Applied Evidence

Each year, Americans save an estimated $8 billion to $10 billion at retail pharmacies by purchasing generic drugs rather than brand-name medications, the US Food and Drug Administration (FDA) reports.¹ The lower cost, of course, is the key advantage of generics. But the very reason for the cost savings—the fact that generic drugs do not have to undergo the large, expensive clinical trials that are required for approval of brand-name medications—gives rise to questions about the quality and safety of generics.

Are these concerns justified? Under what circumstances is it safe to prescribe generics, or to substitute a generic for a brand-name drug? Are brand-name drugs always better? To answer these questions, we conducted a thorough evidence review, which included numerous randomized controlled trials (RCTs) and case reports, as well as a single meta-analysis that assessed the benefits and risks of generics.

Generics: On the positive side

Safety and efficacy. Our literature search yielded little evidence that generic drugs are less safe or less effective than their brand-name equivalents. The meta-analysis, for example,² included 47 studies (38 of 47 were RCTs) covering 9 subclasses of cardiovascular medications. In trials involving beta-blockers, diuretics, calcium channel blockers, antiplatelet agents, statins, angiotensin-converting enzyme inhibitors, and alpha-blockers, no evidence of superiority of brand-name drugs vs generics was found.²

Cost. Generic drugs typically cost 30% to 60% less than their brand-name counterparts,³ and widespread use of generics has the potential to reduce the price of other brand-name drugs by creating more competition.

Another plus: Patients taking generic drugs appear to be more willing to continue therapy than those taking brand-name medications.⁴ Lower co-pays are a key factor. In 1 recent study of patients with hypercholesterolemia or diabetes, those taking generics had greater adherence compared with patients receiving brand-name drugs.⁵

Quality. It is important to note that many generic medications are produced under the license of the manufacturer of the original brand-name product, with the lower-cost equivalent often introduced after the drug’s patent has expired. Even when different manufacturers produce the branded product and the generic, strict standards exist to guarantee the quality of generic drugs.

The journey to market—the similarities, the differences

Both brand-name and generic medications undergo similar new drug application (NDA) procedures. The manufacturers of both are required to submit detailed evidence of the chemistry, manufacturing, controls, labeling, and testing processes. From there, brand-name and generic products take divergent paths to market.

New nongeneric drugs must undergo rigorous animal and human studies, including large RCTs comparing the efficacy of the new product with that of a placebo and carefully tracking side effects. Bioavailability testing is required, as well. For generic drugs, the process is known as an abbreviated new drug application (ANDA), and bioequivalence studies are sufficient.^1,6

The bioequivalence studies required for a new generic are based on pharmacokinetic parameters, most notably, the area under the plasma concentration curve (AUC)—a measure of overall drug exposure—and the maximal plasma concentration (C_max). If AUC and C_max are within an acceptance range (0.80–1.25 of the brand-name product parameters), the therapeutic equivalence of a generic drug is substantiated.^7,8

Concerns about testing, formulation

Opponents of widespread use of generics point out that they are tested on only a few young, healthy individuals, compared with the large numbers of patients who participate in clinical trials of the original drug.

Bioequivalence
According to guidelines from the World Health Organization (WHO), 18 to 24 healthy adult volunteers are considered sufficient for a bioequivalence study.⁹ The number of participants may be greater, however, if absorption or clearance of the drug is highly variable. What’s more, the people who volunteer for generic drug studies cannot smoke or take concurrent medication. To exclude the possibility that food coadministration affects the generic medication being studied, the FDA further recommends bioequivalence testing of oral formulations on volunteers eating standardized meals.⁸ These criteria help minimize the magnitude of intersubject variability and reduce the possibility of bias—which could be caused by the disease process, concurrent conditions, or medication interaction, rather than by formulation differences.⁸

To further minimize the effects of nondrug-related variation, bioequivalence studies typically use a crossover design: Half the subjects receive the test drug first, followed by the brand-name product, with a washout period in between. The other half receive the drugs in reverse order.¹⁰ (The study format is altered, as needed, for extended-release products, topical agents, and drugs that are not absorbed systemically. A generic version of cholestyramine, for example, which acts by sequestering bile salts within the intestine, would be approved on the basis of in vitro studies that quantify the binding of the bile salts.¹⁰)

But does this testing mimic the real world? While possible confounding factors are controlled for in bioequivalence studies of generics, critics point out that this is not the case in the real world. Thus, they worry that when generics are taken by patients with actual illnesses, concurrent use of other medications, medical conditions, and the like may result in differences in treatment that did not occur in the highly controlled environment in which the equivalency studies were conducted.¹¹

Differences in formulation
Another concern centers on formulation differences, which have the potential to affect patients taking generic drugs. A generic copy of a brand-name drug must contain the same active ingredient, in the identical quantity, as the branded product—in the same dose formulation and route of administration. It must also meet standards for strength, purity, quality, and identity.¹¹

However, the inert ingredients in the generic version do not have to be the same as those in the brand-name drug (although the ratio of inert to active compound must be similar).¹² Because drugs tested in bioequivalence studies are administered in single doses, many experts wonder whether the inert compounds used in the generics may affect the distribution, metabolism, or absorption of a drug when it is administered in multiple doses, or whether the serum concentration of the generic drug may be elevated when it is taken for long periods.

Proceed with caution in these situations

For most patients taking most medications, generic drugs pose no problems, and provide an opportunity to obtain the same therapeutic benefit at a considerably lower cost. However, making the switch with certain classes of drugs, and with drugs that have a narrow therapeutic range, poses potential problems and must be done with caution—if at all.

Antiepileptic drugs. The FDA indicates that many people who are on antiseizure medications re-experience seizures despite continued treatment,¹ and that switching to a generic does not increase the risk of treatment failure.^1,13 Nonetheless, there are numerous reports of differences between generic and brand-name antiseizure medications (and small studies indicating improper seizure control after switching patients from a brand-name to a generic antiepileptic drug).¹⁴

For example:

• Researchers compared the pharmacokinetic parameters of Tegretol with 3 generic formulations of carbamazepine, and found that 1 of the 3 was not bioequivalent.¹⁵
• In a crossover study of 18 healthy volunteers, 3 generic formulations of carbamazepine were all within the acceptable bioequivalence range, but were absorbed more rapidly than the brand-name drug.¹⁶
• Differences in the bioavailability of brand-name and generic products have also been reported with phenytoin, primidone, and valproic acid, but the differences were not statistically significant.¹⁷

The American Academy of Neurology has issued a set of recommendations concerning the use of generic antiepileptic drugs (TABLE).¹⁸

Narrow therapeutic ratio. The potential for complications increases in drugs with a narrow therapeutic ratio, defined by the FDA as <2-fold difference between the median lethal dose and the median effective dose, or between the minimum toxic concentration and minimum effective concentration in the blood.¹⁹ The safe and effective use of such drugs—carbamazepine, divalproex, lithium, phenytoin, and warfarin, to name a few—requires careful dosage titration and patient monitoring.

Water solubility and nonlinear pharmacokinetics may present problems in drugs with a narrow therapeutic ratio, especially phenytoin.² The drug’s serum concentration is allowed to range from 8 to 20 mg/L. A concentration above this range increases the risk for acute cerebellar syndrome, delirium, and coma; a concentration below the range may cause seizures.¹²

Warfarin is also of particular concern, as there is always the possibility that a switch from Coumadin to a generic equivalent could result in under- or overcoagulation. However, studies have shown that the use of generic warfarin in patients previously receiving Coumadin did not affect the international normalized ratio more than continued use of the brand-name anticoagulant.^20,21

Psychotropic agents. There has been a number of case reports of problems occurring following a switch from a brand-name antidepressant to a generic—or from 1 generic antidepressant to another. (See “Did a switch to a generic antidepressant cause relapse?” J Fam Pract. 2008;58:109-114.) In fact, the FDA cites some psychotropic drugs for which generic formulations may not be interchangeable—including amitriptyline/ perphenazine and venlafaxine—and others for which generic formulations may not be bioequivalent at all doses.²²

Thyroid medication. There are also concerns about levothyroxine (LT4) administration, and major medical societies debate the use of generic substitution. According to a recent survey from the American Association of Clinical Endocrinologists, the American Thyroid Association, and The Endocrine Society, clinical use of generic LT4 continues to be associated with adverse outcomes.²³ Most of the adverse events (89%) reported by survey respondents were associated with a change, either from a brand-name drug to a generic or from 1 particular generic LT4 to another.

TABLE
Generic substitution of antiepileptic agents: Where the American Academy of Neurology stands¹⁸

Modified-release formulations may also pose a problem
Problems may also occur with generics in modified-release formulations, which may not have the same pharmacokinetic profiles as their brand-named counterparts. The British National Formulary has advised that prescriptions for modified-release diltiazem hydrochloride, nifedipine, and theophylline be filled with the brand-name drug only.^24,25 Morever, a recent study concluded that 2 modified-release products of methylphenidate and nifedipine had concentration profiles that strongly diverged during the period of absorption, although the formulations met the regulatory criteria for bioequivalence. ²⁶

The type of salt used to form a compound is also important. Salt-joining makes a hydrophobic molecule hydrophilic; the result, especially in psychoactive drugs, is improved kinetics, absorption, or physico-chemical properties (eg, stability, hygroscopicity, fluidity).²⁷ This may be the reason for differences identified between generic and brand-name amitriptyline, nortriptyline, desipramine, and trimipramine.²⁸ To avoid problems, physicians should prescribe generics containing the same salt as their brand-name counterparts.

When in doubt …

Brand-name drugs are, and always will be, the best proven therapy, because of the number and extent of clinical trials they go through. In most cases, however, there is no evidence-based reason to avoid generic substitution for patients who cannot afford the brand-name drug. When in doubt, consult the FDA’s Orange Book: Approved Drug Products with Therapeutic Equivalence Evaluations, available at http://www.accessdata.fda.gov/scripts/cder/ob/default.cfm, before making a switch.

CORRESPONDENCE Pawel Lewek, MD, The First Department of Family Medicine, Medical University of Lodz,60 Narutowicza Street, 90-136 Lodz, Poland; [email protected]

Each year, Americans save an estimated $8 billion to $10 billion at retail pharmacies by purchasing generic drugs rather than brand-name medications, the US Food and Drug Administration (FDA) reports.¹ The lower cost, of course, is the key advantage of generics. But the very reason for the cost savings—the fact that generic drugs do not have to undergo the large, expensive clinical trials that are required for approval of brand-name medications—gives rise to questions about the quality and safety of generics.

Are these concerns justified? Under what circumstances is it safe to prescribe generics, or to substitute a generic for a brand-name drug? Are brand-name drugs always better? To answer these questions, we conducted a thorough evidence review, which included numerous randomized controlled trials (RCTs) and case reports, as well as a single meta-analysis that assessed the benefits and risks of generics.

Generics: On the positive side

Safety and efficacy. Our literature search yielded little evidence that generic drugs are less safe or less effective than their brand-name equivalents. The meta-analysis, for example,² included 47 studies (38 of 47 were RCTs) covering 9 subclasses of cardiovascular medications. In trials involving beta-blockers, diuretics, calcium channel blockers, antiplatelet agents, statins, angiotensin-converting enzyme inhibitors, and alpha-blockers, no evidence of superiority of brand-name drugs vs generics was found.²

Cost. Generic drugs typically cost 30% to 60% less than their brand-name counterparts,³ and widespread use of generics has the potential to reduce the price of other brand-name drugs by creating more competition.

Another plus: Patients taking generic drugs appear to be more willing to continue therapy than those taking brand-name medications.⁴ Lower co-pays are a key factor. In 1 recent study of patients with hypercholesterolemia or diabetes, those taking generics had greater adherence compared with patients receiving brand-name drugs.⁵

Quality. It is important to note that many generic medications are produced under the license of the manufacturer of the original brand-name product, with the lower-cost equivalent often introduced after the drug’s patent has expired. Even when different manufacturers produce the branded product and the generic, strict standards exist to guarantee the quality of generic drugs.

The journey to market—the similarities, the differences

Both brand-name and generic medications undergo similar new drug application (NDA) procedures. The manufacturers of both are required to submit detailed evidence of the chemistry, manufacturing, controls, labeling, and testing processes. From there, brand-name and generic products take divergent paths to market.

New nongeneric drugs must undergo rigorous animal and human studies, including large RCTs comparing the efficacy of the new product with that of a placebo and carefully tracking side effects. Bioavailability testing is required, as well. For generic drugs, the process is known as an abbreviated new drug application (ANDA), and bioequivalence studies are sufficient.^1,6

The bioequivalence studies required for a new generic are based on pharmacokinetic parameters, most notably, the area under the plasma concentration curve (AUC)—a measure of overall drug exposure—and the maximal plasma concentration (C_max). If AUC and C_max are within an acceptance range (0.80–1.25 of the brand-name product parameters), the therapeutic equivalence of a generic drug is substantiated.^7,8

Concerns about testing, formulation

Opponents of widespread use of generics point out that they are tested on only a few young, healthy individuals, compared with the large numbers of patients who participate in clinical trials of the original drug.

Bioequivalence
According to guidelines from the World Health Organization (WHO), 18 to 24 healthy adult volunteers are considered sufficient for a bioequivalence study.⁹ The number of participants may be greater, however, if absorption or clearance of the drug is highly variable. What’s more, the people who volunteer for generic drug studies cannot smoke or take concurrent medication. To exclude the possibility that food coadministration affects the generic medication being studied, the FDA further recommends bioequivalence testing of oral formulations on volunteers eating standardized meals.⁸ These criteria help minimize the magnitude of intersubject variability and reduce the possibility of bias—which could be caused by the disease process, concurrent conditions, or medication interaction, rather than by formulation differences.⁸

To further minimize the effects of nondrug-related variation, bioequivalence studies typically use a crossover design: Half the subjects receive the test drug first, followed by the brand-name product, with a washout period in between. The other half receive the drugs in reverse order.¹⁰ (The study format is altered, as needed, for extended-release products, topical agents, and drugs that are not absorbed systemically. A generic version of cholestyramine, for example, which acts by sequestering bile salts within the intestine, would be approved on the basis of in vitro studies that quantify the binding of the bile salts.¹⁰)

But does this testing mimic the real world? While possible confounding factors are controlled for in bioequivalence studies of generics, critics point out that this is not the case in the real world. Thus, they worry that when generics are taken by patients with actual illnesses, concurrent use of other medications, medical conditions, and the like may result in differences in treatment that did not occur in the highly controlled environment in which the equivalency studies were conducted.¹¹

Differences in formulation
Another concern centers on formulation differences, which have the potential to affect patients taking generic drugs. A generic copy of a brand-name drug must contain the same active ingredient, in the identical quantity, as the branded product—in the same dose formulation and route of administration. It must also meet standards for strength, purity, quality, and identity.¹¹

However, the inert ingredients in the generic version do not have to be the same as those in the brand-name drug (although the ratio of inert to active compound must be similar).¹² Because drugs tested in bioequivalence studies are administered in single doses, many experts wonder whether the inert compounds used in the generics may affect the distribution, metabolism, or absorption of a drug when it is administered in multiple doses, or whether the serum concentration of the generic drug may be elevated when it is taken for long periods.

Proceed with caution in these situations

For most patients taking most medications, generic drugs pose no problems, and provide an opportunity to obtain the same therapeutic benefit at a considerably lower cost. However, making the switch with certain classes of drugs, and with drugs that have a narrow therapeutic range, poses potential problems and must be done with caution—if at all.

Antiepileptic drugs. The FDA indicates that many people who are on antiseizure medications re-experience seizures despite continued treatment,¹ and that switching to a generic does not increase the risk of treatment failure.^1,13 Nonetheless, there are numerous reports of differences between generic and brand-name antiseizure medications (and small studies indicating improper seizure control after switching patients from a brand-name to a generic antiepileptic drug).¹⁴

For example:

• Researchers compared the pharmacokinetic parameters of Tegretol with 3 generic formulations of carbamazepine, and found that 1 of the 3 was not bioequivalent.¹⁵
• In a crossover study of 18 healthy volunteers, 3 generic formulations of carbamazepine were all within the acceptable bioequivalence range, but were absorbed more rapidly than the brand-name drug.¹⁶
• Differences in the bioavailability of brand-name and generic products have also been reported with phenytoin, primidone, and valproic acid, but the differences were not statistically significant.¹⁷

The American Academy of Neurology has issued a set of recommendations concerning the use of generic antiepileptic drugs (TABLE).¹⁸

Narrow therapeutic ratio. The potential for complications increases in drugs with a narrow therapeutic ratio, defined by the FDA as <2-fold difference between the median lethal dose and the median effective dose, or between the minimum toxic concentration and minimum effective concentration in the blood.¹⁹ The safe and effective use of such drugs—carbamazepine, divalproex, lithium, phenytoin, and warfarin, to name a few—requires careful dosage titration and patient monitoring.

Water solubility and nonlinear pharmacokinetics may present problems in drugs with a narrow therapeutic ratio, especially phenytoin.² The drug’s serum concentration is allowed to range from 8 to 20 mg/L. A concentration above this range increases the risk for acute cerebellar syndrome, delirium, and coma; a concentration below the range may cause seizures.¹²

Warfarin is also of particular concern, as there is always the possibility that a switch from Coumadin to a generic equivalent could result in under- or overcoagulation. However, studies have shown that the use of generic warfarin in patients previously receiving Coumadin did not affect the international normalized ratio more than continued use of the brand-name anticoagulant.^20,21

Psychotropic agents. There has been a number of case reports of problems occurring following a switch from a brand-name antidepressant to a generic—or from 1 generic antidepressant to another. (See “Did a switch to a generic antidepressant cause relapse?” J Fam Pract. 2008;58:109-114.) In fact, the FDA cites some psychotropic drugs for which generic formulations may not be interchangeable—including amitriptyline/ perphenazine and venlafaxine—and others for which generic formulations may not be bioequivalent at all doses.²²

Thyroid medication. There are also concerns about levothyroxine (LT4) administration, and major medical societies debate the use of generic substitution. According to a recent survey from the American Association of Clinical Endocrinologists, the American Thyroid Association, and The Endocrine Society, clinical use of generic LT4 continues to be associated with adverse outcomes.²³ Most of the adverse events (89%) reported by survey respondents were associated with a change, either from a brand-name drug to a generic or from 1 particular generic LT4 to another.

TABLE
Generic substitution of antiepileptic agents: Where the American Academy of Neurology stands¹⁸

Modified-release formulations may also pose a problem
Problems may also occur with generics in modified-release formulations, which may not have the same pharmacokinetic profiles as their brand-named counterparts. The British National Formulary has advised that prescriptions for modified-release diltiazem hydrochloride, nifedipine, and theophylline be filled with the brand-name drug only.^24,25 Morever, a recent study concluded that 2 modified-release products of methylphenidate and nifedipine had concentration profiles that strongly diverged during the period of absorption, although the formulations met the regulatory criteria for bioequivalence. ²⁶

The type of salt used to form a compound is also important. Salt-joining makes a hydrophobic molecule hydrophilic; the result, especially in psychoactive drugs, is improved kinetics, absorption, or physico-chemical properties (eg, stability, hygroscopicity, fluidity).²⁷ This may be the reason for differences identified between generic and brand-name amitriptyline, nortriptyline, desipramine, and trimipramine.²⁸ To avoid problems, physicians should prescribe generics containing the same salt as their brand-name counterparts.

