UEU-co logo

MD Consult: Books: Goldman: Cecil Medicine: Chapter 275 – DRUG ALLERGY

Goldman: Cecil Medicine, 23rd ed.

Copyright © 2007 Saunders, An Imprint of Elsevier

Chapter 275 – DRUG ALLERGY


Leslie C. Grammer




Adverse drug reactions (ADRs) are recognized as an important public health problem in that they may be life-threatening. An ADR is defined by the World Health Organization (WHO) as an unintended, noxious response to a drug that occurs at a dose usually prescribed for human patients. The classic pharmacologic definition of ADRs by Rawlins and Thompson separates these into two major subtypes: type A reactions, which are predictable and dose dependent, and type B reactions, which are unpredictable and not dose dependent.

Most ADRs are type A reactions. Type B reactions account for 10 to 15% of all ADRs and include drug allergy (i.e., hypersensitivity reactions to drugs). According to the WHO Nomenclature Review Committee, drug allergy refers to a hypersensitivity reaction for which a definite immunologic mechanism, either a B-cell–mediated (antibody) or a T-cell–mediated process, is documented. Most published epidemiologic studies refer to ADRs in general and not to drug allergy specifically because the demonstration of drug-specific B-cell– or T-cell–mediated mechanisms is often quite difficult, and the immunologic culprit may be a drug metabolite.



Drug allergy is responsible for significant mortality, morbidity, and socioeconomic costs that are probably underestimated. Current data must be evaluated carefully, because they involve different populations, different definitions of ADRs/drug allergy, and different methodologies, especially in terms of data analyses. The Boston Collaborative Drug Surveillance Program collected information on all ADRs in 4031 hospitalized patients during a period of 6 months. An incidence of 6.1%, 247 ADRs, was reported, of which 42% were severe; 1% of the severe reactions resulted in the death of a patient. Using an automatic detection system in a Salt Lake City hospital, Claussen and coinvestigators identified 731 ADRs among 36,653 hospitalized patients. Of note, only 12.3% of these were reported by physicians in the hospital. In a meta-analysis of 33 U.S. prospective studies from 1966 to 1996, Lazarou reported that 15% of hospitalized patients incurred an ADR and that the frequency of drug-related hospital admissions varied from 3 to 6%. Most other studies that followed reported similar data. Epidemiologic information on drug allergy in nonhospitalized people and in the general population is even more limited and is mainly confined to studies of antibiotics.

Risk Factors


Some risk factors have been identified for the development of drug allergy. Certain drugs more commonly cause adverse reactions, and some drugs lead to more severe reactions ( Table 275-1 ). The dosage and route of administration of a drug can also be risk factors; intermittent, repeated administrations of a drug can be more sensitizing than uninterrupted therapy. Some ethnic groups appear to be more prone to certain ADRs. White Americans are at a higher risk for hypersensitivity reactions to abacavir than other ethnic groups. In a study of drug allergy caused by angiotensin-converting enzyme (ACE) inhibitors, the vulnerable population was African American.

Anesthetic agents (muscle relaxants, thiopental)
Antiarrhythmic drugs (procainamide, quinidine)
Anticonvulsants (hydantoin, phenobarbital, carbamazepine)
Antihypertensive agents (angiotensin-converting enzyme inhibitors)
Antipsychotic tranquilizers
Antisera (antitoxins, monoclonal antibodies)
Anti-tuberculous drugs (isoniazid, rifampicin)
Aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs)
β-Lactam antibiotics
Enzymes (L-asparaginase, streptokinase, chymopapain)
Heavy metals (gold salts)
Radiocontrast media
Recombinant proteins (insulin, other hormones)
Vaccines (egg-based, gelatin)

In the United States, approximately 10% of individuals who seek health care have a history of penicillin allergy. However, if tested with an appropriate panel of skin tests, fewer than 10% of those individuals would be deemed positive for penicillin allergy. Individuals with a positive history and negative skin tests tolerate penicillin-type antibiotics at the same rate as the general population with a negative history; in addition, there is a very low rate of resensitization.



