ACP medicine, 3rd Edition


Drug Allergies

Mark S. Dykewicz MD, FACP1

Professor of Internal Medicine; Director

Heather Gray MD2


1Allergy and Immunology Fellowship Program, St. Louis University School of Medicine; Medical Staff Physician, St. Louis University Hospital

2Allergy and Immunology, St. Louis University School of Medicine; Fellow, Immunology, St. Louis University Hospital

Mark S. Dykewicz, M.D., F.A.C.P., has received grants for clinical research or educational activities or served as an advisor or consultant to Astra-Zeneca Pharmaceuticals Ltd.; Genentech, Inc.; Novartis AG; GlaxoSmithKline, Inc.; Merck & Co., Inc.; and Schering-Plough Corp.

Heather C. Gray, M.D., has no commercial relationships with manufacturers of products or providers of services discussed in this chapter.

September 2006

Adverse drug reactions are the most common iatrogenic illnesses, occurring in 1% to 15% of drug courses. They are also a frequent presentation in emergency departments. In a large prospective study of 18,820 hospital admissions, 1,225 (6.5%) were related to adverse drug reactions.1 Most adverse drug reactions result from nonimmunologic or unknown mechanisms (e.g., toxic overdose, toxic side effects, intolerance). Approximately 6% to 10% of drug reactions are caused by proven or suspected immunologic mechanisms mediated by specific antibodies, sensitized T cells, or both.2 Immunologic reactions develop in patients who have become sensitized by previous exposure or continuous exposure to the same or an antigenically related drug. Although identification of antibodies or sensitized T cells directed against the drug helps confirm the diagnosis of immunologic drug reactions, diagnosis usually is based on clinical presentation.3 This chapter provides an overview of pathogenesis, discusses recognition of both common and uncommon patterns of allergic drug reactions, and explains the application of diagnostic tests and management techniques.


Hypersensitivity drug reactions can be influenced by intrinsic properties of the drug, its administration, and the host. Drug factors that increase risk include a higher molecular weight, the ability of the drug or its reactive metabolites to readily bind to self-proteins to form antigenic hapten-protein conjugates, higher dose, parenteral (as opposed to oral) administration, and repeated exposure to the drug. Host factors that increase risk include adult age, female gender, concurrent infections, and HIV infection.4 Phenotypic differences in drug metabolism may also influence risk. For example, rashes from sulfonamides are more likely to occur in patients who are slow acetylator phenotypes, because these patients preferentially metabolize sulfa drugs by alternative oxidative pathways that produce highly reactive metabolites that bind to self-protein carriers.

Although β-lactam antibiotics covalently bind to self-protein carriers to form antigenic conjugates, many non-β-lactam antibiotics require enzyme systems, such as cytochrome P-450, to form reactive products that then bind with self-protein carriers. Consequently, testing with a parent drug may not identify sensitivity to reactive intermediate products.

Classification of Drug Reactions

Classifying drug reactions on the basis of either the temporal relation between drug exposure and adverse manifestations or the presumptive immunologic mechanism may aid in evaluation and management. Temporally, drug reactions are classified as immediate, accelerated, or delayed. Immediate reactions occur within 1 hour of administration and include anaphylaxis. Accelerated reactions, such as urticaria and angioedema, occur within 72 hours of administration. Delayed reactions, which occur 72 hours or more after administration, include urticarial and nonurticarial skin rashes; serum sickness-like reactions; fever; and a variety of cardiopulmonary, hematologic, hepatic, renal, and vasculitic effects.

The Gell and Coombs classification system defines four basic immunologic mechanisms for drug reactions [see Figure 1]. However, some clinical presentations may involve several mechanisms, and not all immunologic mechanisms conform to the Gell and Coombs classification. Type I (anaphylactic) reactions occur when drug antigen cross-links adjacent IgE antibodies that are bound to the surfaces of mast cells or basophils, with consequent cell activation and release of mediators such as histamine, tryptase, and leukotrienes. Common causes of type I reactions include antibiotics, vaccines, allergen extracts, and proteins (e.g., antisera, insulin). Reactions occur within seconds to minutes of exposure and range from full anaphylaxis to any component thereof, including pruritus, flushing, angioedema, urticaria, bronchospasm, laryngeal edema, rhinoconjunctivitis, hypotension, tachycardia, nausea, vomiting, diarrhea, and abdominal or uterine cramping. In contrast to anaphylaxis, syncope or vasovagal reactions are typically characterized by blanching rather than flushing and by bradycardia rather than tachycardia.