When in doubt …

Brand-name drugs are, and always will be, the best proven therapy, because of the number and extent of clinical trials they go through. In most cases, however, there is no evidence-based reason to avoid generic substitution for patients who cannot afford the brand-name drug. When in doubt, consult the FDA’s Orange Book: Approved Drug Products with Therapeutic Equivalence Evaluations, available at http://www.accessdata.fda.gov/scripts/cder/ob/default.cfm, before making a switch.

CORRESPONDENCE Pawel Lewek, MD, The First Department of Family Medicine, Medical University of Lodz,60 Narutowicza Street, 90-136 Lodz, Poland; [email protected]

Each year, Americans save an estimated $8 billion to $10 billion at retail pharmacies by purchasing generic drugs rather than brand-name medications, the US Food and Drug Administration (FDA) reports.¹ The lower cost, of course, is the key advantage of generics. But the very reason for the cost savings—the fact that generic drugs do not have to undergo the large, expensive clinical trials that are required for approval of brand-name medications—gives rise to questions about the quality and safety of generics.

Are these concerns justified? Under what circumstances is it safe to prescribe generics, or to substitute a generic for a brand-name drug? Are brand-name drugs always better? To answer these questions, we conducted a thorough evidence review, which included numerous randomized controlled trials (RCTs) and case reports, as well as a single meta-analysis that assessed the benefits and risks of generics.

Generics: On the positive side

Safety and efficacy. Our literature search yielded little evidence that generic drugs are less safe or less effective than their brand-name equivalents. The meta-analysis, for example,² included 47 studies (38 of 47 were RCTs) covering 9 subclasses of cardiovascular medications. In trials involving beta-blockers, diuretics, calcium channel blockers, antiplatelet agents, statins, angiotensin-converting enzyme inhibitors, and alpha-blockers, no evidence of superiority of brand-name drugs vs generics was found.²

Cost. Generic drugs typically cost 30% to 60% less than their brand-name counterparts,³ and widespread use of generics has the potential to reduce the price of other brand-name drugs by creating more competition.

Another plus: Patients taking generic drugs appear to be more willing to continue therapy than those taking brand-name medications.⁴ Lower co-pays are a key factor. In 1 recent study of patients with hypercholesterolemia or diabetes, those taking generics had greater adherence compared with patients receiving brand-name drugs.⁵

Quality. It is important to note that many generic medications are produced under the license of the manufacturer of the original brand-name product, with the lower-cost equivalent often introduced after the drug’s patent has expired. Even when different manufacturers produce the branded product and the generic, strict standards exist to guarantee the quality of generic drugs.

The journey to market—the similarities, the differences

Both brand-name and generic medications undergo similar new drug application (NDA) procedures. The manufacturers of both are required to submit detailed evidence of the chemistry, manufacturing, controls, labeling, and testing processes. From there, brand-name and generic products take divergent paths to market.

New nongeneric drugs must undergo rigorous animal and human studies, including large RCTs comparing the efficacy of the new product with that of a placebo and carefully tracking side effects. Bioavailability testing is required, as well. For generic drugs, the process is known as an abbreviated new drug application (ANDA), and bioequivalence studies are sufficient.^1,6

The bioequivalence studies required for a new generic are based on pharmacokinetic parameters, most notably, the area under the plasma concentration curve (AUC)—a measure of overall drug exposure—and the maximal plasma concentration (C_max). If AUC and C_max are within an acceptance range (0.80–1.25 of the brand-name product parameters), the therapeutic equivalence of a generic drug is substantiated.^7,8

Concerns about testing, formulation

Opponents of widespread use of generics point out that they are tested on only a few young, healthy individuals, compared with the large numbers of patients who participate in clinical trials of the original drug.

Bioequivalence
According to guidelines from the World Health Organization (WHO), 18 to 24 healthy adult volunteers are considered sufficient for a bioequivalence study.⁹ The number of participants may be greater, however, if absorption or clearance of the drug is highly variable. What’s more, the people who volunteer for generic drug studies cannot smoke or take concurrent medication. To exclude the possibility that food coadministration affects the generic medication being studied, the FDA further recommends bioequivalence testing of oral formulations on volunteers eating standardized meals.⁸ These criteria help minimize the magnitude of intersubject variability and reduce the possibility of bias—which could be caused by the disease process, concurrent conditions, or medication interaction, rather than by formulation differences.⁸

To further minimize the effects of nondrug-related variation, bioequivalence studies typically use a crossover design: Half the subjects receive the test drug first, followed by the brand-name product, with a washout period in between. The other half receive the drugs in reverse order.¹⁰ (The study format is altered, as needed, for extended-release products, topical agents, and drugs that are not absorbed systemically. A generic version of cholestyramine, for example, which acts by sequestering bile salts within the intestine, would be approved on the basis of in vitro studies that quantify the binding of the bile salts.¹⁰)

But does this testing mimic the real world? While possible confounding factors are controlled for in bioequivalence studies of generics, critics point out that this is not the case in the real world. Thus, they worry that when generics are taken by patients with actual illnesses, concurrent use of other medications, medical conditions, and the like may result in differences in treatment that did not occur in the highly controlled environment in which the equivalency studies were conducted.¹¹

Differences in formulation
Another concern centers on formulation differences, which have the potential to affect patients taking generic drugs. A generic copy of a brand-name drug must contain the same active ingredient, in the identical quantity, as the branded product—in the same dose formulation and route of administration. It must also meet standards for strength, purity, quality, and identity.¹¹

However, the inert ingredients in the generic version do not have to be the same as those in the brand-name drug (although the ratio of inert to active compound must be similar).¹² Because drugs tested in bioequivalence studies are administered in single doses, many experts wonder whether the inert compounds used in the generics may affect the distribution, metabolism, or absorption of a drug when it is administered in multiple doses, or whether the serum concentration of the generic drug may be elevated when it is taken for long periods.

Proceed with caution in these situations

For most patients taking most medications, generic drugs pose no problems, and provide an opportunity to obtain the same therapeutic benefit at a considerably lower cost. However, making the switch with certain classes of drugs, and with drugs that have a narrow therapeutic range, poses potential problems and must be done with caution—if at all.

Antiepileptic drugs. The FDA indicates that many people who are on antiseizure medications re-experience seizures despite continued treatment,¹ and that switching to a generic does not increase the risk of treatment failure.^1,13 Nonetheless, there are numerous reports of differences between generic and brand-name antiseizure medications (and small studies indicating improper seizure control after switching patients from a brand-name to a generic antiepileptic drug).¹⁴

For example:

• Researchers compared the pharmacokinetic parameters of Tegretol with 3 generic formulations of carbamazepine, and found that 1 of the 3 was not bioequivalent.¹⁵
• In a crossover study of 18 healthy volunteers, 3 generic formulations of carbamazepine were all within the acceptable bioequivalence range, but were absorbed more rapidly than the brand-name drug.¹⁶
• Differences in the bioavailability of brand-name and generic products have also been reported with phenytoin, primidone, and valproic acid, but the differences were not statistically significant.¹⁷

The American Academy of Neurology has issued a set of recommendations concerning the use of generic antiepileptic drugs (TABLE).¹⁸

Narrow therapeutic ratio. The potential for complications increases in drugs with a narrow therapeutic ratio, defined by the FDA as <2-fold difference between the median lethal dose and the median effective dose, or between the minimum toxic concentration and minimum effective concentration in the blood.¹⁹ The safe and effective use of such drugs—carbamazepine, divalproex, lithium, phenytoin, and warfarin, to name a few—requires careful dosage titration and patient monitoring.

Water solubility and nonlinear pharmacokinetics may present problems in drugs with a narrow therapeutic ratio, especially phenytoin.² The drug’s serum concentration is allowed to range from 8 to 20 mg/L. A concentration above this range increases the risk for acute cerebellar syndrome, delirium, and coma; a concentration below the range may cause seizures.¹²

Warfarin is also of particular concern, as there is always the possibility that a switch from Coumadin to a generic equivalent could result in under- or overcoagulation. However, studies have shown that the use of generic warfarin in patients previously receiving Coumadin did not affect the international normalized ratio more than continued use of the brand-name anticoagulant.^20,21

Psychotropic agents. There has been a number of case reports of problems occurring following a switch from a brand-name antidepressant to a generic—or from 1 generic antidepressant to another. (See “Did a switch to a generic antidepressant cause relapse?” J Fam Pract. 2008;58:109-114.) In fact, the FDA cites some psychotropic drugs for which generic formulations may not be interchangeable—including amitriptyline/ perphenazine and venlafaxine—and others for which generic formulations may not be bioequivalent at all doses.²²

Thyroid medication. There are also concerns about levothyroxine (LT4) administration, and major medical societies debate the use of generic substitution. According to a recent survey from the American Association of Clinical Endocrinologists, the American Thyroid Association, and The Endocrine Society, clinical use of generic LT4 continues to be associated with adverse outcomes.²³ Most of the adverse events (89%) reported by survey respondents were associated with a change, either from a brand-name drug to a generic or from 1 particular generic LT4 to another.

TABLE
Generic substitution of antiepileptic agents: Where the American Academy of Neurology stands¹⁸

Modified-release formulations may also pose a problem
Problems may also occur with generics in modified-release formulations, which may not have the same pharmacokinetic profiles as their brand-named counterparts. The British National Formulary has advised that prescriptions for modified-release diltiazem hydrochloride, nifedipine, and theophylline be filled with the brand-name drug only.^24,25 Morever, a recent study concluded that 2 modified-release products of methylphenidate and nifedipine had concentration profiles that strongly diverged during the period of absorption, although the formulations met the regulatory criteria for bioequivalence. ²⁶

The type of salt used to form a compound is also important. Salt-joining makes a hydrophobic molecule hydrophilic; the result, especially in psychoactive drugs, is improved kinetics, absorption, or physico-chemical properties (eg, stability, hygroscopicity, fluidity).²⁷ This may be the reason for differences identified between generic and brand-name amitriptyline, nortriptyline, desipramine, and trimipramine.²⁸ To avoid problems, physicians should prescribe generics containing the same salt as their brand-name counterparts.

When in doubt …

Brand-name drugs are, and always will be, the best proven therapy, because of the number and extent of clinical trials they go through. In most cases, however, there is no evidence-based reason to avoid generic substitution for patients who cannot afford the brand-name drug. When in doubt, consult the FDA’s Orange Book: Approved Drug Products with Therapeutic Equivalence Evaluations, available at http://www.accessdata.fda.gov/scripts/cder/ob/default.cfm, before making a switch.

CORRESPONDENCE Pawel Lewek, MD, The First Department of Family Medicine, Medical University of Lodz,60 Narutowicza Street, 90-136 Lodz, Poland; [email protected]

A 57-year-old man was referred by his general practitioner to our surgical team as an emergency admission. The patient had a 4-month history of post-prandial nausea and vomiting, occasional heartburn, and a change in bowel habits. (His normal bowel habit was 3 times a day, but it had changed to once every 2 to 5 days.) Over the past 2 months, he had lost 16 kg (35.3 lb).

His past medical history included esophagitis, peripheral vascular disease, and an inguinal hernia repair. He was a self-employed baker, a smoker of 8 pack-years, and his alcohol intake was 6.9 oz per week. There was nothing significant about his family history. His regular medications included omeprazole, simvastatin, and aspirin.

On examination, we noted bilateral fingernail clubbing, but no peripheral lymphadenopathy. His abdominal examination revealed mild, left upper quadrant tenderness.

His significant laboratory results were:

• hemoglobin, 11.6 g/dL
• mean corpuscular volume, 93.0 fL
• ferritin, 17 mcg/L
• serum folate, 2.3 mcg/L
• albumin, 29 g/L
• potassium, 3.3 mmol/L
• urea, 13.2 mmol/L with normal creatinine
• sodium, 141 mmol/L

A barium meal test showed a large food residue in the stomach, a markedly dilated duodenal loop, and a stricture at the duodenal-jejunal flexure. Computed tomography scan showed 2 simple liver cysts. Esophagogastroduodenoscopy showed moderate reflux esophagitis, a large residue of food and fluid in the stomach, and a stricture that appeared to be malignant at the duodenal-jejunal flexure.

Biopsies revealed severe, high-grade dysplasia and a possible invasive tumor. Incidentally, one of the biopsies also showed evidence of celiac disease (FIGURE 1A). The patient was started on total parenteral nutrition in light of his significant weight loss. He was transferred to surgical care and underwent laparotomy and resection of a proximal jejunal tumor.

Chemotherapy and a gluten-free diet
Final histopathology revealed a jejunal adenocarcinoma with extension to the serosal surface of the bowel (FIGURE 1B) and celiac lymph node involvement. Histology of the surgical specimen also confirmed the diagnosis of underlying celiac disease. The patient underwent adjuvant chemotherapy and was put on a gluten-free diet.

The patient began to gain weight. On a follow-up visit, he was 20 kg (44.1 lb) heavier than his presurgery weight. He was moving his bowel once a day with formed brown stool.

FIGURE 1
Duodenal biopsy revealed celiac disease and adenocarcinoma

One of the patient’s biopsies showed evidence of celiac disease (A). Another revealed a jejunal adenocarcinoma (B) with extension to the serosal surface of the bowel and celiac lymph node involvement.

Anemic for years, but no one dug deeper
A detailed history revealed that the patient had been anemic since the age of 20, had always had difficulty maintaining his weight, and had been having bowel movements 3 to 4 times a day throughout his life. It was unclear why his condition was never properly investigated.

Progress was short lived
Unfortunately, while receiving adjuvant chemotherapy, our patient developed a gastric outlet obstruction that required further surgery and a palliative gastroenterostomy. Despite good palliation, the patient died several months later from progressive cancer.

A valuable reminder for us all

Celiac disease is not an uncommon cause of malabsorption of 1 or more nutrients in Caucasians—especially those of European decent.^1-4 Colon cancer, gastric cancer, and celiac disease are the most important gastrointestinal causes of iron deficiency anemia.⁵

The seroprevalence of celiac disease is as high as approximately 1% in North America and Western Europe, with a high ratio of un-diagnosed to diagnosed cases.^4,6-8 The most serious complication of celiac disease is the development of a malignancy.^1,9-12 Incidence of both gastrointestinal and nongastrointestinal neoplasms (as well as intestinal lymphoma) is higher in patients with celiac disease.^1,9-11,13 Small-bowel lymphoma appears to be significantly associated with unrecognized celiac disease.⁷

Data from various studies suggest a protective role for a gluten-free diet against malignancy in celiac disease and give further support for advising all patients to adhere to a strict gluten-free diet.^9,10,13,14

Follow the clues: Nutrient deficiencies, including folate and iron deficiency anemia—with all of their associated signs and symptoms—are not disease entities, but clues to diseases. We must remain vigilant in our investigative efforts.

CORRESPONDENCE Arash Assadsangabi, MD, Gastroenterology Department, Bristol Royal Infirmary, Marlborough Street, Bristol, BS2 8HW, UK; [email protected]

A 57-year-old man was referred by his general practitioner to our surgical team as an emergency admission. The patient had a 4-month history of post-prandial nausea and vomiting, occasional heartburn, and a change in bowel habits. (His normal bowel habit was 3 times a day, but it had changed to once every 2 to 5 days.) Over the past 2 months, he had lost 16 kg (35.3 lb).

His past medical history included esophagitis, peripheral vascular disease, and an inguinal hernia repair. He was a self-employed baker, a smoker of 8 pack-years, and his alcohol intake was 6.9 oz per week. There was nothing significant about his family history. His regular medications included omeprazole, simvastatin, and aspirin.

On examination, we noted bilateral fingernail clubbing, but no peripheral lymphadenopathy. His abdominal examination revealed mild, left upper quadrant tenderness.

His significant laboratory results were:

• hemoglobin, 11.6 g/dL
• mean corpuscular volume, 93.0 fL
• ferritin, 17 mcg/L
• serum folate, 2.3 mcg/L
• albumin, 29 g/L
• potassium, 3.3 mmol/L
• urea, 13.2 mmol/L with normal creatinine
• sodium, 141 mmol/L

A barium meal test showed a large food residue in the stomach, a markedly dilated duodenal loop, and a stricture at the duodenal-jejunal flexure. Computed tomography scan showed 2 simple liver cysts. Esophagogastroduodenoscopy showed moderate reflux esophagitis, a large residue of food and fluid in the stomach, and a stricture that appeared to be malignant at the duodenal-jejunal flexure.

Biopsies revealed severe, high-grade dysplasia and a possible invasive tumor. Incidentally, one of the biopsies also showed evidence of celiac disease (FIGURE 1A). The patient was started on total parenteral nutrition in light of his significant weight loss. He was transferred to surgical care and underwent laparotomy and resection of a proximal jejunal tumor.

Chemotherapy and a gluten-free diet
Final histopathology revealed a jejunal adenocarcinoma with extension to the serosal surface of the bowel (FIGURE 1B) and celiac lymph node involvement. Histology of the surgical specimen also confirmed the diagnosis of underlying celiac disease. The patient underwent adjuvant chemotherapy and was put on a gluten-free diet.

The patient began to gain weight. On a follow-up visit, he was 20 kg (44.1 lb) heavier than his presurgery weight. He was moving his bowel once a day with formed brown stool.

FIGURE 1
Duodenal biopsy revealed celiac disease and adenocarcinoma

One of the patient’s biopsies showed evidence of celiac disease (A). Another revealed a jejunal adenocarcinoma (B) with extension to the serosal surface of the bowel and celiac lymph node involvement.

Anemic for years, but no one dug deeper
A detailed history revealed that the patient had been anemic since the age of 20, had always had difficulty maintaining his weight, and had been having bowel movements 3 to 4 times a day throughout his life. It was unclear why his condition was never properly investigated.

Progress was short lived
Unfortunately, while receiving adjuvant chemotherapy, our patient developed a gastric outlet obstruction that required further surgery and a palliative gastroenterostomy. Despite good palliation, the patient died several months later from progressive cancer.

A valuable reminder for us all

Celiac disease is not an uncommon cause of malabsorption of 1 or more nutrients in Caucasians—especially those of European decent.^1-4 Colon cancer, gastric cancer, and celiac disease are the most important gastrointestinal causes of iron deficiency anemia.⁵

The seroprevalence of celiac disease is as high as approximately 1% in North America and Western Europe, with a high ratio of un-diagnosed to diagnosed cases.^4,6-8 The most serious complication of celiac disease is the development of a malignancy.^1,9-12 Incidence of both gastrointestinal and nongastrointestinal neoplasms (as well as intestinal lymphoma) is higher in patients with celiac disease.^1,9-11,13 Small-bowel lymphoma appears to be significantly associated with unrecognized celiac disease.⁷

Data from various studies suggest a protective role for a gluten-free diet against malignancy in celiac disease and give further support for advising all patients to adhere to a strict gluten-free diet.^9,10,13,14

Follow the clues: Nutrient deficiencies, including folate and iron deficiency anemia—with all of their associated signs and symptoms—are not disease entities, but clues to diseases. We must remain vigilant in our investigative efforts.