Hypersensitivity reactions to drugs can be classified according to the type of immunologic reaction. An immunologic response to any antigen may be diverse, and the resulting reaction complex. Drugs are no exception, and they have been associated with each of the mechanisms of immune disorders discussed in Chapter 44 . Drugs that are more frequent perpetrators of significant allergy are listed in Table 275-1 .

Most pharmacologic agents are simple structures with molecular weights of less than 1000 D. Alone, they are unable to induce hypersensitivity-type immunologic responses. However, most of these agents have the ability to covalently bind to proteins, forming hapten–carrier complexes, with the low-molecular-weight agent acting as the hapten and the protein being the carrier. The hapten–carrier complexes are able to induce immunologic responses, with most responses being directed at the hapten.

A well-known example of such an agent is penicillin. Benzylpenicillin has a molecular weight of approximately 300 D and is metabolized into a penicilloyl hapten moiety. The penicilloyl moiety, which constitutes about 95% of all penicillin metabolites, is referred to as the major determinant because it is the major metabolite in terms of quantity. It has been conjugated to poly-D-lysine to form penicilloyl-polylysine, which has been commercially available as Pre-Pen (Hollister Stier, Spokane, WA) for skin testing. However, since 2000 the availability of Pre-Pen has been unpredictable. In the absence of reliable skin test reagents, patients who carry a diagnosis of penicillin allergy are often prescribed antibiotics that are more expensive and have a broader spectrum, contributing to increased health care expenses and to antibiotic resistance.

The other 5% of penicillin metabolites are referred to as the minor determinants. Although they are minor in quantity, these determinants actually are the cause of most of the immediate-type anaphylactic reactions, whereas the major determinant is associated with reactions that are less severe and later in onset. Minor determinant reagents have never been commercially available in the United States. Penicillin skin testing has not been widely used by U.S. physicians; annually, only 40,000 doses of the major determinant have been sold. The potential market for a complete set of penicillin skin test reagents is estimated to be approximately 30 million tests in the United States alone.

In contrast to simple low-molecular-weight drugs, therapeutic agents that are proteins of greater than 5000 D in molecular weight can be recognized by the human immune system and can result in sensitization and hypersensitivity reactions on subsequent exposure. Because these proteins are complete antigens, they can be used as skin testing reagents or as antigens/allergens in in vitro assays. Included among therapeutic protein reagents that have been reported to cause hypersensitivity are porcine insulin, antithymocyte globulin (rabbit or equine), chymopapain, streptokinase, latex, and vaccines such as tetanus toxoid, influenza, yellow fever, pertussis, rubella, and MMR (measles, mumps, rubella). Unexpectedly, a variety of human recombinant proteins, including insulin and monoclonal antibodies, have been reported to cause hypersensitivity reactions. The most plausible explanation is that B cells recognize and mount an immunologic response to an alteration of tertiary or quaternary structure of the recombinant human proteins. These immunologic responses are unlikely to be T-cell driven, because the primary amino acid sequence, which is what T cells recognize, is an exact copy of the endogenously produced protein.

Clinical Manifestations


The clinical manifestations of drug allergy often include a dermatologic component. It is estimated that 80 to 90% of drug allergy results in one of the following cutaneous manifestations: exanthematous or morbilliform eruption; urticaria or angioedema or both; contact dermatitis; fixed drug eruption; erythema multiforme–like eruption; or photosensitivity. There are, however, organ-specific drug hypersensitivity reactions that do not include cutaneous manifestations ( Table 275-2 ).