Figure 1. Gell and Coombs System

The Gell and Coombs system defines four basic immunologic mechanisms for drug reactions.40 Type I reactions (anaphylaxis) result from IgE antibodies binding to drug antigen and cross-linkage of adjacent IgE molecules, leading to mediator release. Type II (cytotoxic) reactions occur when IgG or IgM antibodies recognize drug antigen associated with cell membranes, causing complement activation. Type III (immune complex) reactions involve the formation of antigen-antibody complexes. Type IV (delayed hypersensitivity) reactions are mediated by sensitized lymphocytes.

Type II (cytotoxicity) reactions result in cell destruction mediated by an interaction between IgG or IgM antibodies, complement, and a drug antigen associated with cell membranes. Clinical sequelae include immune hemolytic anemia, thrombocytopenia, and granulocytopenia. Heparin-induced thrombocytopenia is mediated by antibodies directed against antigen complexes of heparin and platelet factor 4 on the surface of platelets [see 5:XIV Thrombotic Disorders].

Type III (immune complex) reactions develop when a drug combines with antibodies to form immune complexes, the deposition of which causes tissue damage. Serum sickness is a type III reaction and may manifest as skin lesions (e.g., urticaria, angioedema, maculopapular or morbilliform rash, palpable purpura), arthralgias and arthritis, lymphadenopathy, fever, nephritis, and hepatitis. Serum sickness usually occurs after 1 to 4 weeks of drug use but may occur sooner in patients with previous exposure. Drug-induced lupus syndromes are also type III reactions. Renal involvement is rare, as is the presence of anti-double-stranded DNA antibodies, but other autoantibodies are common; these autoantibodies include antihistone from procainamide, hydralazine, or phenytoin; perinuclear antineutrophil cytoplasmic antibody (p-ANCA) from minocycline; and SS-A and SS-B from thiazides.5

Type IV (delayed hypersensitivity) reactions are mediated by sensitized CD4+ or CD8+ T cells.6 Allergic contact dermatitis is a classic example and typically develops 24 to 72 hours after topical exposure.


To identify drug reactions, the physician needs to be familiar with general principles of drug reactions [see Table 1] and with the individual drugs the patient has taken.7 Diagnosis depends largely on the nature of the reaction and its timing in relation to drug use. For example, adverse immunologic reactions to drugs usually occur in the early weeks of drug exposure and become less common with continued drug administration. To confirm the clinical diagnosis, skin testing or drug challenges may be valuable in selected cases [see Figure 2].8

Table 1 General Considerations in the Clincial Evaluation of Drug Reactions4

Identify drugs that have a history of causing problems in the patient; determine whether there are cross-reacting agents, and avoid them.

If the patient has a late reaction, such as a drug rash, take a careful history of all drugs used in the past month, because it is possible that the causative drug has been discontinued.

Drugs administered with impunity for prolonged periods (e.g., months to years) are rarely responsible for adverse immunologic reactions. Reactions are more likely to result from drugs introduced more recently.

Have a high index of suspicion for drug reactions whenever a patient experiences adverse clinical manifestations. Bear in mind that drug reactions can involve internal organs (e.g., nephritis, hepatitis, isolated lymphadenopathy), often in the absence of eosinophilia.

If an immunologic drug reaction is suspected, stop all nonessential drugs and substitute non–cross-reactive drugs if possible.


Figure 2. Management of Immune-Mediated Antibiotic Allergy

Management of an immune-mediated antibiotic allergy. The rapidity of a patient's reaction to a drug determines management. An immediate IgE-mediated reaction may require skin testing to identify allergen-specific IgE antibodies; in cases in which the skin test is inconclusive, additional tests may be required. Elevated titers of serum tryptase can confirm the diagnosis in cases of suspected anaphalaxis. Treatment options are determined by the nature and severity of the reaction. In the management of immediate reactions, drug desensitization may be performed if the implicated drug is required for treatment of the patient. For reactions that are not immediate, management depends on the clinical manifestations of the previous reaction. For maculopapular eruptions, readministration of the drug on an incremental-dosing schedule chosen with a specialist (and at dosing intervals ranging from hours to days or even weeks) may allow use of the drug. In all cases, education of the patient and communication with the patient and the referring physician are vital to ensure the success of the management strategy and to prevent a recurrence of antibiotic allergy.8