CORRESPONDENCE Arash Assadsangabi, MD, Gastroenterology Department, Bristol Royal Infirmary, Marlborough Street, Bristol, BS2 8HW, UK; [email protected]

A 57-year-old man was referred by his general practitioner to our surgical team as an emergency admission. The patient had a 4-month history of post-prandial nausea and vomiting, occasional heartburn, and a change in bowel habits. (His normal bowel habit was 3 times a day, but it had changed to once every 2 to 5 days.) Over the past 2 months, he had lost 16 kg (35.3 lb).

His past medical history included esophagitis, peripheral vascular disease, and an inguinal hernia repair. He was a self-employed baker, a smoker of 8 pack-years, and his alcohol intake was 6.9 oz per week. There was nothing significant about his family history. His regular medications included omeprazole, simvastatin, and aspirin.

On examination, we noted bilateral fingernail clubbing, but no peripheral lymphadenopathy. His abdominal examination revealed mild, left upper quadrant tenderness.

His significant laboratory results were:

• hemoglobin, 11.6 g/dL
• mean corpuscular volume, 93.0 fL
• ferritin, 17 mcg/L
• serum folate, 2.3 mcg/L
• albumin, 29 g/L
• potassium, 3.3 mmol/L
• urea, 13.2 mmol/L with normal creatinine
• sodium, 141 mmol/L

A barium meal test showed a large food residue in the stomach, a markedly dilated duodenal loop, and a stricture at the duodenal-jejunal flexure. Computed tomography scan showed 2 simple liver cysts. Esophagogastroduodenoscopy showed moderate reflux esophagitis, a large residue of food and fluid in the stomach, and a stricture that appeared to be malignant at the duodenal-jejunal flexure.

Biopsies revealed severe, high-grade dysplasia and a possible invasive tumor. Incidentally, one of the biopsies also showed evidence of celiac disease (FIGURE 1A). The patient was started on total parenteral nutrition in light of his significant weight loss. He was transferred to surgical care and underwent laparotomy and resection of a proximal jejunal tumor.

Chemotherapy and a gluten-free diet
Final histopathology revealed a jejunal adenocarcinoma with extension to the serosal surface of the bowel (FIGURE 1B) and celiac lymph node involvement. Histology of the surgical specimen also confirmed the diagnosis of underlying celiac disease. The patient underwent adjuvant chemotherapy and was put on a gluten-free diet.

The patient began to gain weight. On a follow-up visit, he was 20 kg (44.1 lb) heavier than his presurgery weight. He was moving his bowel once a day with formed brown stool.

FIGURE 1
Duodenal biopsy revealed celiac disease and adenocarcinoma

One of the patient’s biopsies showed evidence of celiac disease (A). Another revealed a jejunal adenocarcinoma (B) with extension to the serosal surface of the bowel and celiac lymph node involvement.

Anemic for years, but no one dug deeper
A detailed history revealed that the patient had been anemic since the age of 20, had always had difficulty maintaining his weight, and had been having bowel movements 3 to 4 times a day throughout his life. It was unclear why his condition was never properly investigated.

Progress was short lived
Unfortunately, while receiving adjuvant chemotherapy, our patient developed a gastric outlet obstruction that required further surgery and a palliative gastroenterostomy. Despite good palliation, the patient died several months later from progressive cancer.

A valuable reminder for us all

Celiac disease is not an uncommon cause of malabsorption of 1 or more nutrients in Caucasians—especially those of European decent.^1-4 Colon cancer, gastric cancer, and celiac disease are the most important gastrointestinal causes of iron deficiency anemia.⁵

The seroprevalence of celiac disease is as high as approximately 1% in North America and Western Europe, with a high ratio of un-diagnosed to diagnosed cases.^4,6-8 The most serious complication of celiac disease is the development of a malignancy.^1,9-12 Incidence of both gastrointestinal and nongastrointestinal neoplasms (as well as intestinal lymphoma) is higher in patients with celiac disease.^1,9-11,13 Small-bowel lymphoma appears to be significantly associated with unrecognized celiac disease.⁷

Data from various studies suggest a protective role for a gluten-free diet against malignancy in celiac disease and give further support for advising all patients to adhere to a strict gluten-free diet.^9,10,13,14

Follow the clues: Nutrient deficiencies, including folate and iron deficiency anemia—with all of their associated signs and symptoms—are not disease entities, but clues to diseases. We must remain vigilant in our investigative efforts.

CORRESPONDENCE Arash Assadsangabi, MD, Gastroenterology Department, Bristol Royal Infirmary, Marlborough Street, Bristol, BS2 8HW, UK; [email protected]

Although the number of patients undergoing tonsillectomy has gradually declined since the 1970s, it remains one of the most common surgical procedures performed worldwide.¹ The procedure, of course, is fairly routine—but not without risk.

Mortality rates for the operation range from 1 in 10,000 to 1 in 35,000, with morbidity rates ranging from 1.5% to 14%; mortality and morbidity after tonsillectomy are usually the result of postoperative bleeding.^2,3 In addition to bleeding, other common complications include pain, nausea, and vomiting.

The patient’s surgeon plays a central role in minimizing risk. But as a primary care physician, you, too, play an important role in ensuring that your patient’s tonsillectomy is uneventful. This review will help toward that end.

Refine your preop approach

Address informed consent issues. While the surgeon is responsible for obtaining informed consent, a patient may discuss the procedure with you, the family physician. (See TABLE 1 for the indications for tonsillectomy.) Although the procedure is safe and effective, the patient and his or her family need to know that bleeding will most likely occur immediately after the procedure, although it can occur at any time during the first 2 weeks postop. Advise the patient and family that postop pain is similar to that of a throat infection, but is often felt in the ears (“referred otalgia”). Because postop swallowing is painful, the patient may not drink enough fluids. If this problem because severe, he or she may need to be admitted for IV fluid replacement.

Discuss the patient’s length of stay. An increasing number of tonsillectomies are performed on an outpatient basis. However, high-risk patients—those with sleep apnea, coagulation disorders, or other underlying diseases, and anyone younger than 4 years of age or living a long distance from the hospital—should be admitted for overnight observation.⁴

Ask about previous surgeries. Tonsillectomies are performed under general anesthesia in most facilities. Ask about previous surgeries and conditions that may adversely affect the surgery: Does the patient have any upper airway obstruction, difficulty extending his cervical spine, or enzymatic deficiencies? A pseudocholinesterase deficiency, for instance, may cause prolonged paralysis of the respiratory muscles, requiring an extended period of time during which the patient must be mechanically ventilated.

Anesthesiologists use the Mallampati score to predict the ease of intubation.⁵ It’s derived by visualizing the base of the uvula, the tonsillar pillars, and the soft palate. Scoring may be done with or without phonation. A patient is considered to be in 1 of 4 classes, depending upon what can be visualized:

• Class 1: the tonsils, uvula, and soft palate
• Class 2: the hard and soft palate and upper portion of the tonsils and uvula
• Class 3: the soft and hard palate and base of the uvula
• Class 4: the hard palate.

As a general rule, patients scheduled for tonsillectomy are prone to higher scores because of underlying disease.

Order lab tests, stop certain medications. Instruct patients to discontinue aspirin 7 days before surgery, and naproxen and other nonsteroidal anti-inflammatory drugs (NSAIDs) 4 days before surgery.⁶ Antiplatelet aggregation drugs such as clopidogrel should also be stopped 7 days before surgery.⁶

While you should routinely order a complete blood count, experts disagree on the value and cost-effectiveness of routinely running prothrombin time and partial thromboplastin time.⁷

Assess whether the patient is anxious about the surgery. Tonsillectomy and the subsequent hospitalization can make patients anxious, bringing on sleep disturbances, behavioral problems, nightmares, enuresis, and emotional regression. Fortunately, these problems usually disappear without any intervention, but offering patients preop reassurance, demonstrations, and educational materials may help prevent them (TABLE 2).

TABLE 1
Indications for tonsillectomy

TABLE 2
Patient education resources

Address postop complications

Ensure that your patient has proper pain relief. Sore throat and referred otalgia are common, but usually subside within 2 weeks. The pain typically manifests as dysphagia, which results from trauma to the pharyngeal muscles. Keep in mind that pain is not always a minor complication; if prolonged, it may be accompanied by dehydration, fever, and even bleeding. Be sure to rule out otitis media in any patient complaining of otalgia who has had both tonsillectomy and adenoidectomy.

Nearly 50% of children who have had a tonsillectomy experience severe pain, defined as a visual analog score of 8 or more.⁸ According to 1 large-scale study, all children will experience moderate or severe pain while at rest and when swallowing during the first 48 hours after surgery—despite an appropriate around-the-clock dose of acetaminophen with codeine.⁹

Although high-dose acetaminophen, with or without codeine, is the most common analgesic prescribed for children after ambulatory surgery, its analgesic effect varies widely, so it may need to be administered in conjunction with other agents.¹⁰

One meta-analysis of several prospective, randomized, double-blind trials concluded that the risk of posttonsillectomy bleeding was nearly 4 times greater when patients received nonselective NSAIDs. This prompted the recommendation that these agents be avoided.¹¹ However, a more recent analysis that included selective cyclooxygenase-1 (COX-1) and COX-2 inhibitors found that these NSAIDs did not significantly alter perioperative bleeding.¹² With that in mind, selective COX-2 inhibitors such as celecoxib may safely relieve posttonsillectomy pain.¹³

Because so many patients develop postop nausea and vomiting, many surgeons avoid postop opioids such as codeine or tramadol, which can aggravate nausea without significant analgesic benefit. These agents are usually prescribed only after the gastrointestinal adverse effects have subsided, or when combined with antiemetics.

Ensure that nausea and vomiting have been properly addressed. Up to 89% of children undergoing tonsillectomy have postop vomiting and nausea.¹⁴ During the surgery, intravenous dexamethasone is usually administered because of its antiemetic effect—which can last up to 48 hours—and because of its role in preventing postop oropharyngeal edema.¹⁴

Watch for postop bleeding. Bleeding, followed by hypovolemic shock, is the most common cause of morbidity and mortality among patients undergoing tonsillectomy, affecting an estimated 0.5% to 10%. Approximately 1 in 200 patients returns to the operating room (OR) so that the bleeding can be controlled. Mortality from bleeding is 2 in 10,000 tonsillectomies. Most cases of fatal postop bleeding occur within the first 24 hours after surgery.^15-17

Although clinicians should attempt to estimate blood loss during active bleeding, the estimate may be too low in children because they tend to swallow blood. Despite this caveat, consider blood transfusion when the estimated blood loss is 10% or more of the blood volume in infants, 25% in children, and 20% in adults.¹⁸

If transfusion is warranted, 2 large intravenous lines will need to be inserted, blood samples will need to be sent for typing and cross-matching, and a complete blood count and coagulation studies will need to be ordered. The patient’s airway will need to be protected and hemodynamic monitoring maintained.

If the bleeding is not excessive, applying local pressure with gauze on the tonsillar bed may be effective, assuming the patient is cooperative. However, if this doesn’t work, the patient will need to be rushed to an OR to control the bleeding. Anesthetizing a patient with massive oral bleeding is among the most challenging emergencies. Always anticipate aspiration once the bleeding has been controlled.

Be ready for these complications, too

The following complications are occasionally encountered after tonsillectomy, but no large controlled studies have estimated their frequency.

Airway obstruction
Mucosal damage during intubation may cause air to enter parapharyngeal tissues, which in turn can result in laryngeal spasm and cervical emphysema after extubation. It can progress to pneumomediastinum and tension pneumothorax.^19,20 Excessive pressure on the surgical blades and manipulation within the oral cavity during surgery can cause lingual and oropharyngeal edema.²¹ With these potential complications in mind, the primary care physician will need to monitor the patient postoperatively for dyspnea.

Acute airway obstruction can occur if blood or clots accumulate in the hypo pharynx as the patient is awakening, immediately after extubation, or later. The most important way to prevent this complication is to ensure that there are no active bleeding sites. The airway can also be compromised from residual bleeding and an evolving airway edema.

On rare occasions, a patient’s airway can become obstructed if dislodged tonsillar tissue is not fully removed during surgery, or if loose teeth and parts of surgical instrumentation find their way into the airway. Being cognizant of these possibilities will help you, the primary care physician, to make an accurate differential diagnosis postoperatively.

Children suffering from obstructive sleep apnea due to craniofacial abnormalities, hypotonia, obesity, or “kissing” tonsils are at increased risk of developing acute airway maintenance difficulties after tonsillectomy. Prolonged obstruction of the upper airway from the obstructing tonsils can increase intrathoracic pressure and decrease venous return and pulmonary blood volume. After removal of the tonsils, there is a sudden increase in venous return, pulmonary blood volume, and pulmonary hydrostatic pressure. These rapid changes cause a fluid shift from pulmonary vessels into the lungs’ interstitial spaces and alveoli. The complication is sometimes referred to as “negative-pressure pulmonary edema.”^22,23

Appropriate management includes close monitoring and the use of continuous positive airway pressure ventilation, as needed. The same approach is appropriate in children with right-sided cardiac insufficiency or pulmonary hypertension.

Fever and infection
Fever may occur within 18 to 36 hours after tonsillectomy as a result of lung atelectasis, a response to anesthetic agents, or a transient bacteremia. A postop fever lasting more than 24 hours, accompanied by severe throat pain, suggests an emerging pharyngeal infection.

Some surgeons prescribe prophylactic postop antibiotics to reduce the risk of infection in the traumatized region, decrease pain, and enable the patient to return quickly to a normal oral diet.²⁴ A meta-analysis suggests, however, that postop oral antibiotics don’t reduce infection rates, posttonsillectomy pain, or secondary hemorrhage rates; the drugs do facilitate a return to normal activity and diet about a day sooner.^25,26

Pneumonia used to be a common complication due to aspirated blood during the procedure. With the advent of improved surgical and anesthetic techniques, however, this complication rarely occurs.² Neck infection, including suppurative lymphadenitis, is also a rare complication.²⁷

Dehydration
While postop dehydration may be uncommon in the “average” patient, the risk is more common among dysphagic children. Post-tonsillectomy vomiting worsens this situation. With that in mind, encourage patients to resume oral intake of fluids and, when necessary, provide analgesia to facilitate the process. You will need to readmit patients who are unable to drink to achieve better pain relief and restore hydration.²⁸

Some reports suggest that severe post-tonsillectomy hyponatremia can result from excessive hypovolemia-induced antidiuretic hormone secretion, disproportionate administration of hypotonic fluids, or inadequate postop fluid intake.^29,30

Encouraging patients to consume soft and cold foods after tonsillectomy will comfort patients and help restore hemostasis. Because pain is the major obstacle for a return to an oral diet, coupling analgesia and timing of oral intake is crucial for a quick recovery.

Injuries from a forceful surgical technique
Because tonsillectomy requires forceful insertion of instruments into the mouth, a loose tooth may become dislodged, or even aspirated. We don’t recommend routine pre-operative dental consultation, but do urge a watchful eye. (See TABLE 3 for details on tonsillectomy techniques.)

Temporary uvular or tongue edema can develop if the tissues were forcefully manipulated during surgery, causing the patient to feel like he or she is choking. This edema usually subsides within a few days. Additional steroid dosing can be helpful.

If the glossopharyngeal nerve is damaged during dissection or electrocautery, the patient may develop a taste disorder.³¹ Spontaneous healing often occurs within several weeks.

An inadvertent eye injury such as keratitis may occur from exposure or friction from surgical covers. Skin, lip, and buccal mucosa lacerations or burns may also occur, often with the use of electrocautery needles and probes.

Mandibular condyle fracture is a rare complication that can be caused by forceful opening of the mouth. Forceful opening and the use of muscular relaxants lead to dislocation or stress injury of the temporomandibular joint (TMJ).³² This TMJ disorder is treated with analgesia, soft diet, and possible surgical reduction.

Excessive hyperextension of the neck can cause muscular tension and severe neck pain after the surgery. The atlanto-axial joint is at highest risk for dislocation or subluxation, especially in patients with Down syndrome—10% of whom already have atlanto-axial joint hyperlaxity.³³ With that in mind, neck radiographs should be taken prior to surgery to assess the atlanto-axial joint stability in patients with Down syndrome.

In the rare Grisel’s syndrome, patients complain of severe neck pain about a week after surgery, with limitation in neck movements and torticollis. This is due to an impaired venous drainage from the joint, local infection, and injury to the paraspinal ligaments.³⁴ Treatment consists of rest, neck immobilization, and antibiotics.

TABLE 3
Tonsillectomy techniques

Complications when an adenoidectomy is also done
There is an increased risk of complications when an adenoidectomy and tonsillectomy are performed together, simply because there is another operative site. The complications that may occur after adenoidectomy include velopharyngeal insufficiency and nasopharyngeal stenosis. Primary care physicians should watch for nasal or oral bleeding, crusting, halitosis, and velopharyngeal insufficiency. An ear, nose, and throat consultation is warranted when in doubt.

CORRESPONDENCE Yehudah Roth, MD, Department of Otolaryngology, Head & Neck Surgery, The Edith Wolfson Medical Center, P.O. Box 5, 58100 Holon, Israel; [email protected]

Although the number of patients undergoing tonsillectomy has gradually declined since the 1970s, it remains one of the most common surgical procedures performed worldwide.¹ The procedure, of course, is fairly routine—but not without risk.

Mortality rates for the operation range from 1 in 10,000 to 1 in 35,000, with morbidity rates ranging from 1.5% to 14%; mortality and morbidity after tonsillectomy are usually the result of postoperative bleeding.^2,3 In addition to bleeding, other common complications include pain, nausea, and vomiting.

The patient’s surgeon plays a central role in minimizing risk. But as a primary care physician, you, too, play an important role in ensuring that your patient’s tonsillectomy is uneventful. This review will help toward that end.

Refine your preop approach

Address informed consent issues. While the surgeon is responsible for obtaining informed consent, a patient may discuss the procedure with you, the family physician. (See TABLE 1 for the indications for tonsillectomy.) Although the procedure is safe and effective, the patient and his or her family need to know that bleeding will most likely occur immediately after the procedure, although it can occur at any time during the first 2 weeks postop. Advise the patient and family that postop pain is similar to that of a throat infection, but is often felt in the ears (“referred otalgia”). Because postop swallowing is painful, the patient may not drink enough fluids. If this problem because severe, he or she may need to be admitted for IV fluid replacement.

Discuss the patient’s length of stay. An increasing number of tonsillectomies are performed on an outpatient basis. However, high-risk patients—those with sleep apnea, coagulation disorders, or other underlying diseases, and anyone younger than 4 years of age or living a long distance from the hospital—should be admitted for overnight observation.⁴

Ask about previous surgeries. Tonsillectomies are performed under general anesthesia in most facilities. Ask about previous surgeries and conditions that may adversely affect the surgery: Does the patient have any upper airway obstruction, difficulty extending his cervical spine, or enzymatic deficiencies? A pseudocholinesterase deficiency, for instance, may cause prolonged paralysis of the respiratory muscles, requiring an extended period of time during which the patient must be mechanically ventilated.