Reactions Implicated Drugs
Pulmonary infiltrates with eosinophilia Nitrofurantoin
Pneumonitis and fibrosis Bleomycin, amiodarone
Noncardiogenic pulmonary edema Hydrochlorothiazide, cocaine, heroin, methadone
Eosinophilia Gold salts, allopurinol, digitalis
Thrombocytopenia Quinidine, gold salts, sulfonamides, heparin
Hemolytic anemia Penicillin, methyldopa
Agranulocytosis Sulfonamides, propylthiouracil, quinidine, procainamide, phenytoin
HEPATIC MANIFESTATIONS Aminosalicylic acid, dapsone
Cholestasis Phenothiazines, erythromycin
Hepatocellular damage Halothane, isoniazid, phenytoin
Mixed pattern Phenytoin, sulfonamides
Nephrotic syndrome Gold salts, captopril, NSAIDs, penicillamine
Acute interstitial nephritis β -Lactam antibiotics, NSAIDS, sulfonamides
Pseudolymphoma Phenytoin
Infectious mononucleosis-like syndrome Aminosalicylic acid, dapsone
CARDIAC MANIFESTATIONS Sulfonamides, β -lactam antibiotics
NEUROLOGIC MANIFESTATIONS Colchicine, nitrofurantoin, sulfonamides
Peripheral neuritis
NSAIDs = nonsteroidal anti-inflammatory drugs.



The diagnosis of drug allergy requires a complete and exhaustive history, along with a physical examination. It also requires compatible clinical manifestations and temporal eligibility. In vitro tests are rarely clinically useful. In vivo testing such as cutaneous tests and provocative test dosing may be clinically indicated in some situations.

Differential Diagnosis


To distinguish drug allergy from other ADRs, several criteria are helpful. Allergic reactions occur in a tiny fraction of individuals who receive the drug and cannot be predicted. The observed clinical effects do not resemble known pharmacologic actions of the drug. In the absence of prior exposure to the drug, allergic or hypersensitivity symptoms rarely appear before 1 week of continuous therapy. In general, drugs used with no reactions for several months or longer are rarely responsible.

The reaction often resembles other allergic/hypersensitivity reactions such as anaphylaxis, urticaria, and serum sickness–like reactions. Although most drug reactions include cutaneous manifestations, some involve only other organ systems, examples being pulmonary infiltrates with eosinophilia, hepatitis, and acute interstitial nephritis. A list of drugs that cause organ-specific reactions can be found in Table 275-2 . Drug-specific antibodies or T-cell receptors or T lymphocytes have been identified that react with the suspected drugs or relevant drug metabolites. As with ADRs in general, the reaction often subsides after the drug is discontinued. However, a hypersensitivity reaction may persist because of the formation of drug metabolites, which act as haptens and bind to carrier proteins such as human serum albumin.






Evidence-Based Treatments

There is a paucity of evidence-based information regarding drug allergy. One study, evaluating the evidence for premedication before administration of antivenom snake bite toxins, concluded that routine prophylactic adrenaline for polyvalent antivenom, known to have high adverse event rates, seems sensible, based on one clinical trial.[1] Antihistamines appear to be of no obvious benefit in preventing acute reactions from antivenoms. A second study, evaluating treatment for toxic epidermal necrolysis (TEN), concluded that there are no randomized controlled trials of the most commonly used therapies (i.e., systemic steroids, cyclosporin A, and intravenous immunoglobulins). Furthermore, treatment with thalidomide was not shown to be effective and was associated with higher mortality in a randomized controlled trial.

Although not entirely evidence-based, there are published clinical guidelines for management of infusion-related hypersensitivity reactions caused by the administration of chemotherapeutic or biologic therapy. These guidelines were developed as part of a performance improvement initiative and resulted in a standardized approach to management of reactions and reporting of ADRs.



Although the outcome of allergic drug reactions is generally favorable, prevention is the obvious goal. The physician should prescribe medications only if they are clinically appropriate and should, if possible, avoid drugs that are known to produce significant hypersensitivity reactions (see Table 275-1 ). Before prescription or administration of a medication, the patient should be asked about prior ADRs to the medication or to other pharmacologically related medications. If appropriate, oral administration is probably preferable to parenteral administration; anaphylaxis is less likely, as is sensitization. Protocols for skin testing to foreign antisera and for management of medication hypersensitivity reactions (e.g., premedication, test dosing, desensitization) are available in the Suggested Readings. Therapeutic guidelines regarding treatment of the most important and common allergic drug reactions are also reviewed in those references. A general algorithm is provided in Figure 275-1 .