Clinical Presentation

Dermatologic Reactions

Drug reactions involving the skin range from maculopapular and morbilliform rashes to urticaria, angioedema, erythema multiforme, erythema nodosum, bullous eruptions, and exfoliations. Drug eruptions usually occur within days of exposure but may occur after drug cessation. Eruptions are symmetrical and truncal; they are accompanied by pruritus, fever, and, occasionally, eosinophilia. Palm and sole involvement suggests a viral exanthem rather than a drug reaction. Photosensitive drug rashes are of two kinds: phototoxic or photoallergic. Phototoxic reactions are nonimmunologic, generally appear as sunburn 4 to 8 hours after light exposure, and often occur with the first exposure to the drug; tetracycline has been associated with phototoxic reactions. Photoallergic reactions are typically eczematous rashes that occur after days or months of exposure; photoallergic reactions have been associated with some sulfa drugs. Alteration of the drug by ultraviolet light enables conjugation of the drug to self-proteins and T cell-mediated immune responses. Neither type of photosensitive reaction is predictive of other types of adverse reactions to a drug.


Drug fever may occur by itself or in association with other allergic manifestations, such as a rash. The fever stems from the release of pyrogens by phagocytic cells that have engulfed drug-IgG immune complexes or from cytokine release and other incompletely established processes associated with specific T cell activation. Drug fever usually occurs 7 to 10 days into a treatment course, with prompt defervescence within 48 hours of cessation of the responsible agent.

Systemic Manifestations

Drug reactions can result in systemic involvement, such as interstitial nephritis, nephrotic syndrome, hepatic reactions, myocarditis, and vasculitis. Lung involvement may present as part of a syndrome consisting of malaise, nonproductive cough, chest discomfort, and migratory infiltrates, with or without peripheral eosinophilia (Löffler syndrome). Long-term treatment with penicillin, sulfonamides, or phenytoin can result in generalized lymphadenopathy. Aseptic meningitis has been reported from nonsteroidal anti-inflammatory drugs (NSAIDs), radiocontrast media, and other agents.4,9,10


Skin Testing

Drug reactions can be identified through immediate-type skin testing only when the process is IgE mediated [see Figure 2]. Non-IgE-mediated reactions, such as nonurticarial skin rashes, cannot be identified in this manner. Skin testing is reliable for protein agents and small-molecular-weight drugs whose allergenic metabolites have been identified and made available for skin testing (e.g., penicillin). Skin testing requires knowledgeable personnel and the use of appropriate concentrations (i.e., concentrations high enough to provoke a reaction but low enough to avoid causing a systemic response).

In Vitro Testing

There are a limited number of radioallergosorbent tests (RASTs) available for detection of IgE antibodies against drugs, including β-lactam antibiotics and anesthetic agents. In vitro tests for drug allergy are generally less sensitive than skin tests but may be useful in certain cases in which skin testing is not possible (e.g., in patients with severe, generalized eczema or in those who must take medications that can suppress skin-test responses).11

Drug Challenges

When the probability of a true allergy is remote, a graded challenge can be used to confirm the clinical diagnosis of drug reaction. In graded challenges, the patient is given a test dose at a dose lower than would cause a serious reaction. Subsequent doses are then escalated in large increments, and the patient is observed between doses.8

Reactions to Specific Drugs


Penicillin, a β-lactam antibiotic, is among the most common causes of immunologic drug reactions. Most deaths from penicillin reactions occur in patients with no history of penicillin allergy. Nonimmunologic rashes are frequently seen with ampicillin or amoxicillin in patients who have concomitant viral infections, chronic lymphocytic leukemia, or hyperuricemia and in patients taking allopurinol. These rashes are typically nonpruritic and are not associated with an increased risk of future intolerance of penicillin antibiotics.

Most immunologic reactions to penicillins are directed against β-lactam core determinants. Less than 5% of penicillin is metabolized to the minor determinants, which include benzyl penicillin G, penicilloates, and benzylpenicilloylamine. IgE antibodies to the minor determinants are usually responsible for severe immediate-type reactions to penicillin. The benzylpenicilloyl moiety, the so-called major determinant, makes up 95% of penicillin metabolites but is less commonly responsible for severe immediate reactions.