Anesthesiologists use the Mallampati score to predict the ease of intubation.⁵ It’s derived by visualizing the base of the uvula, the tonsillar pillars, and the soft palate. Scoring may be done with or without phonation. A patient is considered to be in 1 of 4 classes, depending upon what can be visualized:

• Class 1: the tonsils, uvula, and soft palate
• Class 2: the hard and soft palate and upper portion of the tonsils and uvula
• Class 3: the soft and hard palate and base of the uvula
• Class 4: the hard palate.

As a general rule, patients scheduled for tonsillectomy are prone to higher scores because of underlying disease.

Order lab tests, stop certain medications. Instruct patients to discontinue aspirin 7 days before surgery, and naproxen and other nonsteroidal anti-inflammatory drugs (NSAIDs) 4 days before surgery.⁶ Antiplatelet aggregation drugs such as clopidogrel should also be stopped 7 days before surgery.⁶

While you should routinely order a complete blood count, experts disagree on the value and cost-effectiveness of routinely running prothrombin time and partial thromboplastin time.⁷

Assess whether the patient is anxious about the surgery. Tonsillectomy and the subsequent hospitalization can make patients anxious, bringing on sleep disturbances, behavioral problems, nightmares, enuresis, and emotional regression. Fortunately, these problems usually disappear without any intervention, but offering patients preop reassurance, demonstrations, and educational materials may help prevent them (TABLE 2).

TABLE 1
Indications for tonsillectomy

TABLE 2
Patient education resources

Address postop complications

Ensure that your patient has proper pain relief. Sore throat and referred otalgia are common, but usually subside within 2 weeks. The pain typically manifests as dysphagia, which results from trauma to the pharyngeal muscles. Keep in mind that pain is not always a minor complication; if prolonged, it may be accompanied by dehydration, fever, and even bleeding. Be sure to rule out otitis media in any patient complaining of otalgia who has had both tonsillectomy and adenoidectomy.

Nearly 50% of children who have had a tonsillectomy experience severe pain, defined as a visual analog score of 8 or more.⁸ According to 1 large-scale study, all children will experience moderate or severe pain while at rest and when swallowing during the first 48 hours after surgery—despite an appropriate around-the-clock dose of acetaminophen with codeine.⁹

Although high-dose acetaminophen, with or without codeine, is the most common analgesic prescribed for children after ambulatory surgery, its analgesic effect varies widely, so it may need to be administered in conjunction with other agents.¹⁰

One meta-analysis of several prospective, randomized, double-blind trials concluded that the risk of posttonsillectomy bleeding was nearly 4 times greater when patients received nonselective NSAIDs. This prompted the recommendation that these agents be avoided.¹¹ However, a more recent analysis that included selective cyclooxygenase-1 (COX-1) and COX-2 inhibitors found that these NSAIDs did not significantly alter perioperative bleeding.¹² With that in mind, selective COX-2 inhibitors such as celecoxib may safely relieve posttonsillectomy pain.¹³

Because so many patients develop postop nausea and vomiting, many surgeons avoid postop opioids such as codeine or tramadol, which can aggravate nausea without significant analgesic benefit. These agents are usually prescribed only after the gastrointestinal adverse effects have subsided, or when combined with antiemetics.

Ensure that nausea and vomiting have been properly addressed. Up to 89% of children undergoing tonsillectomy have postop vomiting and nausea.¹⁴ During the surgery, intravenous dexamethasone is usually administered because of its antiemetic effect—which can last up to 48 hours—and because of its role in preventing postop oropharyngeal edema.¹⁴

Watch for postop bleeding. Bleeding, followed by hypovolemic shock, is the most common cause of morbidity and mortality among patients undergoing tonsillectomy, affecting an estimated 0.5% to 10%. Approximately 1 in 200 patients returns to the operating room (OR) so that the bleeding can be controlled. Mortality from bleeding is 2 in 10,000 tonsillectomies. Most cases of fatal postop bleeding occur within the first 24 hours after surgery.^15-17

Although clinicians should attempt to estimate blood loss during active bleeding, the estimate may be too low in children because they tend to swallow blood. Despite this caveat, consider blood transfusion when the estimated blood loss is 10% or more of the blood volume in infants, 25% in children, and 20% in adults.¹⁸

If transfusion is warranted, 2 large intravenous lines will need to be inserted, blood samples will need to be sent for typing and cross-matching, and a complete blood count and coagulation studies will need to be ordered. The patient’s airway will need to be protected and hemodynamic monitoring maintained.

If the bleeding is not excessive, applying local pressure with gauze on the tonsillar bed may be effective, assuming the patient is cooperative. However, if this doesn’t work, the patient will need to be rushed to an OR to control the bleeding. Anesthetizing a patient with massive oral bleeding is among the most challenging emergencies. Always anticipate aspiration once the bleeding has been controlled.

Be ready for these complications, too

The following complications are occasionally encountered after tonsillectomy, but no large controlled studies have estimated their frequency.

Airway obstruction
Mucosal damage during intubation may cause air to enter parapharyngeal tissues, which in turn can result in laryngeal spasm and cervical emphysema after extubation. It can progress to pneumomediastinum and tension pneumothorax.^19,20 Excessive pressure on the surgical blades and manipulation within the oral cavity during surgery can cause lingual and oropharyngeal edema.²¹ With these potential complications in mind, the primary care physician will need to monitor the patient postoperatively for dyspnea.

Acute airway obstruction can occur if blood or clots accumulate in the hypo pharynx as the patient is awakening, immediately after extubation, or later. The most important way to prevent this complication is to ensure that there are no active bleeding sites. The airway can also be compromised from residual bleeding and an evolving airway edema.

On rare occasions, a patient’s airway can become obstructed if dislodged tonsillar tissue is not fully removed during surgery, or if loose teeth and parts of surgical instrumentation find their way into the airway. Being cognizant of these possibilities will help you, the primary care physician, to make an accurate differential diagnosis postoperatively.

Children suffering from obstructive sleep apnea due to craniofacial abnormalities, hypotonia, obesity, or “kissing” tonsils are at increased risk of developing acute airway maintenance difficulties after tonsillectomy. Prolonged obstruction of the upper airway from the obstructing tonsils can increase intrathoracic pressure and decrease venous return and pulmonary blood volume. After removal of the tonsils, there is a sudden increase in venous return, pulmonary blood volume, and pulmonary hydrostatic pressure. These rapid changes cause a fluid shift from pulmonary vessels into the lungs’ interstitial spaces and alveoli. The complication is sometimes referred to as “negative-pressure pulmonary edema.”^22,23

Appropriate management includes close monitoring and the use of continuous positive airway pressure ventilation, as needed. The same approach is appropriate in children with right-sided cardiac insufficiency or pulmonary hypertension.

Fever and infection
Fever may occur within 18 to 36 hours after tonsillectomy as a result of lung atelectasis, a response to anesthetic agents, or a transient bacteremia. A postop fever lasting more than 24 hours, accompanied by severe throat pain, suggests an emerging pharyngeal infection.

Some surgeons prescribe prophylactic postop antibiotics to reduce the risk of infection in the traumatized region, decrease pain, and enable the patient to return quickly to a normal oral diet.²⁴ A meta-analysis suggests, however, that postop oral antibiotics don’t reduce infection rates, posttonsillectomy pain, or secondary hemorrhage rates; the drugs do facilitate a return to normal activity and diet about a day sooner.^25,26

Pneumonia used to be a common complication due to aspirated blood during the procedure. With the advent of improved surgical and anesthetic techniques, however, this complication rarely occurs.² Neck infection, including suppurative lymphadenitis, is also a rare complication.²⁷

Dehydration
While postop dehydration may be uncommon in the “average” patient, the risk is more common among dysphagic children. Post-tonsillectomy vomiting worsens this situation. With that in mind, encourage patients to resume oral intake of fluids and, when necessary, provide analgesia to facilitate the process. You will need to readmit patients who are unable to drink to achieve better pain relief and restore hydration.²⁸

Some reports suggest that severe post-tonsillectomy hyponatremia can result from excessive hypovolemia-induced antidiuretic hormone secretion, disproportionate administration of hypotonic fluids, or inadequate postop fluid intake.^29,30

Encouraging patients to consume soft and cold foods after tonsillectomy will comfort patients and help restore hemostasis. Because pain is the major obstacle for a return to an oral diet, coupling analgesia and timing of oral intake is crucial for a quick recovery.

Injuries from a forceful surgical technique
Because tonsillectomy requires forceful insertion of instruments into the mouth, a loose tooth may become dislodged, or even aspirated. We don’t recommend routine pre-operative dental consultation, but do urge a watchful eye. (See TABLE 3 for details on tonsillectomy techniques.)

Temporary uvular or tongue edema can develop if the tissues were forcefully manipulated during surgery, causing the patient to feel like he or she is choking. This edema usually subsides within a few days. Additional steroid dosing can be helpful.

If the glossopharyngeal nerve is damaged during dissection or electrocautery, the patient may develop a taste disorder.³¹ Spontaneous healing often occurs within several weeks.

An inadvertent eye injury such as keratitis may occur from exposure or friction from surgical covers. Skin, lip, and buccal mucosa lacerations or burns may also occur, often with the use of electrocautery needles and probes.

Mandibular condyle fracture is a rare complication that can be caused by forceful opening of the mouth. Forceful opening and the use of muscular relaxants lead to dislocation or stress injury of the temporomandibular joint (TMJ).³² This TMJ disorder is treated with analgesia, soft diet, and possible surgical reduction.

Excessive hyperextension of the neck can cause muscular tension and severe neck pain after the surgery. The atlanto-axial joint is at highest risk for dislocation or subluxation, especially in patients with Down syndrome—10% of whom already have atlanto-axial joint hyperlaxity.³³ With that in mind, neck radiographs should be taken prior to surgery to assess the atlanto-axial joint stability in patients with Down syndrome.

In the rare Grisel’s syndrome, patients complain of severe neck pain about a week after surgery, with limitation in neck movements and torticollis. This is due to an impaired venous drainage from the joint, local infection, and injury to the paraspinal ligaments.³⁴ Treatment consists of rest, neck immobilization, and antibiotics.

TABLE 3
Tonsillectomy techniques

Complications when an adenoidectomy is also done
There is an increased risk of complications when an adenoidectomy and tonsillectomy are performed together, simply because there is another operative site. The complications that may occur after adenoidectomy include velopharyngeal insufficiency and nasopharyngeal stenosis. Primary care physicians should watch for nasal or oral bleeding, crusting, halitosis, and velopharyngeal insufficiency. An ear, nose, and throat consultation is warranted when in doubt.

CORRESPONDENCE Yehudah Roth, MD, Department of Otolaryngology, Head & Neck Surgery, The Edith Wolfson Medical Center, P.O. Box 5, 58100 Holon, Israel; [email protected]

Although the number of patients undergoing tonsillectomy has gradually declined since the 1970s, it remains one of the most common surgical procedures performed worldwide.¹ The procedure, of course, is fairly routine—but not without risk.

Mortality rates for the operation range from 1 in 10,000 to 1 in 35,000, with morbidity rates ranging from 1.5% to 14%; mortality and morbidity after tonsillectomy are usually the result of postoperative bleeding.^2,3 In addition to bleeding, other common complications include pain, nausea, and vomiting.

The patient’s surgeon plays a central role in minimizing risk. But as a primary care physician, you, too, play an important role in ensuring that your patient’s tonsillectomy is uneventful. This review will help toward that end.

Refine your preop approach

Address informed consent issues. While the surgeon is responsible for obtaining informed consent, a patient may discuss the procedure with you, the family physician. (See TABLE 1 for the indications for tonsillectomy.) Although the procedure is safe and effective, the patient and his or her family need to know that bleeding will most likely occur immediately after the procedure, although it can occur at any time during the first 2 weeks postop. Advise the patient and family that postop pain is similar to that of a throat infection, but is often felt in the ears (“referred otalgia”). Because postop swallowing is painful, the patient may not drink enough fluids. If this problem because severe, he or she may need to be admitted for IV fluid replacement.

Discuss the patient’s length of stay. An increasing number of tonsillectomies are performed on an outpatient basis. However, high-risk patients—those with sleep apnea, coagulation disorders, or other underlying diseases, and anyone younger than 4 years of age or living a long distance from the hospital—should be admitted for overnight observation.⁴

Ask about previous surgeries. Tonsillectomies are performed under general anesthesia in most facilities. Ask about previous surgeries and conditions that may adversely affect the surgery: Does the patient have any upper airway obstruction, difficulty extending his cervical spine, or enzymatic deficiencies? A pseudocholinesterase deficiency, for instance, may cause prolonged paralysis of the respiratory muscles, requiring an extended period of time during which the patient must be mechanically ventilated.

Anesthesiologists use the Mallampati score to predict the ease of intubation.⁵ It’s derived by visualizing the base of the uvula, the tonsillar pillars, and the soft palate. Scoring may be done with or without phonation. A patient is considered to be in 1 of 4 classes, depending upon what can be visualized:

• Class 1: the tonsils, uvula, and soft palate
• Class 2: the hard and soft palate and upper portion of the tonsils and uvula
• Class 3: the soft and hard palate and base of the uvula
• Class 4: the hard palate.

As a general rule, patients scheduled for tonsillectomy are prone to higher scores because of underlying disease.

Order lab tests, stop certain medications. Instruct patients to discontinue aspirin 7 days before surgery, and naproxen and other nonsteroidal anti-inflammatory drugs (NSAIDs) 4 days before surgery.⁶ Antiplatelet aggregation drugs such as clopidogrel should also be stopped 7 days before surgery.⁶

While you should routinely order a complete blood count, experts disagree on the value and cost-effectiveness of routinely running prothrombin time and partial thromboplastin time.⁷

Assess whether the patient is anxious about the surgery. Tonsillectomy and the subsequent hospitalization can make patients anxious, bringing on sleep disturbances, behavioral problems, nightmares, enuresis, and emotional regression. Fortunately, these problems usually disappear without any intervention, but offering patients preop reassurance, demonstrations, and educational materials may help prevent them (TABLE 2).

TABLE 1
Indications for tonsillectomy

TABLE 2
Patient education resources

Address postop complications

Ensure that your patient has proper pain relief. Sore throat and referred otalgia are common, but usually subside within 2 weeks. The pain typically manifests as dysphagia, which results from trauma to the pharyngeal muscles. Keep in mind that pain is not always a minor complication; if prolonged, it may be accompanied by dehydration, fever, and even bleeding. Be sure to rule out otitis media in any patient complaining of otalgia who has had both tonsillectomy and adenoidectomy.

Nearly 50% of children who have had a tonsillectomy experience severe pain, defined as a visual analog score of 8 or more.⁸ According to 1 large-scale study, all children will experience moderate or severe pain while at rest and when swallowing during the first 48 hours after surgery—despite an appropriate around-the-clock dose of acetaminophen with codeine.⁹

Although high-dose acetaminophen, with or without codeine, is the most common analgesic prescribed for children after ambulatory surgery, its analgesic effect varies widely, so it may need to be administered in conjunction with other agents.¹⁰

One meta-analysis of several prospective, randomized, double-blind trials concluded that the risk of posttonsillectomy bleeding was nearly 4 times greater when patients received nonselective NSAIDs. This prompted the recommendation that these agents be avoided.¹¹ However, a more recent analysis that included selective cyclooxygenase-1 (COX-1) and COX-2 inhibitors found that these NSAIDs did not significantly alter perioperative bleeding.¹² With that in mind, selective COX-2 inhibitors such as celecoxib may safely relieve posttonsillectomy pain.¹³

Because so many patients develop postop nausea and vomiting, many surgeons avoid postop opioids such as codeine or tramadol, which can aggravate nausea without significant analgesic benefit. These agents are usually prescribed only after the gastrointestinal adverse effects have subsided, or when combined with antiemetics.

Ensure that nausea and vomiting have been properly addressed. Up to 89% of children undergoing tonsillectomy have postop vomiting and nausea.¹⁴ During the surgery, intravenous dexamethasone is usually administered because of its antiemetic effect—which can last up to 48 hours—and because of its role in preventing postop oropharyngeal edema.¹⁴

Watch for postop bleeding. Bleeding, followed by hypovolemic shock, is the most common cause of morbidity and mortality among patients undergoing tonsillectomy, affecting an estimated 0.5% to 10%. Approximately 1 in 200 patients returns to the operating room (OR) so that the bleeding can be controlled. Mortality from bleeding is 2 in 10,000 tonsillectomies. Most cases of fatal postop bleeding occur within the first 24 hours after surgery.^15-17

Although clinicians should attempt to estimate blood loss during active bleeding, the estimate may be too low in children because they tend to swallow blood. Despite this caveat, consider blood transfusion when the estimated blood loss is 10% or more of the blood volume in infants, 25% in children, and 20% in adults.¹⁸

If transfusion is warranted, 2 large intravenous lines will need to be inserted, blood samples will need to be sent for typing and cross-matching, and a complete blood count and coagulation studies will need to be ordered. The patient’s airway will need to be protected and hemodynamic monitoring maintained.

If the bleeding is not excessive, applying local pressure with gauze on the tonsillar bed may be effective, assuming the patient is cooperative. However, if this doesn’t work, the patient will need to be rushed to an OR to control the bleeding. Anesthetizing a patient with massive oral bleeding is among the most challenging emergencies. Always anticipate aspiration once the bleeding has been controlled.

Be ready for these complications, too

The following complications are occasionally encountered after tonsillectomy, but no large controlled studies have estimated their frequency.

Airway obstruction
Mucosal damage during intubation may cause air to enter parapharyngeal tissues, which in turn can result in laryngeal spasm and cervical emphysema after extubation. It can progress to pneumomediastinum and tension pneumothorax.^19,20 Excessive pressure on the surgical blades and manipulation within the oral cavity during surgery can cause lingual and oropharyngeal edema.²¹ With these potential complications in mind, the primary care physician will need to monitor the patient postoperatively for dyspnea.

Acute airway obstruction can occur if blood or clots accumulate in the hypo pharynx as the patient is awakening, immediately after extubation, or later. The most important way to prevent this complication is to ensure that there are no active bleeding sites. The airway can also be compromised from residual bleeding and an evolving airway edema.

On rare occasions, a patient’s airway can become obstructed if dislodged tonsillar tissue is not fully removed during surgery, or if loose teeth and parts of surgical instrumentation find their way into the airway. Being cognizant of these possibilities will help you, the primary care physician, to make an accurate differential diagnosis postoperatively.

