INTRODUCTION — The onset of hives within a few hours of taking a new medication is easily recognized as possible drug hypersensitivity. However, many clinical presentations of drug hypersensitivity are more complex or take place in the setting of illness and/or polypharmacotherapy. The patient in intensive care who develops a rash while receiving multiple medications or the ambulatory patient with complex chronic diseases who develops a new and unexplained symptom while taking many medications, illustrates two common presentations of drug allergy.
The main questions that arise in such situations include the following:
●Is the adverse event related to a drug?
●If so, which drug is responsible?
●Is the reaction due to an immune response to the drug (ie, true drug allergy) or due to another mechanism, such as pseudoallergy? (See 'Drug pseudoallergy' below.)
●If the reaction is a drug allergy, what is the likely mechanism?
●How can necessary therapy for the underlying medical problem be continued?
A systematic approach to the evaluation of patients with suspected drug allergy, particularly those receiving multiple medications simultaneously, will be presented here. Evaluation of patients with specific symptoms or laboratory findings that may represent drug allergy is discussed separately. (See "Drug eruptions" and "Fever and rash in the immunocompetent patient" and "Fever and rash in patients with HIV" and "Approach to the patient with unexplained eosinophilia".)
DEFINITION AND CLASSIFICATION — "Drug hypersensitivity" is a general term that includes both allergic and pseudoallergic drug reactions, which are a subset of idiosyncratic drug reactions.
Drug allergy — A drug allergy is an adverse drug reaction that is caused by an immunologic reaction elicited by a drug [1]. There are different classification systems for drug allergy, which are briefly reviewed here and discussed in detail elsewhere. (See "Drug hypersensitivity: Classification and clinical features" and "Drug allergy: Pathogenesis".)
Classification — Immunologic drug reactions are divided into four categories according to the Gell and Coombs system (table 1):
●Type I – Immediate in onset and caused by immunoglobulin (Ig)E-mediated activation of mast cells and basophils
●Type II – Delayed in onset and caused by antibody (usually IgG-mediated) cell destruction
●Type III – Delayed in onset and caused by immune complex (IgG:drug) deposition and complement activation
●Type IV – Delayed in onset and T cell-mediated
The classification of a drug-allergic reaction is important in determining appropriate diagnostic procedures, options for further treatment, and possible cross-reactivity with similar medications.
The World Allergy Organization (WAO) has recommended dividing immunologic drug reactions into two types:
●Immediate reactions, occurring within one hour of the first administered dose
●Delayed reactions, occurring after one hour but usually more than six hours and occasionally weeks to months after the start of administration
The WAO distinction between immediate and delayed drug reactions is helpful in distinguishing type I, IgE-mediated reactions from the other types. Type I reactions carry the risk of immediate life-threatening anaphylaxis if the drug is readministered. These reactions most commonly appear within minutes after exposure but may begin after one hour following oral administration, especially if the drug is taken with food, which further slows absorption.
Delayed reactions are usually not IgE-mediated and result instead from types II, III, and mainly type IV hypersensitivity or from multiple or unknown pathophysiologic mechanisms. Some of these reactions can also be life-threatening, and as a group, delayed reactions probably account for more deaths than anaphylaxis.
Drug pseudoallergy — A pseudoallergic drug reaction is a reaction that is similar or identical in presentation to an immunologic reaction but is not mediated by the immune system (table 2). The term "nonimmune-mediated hypersensitivity" is also used. Pseudoallergic reactions, such as those to nonsteroidal anti-inflammatory drugs, can be as severe as IgE-mediated anaphylaxis and are managed acutely in the same manner (ie, with epinephrine). However, the evaluation and future prevention of these reactions is different from that of immunologic reactions. Pseudoallergic drug reactions are discussed in more detail elsewhere. (See "Drug hypersensitivity: Classification and clinical features", section on 'Pseudoallergy'.)
TREATMENT OF ACUTE REACTIONS — The treatment of various types of drug reactions is discussed in specific topic reviews. The suspected culprit drug is immediately discontinued in most cases, except for some of the milder delayed reactions in situations in which the drug is essential.
Immediate reactions (ie, urticaria, angioedema, or anaphylaxis):
●(See "Anaphylaxis: Emergency treatment".)
●(See "New-onset urticaria", section on 'Treatment'.)
Delayed reactions:
●(See "Exanthematous (maculopapular) drug eruption", section on 'Management'.)
●(See "Drug fever".)
●(See "Drug reaction with eosinophilia and systemic symptoms (DRESS)", section on 'Management'.)
●(See "Serum sickness and serum sickness-like reactions", section on 'Treatment'.)