Patients who have suffered IgE-mediated penicillin reactions tend to lose their sensitivity over time if penicillin is avoided. By 5 years after an immediate reaction, 50% of such patients have negative results on skin testing. There is controversy about whether patients who have lost their sensitivity to penicillin may be more likely than others to be sensitized with subsequent penicillin exposure; accumulating evidence indicates that penicillin resensitization is low.4,12,13 Skin testing with a major determinant preparation and penicillin G identifies at least 90% to 93% of patients at risk for immediate reaction to penicillin. The negative predictive value of penicillin skin testing is significantly increased by the addition of a minor determinant mixture, but that is not currently commercially available. Penicillin skin testing is usually reliable in identifying patients at risk for immediate reactions to the semisynthetic penicillins,4,14 but reactions to unique semisynthetic side-chain determinants may occur in special-risk populations, such as cystic fibrosis patients.15 Skin testing with the side-chain determinants is not commonly done.

Not everyone with a history of a reaction to penicillin should undergo skin testing, but it is important to perform a skin test on patients with a history of anaphylaxis or urticaria associated with penicillin use before they are given penicillin again [see Figure 2]. Patients who report a history of maculopapular or morbilliform skin rashes from the use of penicillin are not at higher risk for immediate-type reactions, but skin testing may be considered for these patients because studies have demonstrated that patient histories can be unreliable. A carefully taken patient history can often distinguish patients with credible histories of penicillin allergy (who might benefit from skin testing) from patients whose histories are not credible (who might then be treated with penicillin).16 However, a large proportion of patients who report vague histories of penicillin allergy have penicillin-specific IgE antibodies; consequently, patients with vague histories of allergy should undergo penicillin skin testing.17 Penicillin should not be readministered to patients with a history of penicillin-induced Stevens-Johnson syndrome, toxic epidermal necrolysis (TEN), other exfoliative dermatitis, or bullous skin lesions; therefore, skin testing is not indicated in these cases. Patients with a family history of penicillin allergy but no personal history of penicillin allergy are not at increased risk for allergy to penicillin and, therefore, do not require skin testing. Desensitization is not required in a patient with a history of an immediate reaction and a negative skin test result, but a small test dose may be given as an additional precaution if the previous reaction was life threatening. If there is a compelling indication for a penicillin antibiotic (e.g., neurosyphilis), a rapid desensitization protocol should be used [see Table 2].9,18,19,20

Table 2 Intravenous Desensitization Protocol for β-Lactam Antibiotics31


β-Lactam Stock Concentration (mg/ml)

Cumulative Dose Given (ml)




















































Note: observe patient for approximately 15–40 min after each interval dose and for 30 min after final dose.

Cephalosporins and penicillin have similar bicyclic β-lactam structures and amide side chains [see Figure 3]. The degree of immunologic cross-reactivity between these β-lactams is controversial, but patients with penicillin allergy are more likely than the general population to have a reaction to another β-lactam drug. In a large study of patients receiving penicillin followed by a cephalosporin, the unadjusted risk ratio for an allergic-like event for patients who had experienced a prior event, as compared with those who had no such prior event, was 10.1 (confidence interval, 7.4 to 13.8); however, the researchers concluded that cross-reactivity alone is not an adequate explanation for this increased risk.21 Most experts agree that cephalosporins should be avoided in patients who have a history of an immediate-type reaction to penicillin.20,22 Immunologic reactions to cephalosporins are more often related to side chains of the cephalosporin antibiotics than to β-lactam core determinants.23 Cephalosporin skin testing is experimental and has uncertain negative predictive value. There is a lower incidence of immediate-type reactions to third-generation cephalosporins than to the first- and second-generation compounds.24Because of its 3-methylthiotetrazole side chain, cefoperazone can cause a nonimmunologic disulfiram-like reaction if taken after ingestion of alcohol. Carbapenems (imipenem) and carbacephems (loracarbef) contain bicyclic β-lactam rings that cause significant cross-reactivity with penicillin.25 Reactions to monobactams (e.g., aztreonam) are typically directed against side-chain determinants rather than the monocyclic β-lactam nucleus, but cross-reactivity occurs with ceftazidime, which shares an identical side chain with penicillin.26,27


Figure 3. Structure of β-lactam Antibiotics

Structure of β-lactam antibiotics. Substitutions at the R position of 6-aminopenicillanic acid create penicillin derivatives. Substitutions at positions 1, R1, R2, and C7 of 7-aminocephalosporanic acid create cephalosporin derivatives.