Children suffering from obstructive sleep apnea due to craniofacial abnormalities, hypotonia, obesity, or “kissing” tonsils are at increased risk of developing acute airway maintenance difficulties after tonsillectomy. Prolonged obstruction of the upper airway from the obstructing tonsils can increase intrathoracic pressure and decrease venous return and pulmonary blood volume. After removal of the tonsils, there is a sudden increase in venous return, pulmonary blood volume, and pulmonary hydrostatic pressure. These rapid changes cause a fluid shift from pulmonary vessels into the lungs’ interstitial spaces and alveoli. The complication is sometimes referred to as “negative-pressure pulmonary edema.”^22,23

Appropriate management includes close monitoring and the use of continuous positive airway pressure ventilation, as needed. The same approach is appropriate in children with right-sided cardiac insufficiency or pulmonary hypertension.

Fever and infection
Fever may occur within 18 to 36 hours after tonsillectomy as a result of lung atelectasis, a response to anesthetic agents, or a transient bacteremia. A postop fever lasting more than 24 hours, accompanied by severe throat pain, suggests an emerging pharyngeal infection.

Some surgeons prescribe prophylactic postop antibiotics to reduce the risk of infection in the traumatized region, decrease pain, and enable the patient to return quickly to a normal oral diet.²⁴ A meta-analysis suggests, however, that postop oral antibiotics don’t reduce infection rates, posttonsillectomy pain, or secondary hemorrhage rates; the drugs do facilitate a return to normal activity and diet about a day sooner.^25,26

Pneumonia used to be a common complication due to aspirated blood during the procedure. With the advent of improved surgical and anesthetic techniques, however, this complication rarely occurs.² Neck infection, including suppurative lymphadenitis, is also a rare complication.²⁷

Dehydration
While postop dehydration may be uncommon in the “average” patient, the risk is more common among dysphagic children. Post-tonsillectomy vomiting worsens this situation. With that in mind, encourage patients to resume oral intake of fluids and, when necessary, provide analgesia to facilitate the process. You will need to readmit patients who are unable to drink to achieve better pain relief and restore hydration.²⁸

Some reports suggest that severe post-tonsillectomy hyponatremia can result from excessive hypovolemia-induced antidiuretic hormone secretion, disproportionate administration of hypotonic fluids, or inadequate postop fluid intake.^29,30

Encouraging patients to consume soft and cold foods after tonsillectomy will comfort patients and help restore hemostasis. Because pain is the major obstacle for a return to an oral diet, coupling analgesia and timing of oral intake is crucial for a quick recovery.

Injuries from a forceful surgical technique
Because tonsillectomy requires forceful insertion of instruments into the mouth, a loose tooth may become dislodged, or even aspirated. We don’t recommend routine pre-operative dental consultation, but do urge a watchful eye. (See TABLE 3 for details on tonsillectomy techniques.)

Temporary uvular or tongue edema can develop if the tissues were forcefully manipulated during surgery, causing the patient to feel like he or she is choking. This edema usually subsides within a few days. Additional steroid dosing can be helpful.

If the glossopharyngeal nerve is damaged during dissection or electrocautery, the patient may develop a taste disorder.³¹ Spontaneous healing often occurs within several weeks.

An inadvertent eye injury such as keratitis may occur from exposure or friction from surgical covers. Skin, lip, and buccal mucosa lacerations or burns may also occur, often with the use of electrocautery needles and probes.

Mandibular condyle fracture is a rare complication that can be caused by forceful opening of the mouth. Forceful opening and the use of muscular relaxants lead to dislocation or stress injury of the temporomandibular joint (TMJ).³² This TMJ disorder is treated with analgesia, soft diet, and possible surgical reduction.

Excessive hyperextension of the neck can cause muscular tension and severe neck pain after the surgery. The atlanto-axial joint is at highest risk for dislocation or subluxation, especially in patients with Down syndrome—10% of whom already have atlanto-axial joint hyperlaxity.³³ With that in mind, neck radiographs should be taken prior to surgery to assess the atlanto-axial joint stability in patients with Down syndrome.

In the rare Grisel’s syndrome, patients complain of severe neck pain about a week after surgery, with limitation in neck movements and torticollis. This is due to an impaired venous drainage from the joint, local infection, and injury to the paraspinal ligaments.³⁴ Treatment consists of rest, neck immobilization, and antibiotics.

TABLE 3
Tonsillectomy techniques

Complications when an adenoidectomy is also done
There is an increased risk of complications when an adenoidectomy and tonsillectomy are performed together, simply because there is another operative site. The complications that may occur after adenoidectomy include velopharyngeal insufficiency and nasopharyngeal stenosis. Primary care physicians should watch for nasal or oral bleeding, crusting, halitosis, and velopharyngeal insufficiency. An ear, nose, and throat consultation is warranted when in doubt.

CORRESPONDENCE Yehudah Roth, MD, Department of Otolaryngology, Head & Neck Surgery, The Edith Wolfson Medical Center, P.O. Box 5, 58100 Holon, Israel; [email protected]

CASE Sam J. is a 56-year-old man with a 2-day history of severe pain in the front of his neck. He aches all over and says he had “a bad cold” a week ago, but no fever, chills, cough, or headache. He hasn’t had a toothache or difficulty swallowing, but he does complain of a sore throat and muscle aches. His temperature is 100.7°f. He has no respiratory distress or stridor. there are no tonsillar exudates. the pharynx is not red or inflamed.

When you palpate his thyroid gland, you find diffuse tenderness. What could be causing his pain?

The patient with a painful thyroid is uncommon in clinical practice. In most cases, the culprit is a viral infection, radiation, or a traumatic injury to the gland. Bacterial, fungal, parasitic, or mycobacterial infections also occur, but these are rare, making up only 0.1% to 0.7% of all thyroid disease.¹ Some thyroid infections can be devastating, necessitating quick interventions including airway protection, intravenous antibiotics, and surgical drainage in certain situations.

Subacute thyroiditis (ST), the most common source of thyroid pain, is less severe, but can be disabling if not recognized and appropriately treated.² Occasionally, thyroid cysts can rupture, mimicking a presentation of ST.

This article will focus on how to evaluate the painful thyroid, so as to differentiate among the various conditions that can produce this uncommon presentation.

Classifying thyroiditis: Pain or no pain

Thyroiditis can be classified in several ways, but the simplest classification is in terms of pain. Conditions that can cause painless thyroiditis include Hashimoto’s disease, postpartum thyroiditis, subacute lymphocytic thyroiditis, Reidel’s thyroiditis, and thyroiditis induced by drugs such as amiodarone, interleukin-2, lithium, and interferon-alfa.³ Conditions that can cause painful thyroiditis include acute suppurative thyroiditis (AST), radiation-induced thyroiditis, trauma, ST, and ruptured cysts.¹

Acute suppurative thyroiditis
AST is a rare condition most commonly caused by bacterial, mycobacterial, fungal, or parasitic infection. The thyroid is remarkably resistant to infection because the gland is encapsulated, rich in iodide content, and has an abundant blood supply and lymphatic drainage.^1,2 AST usually occurs in patients with underlying thyroid disease, such as thyroid cancer, Hashimoto’s thyroiditis, multinodular goiter, congenital thyroid abnormalities such as pyriform sinus fistulas in children, and in patients who are immunocompromised.^1,2

Patients with AST are acutely ill and will complain of fever, dysphagia, and severe anterior neck pain. Stridor and tracheal obstruction may occur, requiring intubation or tracheotomy. Thyroid hormone levels are sometimes increased secondary to the release of preformed hormone from the intensely inflamed gland. Computed tomography (CT) or magnetic resonance imaging (MRI) thyroid scans can identify abscesses and guide biopsies that clinch the diagnosis.

Management consists of airway support, broad-spectrum intravenous antibiotics until gram stains and cultures return (with particular attention to coverage against staphylococcal, streptococcal, and anaerobic organisms), and surgical drainage of abscesses.

Thyroiditis caused by radiation or trauma
One percent of patients develop thyroiditis after receiving radioactive iodine ablation for hyperthyroidism (Graves’ disease).⁴ This condition usually occurs 5 to 10 days after the treatment. Radiation triggers thyroid gland destruction, causing transient additional release of thyroid hormone. Similar damage occurs after external beam radiation for lymphomas and head and neck cancers. In such cases, use nonsteroidal anti-inflammatory drugs (NSAIDs) and prednisone to reduce inflammation and beta-blockers to control hyperthyroid symptoms.

Trauma to the thyroid—from motor vehicle accidents, falls, or physical assaults—can cause pain, tenderness, and abnormalities in thyroid hormone levels. Effects are usually self-limited, and management centers on airway protection and prevention of infection.

Subacute thyroiditis
The cause of ST is not clear. However, it typically follows an upper respiratory infection. Many viruses have been implicated in the disease, including Coxsackie, echo, adeno, and influenza viruses.⁵ It occurs more frequently during summer months, coinciding with the peak incidence of enterovirus.² ST is 4 times more common in women than men, and the median age for occurrence is 45.⁶

Patients with ST usually seek care for such symptoms as fever, myalgias, and malaise, which suggest a viral infection.^2,7-9 Neck pain is significant and may radiate to the jaw or ear.⁹ Patients may have pain on swallowing and a sore throat. It’s possible to confuse ST with severe pharyngitis or even epiglottitis if you fail to palpate the thyroid gland. When you examine the thyroid, the gland is exquisitely tender with ST. Patients with ST appear less ill than those with AST.

In 50% of ST patients, thyroid-stimulating hormone (TSH) is suppressed, and thyroid hormone levels are elevated.⁴ Patients with hyperthyroidism exhibit the usual signs and symptoms associated with high levels of the hormone: tremor, palpitations, heat intolerance, and diarrhea. After a few weeks, thyroid hormone levels normalize and then drop to below normal in the 4 to 6 months that follow. Levels return to normal in 95% of patients after 6 to 12 months. ST recurs in only 2% of patients.⁴

Sedimentation rates are almost always elevated in ST, reflecting the significant inflammation associated with the disease. Cytotoxic T lymphocytes damage thyroid follicles, causing release of preformed thyroid hormone, in turn suppressing TSH.⁴ Radioactive iodine scans show low uptake because TSH is needed for the uptake of iodine.¹⁰

Treatment of ST should focus on reducing inflammation and pain. NSAIDs may be sufficient, but patients with moderate to severe symptoms may require prednisone 40 mg daily, tapered over 4 to 6 weeks.² If the patient is hyperthyroid, you’ll need to control symptoms with beta-blockers.

Ruptured thyroid cyst: An ST look-alike
The clinical presentation of a ruptured cyst closely mimics that of ST: thyroid pain, history of a recent cold, sore throat, aches, and pains. Laboratory and nuclear medicine evaluations, however, reveal significant differences:

• TSH is not suppressed;
• ESR is not elevated;
• Radioactive iodine uptake is normal.

Palpation may reveal a thyroid nodule. You can confirm the diagnosis by ultrasound and a needle aspiration.

CASE The right diagnosis for Sam

Sam’s initial clinical presentation suggested St, and he was started on prednisone 40 mg daily to control his pain. but when laboratory and nuclear medicine evaluations became available the next day, the St diagnosis didn’t hold up: TSH, ESR, and radioactive iodine uptake were all normal.

When Sam came in for a follow-up visit 3 days later, his thyroid pain was gone and he was feeling much better. palpation of his thyroid revealed a slightly tender nodule that went undetected in the initial exam. Subsequent thyroid ultrasound showed a 2x2 cm nodule in the right lobe. a fine needle aspiration revealed a colloid cyst. the cyst had ruptured, causing acute pain from hemorrhage and inflammation.

CORRESPONDENCE Dean Gianakos, MD, 2323 Memorial Drive, #10, Lynchburg Family Medicine Residency, Lynchburg, VA 24501; [email protected]

CASE Sam J. is a 56-year-old man with a 2-day history of severe pain in the front of his neck. He aches all over and says he had “a bad cold” a week ago, but no fever, chills, cough, or headache. He hasn’t had a toothache or difficulty swallowing, but he does complain of a sore throat and muscle aches. His temperature is 100.7°f. He has no respiratory distress or stridor. there are no tonsillar exudates. the pharynx is not red or inflamed.

When you palpate his thyroid gland, you find diffuse tenderness. What could be causing his pain?

The patient with a painful thyroid is uncommon in clinical practice. In most cases, the culprit is a viral infection, radiation, or a traumatic injury to the gland. Bacterial, fungal, parasitic, or mycobacterial infections also occur, but these are rare, making up only 0.1% to 0.7% of all thyroid disease.¹ Some thyroid infections can be devastating, necessitating quick interventions including airway protection, intravenous antibiotics, and surgical drainage in certain situations.

Subacute thyroiditis (ST), the most common source of thyroid pain, is less severe, but can be disabling if not recognized and appropriately treated.² Occasionally, thyroid cysts can rupture, mimicking a presentation of ST.

This article will focus on how to evaluate the painful thyroid, so as to differentiate among the various conditions that can produce this uncommon presentation.

Classifying thyroiditis: Pain or no pain

Thyroiditis can be classified in several ways, but the simplest classification is in terms of pain. Conditions that can cause painless thyroiditis include Hashimoto’s disease, postpartum thyroiditis, subacute lymphocytic thyroiditis, Reidel’s thyroiditis, and thyroiditis induced by drugs such as amiodarone, interleukin-2, lithium, and interferon-alfa.³ Conditions that can cause painful thyroiditis include acute suppurative thyroiditis (AST), radiation-induced thyroiditis, trauma, ST, and ruptured cysts.¹

Acute suppurative thyroiditis
AST is a rare condition most commonly caused by bacterial, mycobacterial, fungal, or parasitic infection. The thyroid is remarkably resistant to infection because the gland is encapsulated, rich in iodide content, and has an abundant blood supply and lymphatic drainage.^1,2 AST usually occurs in patients with underlying thyroid disease, such as thyroid cancer, Hashimoto’s thyroiditis, multinodular goiter, congenital thyroid abnormalities such as pyriform sinus fistulas in children, and in patients who are immunocompromised.^1,2

Patients with AST are acutely ill and will complain of fever, dysphagia, and severe anterior neck pain. Stridor and tracheal obstruction may occur, requiring intubation or tracheotomy. Thyroid hormone levels are sometimes increased secondary to the release of preformed hormone from the intensely inflamed gland. Computed tomography (CT) or magnetic resonance imaging (MRI) thyroid scans can identify abscesses and guide biopsies that clinch the diagnosis.

Management consists of airway support, broad-spectrum intravenous antibiotics until gram stains and cultures return (with particular attention to coverage against staphylococcal, streptococcal, and anaerobic organisms), and surgical drainage of abscesses.

Thyroiditis caused by radiation or trauma
One percent of patients develop thyroiditis after receiving radioactive iodine ablation for hyperthyroidism (Graves’ disease).⁴ This condition usually occurs 5 to 10 days after the treatment. Radiation triggers thyroid gland destruction, causing transient additional release of thyroid hormone. Similar damage occurs after external beam radiation for lymphomas and head and neck cancers. In such cases, use nonsteroidal anti-inflammatory drugs (NSAIDs) and prednisone to reduce inflammation and beta-blockers to control hyperthyroid symptoms.

Trauma to the thyroid—from motor vehicle accidents, falls, or physical assaults—can cause pain, tenderness, and abnormalities in thyroid hormone levels. Effects are usually self-limited, and management centers on airway protection and prevention of infection.

Subacute thyroiditis
The cause of ST is not clear. However, it typically follows an upper respiratory infection. Many viruses have been implicated in the disease, including Coxsackie, echo, adeno, and influenza viruses.⁵ It occurs more frequently during summer months, coinciding with the peak incidence of enterovirus.² ST is 4 times more common in women than men, and the median age for occurrence is 45.⁶

Patients with ST usually seek care for such symptoms as fever, myalgias, and malaise, which suggest a viral infection.^2,7-9 Neck pain is significant and may radiate to the jaw or ear.⁹ Patients may have pain on swallowing and a sore throat. It’s possible to confuse ST with severe pharyngitis or even epiglottitis if you fail to palpate the thyroid gland. When you examine the thyroid, the gland is exquisitely tender with ST. Patients with ST appear less ill than those with AST.

In 50% of ST patients, thyroid-stimulating hormone (TSH) is suppressed, and thyroid hormone levels are elevated.⁴ Patients with hyperthyroidism exhibit the usual signs and symptoms associated with high levels of the hormone: tremor, palpitations, heat intolerance, and diarrhea. After a few weeks, thyroid hormone levels normalize and then drop to below normal in the 4 to 6 months that follow. Levels return to normal in 95% of patients after 6 to 12 months. ST recurs in only 2% of patients.⁴

Sedimentation rates are almost always elevated in ST, reflecting the significant inflammation associated with the disease. Cytotoxic T lymphocytes damage thyroid follicles, causing release of preformed thyroid hormone, in turn suppressing TSH.⁴ Radioactive iodine scans show low uptake because TSH is needed for the uptake of iodine.¹⁰

Treatment of ST should focus on reducing inflammation and pain. NSAIDs may be sufficient, but patients with moderate to severe symptoms may require prednisone 40 mg daily, tapered over 4 to 6 weeks.² If the patient is hyperthyroid, you’ll need to control symptoms with beta-blockers.

Ruptured thyroid cyst: An ST look-alike
The clinical presentation of a ruptured cyst closely mimics that of ST: thyroid pain, history of a recent cold, sore throat, aches, and pains. Laboratory and nuclear medicine evaluations, however, reveal significant differences:

• TSH is not suppressed;
• ESR is not elevated;
• Radioactive iodine uptake is normal.

Palpation may reveal a thyroid nodule. You can confirm the diagnosis by ultrasound and a needle aspiration.

CASE The right diagnosis for Sam

Sam’s initial clinical presentation suggested St, and he was started on prednisone 40 mg daily to control his pain. but when laboratory and nuclear medicine evaluations became available the next day, the St diagnosis didn’t hold up: TSH, ESR, and radioactive iodine uptake were all normal.

When Sam came in for a follow-up visit 3 days later, his thyroid pain was gone and he was feeling much better. palpation of his thyroid revealed a slightly tender nodule that went undetected in the initial exam. Subsequent thyroid ultrasound showed a 2x2 cm nodule in the right lobe. a fine needle aspiration revealed a colloid cyst. the cyst had ruptured, causing acute pain from hemorrhage and inflammation.

CORRESPONDENCE Dean Gianakos, MD, 2323 Memorial Drive, #10, Lynchburg Family Medicine Residency, Lynchburg, VA 24501; [email protected]

CASE Sam J. is a 56-year-old man with a 2-day history of severe pain in the front of his neck. He aches all over and says he had “a bad cold” a week ago, but no fever, chills, cough, or headache. He hasn’t had a toothache or difficulty swallowing, but he does complain of a sore throat and muscle aches. His temperature is 100.7°f. He has no respiratory distress or stridor. there are no tonsillar exudates. the pharynx is not red or inflamed.

When you palpate his thyroid gland, you find diffuse tenderness. What could be causing his pain?

The patient with a painful thyroid is uncommon in clinical practice. In most cases, the culprit is a viral infection, radiation, or a traumatic injury to the gland. Bacterial, fungal, parasitic, or mycobacterial infections also occur, but these are rare, making up only 0.1% to 0.7% of all thyroid disease.¹ Some thyroid infections can be devastating, necessitating quick interventions including airway protection, intravenous antibiotics, and surgical drainage in certain situations.