●(See "Acute generalized exanthematous pustulosis (AGEP)", section on 'Management'.)
●(See "Fixed drug eruption".)
RISK FACTORS FOR DRUG ALLERGY — Risk factors for the development of drug allergy include the female sex, prior history of drug-allergic reactions, recurrent drug exposure, genetic factors, and certain disease states.
Female sex — Women are at greater risk than men for allergic reactions to some drugs, for reasons that are not well-defined. Women can experience higher rates of both immediate and delayed reactions. As an example of the latter, women of childbearing age developed delayed-onset rashes to gemifloxacin (a quinolone) in 30 percent of administrations, compared with 3 percent in men [2]. Intraoperative reactions to neuromuscular-blocking agents are also much more frequent in women. In contrast, neither immediate nor delayed allergic reactions to penicillins differ by gender.
Prior history of allergic drug reactions — A prior history of allergic reactions to one or more drugs increases the risk of developing additional drug allergies. This is obviously true in patients with past drug reactions who are subsequently given the same or a related medication. It is also true with pseudoallergic reactions due to nonsteroidal anti-inflammatory drugs (NSAIDs), since structurally different NSAIDs may cause similar reactions. Pseudoallergic reactions are reviewed separately. (See "Drug hypersensitivity: Classification and clinical features", section on 'Pseudoallergy'.)
Multiple drug allergy syndrome — The term "multiple drug allergy syndrome" has been used to describe different conditions:
●It was initially used to describe patients with past drug-allergic reactions, especially IgE-dependent reactions, who were at increased risk for reactions to unrelated drugs [3]. One prospective study showed that patients with antibiotic sensitivity by history had a 10-fold increased risk for allergic reactions to unrelated antibiotics [4]. A wide variety of clinical reactions were observed in these patients.
●Other researchers applied the term "multiple drug allergy syndrome" to patients who reacted to antibiotics and often various NSAIDs [5]. These reactions were typically urticaria or mild, transient rashes. Some of these patients had underlying chronic urticaria, which is a main differential diagnosis in patients with recurrent "rashes" after various, chemically unrelated medications. The majority of such patients are classified as having pseudoallergic reactions to NSAIDs or NSAID intolerance, but some appear to react to other drugs as well. (See "NSAIDs (including aspirin): Allergic and pseudoallergic reactions".)
●Subsequently, the term "multiple drug allergy syndrome" has been used more narrowly to describe persons who had experienced well-documented immunologic drug reactions (of any Gell and Coombs type) to two or more chemically unrelated medications [6,7]. This appears to arise because of an enhanced responsiveness of the patient's T cells to pharmaceutical substances [8,9]. The implicated drugs are often antibiotics, although hypnotics, antidepressants, local anesthetics, glucocorticoids, and other drug classes were also reported. In a typical clinical scenario, a patient might initially develop drug-induced hypersensitivity syndrome/drug reaction with eosinophilia and systemic symptoms (eg, DiHS/DRESS, a type IV reaction) to an antiepileptic drug. After this event, the patient develops additional drug allergies. The nature of subsequent reactions depends upon the structure/type of allergy normally associated with the drug in question. For example, if the same patient is later given amoxicillin, the reaction is typically an exanthema. If given phenytoin, another DiHS/DRESS reaction may ensue. DiHS/DRESS reactions are described separately. (See "Drug reaction with eosinophilia and systemic symptoms (DRESS)".)
Multiple drug allergy syndrome should not be applied to the numerous patients who report adverse reactions to many drugs, in whom symptoms are often not typical of immunologic reactions. Rather, "multiple drug intolerance" may be a more appropriate description.
When multiple unrelated drugs elicit very similar symptoms, with immediate or slightly delayed onset, a role for conditioned responses needs to be considered. It is not immunologically plausible that structurally unrelated drugs should produce exactly the same syndrome, so when this is the history given by the patients (often with more than two drugs), classical Pavlovian conditioning is the most likely diagnosis, especially if the original reaction was frightening or traumatic in any way. Another clue to this diagnosis is a history of allergic-like symptoms to many drugs, including some that uncommonly cause true immunologic drug reactions (eg, macrolide antibiotics, glucocorticoids, and local anesthetics).
In conclusion, multiple drug allergy syndrome should only be used to describe patients with immunologic reactions to at least two (usually more) unrelated drugs because such patients often need special attention and management.
Recurrent drug exposure — Repeated courses of therapy with the same drugs or related drugs are associated with higher rates of drug allergy [10,11]. As an example, patients with cystic fibrosis who require recurrent courses of the same or related antibiotics are at increased risk for developing allergies to those drugs.