Drug exanthems from sulfonamide antibiotics are more common than immediate-type reactions, and there is significant cross-reactivity among sulfa compounds. Adverse effects from sulfa antibiotics occur in 2% to 10% of patients who do not have AIDS, whereas 50% of AIDS patients suffer ill effects.4 The incidence of reactions may be related to the degree of immunodeficiency from HIV infection. The sulfapyridine moiety of sulfasalazine is responsible for most skin rashes from that agent. A 1-month graded challenge protocol is usually successful at inducing tolerance to sulfasalazine in patients who require this agent for treatment of inflammatory bowel disease.28Desensitization to sulfonamide antibiotics may be considered to permit the use of these agents for prophylaxis against Pneumocystis jirovecipneumonia in patients with AIDS, toxoplasmosis, and other infections for which there are no good alternatives. A number of protocols have been published. Patients with acute Pneumocystis pneumonia may require rapid desensitization protocols that permit therapeutic use of the medication within 6 to 8 hours29; those patients receiving a sulfonamide prophylactically can be desensitized with slowly increasing doses. Desensitization protocols should not be attempted in patients with a history of severe drug reactions, such as Stevens-Johnson syndrome or TEN.20 There is no immunologic cross-reactivity between sulfonamide antibiotics and nonantibiotic agents that have sulfonamide moieties (e.g., thiazides, celecoxib, glyburide, and triptans). However, patients with a history of allergic reactions to sulfonamide or penicillin antibiotics are at increased risk for developing reactions to nonantibiotic sulfonamides.30 This evidence further demonstrates that patients with a history of drug allergy to some agents are at greater risk than the general population for developing allergic reactions to structurally distinct drugs.


Vancomycin infusions are commonly associated with the red man syndrome (hypotension, flushing, erythema, pruritus, urticaria, and pain or muscle spasms of the chest and back). The syndrome is caused by non-IgE-mediated histamine release that is more likely with rapid infusion rates (> 10 mg/min). Tolerance of readministration is promoted by reduction of the infusion rate and pretreatment with H1 (but not H2) antihistamines. Rarer IgE-mediated reactions to vancomycin can be identified by skin tests.3,4,9

ACE Inhibitors

Nonimmunologic adverse effects of angiotensin-converting enzyme (ACE) inhibitors are thought to stem from an accumulation of bradykinin and other vasoactive peptides. The most frequently documented adverse reactions include cough (10% to 25%), rhinitis, and angioedema (0.1% to 0.2%). The onset of cough can occur from 1 day to up to 12 months after starting these drugs and may be associated with increased bronchial reactivity to methacholine. Cough usually resolves within several weeks after drug cessation but may persist for more than a month. In about 60% of cases, angioedema occurs within 2 weeks after patients start the drug, but it can occur after months. Angioedema usually involves the face and oropharyngeal tissue. It can result in life-threatening upper airway obstruction; for patients with angioedema, particularly of the head and neck, that is unresponsive to usual measures [see 6:XIII Urticaria, Angioedema, and Anaphylaxis], fresh frozen plasma administration has been reported to be beneficial.31 Visceral angioedema can cause abdominal pain. Cough and angioedema do not usually occur in the same patient. Skin testing is of no value. Intolerance to one ACE inhibitor usually predicts intolerance to all drugs of this class. Patients with idiopathic angioedema and urticaria are susceptible to more severe and frequent episodes when given ACE inhibitors. The angiotensin II receptor blockers are generally well tolerated in patients with idiopathic angioedema and urticaria or with ACE inhibitor-related angioedema.3,9,32