Subacute thyroiditis (ST), the most common source of thyroid pain, is less severe, but can be disabling if not recognized and appropriately treated.² Occasionally, thyroid cysts can rupture, mimicking a presentation of ST.

This article will focus on how to evaluate the painful thyroid, so as to differentiate among the various conditions that can produce this uncommon presentation.

Classifying thyroiditis: Pain or no pain

Thyroiditis can be classified in several ways, but the simplest classification is in terms of pain. Conditions that can cause painless thyroiditis include Hashimoto’s disease, postpartum thyroiditis, subacute lymphocytic thyroiditis, Reidel’s thyroiditis, and thyroiditis induced by drugs such as amiodarone, interleukin-2, lithium, and interferon-alfa.³ Conditions that can cause painful thyroiditis include acute suppurative thyroiditis (AST), radiation-induced thyroiditis, trauma, ST, and ruptured cysts.¹

Acute suppurative thyroiditis
AST is a rare condition most commonly caused by bacterial, mycobacterial, fungal, or parasitic infection. The thyroid is remarkably resistant to infection because the gland is encapsulated, rich in iodide content, and has an abundant blood supply and lymphatic drainage.^1,2 AST usually occurs in patients with underlying thyroid disease, such as thyroid cancer, Hashimoto’s thyroiditis, multinodular goiter, congenital thyroid abnormalities such as pyriform sinus fistulas in children, and in patients who are immunocompromised.^1,2

Patients with AST are acutely ill and will complain of fever, dysphagia, and severe anterior neck pain. Stridor and tracheal obstruction may occur, requiring intubation or tracheotomy. Thyroid hormone levels are sometimes increased secondary to the release of preformed hormone from the intensely inflamed gland. Computed tomography (CT) or magnetic resonance imaging (MRI) thyroid scans can identify abscesses and guide biopsies that clinch the diagnosis.

Management consists of airway support, broad-spectrum intravenous antibiotics until gram stains and cultures return (with particular attention to coverage against staphylococcal, streptococcal, and anaerobic organisms), and surgical drainage of abscesses.

Thyroiditis caused by radiation or trauma
One percent of patients develop thyroiditis after receiving radioactive iodine ablation for hyperthyroidism (Graves’ disease).⁴ This condition usually occurs 5 to 10 days after the treatment. Radiation triggers thyroid gland destruction, causing transient additional release of thyroid hormone. Similar damage occurs after external beam radiation for lymphomas and head and neck cancers. In such cases, use nonsteroidal anti-inflammatory drugs (NSAIDs) and prednisone to reduce inflammation and beta-blockers to control hyperthyroid symptoms.

Trauma to the thyroid—from motor vehicle accidents, falls, or physical assaults—can cause pain, tenderness, and abnormalities in thyroid hormone levels. Effects are usually self-limited, and management centers on airway protection and prevention of infection.

Subacute thyroiditis
The cause of ST is not clear. However, it typically follows an upper respiratory infection. Many viruses have been implicated in the disease, including Coxsackie, echo, adeno, and influenza viruses.⁵ It occurs more frequently during summer months, coinciding with the peak incidence of enterovirus.² ST is 4 times more common in women than men, and the median age for occurrence is 45.⁶

Patients with ST usually seek care for such symptoms as fever, myalgias, and malaise, which suggest a viral infection.^2,7-9 Neck pain is significant and may radiate to the jaw or ear.⁹ Patients may have pain on swallowing and a sore throat. It’s possible to confuse ST with severe pharyngitis or even epiglottitis if you fail to palpate the thyroid gland. When you examine the thyroid, the gland is exquisitely tender with ST. Patients with ST appear less ill than those with AST.

In 50% of ST patients, thyroid-stimulating hormone (TSH) is suppressed, and thyroid hormone levels are elevated.⁴ Patients with hyperthyroidism exhibit the usual signs and symptoms associated with high levels of the hormone: tremor, palpitations, heat intolerance, and diarrhea. After a few weeks, thyroid hormone levels normalize and then drop to below normal in the 4 to 6 months that follow. Levels return to normal in 95% of patients after 6 to 12 months. ST recurs in only 2% of patients.⁴

Sedimentation rates are almost always elevated in ST, reflecting the significant inflammation associated with the disease. Cytotoxic T lymphocytes damage thyroid follicles, causing release of preformed thyroid hormone, in turn suppressing TSH.⁴ Radioactive iodine scans show low uptake because TSH is needed for the uptake of iodine.¹⁰

Treatment of ST should focus on reducing inflammation and pain. NSAIDs may be sufficient, but patients with moderate to severe symptoms may require prednisone 40 mg daily, tapered over 4 to 6 weeks.² If the patient is hyperthyroid, you’ll need to control symptoms with beta-blockers.

Ruptured thyroid cyst: An ST look-alike
The clinical presentation of a ruptured cyst closely mimics that of ST: thyroid pain, history of a recent cold, sore throat, aches, and pains. Laboratory and nuclear medicine evaluations, however, reveal significant differences:

• TSH is not suppressed;
• ESR is not elevated;
• Radioactive iodine uptake is normal.

Palpation may reveal a thyroid nodule. You can confirm the diagnosis by ultrasound and a needle aspiration.

CASE The right diagnosis for Sam

Sam’s initial clinical presentation suggested St, and he was started on prednisone 40 mg daily to control his pain. but when laboratory and nuclear medicine evaluations became available the next day, the St diagnosis didn’t hold up: TSH, ESR, and radioactive iodine uptake were all normal.

When Sam came in for a follow-up visit 3 days later, his thyroid pain was gone and he was feeling much better. palpation of his thyroid revealed a slightly tender nodule that went undetected in the initial exam. Subsequent thyroid ultrasound showed a 2x2 cm nodule in the right lobe. a fine needle aspiration revealed a colloid cyst. the cyst had ruptured, causing acute pain from hemorrhage and inflammation.

CORRESPONDENCE Dean Gianakos, MD, 2323 Memorial Drive, #10, Lynchburg Family Medicine Residency, Lynchburg, VA 24501; [email protected]

Late last year, the US Preventive Services Task Force (USPSTF) sparked a nationwide controversy when it announced that it was recommending against routine screening mammography for women younger than age 50.¹ Indeed, that’s a recommendation that many other organizations, including the American Cancer Society (ACS),² the American College of Obstetricians and Gynecologists (ACOG),³ and the National Comprehensive Cancer Network (NCCN),⁴ disagree with. But the age at which women should begin routine mammography isn’t the only controversial question. Experts disagree on the benefits of breast self-examination, the optimal frequency of clinical breast exams, and the use of digital mammography—among other issues. This evidence-based review can help you cut through the confusion.

CASE Carrie, a 39-year-old woman who has never been pregnant, comes in for an annual Pap smear and gynecologic exam. She has a negative past medical history, but a positive family history for breast cancer—both her mother and 1 of her sisters had the disease. How would you assess Carrie’s risk of breast cancer, and what preventive measures would you recommend?

Use this predictive model to pinpoint your patient’s risk

When making decisions regarding primary prevention of and screening for breast cancer, an accurate assessment of risk is critical. Many predictive models have been developed with that in mind. The most widely studied, the Gail model, incorporates a number of important risk factors (TABLE 1), including age; race; family history; reproductive factors such as age of menarche, menopause, and first childbirth; and previous history of breast biopsy and atypical findings, to calculate a woman’s 5-year risk.⁵

A risk calculator (the Breast Cancer Risk Assessment Tool) based on the Gail model is available on the National Cancer Institute’s Web site, at http://www.cancer.gov/bcrisktool. Generally, a score ≥1.66%,⁵ which indicates that a patient has at least a 1.66% chance of developing breast cancer over the next 5 years, is considered high risk.^6,7

CASE Carrie’s 2 first-degree relatives affected by breast cancer and her nulliparous status place her at increased risk. Further questioning reveals a particularly strong family history, as both relatives were diagnosed before the age of 50 (her mom at 45 years of age and her sister, at 39). Carrie’s 5-year risk is 1.8%.

TABLE 1
Risk factors for breast cancer^5,29

All women can benefit from these preventive measures

As primary care physicians, we have a responsibility to stress lifestyle modification as the mainstay of breast cancer prevention. Whether or not a woman is at high risk, advise her that maintaining a normal weight, exercising vigorously, limiting alcohol consumption, and breastfeeding are evidence-based methods of primary prevention. Diets low in fat and high in fiber may be associated with a lower risk of invasive breast cancer, but there is no conclusive evidence to support specific dietary interventions to reduce the risk.^8-11 Nor has a link between active or passive smoking, antioxidants, or fruit and vegetable intake been firmly established.¹²

There is a clear association between prolonged estrogen exposure and breast cancer, however. Many reproductive factors, such as early menarche, late menopause, later age at time of first full-term pregnancy, and nulliparity, increase a woman’s exposure to endogenous estrogen—and her risk of developing breast cancer.^12,13

Exposure to exogenous estrogen is also linked to the development of breast cancer. In 2002, the Women’s Health Initiative (WHI) was stopped early after a report was released stating that the risks of hormone replacement therapy (HRT)—a higher incidence of cardiovascular events, stroke, and venous thromboembolism, as well as breast cancer—outweighed the benefits.¹⁴ Subsequent analyses have found a relationship between the declining incidence of breast cancer and the marked decrease in HRT use prompted by the WHI report. While causality has not been firmly established, multiple studies strongly suggest it.^15,16

The association between oral contraceptives (OCs) and breast cancer is more controversial. Some studies have found an increased breast cancer risk among OC users, but both the relative risk and absolute risk were found to be very small and to dissipate 10 years after stopping OC use. More recent studies with newer formulations containing lower doses of estrogen have failed to show an increased risk.⁸

Breast cancer screening: The parameters have changed

Various organizations have published guidelines for breast cancer screening (TABLE 2), and all are somewhat different. Here’s what you need to know.

Breast self-examination (BSE), which women were previously advised to perform monthly, has not been shown to improve mortality in any age group, and is no longer routinely recommended.¹⁷ While both the USPSTF and the Canadian Task Force on Preventive Health Care recommend against teaching women BSE, the ACS, ACOG, and NCCN encourage self-examination—particularly among women older than 40 years.^1-4,17,18

Clinical breast examination has an average sensitivity of 50% and detects approximately 5% of mammographically occult cancers.¹⁹ It is still not clear whether clinical breast exams save lives, however—a finding that is reflected in the USPSTF’s “I” (insufficient evidence to assess the benefits and harms) recommendation.¹ Other consensus guidelines still recommend clinical breast examination, albeit at varying frequencies.

Screening mammography decreases mortality rates by anywhere from 28% to 65%, depending on the statistical model used.²⁰ The benefit is greatest in women between the ages of 50 and 69 years, however, and most groups agree that mammography every 1 to 2 years is advisable for this age group. (There is limited data on the value of mammography for women 70 years of age and older, and no consensus on the age at which to stop screening.^1,21) But because the mortality benefit from screening mammography is lower for women aged 40 to 49, guidelines for this age group are more controversial.

Mammography’s sensitivity is affected by a variety of factors, including age and menopausal status, prior breast surgery or radiation, breast density, and the experience of the radiologist. Women in their 40s have denser breast tissue than older women, making mammography less sensitive for this age group. Because of that, and because the overall incidence of breast cancer is lower for women younger than 50, some argue that screening mammography for women between the ages of 40 and 49 years leads to unacceptably high false-positive rates (9.8% annually²²) and that the harm associated with mammography may outweigh the benefit. Others counter that tumors in younger women tend to be more aggressive and faster growing, making early detection even more critical than for older women.

What should you advise women in this age group? You might point out that the USPSTF recommends against routine screening, but indicates that the decision to begin (or defer) routine mammography before age 50 should be individualized, based on the needs and values of each patient.¹

Digital mammography. A recent study of more than 43,000 women demonstrated that digital mammography is more accurate than film—but only for certain groups: These include women <50 years of age, women with dense breasts, and pre- and perimenopausal women. ²³ Because it is still not clear whether the increased accuracy will translate into a mortality benefit, more research is needed before digital mammography is widely adopted. The USPSTF maintains that there is insufficient evidence to assess the benefits and harms of using either digital mammography or magnetic resonance imaging (MRI) rather than film mammography to screen for breast cancer.¹

MRI. In 2007, the ACS published guidelines on the use of MRI as an adjunct to mammography for breast cancer screening in high-risk women.²⁴ According to ACS guidelines, screening MRI should be offered to patients with a known BRCA 1 or 2 mutation (5%-10% of all breast cancers are associated with a mutation in the BRCA 1 or BRCA 2 gene, which is transmitted in an autosomal dominant pattern⁶). It also should be offered to those with a strong family history, or a lifetime risk of developing breast cancer that is >20% to 25%. And finally, MRI should be offered to women who had chest wall radiation when they were between the ages of 10 and 30 years—another significant risk factor for breast cancer—and those with other genetic syndromes that increase their lifetime risk of breast cancer.²⁴

Evidence is insufficient for or against MRI screening for women with a personal history of breast cancer, atypical hyperplasia, or lobular carcinoma in situ, however, and neither breast ultrasound (which is generally used diagnostically, not for screening purposes) nor MRI has been shown to be helpful as a screening tool in women with <15% lifetime risk of developing breast cancer.^24,25

TABLE 2
Guidelines for breast cancer screening for women with average risk

When to consider chemoprevention

For women like Carrie, who are at high risk of developing breast cancer, selective estrogen receptor modulator (SERM) therapy and surgical interventions may be options to consider. The Breast Cancer Prevention Trial demonstrated the efficacy of tamoxifen as a preventive agent. This landmark trial showed that for high-risk women older than 35, 5 years of tamoxifen therapy can reduce the incidence of invasive breast cancer by nearly 50%.²⁶

Women with the BRCA 1 or 2 mutation—all of whom should be offered genetic counseling—were included in the study. Tamoxifen reduced the incidence of breast cancer in BRCA 2 carriers by 62%, the researchers found, but did not reduce risk in carriers of the BRCA 1 gene. This is likely due to the high prevalence of estrogen receptor-negative breast cancers among BRCA 1 carriers.²⁶

More recently, the Study of Tamoxifen and Raloxifene (STAR) trial compared the efficacy of tamoxifen and raloxifene, a second-generation SERM, in high-risk postmenopausal women ages 35 and older. The drugs were found to be equally effective in reducing the risk of invasive breast cancer, but raloxifene had a better side effect profile, with a lower incidence of thromboembolism and cataracts. ²⁷

What the guidelines call for. In 2003, the USPSTF recommended that clinicians discuss chemoprevention with women at high risk for breast cancer and low risk for adverse effects of SERMs.²⁸

The most recent update to the NCCN breast cancer risk reduction guidelines recommends that clinicians offer tamoxifen to premenopausal women with a 5-year projected breast cancer risk ≥1.7% and offer tamoxifen or raloxifene to high-risk postmenopausal women.²⁹ It is worth noting, however, that SERMs can have significant adverse effects, including venous thromboembolism, stroke, cataracts, uterine malignancy, and hot flashes, while lifestyle modifications and the avoidance of HRT have few, if any, negative effects.

CASE After consultation with a genetic counselor, Carrie underwent testing for both the BRCA 1 and BRCA 2 mutations. She tested negative for both. She declined chemoprevention and prophylactic surgery, opting for enhanced screening with yearly mammography and MRI and lifestyle modification instead.

When a mass is found

For women ages 30 or older with palpable masses or solid masses ≥2 cm found on imaging, core needle biopsy is recommended.^30,31 Biopsy is indicated for women younger than 30 as well, if the mass is >2 cm or imaging is suspicious. In general, a needle biopsy read as benign is considered adequate for diagnostic purposes only if the lesion appeared benign on imaging.

For lesions shown to be cystic on imaging, recommendations for follow-up or additional testing are based on the characteristics of the cyst. For simple cysts, 2- to 4-month follow up for stability, followed by routine screening, is adequate.²¹ Additional evaluation of complex cysts is indicated, including aspiration for complicated cysts and biopsy for complex cysts. After aspiration, surgical excision of bloody aspirates or persistent masses is recommended.^30,31

Staging using the TNM system
The TNM (tumor, node, metastases) classification system is used for the staging of breast cancer:

• T refers to the tumor type, size, and extent of local involvement
• N describes regional lymph node involvement
• M refers to distant metastases.

The TNM classifications are also grouped by stage (I through IV).,

Lumpectomy and sentinel node mapping with excision is the preferred method for staging of early-stage breast cancer without palpable lymphadenopathy—provided that the surgical team has documented experience with sentinel node biopsy.³² Sentinel node biopsy is preferred because of its safety, low (<10%) false negative rate, and decreased morbidity compared with full axillary dissection, although dissection is recommended for patients with more advanced cancer or a positive sentinel node.³² The comparative effects of sentinel node biopsy vs axillary node dissection on tumor recurrence and patient survival are not known.³³

Testing for tumor markers such as estrogen and progesterone receptors and human epidermal growth factor receptor 2 (HER2) expression status in biopsy-proven breast cancer is now the standard of care. Seventy percent of breast cancers are estrogen receptor-positive, with increasing frequency associated with older age.³⁴ Estrogen/progesterone receptor positivity is associated with a more favorable outcome, and multiple hormonal therapies can be aimed at these receptors.³⁴ While HER2 overexpression—which occurs in 15% to 30% of newly diagnosed breast cancers³⁵—is associated with more aggressive tumors, women with this type of tumor cell can benefit from trastuzumab, an anti-HER2 drug.³⁶

Key factors that affect prognosis

Important factors affecting prognosis and treatment of localized breast cancer are tumor size, age and menopausal status, tumor expression of hormone receptors and/ or the HER2 protein, as well as the status of the draining axillary nodes. Factors that predict a greater chance of recurrence include the spread of disease to axillary nodes, larger tumor size, invasive histology, inflammatory pathology, lack of estrogen/progesterone receptors, and age <50 years or premenopausal status.

Treatment options include surgical resection, radiation, and systemic adjuvant therapy in the form of chemotherapy, endocrine therapy, or anti-HER2 monoclonal antibodies.³⁷ (For more on treatment, see “Surgery, radiation, and systemic therapy: Making the most of what’s in our arsenal” at jfponline.com.)

Don’t overlook quality-of-life issues
Follow-up of breast cancer patients should go beyond treatment and work-up for recurrence and metastatic disease to focus on health and lifestyle issues, such as stress reduction, mood, smoking cessation, diet and exercise, treatment of hot flashes, sexual dysfunction, and bone health. A recent study found both reduced recurrence and increased survival in women receiving psychological interventions to improve quality-of-life measures after an 11-year follow-up.³⁸

Refer women to targeted Web sites such as the National Breast Cancer Awareness Month organization (http://www.nbcam.org/), the National Breast Cancer Foundation (http://community.nationalbreastcancer.org/), and the Susan G. Komen Breast Cancer Foundation (http://ww5.komen.org/). Offer treatment for bothersome symptoms. Hot flashes and depression, for example, often related to endocrine therapy, can be treated with selective serotonin reuptake inhibitors (SSRIs). That said, some SSRIs decrease the active metabolite of tamoxifen by inhibiting CYP2D6 enzyme and must, therefore, be used with caution. However, venlafaxine and citalopram are less likely to alter tamoxifen metabolism than other SSRIs.³⁹

CASE When Carrie was 47, she had an abnormal MRI of the left breast. Core needle biopsy and pathology of the lesion revealed an estrogen and progesterone receptor-positive tumor that was negative for HER2 overexpression. She underwent lumpectomy, which revealed a 1.5 cm tumor, followed by a negative sentinel node biopsy, and was diagnosed with stage I (T1N0M0) breast cancer. Carrie had radiation after surgery; she did not require chemotherapy, but was told to take tamoxifen for 5 years. This adjuvant endocrine therapy led to hot flashes and depression, both of which were successfully treated with venlafaxine. Carrie is currently cancer-free and participates in a breast cancer survivor program that includes regular visits with her primary physician and her oncologist.