HLA type — There is evidence for a familial propensity to develop immunologic drug reactions [12]. Data on a variety of drugs and severe drug allergies demonstrated that certain human leukocyte antigen (HLA)-B alleles represent highly significant risk factors for severe side effects to a particular drug and are also involved in presenting the drug to the immune system [13-15].
●Stevens-Johnson syndrome/toxic epidermal necrolysis (SJS/TEN) caused by carbamazepine in Han Chinese shows a strong association with HLA-B*15:02 [13]. However, HLA-B*15:02 is not increased among Han Chinese patients with milder, maculopapular drug eruptions, nor is it associated with SJS/TEN in European Caucasians [16]. Instead, European carriers of the HLA-A*31:01 allele have an increased risk of developing DiHS/DRESS [17,18]. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Carbamazepine'.)
●SJS/TEN and possibly DiHS/DRESS caused by allopurinol also shows a strong association with HLA-B*5801 [19,20].
●A severe hypersensitivity syndrome is caused by the reverse-transcriptase inhibitor abacavir and is highly associated with HLA-B*57:01 in Caucasians [21,22]. Typing of HLA-B*57:01 reduced the incidence of side effects dramatically [23]. The same allele is also implicated in flucloxacillin-induced hepatitis but with a very low penetrance of the disease. Thus, typing is not recommended [24]. This syndrome is discussed in detail elsewhere. (See "Abacavir hypersensitivity reaction".)
There are multiple instances of associations between HLA alleles (mostly class I) and hypersensitivity reactions to specific medications [25]. Some associations have a high specificity (eg, dapsone hypersensitivity, HLA-B*13:01, and the dapsone hypersensitivity syndrome) [26], although for the majority, the odds ratio is rather low (<10), implying that besides HLA alleles, other still unknown factors play a crucial role. The pathophysiologic mechanisms that may underlie these associations are discussed separately. (See "Drug allergy: Pathogenesis", section on 'Type IV (T cell-mediated)'.)
Certain disease states — Some allergic drug reactions are much more likely to occur if the drug is administered to patients with particular underlying conditions.
●Reactions to aminopenicillins occur more commonly in patients with atypical or abnormal lymphocytes (eg, in massive immune stimulations, such as Epstein-Barr virus infection or leukemia) [27,28].
●Patients with acquired immunodeficiency syndrome (AIDS) have very high rates of dermatologic reactions to sulfonamides and other drugs [13,29-32]. HIV infection may create an inflammatory milieu that enhances drug allergy [4,33]. A similar phenomenon is seen in generalized herpes virus infections. Indeed, activated, atypical lymphocytes are frequently found in the circulation of persons infected with either of these viruses. Other contributing factors may include reduced hepatic glutathione levels due to polypharmacy and immunologic dysfunction associated with AIDS [31,34].
●A common problem is exanthematous skin reactions in small children treated with antibiotics, in particular penicillins. These mostly mild and transient skin reactions appear to be due to a combined effect of the drug use and various viral infections causing immune stimulation [35]. The vast majority of children will tolerate penicillins if given them again later.
Drug metabolism — It is unclear if genetic differences in drug metabolism are a risk factor for drug allergy. Although some initial studies suggested that slow acetylation or low glutathione levels might influence the frequency of drug hypersensitivity to sulfonamides and other drugs, later studies did not confirm these theories. Therefore, typing for pharmacogenetic predisposition (eg, cytochrome p450 polymorphism) has not entered clinical practice for the purposes of assessing risk of drug allergy.
Atopy — Atopy may be defined as an inherited phenotype that includes the predisposition to develop allergen-specific IgE to inhaled or ingested allergens. Patients with atopic/allergic disorders, such as allergic asthma or food allergy, are not at increased risk for the development of drug allergy. However, atopic conditions may aggravate the clinical manifestations of IgE-mediated drug allergies. As an example, positive penicillin skin tests do not occur more frequently in atopic individuals [36], but an atopic background is a substantial risk factor for severe and fatal penicillin anaphylaxis [37].
Age — Age is not known to be an independent risk factor for drug allergy, but data are confounded by very different patterns of drug use in various age groups. Pediatric drug reactions (in particular mild exanthems) appear to be transient (ie, not recurrent upon repeated exposure) more often than reactions that occur in adulthood.
SYSTEMATIC APPROACH TO THE PATIENT — A systematic approach to the evaluation of putative drug allergy in patients receiving multiple medications begins with the following questions:
●Is the presentation consistent with drug hypersensitivity? If so, what type? Are the signs and symptoms suggestive of an immunologic response (ie, drug allergy)? Is the drug to which the patient was recently exposed known to cause such symptoms or does it usually cause pseudoallergic reactions (eg, nonsteroidal anti-inflammatory drugs)? The different presentations of drug allergy and pseudoallergic reactions are summarized in the table and presented separately (table 1). (See "Drug hypersensitivity: Classification and clinical features".)