Aspirin and NSAIDS

Aspirin and other NSAIDs may induce a variety of reactions, ranging from bronchospasm, rhinorrhea, urticaria, and angioedema to anaphylaxis. Aspirin-sensitive respiratory reactions are likely caused by derangement of arachidonic acid metabolism with increased leukotriene production. Mast cell activation may also occur. Patients with respiratory sensitivity to aspirin generally develop dose-dependent reactions to aspirin or structurally distinct NSAIDs that are significant inhibitors of cyclooxygenase-1 (COX-1) but often tolerate agents that have less effect on COX-1 (e.g., salsalate, acetaminophen, sodium or magnesium salicylate). Selective COX-2 inhibitors (e.g., celecoxib, rofecoxib) are well tolerated in these patients.33 Patients may present with the so-called aspirin triad of concomitant asthma, nasal polyps, and aspirin sensitivity. Between 30% and 40% of patients with polyps and sinusitis and between 8% and 21% of adults with asthma have positive bronchial responses to aspirin.9,34 Urticarial reactions from aspirin typically occur in a patient subset different from that in which respiratory reactions occur.35 Some patients with skin reactions to aspirin (often those with a history of idiopathic urticaria and angioedema) have cross-reactivity with NSAIDs, whereas other skin reactors and those who develop anaphylaxis have only specific sensitivity to aspirin or a particular NSAID.34,36 However, skin testing is not helpful. With appropriate precautions, oral desensitization with aspirin by experienced clinicians can be performed over several days. It is usually effective in patients with respiratory sensitivity but not in those with skin reactions.37

Anesthetic Agents

Adverse reactions to local anesthetic agents are rarely IgE mediated. More commonly, such reactions are toxic responses from inadvertent intravenous administration, overdose, rapid absorption, or anxiety. Symptoms often involve the central nervous system or the cardiovascular system and include hypotension, convulsions, and cardiorespiratory failure. Concurrent administration of epinephrine may be responsible for shakiness and tachycardia. Allergic contact dermatitis and some large local reactions do occur through delayed-type immunologic responses [see 2:V Contact Dermatitis and Related Disorders]. Local anesthetics are either benzoid acid esters (type I [e.g., procaine, benzocaine]) or nonesters and amides (type II [e.g., lidocaine, bupivacaine, mepivacaine]). There is no cross-reactivity between the two classes, but type I agents cross-react with each other. Management of suspected local anesthetic allergy includes subcutaneous test dosing with the local anesthetic without epinephrine.

Histamine release has been implicated in some reactions from anesthesia-induction agents and muscle relaxants, but the responsible mechanisms for many reactions are not established. During anesthesia, generalized reactions may be caused by muscle relaxants (e.g., succinylcholine, alcuronium, pancuronium), induction agents (e.g., thiopental), opiates, or antibiotics. Narcotics stimulate mast cells directly without an IgE mechanism.

Radiographic Contrast Media

Radiographic contrast media cause non-IgE-mediated anaphylactoid reactions that involve direct mast cell and perhaps complement activation; therefore, immediate-type skin testing and test dosing are not helpful. Shellfish allergy results from IgE-mediated reactions to shellfish proteins, and therefore, it is not predictive of risk for contrast reactions. A previous anaphylactoid reaction to contrast at any time in a patient's history is predictive of persistently increased risk of a repeat anaphylactoid reaction, even though the patient may have tolerated contrast without a reaction in the interim. Asthma and allergies are also associated with increased risk.38

The use of nonionic contrast media and medication pretreatment can reduce the risk of reaction. One commonly used pretreatment regimen consists of corticosteroids (prednisone, 50 mg, given 13 hours, 7 hours, and 1 hour before contrast administration), H1 antihistamines (diphenhydramine, 50 mg orally, given 1 hour before administration), and oral adrenergic agents (ephedrine, 25 mg, or albuterol, 4 mg, given orally 1 hour before administration). H2 receptor blockers are sometimes added. Despite an adequate pretreatment regimen, reactions can still occur. Corticosteroids administered only 1 to 2 hours before administration of contrast do not reliably prevent reactions.