CORRESPONDENCE Denise Sur, MD, 1920 Colorado Avenue, Santa Monica, CA 90404; [email protected] 

Late last year, the US Preventive Services Task Force (USPSTF) sparked a nationwide controversy when it announced that it was recommending against routine screening mammography for women younger than age 50.¹ Indeed, that’s a recommendation that many other organizations, including the American Cancer Society (ACS),² the American College of Obstetricians and Gynecologists (ACOG),³ and the National Comprehensive Cancer Network (NCCN),⁴ disagree with. But the age at which women should begin routine mammography isn’t the only controversial question. Experts disagree on the benefits of breast self-examination, the optimal frequency of clinical breast exams, and the use of digital mammography—among other issues. This evidence-based review can help you cut through the confusion.

CASE Carrie, a 39-year-old woman who has never been pregnant, comes in for an annual Pap smear and gynecologic exam. She has a negative past medical history, but a positive family history for breast cancer—both her mother and 1 of her sisters had the disease. How would you assess Carrie’s risk of breast cancer, and what preventive measures would you recommend?

Use this predictive model to pinpoint your patient’s risk

When making decisions regarding primary prevention of and screening for breast cancer, an accurate assessment of risk is critical. Many predictive models have been developed with that in mind. The most widely studied, the Gail model, incorporates a number of important risk factors (TABLE 1), including age; race; family history; reproductive factors such as age of menarche, menopause, and first childbirth; and previous history of breast biopsy and atypical findings, to calculate a woman’s 5-year risk.⁵

A risk calculator (the Breast Cancer Risk Assessment Tool) based on the Gail model is available on the National Cancer Institute’s Web site, at http://www.cancer.gov/bcrisktool. Generally, a score ≥1.66%,⁵ which indicates that a patient has at least a 1.66% chance of developing breast cancer over the next 5 years, is considered high risk.^6,7

CASE Carrie’s 2 first-degree relatives affected by breast cancer and her nulliparous status place her at increased risk. Further questioning reveals a particularly strong family history, as both relatives were diagnosed before the age of 50 (her mom at 45 years of age and her sister, at 39). Carrie’s 5-year risk is 1.8%.

TABLE 1
Risk factors for breast cancer^5,29

All women can benefit from these preventive measures

As primary care physicians, we have a responsibility to stress lifestyle modification as the mainstay of breast cancer prevention. Whether or not a woman is at high risk, advise her that maintaining a normal weight, exercising vigorously, limiting alcohol consumption, and breastfeeding are evidence-based methods of primary prevention. Diets low in fat and high in fiber may be associated with a lower risk of invasive breast cancer, but there is no conclusive evidence to support specific dietary interventions to reduce the risk.^8-11 Nor has a link between active or passive smoking, antioxidants, or fruit and vegetable intake been firmly established.¹²

There is a clear association between prolonged estrogen exposure and breast cancer, however. Many reproductive factors, such as early menarche, late menopause, later age at time of first full-term pregnancy, and nulliparity, increase a woman’s exposure to endogenous estrogen—and her risk of developing breast cancer.^12,13

Exposure to exogenous estrogen is also linked to the development of breast cancer. In 2002, the Women’s Health Initiative (WHI) was stopped early after a report was released stating that the risks of hormone replacement therapy (HRT)—a higher incidence of cardiovascular events, stroke, and venous thromboembolism, as well as breast cancer—outweighed the benefits.¹⁴ Subsequent analyses have found a relationship between the declining incidence of breast cancer and the marked decrease in HRT use prompted by the WHI report. While causality has not been firmly established, multiple studies strongly suggest it.^15,16

The association between oral contraceptives (OCs) and breast cancer is more controversial. Some studies have found an increased breast cancer risk among OC users, but both the relative risk and absolute risk were found to be very small and to dissipate 10 years after stopping OC use. More recent studies with newer formulations containing lower doses of estrogen have failed to show an increased risk.⁸

Breast cancer screening: The parameters have changed

Various organizations have published guidelines for breast cancer screening (TABLE 2), and all are somewhat different. Here’s what you need to know.

Breast self-examination (BSE), which women were previously advised to perform monthly, has not been shown to improve mortality in any age group, and is no longer routinely recommended.¹⁷ While both the USPSTF and the Canadian Task Force on Preventive Health Care recommend against teaching women BSE, the ACS, ACOG, and NCCN encourage self-examination—particularly among women older than 40 years.^1-4,17,18

Clinical breast examination has an average sensitivity of 50% and detects approximately 5% of mammographically occult cancers.¹⁹ It is still not clear whether clinical breast exams save lives, however—a finding that is reflected in the USPSTF’s “I” (insufficient evidence to assess the benefits and harms) recommendation.¹ Other consensus guidelines still recommend clinical breast examination, albeit at varying frequencies.

Screening mammography decreases mortality rates by anywhere from 28% to 65%, depending on the statistical model used.²⁰ The benefit is greatest in women between the ages of 50 and 69 years, however, and most groups agree that mammography every 1 to 2 years is advisable for this age group. (There is limited data on the value of mammography for women 70 years of age and older, and no consensus on the age at which to stop screening.^1,21) But because the mortality benefit from screening mammography is lower for women aged 40 to 49, guidelines for this age group are more controversial.

Mammography’s sensitivity is affected by a variety of factors, including age and menopausal status, prior breast surgery or radiation, breast density, and the experience of the radiologist. Women in their 40s have denser breast tissue than older women, making mammography less sensitive for this age group. Because of that, and because the overall incidence of breast cancer is lower for women younger than 50, some argue that screening mammography for women between the ages of 40 and 49 years leads to unacceptably high false-positive rates (9.8% annually²²) and that the harm associated with mammography may outweigh the benefit. Others counter that tumors in younger women tend to be more aggressive and faster growing, making early detection even more critical than for older women.

What should you advise women in this age group? You might point out that the USPSTF recommends against routine screening, but indicates that the decision to begin (or defer) routine mammography before age 50 should be individualized, based on the needs and values of each patient.¹

Digital mammography. A recent study of more than 43,000 women demonstrated that digital mammography is more accurate than film—but only for certain groups: These include women <50 years of age, women with dense breasts, and pre- and perimenopausal women. ²³ Because it is still not clear whether the increased accuracy will translate into a mortality benefit, more research is needed before digital mammography is widely adopted. The USPSTF maintains that there is insufficient evidence to assess the benefits and harms of using either digital mammography or magnetic resonance imaging (MRI) rather than film mammography to screen for breast cancer.¹

MRI. In 2007, the ACS published guidelines on the use of MRI as an adjunct to mammography for breast cancer screening in high-risk women.²⁴ According to ACS guidelines, screening MRI should be offered to patients with a known BRCA 1 or 2 mutation (5%-10% of all breast cancers are associated with a mutation in the BRCA 1 or BRCA 2 gene, which is transmitted in an autosomal dominant pattern⁶). It also should be offered to those with a strong family history, or a lifetime risk of developing breast cancer that is >20% to 25%. And finally, MRI should be offered to women who had chest wall radiation when they were between the ages of 10 and 30 years—another significant risk factor for breast cancer—and those with other genetic syndromes that increase their lifetime risk of breast cancer.²⁴

Evidence is insufficient for or against MRI screening for women with a personal history of breast cancer, atypical hyperplasia, or lobular carcinoma in situ, however, and neither breast ultrasound (which is generally used diagnostically, not for screening purposes) nor MRI has been shown to be helpful as a screening tool in women with <15% lifetime risk of developing breast cancer.^24,25

TABLE 2
Guidelines for breast cancer screening for women with average risk

When to consider chemoprevention

For women like Carrie, who are at high risk of developing breast cancer, selective estrogen receptor modulator (SERM) therapy and surgical interventions may be options to consider. The Breast Cancer Prevention Trial demonstrated the efficacy of tamoxifen as a preventive agent. This landmark trial showed that for high-risk women older than 35, 5 years of tamoxifen therapy can reduce the incidence of invasive breast cancer by nearly 50%.²⁶

Women with the BRCA 1 or 2 mutation—all of whom should be offered genetic counseling—were included in the study. Tamoxifen reduced the incidence of breast cancer in BRCA 2 carriers by 62%, the researchers found, but did not reduce risk in carriers of the BRCA 1 gene. This is likely due to the high prevalence of estrogen receptor-negative breast cancers among BRCA 1 carriers.²⁶

More recently, the Study of Tamoxifen and Raloxifene (STAR) trial compared the efficacy of tamoxifen and raloxifene, a second-generation SERM, in high-risk postmenopausal women ages 35 and older. The drugs were found to be equally effective in reducing the risk of invasive breast cancer, but raloxifene had a better side effect profile, with a lower incidence of thromboembolism and cataracts. ²⁷

What the guidelines call for. In 2003, the USPSTF recommended that clinicians discuss chemoprevention with women at high risk for breast cancer and low risk for adverse effects of SERMs.²⁸

The most recent update to the NCCN breast cancer risk reduction guidelines recommends that clinicians offer tamoxifen to premenopausal women with a 5-year projected breast cancer risk ≥1.7% and offer tamoxifen or raloxifene to high-risk postmenopausal women.²⁹ It is worth noting, however, that SERMs can have significant adverse effects, including venous thromboembolism, stroke, cataracts, uterine malignancy, and hot flashes, while lifestyle modifications and the avoidance of HRT have few, if any, negative effects.

CASE After consultation with a genetic counselor, Carrie underwent testing for both the BRCA 1 and BRCA 2 mutations. She tested negative for both. She declined chemoprevention and prophylactic surgery, opting for enhanced screening with yearly mammography and MRI and lifestyle modification instead.

When a mass is found

For women ages 30 or older with palpable masses or solid masses ≥2 cm found on imaging, core needle biopsy is recommended.^30,31 Biopsy is indicated for women younger than 30 as well, if the mass is >2 cm or imaging is suspicious. In general, a needle biopsy read as benign is considered adequate for diagnostic purposes only if the lesion appeared benign on imaging.

For lesions shown to be cystic on imaging, recommendations for follow-up or additional testing are based on the characteristics of the cyst. For simple cysts, 2- to 4-month follow up for stability, followed by routine screening, is adequate.²¹ Additional evaluation of complex cysts is indicated, including aspiration for complicated cysts and biopsy for complex cysts. After aspiration, surgical excision of bloody aspirates or persistent masses is recommended.^30,31

Staging using the TNM system
The TNM (tumor, node, metastases) classification system is used for the staging of breast cancer:

• T refers to the tumor type, size, and extent of local involvement
• N describes regional lymph node involvement
• M refers to distant metastases.

The TNM classifications are also grouped by stage (I through IV).,

Lumpectomy and sentinel node mapping with excision is the preferred method for staging of early-stage breast cancer without palpable lymphadenopathy—provided that the surgical team has documented experience with sentinel node biopsy.³² Sentinel node biopsy is preferred because of its safety, low (<10%) false negative rate, and decreased morbidity compared with full axillary dissection, although dissection is recommended for patients with more advanced cancer or a positive sentinel node.³² The comparative effects of sentinel node biopsy vs axillary node dissection on tumor recurrence and patient survival are not known.³³

Testing for tumor markers such as estrogen and progesterone receptors and human epidermal growth factor receptor 2 (HER2) expression status in biopsy-proven breast cancer is now the standard of care. Seventy percent of breast cancers are estrogen receptor-positive, with increasing frequency associated with older age.³⁴ Estrogen/progesterone receptor positivity is associated with a more favorable outcome, and multiple hormonal therapies can be aimed at these receptors.³⁴ While HER2 overexpression—which occurs in 15% to 30% of newly diagnosed breast cancers³⁵—is associated with more aggressive tumors, women with this type of tumor cell can benefit from trastuzumab, an anti-HER2 drug.³⁶

Key factors that affect prognosis

Important factors affecting prognosis and treatment of localized breast cancer are tumor size, age and menopausal status, tumor expression of hormone receptors and/ or the HER2 protein, as well as the status of the draining axillary nodes. Factors that predict a greater chance of recurrence include the spread of disease to axillary nodes, larger tumor size, invasive histology, inflammatory pathology, lack of estrogen/progesterone receptors, and age <50 years or premenopausal status.

Treatment options include surgical resection, radiation, and systemic adjuvant therapy in the form of chemotherapy, endocrine therapy, or anti-HER2 monoclonal antibodies.³⁷ (For more on treatment, see “Surgery, radiation, and systemic therapy: Making the most of what’s in our arsenal” at jfponline.com.)

Don’t overlook quality-of-life issues
Follow-up of breast cancer patients should go beyond treatment and work-up for recurrence and metastatic disease to focus on health and lifestyle issues, such as stress reduction, mood, smoking cessation, diet and exercise, treatment of hot flashes, sexual dysfunction, and bone health. A recent study found both reduced recurrence and increased survival in women receiving psychological interventions to improve quality-of-life measures after an 11-year follow-up.³⁸

Refer women to targeted Web sites such as the National Breast Cancer Awareness Month organization (http://www.nbcam.org/), the National Breast Cancer Foundation (http://community.nationalbreastcancer.org/), and the Susan G. Komen Breast Cancer Foundation (http://ww5.komen.org/). Offer treatment for bothersome symptoms. Hot flashes and depression, for example, often related to endocrine therapy, can be treated with selective serotonin reuptake inhibitors (SSRIs). That said, some SSRIs decrease the active metabolite of tamoxifen by inhibiting CYP2D6 enzyme and must, therefore, be used with caution. However, venlafaxine and citalopram are less likely to alter tamoxifen metabolism than other SSRIs.³⁹

CASE When Carrie was 47, she had an abnormal MRI of the left breast. Core needle biopsy and pathology of the lesion revealed an estrogen and progesterone receptor-positive tumor that was negative for HER2 overexpression. She underwent lumpectomy, which revealed a 1.5 cm tumor, followed by a negative sentinel node biopsy, and was diagnosed with stage I (T1N0M0) breast cancer. Carrie had radiation after surgery; she did not require chemotherapy, but was told to take tamoxifen for 5 years. This adjuvant endocrine therapy led to hot flashes and depression, both of which were successfully treated with venlafaxine. Carrie is currently cancer-free and participates in a breast cancer survivor program that includes regular visits with her primary physician and her oncologist.

CORRESPONDENCE Denise Sur, MD, 1920 Colorado Avenue, Santa Monica, CA 90404; [email protected] 

Late last year, the US Preventive Services Task Force (USPSTF) sparked a nationwide controversy when it announced that it was recommending against routine screening mammography for women younger than age 50.¹ Indeed, that’s a recommendation that many other organizations, including the American Cancer Society (ACS),² the American College of Obstetricians and Gynecologists (ACOG),³ and the National Comprehensive Cancer Network (NCCN),⁴ disagree with. But the age at which women should begin routine mammography isn’t the only controversial question. Experts disagree on the benefits of breast self-examination, the optimal frequency of clinical breast exams, and the use of digital mammography—among other issues. This evidence-based review can help you cut through the confusion.

CASE Carrie, a 39-year-old woman who has never been pregnant, comes in for an annual Pap smear and gynecologic exam. She has a negative past medical history, but a positive family history for breast cancer—both her mother and 1 of her sisters had the disease. How would you assess Carrie’s risk of breast cancer, and what preventive measures would you recommend?

Use this predictive model to pinpoint your patient’s risk

When making decisions regarding primary prevention of and screening for breast cancer, an accurate assessment of risk is critical. Many predictive models have been developed with that in mind. The most widely studied, the Gail model, incorporates a number of important risk factors (TABLE 1), including age; race; family history; reproductive factors such as age of menarche, menopause, and first childbirth; and previous history of breast biopsy and atypical findings, to calculate a woman’s 5-year risk.⁵

A risk calculator (the Breast Cancer Risk Assessment Tool) based on the Gail model is available on the National Cancer Institute’s Web site, at http://www.cancer.gov/bcrisktool. Generally, a score ≥1.66%,⁵ which indicates that a patient has at least a 1.66% chance of developing breast cancer over the next 5 years, is considered high risk.^6,7

CASE Carrie’s 2 first-degree relatives affected by breast cancer and her nulliparous status place her at increased risk. Further questioning reveals a particularly strong family history, as both relatives were diagnosed before the age of 50 (her mom at 45 years of age and her sister, at 39). Carrie’s 5-year risk is 1.8%.

TABLE 1
Risk factors for breast cancer^5,29

All women can benefit from these preventive measures

As primary care physicians, we have a responsibility to stress lifestyle modification as the mainstay of breast cancer prevention. Whether or not a woman is at high risk, advise her that maintaining a normal weight, exercising vigorously, limiting alcohol consumption, and breastfeeding are evidence-based methods of primary prevention. Diets low in fat and high in fiber may be associated with a lower risk of invasive breast cancer, but there is no conclusive evidence to support specific dietary interventions to reduce the risk.^8-11 Nor has a link between active or passive smoking, antioxidants, or fruit and vegetable intake been firmly established.¹²

There is a clear association between prolonged estrogen exposure and breast cancer, however. Many reproductive factors, such as early menarche, late menopause, later age at time of first full-term pregnancy, and nulliparity, increase a woman’s exposure to endogenous estrogen—and her risk of developing breast cancer.^12,13

Exposure to exogenous estrogen is also linked to the development of breast cancer. In 2002, the Women’s Health Initiative (WHI) was stopped early after a report was released stating that the risks of hormone replacement therapy (HRT)—a higher incidence of cardiovascular events, stroke, and venous thromboembolism, as well as breast cancer—outweighed the benefits.¹⁴ Subsequent analyses have found a relationship between the declining incidence of breast cancer and the marked decrease in HRT use prompted by the WHI report. While causality has not been firmly established, multiple studies strongly suggest it.^15,16

The association between oral contraceptives (OCs) and breast cancer is more controversial. Some studies have found an increased breast cancer risk among OC users, but both the relative risk and absolute risk were found to be very small and to dissipate 10 years after stopping OC use. More recent studies with newer formulations containing lower doses of estrogen have failed to show an increased risk.⁸

Breast cancer screening: The parameters have changed

Various organizations have published guidelines for breast cancer screening (TABLE 2), and all are somewhat different. Here’s what you need to know.