FIGURE 275-1  Guidelines for the treatment of patients with a history of a drug allergy. In patients with a history of a suspected or known drug allergy, the first choice would be to use an appropriate non–cross-reacting drug. If such a drug is not available, or if the patient does not respond to it, a schema is presented for further evaluation, based on the availability of a reliable immunologic test to detect drug hypersensitivity.

The risk of an anaphylactic reaction to a drug such as penicillin is a function of the history of onset, severity, and proximity ( Table 275-3 ). If an individual experienced an immediate-type reaction that was rapid in onset, involved life-threatening symptoms or signs, and occurred relatively recently, that individual is at high risk for a severe anaphylactic reaction on subsequent exposure.

Risk Low High
Onset of previous reaction >24 hr < 30 min
Signs and symptoms of previous reaction Morbilliform eruption Life-threatening symptom: hypotension, upper airway angioedema, bronchospasm
Urticaria alone
Time elapsed since reaction >20 yr < 1 yr

Because there are no commercially available reagents for penicillin skin testing, the approach to a patient who needs a β-lactam antibiotic, such as a patient with neutropenic fever requiring Pseudomonas coverage, usually depends on the risk. Risks and benefits should be thoughtfully discussed and documented. In a low-risk individual, cautious test dosing could be performed. In a high-risk individual, desensitization could be considered if the clinical risks and benefits so warrant.



Most drug allergies involve cutaneous eruptions that are self-limited and resolve shortly after the offending agent has been discontinued. However, severe, life-threatening reactions occur in approximately 1 in every 1000 hospitalized patients. In the United States in 1994, 0.32% of hospitalized patients (106,000 people) died from ADRs. The proportion of ADRs that were allergic reactions was not determined in this study but could be estimated to be about one fourth. The incidence of adverse cutaneous reactions to drugs is higher in women than in men. There also is an increased incidence of ADRs in the elderly.

One of the most severe reactions associated with drug allergy fatalities is anaphylactic shock. It is usually immunoglobulin E (IgE) mediated, but it may occur with non–IgE-mediated reactions to drugs such as nonsteroidal anti-inflammatory drugs (NSAIDs) or radiocontrast media. It is estimated that approximately 1500 deaths occur annually in the United States due to anaphylaxis from medications. In the United Kingdom, drugs are the leading cause of anaphylactic fatalities (88 of 202 fatalities in 2001).

Anaphylaxis is not the major cause of mortality due to allergic drug reactions. Erythema multiforme major, also known as Stevens-Johnson syndrome, has a mortality rate of about 5%, and TEN has a mortality rate of about 30%, mostly due to sepsis. There is an estimated incidence of 0.4 to 1.2 cases per million people per year for Stevens-Johnson syndrome, and 1.2 to 6 cases per million annually for TEN.

Future Directions


Pharmacogenomics is an important future direction to identify those individuals who are at risk for a significant allergic reaction to a given drug. Human leukocyte antigen (HLA) genotyping has been reported to identify individuals who are at increased risk for drug hypersensitivity. As an example, individuals with HLA-B☆5701, are at a greater risk (odds ratio, 117; confidence interval, 29 to 481) for a drug hypersensitivity reaction to abacavir, a human immunodeficiency virus (HIV) transcriptase inhibitor. In another example, severe cutaneous adverse reactions to allopurinol are highly associated with the genetic marker HLA-B☆5801. Other avenues by which individuals with specific susceptibility to developing allergy to a given medication may be identified include polymorphisms in immune recognition molecules, drug-metabolizing enzymes (DME), and macromolecular adduct repair systems.

Copyright © 2007 Elsevier Inc. All rights reserved. –

Leave a Reply

Time limit is exhausted. Please reload the CAPTCHA.


apply_now Pepperstone Group Limited