Management of Acute Reactions

Treatment of acute adverse immunologic drug reactions includes stopping all nonessential suspect drugs and, if necessary, substituting new drugs that should not have cross-reactivity with any of the suspect drugs. Epinephrine, antihistamines, and corticosteroids are the mainstays of treatment of anaphylaxis, and other resuscitative measures may be required [see 6:XIII Urticaria, Angioedema, and Anaphylaxis]. Mild maculopapular rashes may respond to antihistamines alone, but progressing rashes or rashes associated with fever, nausea, or arthralgias should also be treated with systemic corticosteroids. For prolonged, severe reactions, several weeks of prednisone therapy may be required.3,4,20

Desensitization and other Specialized Approaches

If the probability of a drug allergy is high and drug administration is essential, one may consider desensitization, in which the drug is administered in increasing doses in small increments.39 Because of the risk of adverse reactions, only experienced physicians should perform desensitization. Once desensitization is achieved, the drug must be continued; otherwise, desensitization will be lost, and the patient will require repeat desensitization before readministration. Pretreatment with antihistamines and corticosteroids is not reliable for preventing IgE-mediated anaphylaxis but can be useful when an anaphylactoid reaction is of concern (as in the use of radiographic contrast media). In extreme circumstances, when continued administration of a drug is essential but the patient is experiencing a reaction such as a late drug rash or interstitial nephritis, continued drug administration may be tolerated if corticosteroids and antihistamines are given to suppress the immunologic reaction. However, in such cases, there is the risk of progression to an exfoliative skin rash, mucocutaneous disorders (e.g., Stevens-Johnson syndrome, TEN), nephritis, hepatitis, or serum sickness.


Figures 1 and 3 Seward Hung.