Breast self-examination (BSE), which women were previously advised to perform monthly, has not been shown to improve mortality in any age group, and is no longer routinely recommended.¹⁷ While both the USPSTF and the Canadian Task Force on Preventive Health Care recommend against teaching women BSE, the ACS, ACOG, and NCCN encourage self-examination—particularly among women older than 40 years.^1-4,17,18

Clinical breast examination has an average sensitivity of 50% and detects approximately 5% of mammographically occult cancers.¹⁹ It is still not clear whether clinical breast exams save lives, however—a finding that is reflected in the USPSTF’s “I” (insufficient evidence to assess the benefits and harms) recommendation.¹ Other consensus guidelines still recommend clinical breast examination, albeit at varying frequencies.

Screening mammography decreases mortality rates by anywhere from 28% to 65%, depending on the statistical model used.²⁰ The benefit is greatest in women between the ages of 50 and 69 years, however, and most groups agree that mammography every 1 to 2 years is advisable for this age group. (There is limited data on the value of mammography for women 70 years of age and older, and no consensus on the age at which to stop screening.^1,21) But because the mortality benefit from screening mammography is lower for women aged 40 to 49, guidelines for this age group are more controversial.

Mammography’s sensitivity is affected by a variety of factors, including age and menopausal status, prior breast surgery or radiation, breast density, and the experience of the radiologist. Women in their 40s have denser breast tissue than older women, making mammography less sensitive for this age group. Because of that, and because the overall incidence of breast cancer is lower for women younger than 50, some argue that screening mammography for women between the ages of 40 and 49 years leads to unacceptably high false-positive rates (9.8% annually²²) and that the harm associated with mammography may outweigh the benefit. Others counter that tumors in younger women tend to be more aggressive and faster growing, making early detection even more critical than for older women.

What should you advise women in this age group? You might point out that the USPSTF recommends against routine screening, but indicates that the decision to begin (or defer) routine mammography before age 50 should be individualized, based on the needs and values of each patient.¹

Digital mammography. A recent study of more than 43,000 women demonstrated that digital mammography is more accurate than film—but only for certain groups: These include women <50 years of age, women with dense breasts, and pre- and perimenopausal women. ²³ Because it is still not clear whether the increased accuracy will translate into a mortality benefit, more research is needed before digital mammography is widely adopted. The USPSTF maintains that there is insufficient evidence to assess the benefits and harms of using either digital mammography or magnetic resonance imaging (MRI) rather than film mammography to screen for breast cancer.¹

MRI. In 2007, the ACS published guidelines on the use of MRI as an adjunct to mammography for breast cancer screening in high-risk women.²⁴ According to ACS guidelines, screening MRI should be offered to patients with a known BRCA 1 or 2 mutation (5%-10% of all breast cancers are associated with a mutation in the BRCA 1 or BRCA 2 gene, which is transmitted in an autosomal dominant pattern⁶). It also should be offered to those with a strong family history, or a lifetime risk of developing breast cancer that is >20% to 25%. And finally, MRI should be offered to women who had chest wall radiation when they were between the ages of 10 and 30 years—another significant risk factor for breast cancer—and those with other genetic syndromes that increase their lifetime risk of breast cancer.²⁴

Evidence is insufficient for or against MRI screening for women with a personal history of breast cancer, atypical hyperplasia, or lobular carcinoma in situ, however, and neither breast ultrasound (which is generally used diagnostically, not for screening purposes) nor MRI has been shown to be helpful as a screening tool in women with <15% lifetime risk of developing breast cancer.^24,25

TABLE 2
Guidelines for breast cancer screening for women with average risk

When to consider chemoprevention

For women like Carrie, who are at high risk of developing breast cancer, selective estrogen receptor modulator (SERM) therapy and surgical interventions may be options to consider. The Breast Cancer Prevention Trial demonstrated the efficacy of tamoxifen as a preventive agent. This landmark trial showed that for high-risk women older than 35, 5 years of tamoxifen therapy can reduce the incidence of invasive breast cancer by nearly 50%.²⁶

Women with the BRCA 1 or 2 mutation—all of whom should be offered genetic counseling—were included in the study. Tamoxifen reduced the incidence of breast cancer in BRCA 2 carriers by 62%, the researchers found, but did not reduce risk in carriers of the BRCA 1 gene. This is likely due to the high prevalence of estrogen receptor-negative breast cancers among BRCA 1 carriers.²⁶

More recently, the Study of Tamoxifen and Raloxifene (STAR) trial compared the efficacy of tamoxifen and raloxifene, a second-generation SERM, in high-risk postmenopausal women ages 35 and older. The drugs were found to be equally effective in reducing the risk of invasive breast cancer, but raloxifene had a better side effect profile, with a lower incidence of thromboembolism and cataracts. ²⁷

What the guidelines call for. In 2003, the USPSTF recommended that clinicians discuss chemoprevention with women at high risk for breast cancer and low risk for adverse effects of SERMs.²⁸

The most recent update to the NCCN breast cancer risk reduction guidelines recommends that clinicians offer tamoxifen to premenopausal women with a 5-year projected breast cancer risk ≥1.7% and offer tamoxifen or raloxifene to high-risk postmenopausal women.²⁹ It is worth noting, however, that SERMs can have significant adverse effects, including venous thromboembolism, stroke, cataracts, uterine malignancy, and hot flashes, while lifestyle modifications and the avoidance of HRT have few, if any, negative effects.

CASE After consultation with a genetic counselor, Carrie underwent testing for both the BRCA 1 and BRCA 2 mutations. She tested negative for both. She declined chemoprevention and prophylactic surgery, opting for enhanced screening with yearly mammography and MRI and lifestyle modification instead.

When a mass is found

For women ages 30 or older with palpable masses or solid masses ≥2 cm found on imaging, core needle biopsy is recommended.^30,31 Biopsy is indicated for women younger than 30 as well, if the mass is >2 cm or imaging is suspicious. In general, a needle biopsy read as benign is considered adequate for diagnostic purposes only if the lesion appeared benign on imaging.

For lesions shown to be cystic on imaging, recommendations for follow-up or additional testing are based on the characteristics of the cyst. For simple cysts, 2- to 4-month follow up for stability, followed by routine screening, is adequate.²¹ Additional evaluation of complex cysts is indicated, including aspiration for complicated cysts and biopsy for complex cysts. After aspiration, surgical excision of bloody aspirates or persistent masses is recommended.^30,31

Staging using the TNM system
The TNM (tumor, node, metastases) classification system is used for the staging of breast cancer:

• T refers to the tumor type, size, and extent of local involvement
• N describes regional lymph node involvement
• M refers to distant metastases.

The TNM classifications are also grouped by stage (I through IV).,

Lumpectomy and sentinel node mapping with excision is the preferred method for staging of early-stage breast cancer without palpable lymphadenopathy—provided that the surgical team has documented experience with sentinel node biopsy.³² Sentinel node biopsy is preferred because of its safety, low (<10%) false negative rate, and decreased morbidity compared with full axillary dissection, although dissection is recommended for patients with more advanced cancer or a positive sentinel node.³² The comparative effects of sentinel node biopsy vs axillary node dissection on tumor recurrence and patient survival are not known.³³

Testing for tumor markers such as estrogen and progesterone receptors and human epidermal growth factor receptor 2 (HER2) expression status in biopsy-proven breast cancer is now the standard of care. Seventy percent of breast cancers are estrogen receptor-positive, with increasing frequency associated with older age.³⁴ Estrogen/progesterone receptor positivity is associated with a more favorable outcome, and multiple hormonal therapies can be aimed at these receptors.³⁴ While HER2 overexpression—which occurs in 15% to 30% of newly diagnosed breast cancers³⁵—is associated with more aggressive tumors, women with this type of tumor cell can benefit from trastuzumab, an anti-HER2 drug.³⁶

Key factors that affect prognosis

Important factors affecting prognosis and treatment of localized breast cancer are tumor size, age and menopausal status, tumor expression of hormone receptors and/ or the HER2 protein, as well as the status of the draining axillary nodes. Factors that predict a greater chance of recurrence include the spread of disease to axillary nodes, larger tumor size, invasive histology, inflammatory pathology, lack of estrogen/progesterone receptors, and age <50 years or premenopausal status.

Treatment options include surgical resection, radiation, and systemic adjuvant therapy in the form of chemotherapy, endocrine therapy, or anti-HER2 monoclonal antibodies.³⁷ (For more on treatment, see “Surgery, radiation, and systemic therapy: Making the most of what’s in our arsenal” at jfponline.com.)

Don’t overlook quality-of-life issues
Follow-up of breast cancer patients should go beyond treatment and work-up for recurrence and metastatic disease to focus on health and lifestyle issues, such as stress reduction, mood, smoking cessation, diet and exercise, treatment of hot flashes, sexual dysfunction, and bone health. A recent study found both reduced recurrence and increased survival in women receiving psychological interventions to improve quality-of-life measures after an 11-year follow-up.³⁸

Refer women to targeted Web sites such as the National Breast Cancer Awareness Month organization (http://www.nbcam.org/), the National Breast Cancer Foundation (http://community.nationalbreastcancer.org/), and the Susan G. Komen Breast Cancer Foundation (http://ww5.komen.org/). Offer treatment for bothersome symptoms. Hot flashes and depression, for example, often related to endocrine therapy, can be treated with selective serotonin reuptake inhibitors (SSRIs). That said, some SSRIs decrease the active metabolite of tamoxifen by inhibiting CYP2D6 enzyme and must, therefore, be used with caution. However, venlafaxine and citalopram are less likely to alter tamoxifen metabolism than other SSRIs.³⁹

CASE When Carrie was 47, she had an abnormal MRI of the left breast. Core needle biopsy and pathology of the lesion revealed an estrogen and progesterone receptor-positive tumor that was negative for HER2 overexpression. She underwent lumpectomy, which revealed a 1.5 cm tumor, followed by a negative sentinel node biopsy, and was diagnosed with stage I (T1N0M0) breast cancer. Carrie had radiation after surgery; she did not require chemotherapy, but was told to take tamoxifen for 5 years. This adjuvant endocrine therapy led to hot flashes and depression, both of which were successfully treated with venlafaxine. Carrie is currently cancer-free and participates in a breast cancer survivor program that includes regular visits with her primary physician and her oncologist.

CORRESPONDENCE Denise Sur, MD, 1920 Colorado Avenue, Santa Monica, CA 90404; [email protected] 

Many systemic conditions are accompanied by skin manifestations. This is especially true for connective tissue disorders, for which dermatologic findings are often the key to diagnosis.

In this review, we describe the dermatologic findings of some well-known connective tissue disorders. The text and photographs in the pages that follow will help you hone your diagnostic skills, leading to earlier treatment and, possibly, better outcomes.

Lupus erythematosus: Cutaneous and systemic disease often overlap

Lupus erythematosus (LE), a chronic, inflammatory autoimmune condition that primarily affects women in their 20s and 30s, may initially present as a systemic disease or in a purely cutaneous form. However, most patients with systemic LE have some skin manifestations, and those with cutaneous LE often have—or subsequently develop—systemic involvement.¹ Thus, recognizing the cutaneous manifestations of LE will not only aid in diagnosis, but will help you identify patients at risk for systemic disease.

Cutaneous LE has 4 subtypes
There are 4 subcategories of cutaneous LE—acute, subacute, chronic, and intermittent.² Each is differentiated by the appearance of the lesions (TABLE 1), histology, and serological markers.¹ Photosensitivity is common to all the subcategories to varying degrees.

Acute cutaneous lupus erythematosus (ACLE) is typically characterized by the classic malar “butterfly” rash, an erythematous eruption of macules or edematous papules over the bridge of the nose and cheek.³ Although this presentation is most common, there are variations—1 in which the lesions cover other exposed areas (commonly including the “V” of the chest, the extensor surface of the arms, and the hands), and a rare form in which toxic epidermal necrolysis-like blistering occurs.^1,4 These skin changes—which generally last anywhere from a few hours to several weeks—typically resolve without scarring, although pigment changes can occur.⁵

Subacute cutaneous lupus erythematosus (SCLE), which usually affects middle-aged Caucasian women, is characterized by erythematous papulosquamous (psoriasis-like) eruptions or annular lesions with raised red borders and central clearing—or both. These lesions, which are nonscarring, lack in-duration, and rarely affect the scalp or face, appear suddenly, usually after exposure to sunlight (FIGURE 1)^5,7-9 or certain drugs. Hydrochlorothiazide, terbinafine, calcium channel blockers, and angiotensin-converting enzyme inhibitors are common offenders.^1,6

FIGURE 1
Annular lesions in subacute cutaneous lupus erythematosus

SCLE is often associated with extracutaneous symptoms such as arthritis and myalgias,^1,8 but patients are at relatively low risk for severe systemic manifestations.^5,8,10 Serology is often notable, with anti-Ro (SS-A) antibodies present in 70% to 90% of patients and anti-La (SS-B) autoantibodies found in 30% to 50%.^1,11

Chronic cutaneous lupus erythematosus (CCLE) also occurs predominantly in females, at a ratio as high as 5 to 1.¹² There are 3 variations of CCLE: discoid lupus erythematosus (DLE), LE profundus, and chilblain LE (TABLE 1). DLE, characterized by alopecia, skin atrophy, and dyspigmentation, is the most common and affects patients of all ages and ethnic groups.^9,13,14 (See image above.)

DLE lesions typically begin as erythematous papules and plaques with scale. As the disorder progresses, the lesions spread, causing follicular plugging, peripheral hyperpigmentation and central hypopigmentation, telangiectasia, and atrophy.^9,15 In some cases, patients develop thickened, scarred skin and permanent scarring alopecia.¹⁵ Prompt recognition of DLE is particularly important, as early referral and treatment may reduce the likelihood of permanent scarring alopecia and pigment changes.¹¹

Intermittent cutaneous lupus erythematosus (ICLE), a relatively new subtype of cutaneous LE, is represented by a rare condition—lupus erythematosus tumidus (LET)—reported in <100 cases worldwide. LET is characterized by succulent, erythematous, and edematous plaques on sun-exposed parts of the body.^1,16

TABLE 1
Cutaneous lupus erythematosus: Recognizing the subtypes¹

Cutaneous LE diagnosis and treatment: Start with ACR criteria
When evaluating patients with suspected cutaneous LE, it is important not only to identify the subtype, but also to rule out systemic disease using criteria established by the American College of Rheumatology (ACR).¹⁷ Notably, 4 of the 11 diagnostic criteria for systemic disease involve visual clues, including malar rash, discoid rash, photosensitivity, and oral ulcerations. Laboratory evidence of systemic disease may include a positive antinuclear antibody, anti-dsDNA, or anti-Sm autoantibody test results, as well as hematologic abnormalities described in the ACR guidelines.

Ultimately, a diagnosis of cutaneous LE should be based on the patient’s history and physical exam, autoantibody profile, and histologic and immunofluorescent biopsy findings. A rheumatologic evaluation may help to determine which patients have systemic disease, as the ACR criteria may overdiagnose systemic LE in those with predominantly skin changes.⁶

Treatment of cutaneous LE is based on the subtype and extent of disease, with potent topical corticosteroids, in combination with antimalarial agents, being the primary therapies. ACLE skin lesions generally respond best to systemic corticosteroids and immunosuppressive agents (such as azathioprine or cyclophosphamide) that are used to control underlying systemic disease. SCLE can be managed with topical corticosteroids; however, patients typically also require systemic treatment, often with hydroxychloroquine, for optimal control. DLE, the most common form of CCLE, is managed in a similar fashion, with a greater role for intralesional corticosteroid injections. Lesions associated with ICLE, which often resolve spontaneously, may be treated with topical corticosteroids and antimalarials.¹⁶ It is also important to advise all patients with cutaneous LE to use a broad-spectrum sunscreen, as ultraviolet (UV) exposure can induce or exacerbate the lesions.⁹

Dermatomyositis: Rare but serious

Dermatomyositis is an idiopathic inflammatory myopathy characterized by chronic muscle inflammation, symmetric proximal muscle weakness, and distinct cutaneous findings. In addition to possible cardiac and pulmonary complications, dysphagia, and joint contractures, dermatomyositis is associated with cancer, with up to 25% of affected adults having an underlying occult malignancy.¹⁸

Dermatomyositis is a rare condition, with a prevalence of only 1 to 10 cases per million adults and 3.2 cases per million children.¹⁹ It has a bimodal age distribution, with most juvenile cases affecting children between the ages of 5 and 14 years and most adult cases developing in the fifth and sixth decades of life. Women are affected twice as often as men.¹⁸

Suspect dermatomyositis when you see any of the following signs:

• A heliotrope rash: violaceous macules and patches, with or without edema, symmetrically on the periorbital skin, present early in the disease course in 30% to 60% of patients.¹⁹ (See image)
• Gottron’s papules: violaceous papules on the dorsal interphalangeal and metacarpophalangeal joints of the hands, elbows, and knees, occurring in as many as 70% of patients (FIGURE 2).^19,20
• Gottron’s sign: nonscaling, violaceous erythematous macules and plaques occurring symmetrically in the same distribution as Gottron’s papules, but with sparing of the interphalangeal spaces.²⁰
• Periungual erythema and telangiectasias: redness and dilation of the blood vessels in the skin surrounding the nail plate.
• The shawl and V-signs: erythematous macular eruptions occurring in a “shawl” pattern on the shoulders, arms, and upper back and in a V-shaped pattern on the anterior neck and chest.
• Mechanic’s hand: extensive scaling, fissuring, and roughening of the palmar aspect of the hand.^20,21
• Poikiloderma vasculare atrophicans: circumscribed violaceous erythema with thinning of the skin, prominent telangiectasias, and a mottled pattern of hypo- and hyperpigmentation, typically occurring on the anterior neck, chest, posterior shoulders, back, and buttocks, years after onset of the disease.^21-24
• Cutaneous calcification (calcinosis cutis): usually on the buttocks, elbows, knees, and traumatized areas, affecting 30% to 70% of children with dermatomyositis, but only 10% of adults with the disorder.^20,25-27

FIGURE 2
Gottron’s papules in dermatomyositis

Bohan and Peter criteria help with dermatomyositis diagnosis
A diagnosis of dermatomyositis can be established using the following criteria developed by Bohan and Peter in 1975: (1) symmetric muscle weakness; (2) elevation of muscle enzymes, most notably creatinine phosphokinase; (3) evidence of inflammation on muscle biopsy; (4) electromyographic features of myositis; and (5) characteristic dermatologic signs, including a heliotrope rash and Gottron’s papules.^28,29 A definitive diagnosis can be made when the patient meets 3 of the first 4 criteria, as well as the fifth.

Corticosteroids are the mainstay of treatment for dermatomyositis, although the dose and duration are subject to debate. Cutaneous manifestations of dermatomyositis are commonly treated with topical corticosteroids and oral hydroxychloroquine, as well as emollients and antipruritic agents.

Cancer screening. All adults with signs and symptoms of dermatomyositis should undergo cancer screening. The most common malignancies are ovarian cancer, gastric cancer, and lymphoma.²¹

Photoprotection. As with cutaneous LE, UV exposure can exacerbate dermatomyositis, and patients should use a broad-spectrum sunscreen.