.⁵

The three D’s of niacin deficiency

The clinical picture of pellagra includes the three “D’s” of niacin deficiency—diarrhea, dementia, and dermatitis—representing the organ systems most commonly affected (ie, the gastrointestinal tract, nervous system, and skin). Gastrointestinal signs and symptoms may include anorexia, diarrhea, stomatitis, and abdominal discomfort.⁴ Neurologic signs and symptoms are the most frequent clinical manifestations of pellagra but are often overlooked in alcoholics with altered mental status.^5,6 Early neurologic findings may include apathy, depression, paresthesias, dizziness, and falling.^4,5 Later findings may include hallucinations, seizures, and psychosis.^4,5 If severe enough, this may progress to coma and to a fourth “D”—death—if left untreated.

Skin findings, if present, are typically the most characteristic of pellagra.^5,6 They include a photosensitive eruption that most commonly appears on the face, arms, and the “V” of the neck (“Casal’s necklace”).^3,6 Erythema develops on areas of sun-exposed skin and may be painful or pruritic.⁶ It may take up to four times as long to resolve as a normal sunburn.⁶ On repeated exposure, the skin becomes thickened, hyperkeratotic, and hyperpigmented.⁶

DIAGNOSIS AND TREATMENT OF PELLAGRA

Suspicion of pellagra should be high in any patient with chronic alcohol abuse with changes in mental status.⁵ Pellagra is typically diagnosed clinically. Although testing of serum levels of niacin or urine levels of niacin metabolites can be performed,^4,5 these tests are not commonly used, as they are neither widely available nor reliable.^3,4 Because there is little downside to giving niacin, treatment should be started if there is clinical suspicion of pellagra.^3,4 The diagnosis is confirmed by clinical improvement after niacin replacement begins.⁴

Pellagra is easy and inexpensive to treat once the diagnosis has been established. Supplementation can be given orally at 50 mg to 100 mg three times daily, with a maximum of 500 mg daily for 3 to 4 weeks.^3,5 The multivitamin our patient initially received provided only 30 mg of niacin, which was insufficient to treat his deficiency.

Other micronutrient deficiencies are also common in pellagra, such as the suspected thiamine deficiency in this patient. Although rare in adults, pyridoxine deficiency should be considered because pyridoxine plays an important role in niacin metabolism, and in patients such as this, the signs and symptoms of pellagra will not resolve with niacin administration alone.^3,4

Neurologic symptoms can begin to resolve as soon as 24 to 48 hours after replacement therapy is started, although skin symptoms take longer to respond.⁶

OUR PATIENT’S MANAGEMENT AND LESSONS LEARNED

Our patient was treated with extended-release niacin 500 mg daily. He was significantly less confused within 24 hours, and 2 days later he was no longer delirious. His diarrhea resolved, but his peripheral neuropathy improved only modestly, and we believe that this was the result of his long history of alcohol abuse.

Although there are guidelines to help with the management of patients undergoing alcohol withdrawal, our experience with this patient illustrates that important clinical conditions, including micronutrient deficiencies, can easily be overlooked or misattributed, with potentially dangerous consequences. It is crucial to be mindful of these considerations.

Managing a patient with chronic alcohol abuse who is beginning to withdraw is a situation in which to expect the unexpected.

A 61-year-old man with a 40-year history of alcohol abuse was admitted to the hospital after presenting with a 6-month history of weakness and increasing frequency of falls, as well as a 2-day history of myalgias and fatigue. He had anorexia and chronic diarrhea, and he reported an unintentional 100-lb weight loss over the past year. He consumed at least three to five alcoholic drinks daily, including a daily “eye-opener,” and his diet was essentially devoid of meat, bread, fruits, and vegetables. His last drink had been on the morning of admission.

On physical examination, his vital signs were notable for marked orthostatic hypotension. He had angular cheilitis, an erythematous, scaly facial rash, and poorly healing severe sunburn on the dorsal surface of the left forearm, acquired 1 month earlier after minimal sun exposure while driving his car (Figure 1).

The neurologic examination noted decreased sensation in a stocking distribution, reduced proprioception in both great toes, and an intention tremor. His upper extremities were slightly hyperreflexic, he had down-going toes, he did not have clonus, and his gait was wide-based.

Laboratory results were notable for mild anemia, with an elevated mean corpuscular volume of 109.2 fL and a normal thyrotropin level. Urinalysis and urine culture were normal. Stool testing for Clostridium difficile toxin was negative. Computed tomography of the head was unremarkable.

He was placed on the Clinical Institute Withdrawal Assessment for Alcohol revised protocol,1 and he was started on careful fluid, electrolyte, and nutritional management, including intravenous supplementation with thiamine 100 mg daily, folic acid 1 mg twice daily, and an oral multivitamin. Despite these interventions, he became delirious and developed horizontal nystagmus, visual hallucinations, dizziness, and increasing weakness. Benzodiazepine toxicity, which may cause similar findings, was thought unlikely, as he had received only a maximum of 4 mg of lorazepam in any single 24-hour period. His ongoing confusion, orthostatic hypotension, ataxia, and nystagmus raised concern for Wernicke encephalopathy. Thiamine was increased to 500 mg intravenously three times daily for 2 days, followed by 500 mg daily for 2 days, as recommended for acute Wernicke encephalopathy by the Royal College of Physicians.2

Although his ocular findings resolved and his blood pressure improved with high-dose thiamine, he remained confused. Other causes of delirium were considered, including other micronutrient deficiencies. He was evaluated for magnesium deficiency, as magnesium is a necessary cofactor in thiamine metabolism,³ but his level was within normal limits. The constellation of symptoms, including persistent delirium, diarrhea, and skin findings, appeared to be consistent with pellagra, and niacin deficiency became a concern.

PELLAGRA AND ALCOHOL ABUSE

Niacin is a water-soluble vitamin converted to two important coenzymes involved in carbohydrate metabolism and fatty-acid synthesis.⁴ Body stores are minimal, although the liver can convert tryptophan to niacin.⁴

In the United States, pellagra has become rare since niacin enrichment of bread and flour began in the 1940s. However, the risk is increased in those with chronic alcohol dependency, metabolic disease such as carcinoid syndrome, and malabsorptive states such as Crohn disease.⁵ Alcohol abuse is commonly associated with micronutrient deficiencies. Even in alcoholics with sufficient food intake, their ability to absorb niacin may be impaired both directly and indirectly because of the toxic effects of alcohol on the gastrointestinal tract.⁵

The three D’s of niacin deficiency