  1. Pirmohamed M, James S, Meakin S, et al: Adverse drug reactions as cause of admission to hospital: prospective analysis of 18,820 patients. BMJ 329:15, 2004
  2. Gruchalla RS: Drug allergy. J Allergy Clin Immunol 111(2 suppl):548, 2003
  3. Dykewicz MS: Drug allergy. Comprehensive Therapy 22:353, 1996
  4. DeSwarte RD: Drug allergy. Allergic Diseases: Diagnosis and Management, 6th ed. Patterson R, Greenberger PA, Grammer LC, Eds. Lippincott, Philadelphia, 2002, p 295
  5. Srivastava M, Rencic A, Diglio G, et al: Drug-induced, Ro/SSA-positive cutaneous lupus erythematosus. Arch Dermatol 139:45, 2003
  6. Pichler WJ: Delayed drug hypersensitivity reactions. Ann Intern Med 139:683, 2003
  7. Greenberger PA: Drug allergy. J Allergy Clin Immunol 117:S464, 2006
  8. Gruchalla RS, Pirmohamed M: Antibiotic allergy. N Engl J Med 354:601, 2006
  9. Dykewicz MS: Drug allergy. An Expert Guide to Allergy and Immunology. Slavin RG, Reisman RE, Eds. American College of Physicians, Philadelphia, 1999, p 127
  10. DeShazo RD, Kemp SF: Allergic reactions to drugs and biologic agents. JAMA 278:1895, 1997
  11. Primeau MN, Adkinson NF Jr: Recent advances in the diagnosis of drug allergy. Curr Opin Allergy Clin Immunol 1:337, 2001
  12. Bittner A, Greenberger PA: Incidence of resensitization after tolerating penicillin treatment in penicillin-allergic patients. Allergy Asthma Proc 25:161, 2004
  13. Solensky R, Earl HS, Gruchalla RS: Lack of penicillin resensitization in patients with a history of penicillin allergy after receiving repeated penicillin courses. Arch Intern Med 162:822, 2002
  14. Sogn DD, Evans R III, Shepherd GM, et al: Results of the National Institute of Allergy and Infectious Diseases collaborative clinical trial to test the predictive value of skin testing with major and minor penicillin determinant derivatives in hospitalized patients. Arch Intern Med 152:1025, 1992
  15. Ramesh S: Antibiotic hypersensitivity in patients with CF. Clin Rev Allergy Immunol 23:123, 2002
  16. Stember RH: Prevalence of skin test reactivity in patients with convincing, vague, and unacceptable histories of penicillin allergy. Allergy Asthma Proc 26:59, 2005
  17. Solensky R, Earl HS, Gruchalla RS: Penicillin allergy: prevalence of vague history in skin test-positive patients. Ann Allergy Asthma Immunol 85:195, 2000
  18. Sullivan T, Yecies L, Shatz G, et al: Desensitization of patients allergic to penicillin using orally administered beta-lactam antibiotics. J Allergy Clin Immunol 69:275, 1982
  19. Wickern GM, Nish WA, Bitner AS, et al: Allergy to beta-lactams: a survey of current practices. J Allergy Clin Immunol 94:725, 1994
  20. Executive summary of disease management of drug hypersensitivity: a practice parameter. Joint Task Force on Practice Parameters, and the Joint Council of Allergy, Asthma and Immunology. Ann Allergy Asthma Immunol 83:665, 1999
  21. Apter AJ, Kinman JL, Bilker WB, et al: Is there cross-reactivity between penicillins and cephalosporins? Am J Med 119:354, 2006
  22. Romano A, Gueant-Rogriguez RM, Viola M, et al: Cross-reactivity and tolerability of cephalosporins in patients with immediate hypersensitivity to penicillins. Ann Intern Med 141:16, 2004
  23. Kelkar PS, Li JT: Current concepts: cephalosporin allergy. N Engl J Med 345:804, 2001
  24. Fonacier L, Hirschberg R, Gerson S: Adverse drug reactions to a cephalosporin in hospitalized patients with a history of penicillin allergy. Allergy Asthma Proc 26:135, 2005
  25. Prescott WA Jr, DePestel DD, Ellis JJ, et al: Incidence of carbapenem-associated allergic-type reactions among patients with versus patients without a reported penicillin allergy. Clin Infect Dis 38:1102, 2004
  26. Saxon A, Adelman DC, Patel A, et al: Imipenem cross-reactivity with penicillin in humans. J Allergy Clin Immunol 82:213, 1988
  27. Adkinson NF Jr: Immunogenicity and cross-allergenicity of aztreonam. Am J Med 88:12S, 1990
  28. Purdy BH, Philips DM, Summers RW: Desensitization for sulfasalazine skin rash. Ann Intern Med 100:512, 1984
  29. Leoung GS, Stanford JF, Giordano MR, et al: Trimethoprim-sulfamethoxazole (TMP-SMZ) dose escalation versus direct rechallenge forPneumocystis cariniipneumonia prophylaxis in human immunodeficiency virus-infected patients with previous adverse reaction to TMP-SMZ. J Infect Dis 184:992, 2001
  30. Strom BL, Schinnar R, Apter AJ: Absence of cross-reactivity between sulfonamide antibiotics and sulfonamide nonantibiotics. N Engl J Med 349:1628, 2003
  31. Warrier MR, Copilevitz CA, Dykewicz MS, et al: Fresh frozen plasma (FFP) in the treatment of resistant ACE inhibitor (ACE-I) angioedema. Ann Allergy Asthma Immunol 92:573, 2004
  32. Dykewicz MS: Cough and angioedema from ACE inhibitors: new insights into mechanisms and management. Current Opin Allergy Clin Immunol 4:267, 2004
  33. Pacor ML, Di Lorenzo G, Biasi D, et al: Safety of rofecoxib in subjects with a history of adverse cutaneous reactions to aspirin and/or non-steroidal anti-inflammatory drugs. Clin Exp Allergy 32:397, 2002
  34. Jenkins C, Costello J, Hodge L: Systematic review of prevalence of aspirin induced asthma and its implications for clinical practice. BMJ 328:434, 2004
  35. Szczeklik A, Sanak M: The broken balance in aspirin hypersensitivity. Eur J Pharmacol 533:145, 2006
  36. Namazy JA, Simon RA: Sensitivity to nonsteroidal anti-inflammatory drugs. Ann Allergy Asthma Immunol 89:542, 2002
  37. Berges-Gimeno MP, Simon RA, Stevenson DD: Long-term treatment with aspirin desensitization in asthmatic patients with aspirin-exacerbated respiratory disease. J Allergy Clin Immunol 111:180, 2003
  38. Maddox TG: Adverse reactions to contrast material: recognition, prevention, and treatment. Am Fam Physician 66:1229, 2002
  39. Solensky R: Drug desensitization. Immunol Allergy Clin North Am 24:425, 2004
  40. Grammer LC: Drug allergy. Atlas of Allergies. Slavin RG, Fireman P, Eds. Mosby-Wolfe Publishers, Barcelona, Spain, 1996, p 286
  41. Sullivan TJ: Drug allergy. Middleton's Allergy: Principles and Practice, 6th ed. Adkinson NF Jr, Busse WW, Holgate ST, et al, Eds. CV Mosby, Philadelphia, 2003, p 1679

Editors: Dale, David C.; Federman, Daniel D.