●How severe is the reaction, and what organ systems are involved? This can be assessed with a meticulous skin examination, followed by further evaluation of any other organ system that appears to be affected. A complete blood count and differential and tests of liver and renal function should be obtained. This evaluation is indicated if the exanthema is substantial, bullous or pustular, confluent, involves a substantial portion of the body surface area, or if general symptoms (malaise, skin pain, lymphadenopathy) are present.
●Which drug is responsible? This assessment can be approached in a stepwise manner:
•Step 1 – Gather information, as discussed in detail below. (See 'Clinical history' below and 'Review the medical record' below.)
•Step 2 – Examine the medical record to see if temporal relationships between symptoms and the administration of specific drugs are apparent. (See 'Review the medical record' below.)
•Step 3 – For each suspect drug, consider the likelihood of its causing the type of allergic reaction in question, based upon the published literature. (See 'Drugs with high allergenic potential' below.)
•Step 4 – Stop or substitute any/all suspect drugs with known allergic potential that demonstrate a temporal relationship with the symptoms. Observe the consequences of stopping these medications.
●Is testing possible?
•Step 5 – For reactions that are suspected to be IgE-mediated, consider skin testing (see 'Skin testing for drug-specific IgE' below). Both skin and in vitro tests (if available) to identify the eliciting drug are normally done after the reaction has resolved completely [38]. In vitro tests are available for some type II, III, and IV reactions.
•Step 6 – If there is ambiguity surrounding which drug induced an immunologic reaction and multiple drugs were stopped, plans should be made to continue the evaluation after the patient's convalescence, particularly for drugs that the individual may need in the future. Entering multiple "possible" drug allergies in the medical record will lead to confusion and is not in the patient's best interest.
IDENTIFICATION OF THE SUSPECT DRUG — The suspect drug is identified using the patient's clinical history of present and past drug reactions and review of the medical record for temporal associations between administration and symptoms, combined with an understanding of what types of drugs cause various reactions. If available, objective tests (eg, skin testing, in vitro testing) may also support the diagnosis.
Clinical history — The initial step involves determining all the medications to which the patient has had adverse reactions in the past. Additional history can sometimes be obtained from relatives or primary care providers. This information should be documented in the patient's permanent medical record.
The following specific questions should then be answered for each of the reported drugs in question:
●What were the symptoms and signs? What is known about the reaction?
●How long ago did the reaction take place?
●Did the patient require medical treatment or hospitalization because of the reaction?
●Why was the medication taken (indication for use)?
●The dose and route of medication taken (if known).
●Had the patient taken the medication on a previous occasion?
●Was the patient taking other concurrent medications when the reaction occurred? Were any of these newly started?
●The timing of onset of the reaction (from both the precipitating dose, as well as from the initiation of that course of therapy).
●Any treatment given and response to that treatment (including the duration of reaction).
●Has the patient received that medication or a related medication again since the reaction? If so, were there recurrent symptoms?
●Any prior or subsequent history of exposure to other agents in that class and presence or absence of any associated reaction(s).
Review the medical record — The medical record should be reviewed if possible. In critically ill patients who cannot communicate, it may be the only available source of information.
One approach is the creation of a timeline of medication starts and stops, on which the appearance and resolution of symptoms is also charted. It is important to note the onset of fever or the appearance of relevant laboratory abnormalities, such as blood eosinophilia or urine eosinophilia or hematuria.
Limitations of the clinical history — History alone is often not sufficient for establishing current drug sensitivity. Studies performed on large series of patients with histories of drug allergy have shown that less than 20 percent actually react to the offending drug(s) upon direct challenge [39,40]. There are multiple reasons why a previously reactive patient may tolerate a drug upon subsequent administration:
●The suspect drug may never have caused symptoms but was implicated based upon a temporal association. This is probably the most common reason.
●The patient could have had a true IgE-mediated reaction in the past but has lost sensitivity to the drug over time. Approximately 80 percent of patients with IgE-mediated penicillin allergy have lost the sensitivity after 10 years. (See "Penicillin allergy: Immediate reactions".)
●Biologic states required for reactivity may wane over time and render the patient less vulnerable. For example, aspirin and nonsteroidal anti-inflammatory drugs often aggravate chronic idiopathic urticaria but are well-tolerated once the urticaria remits.
●IgE-mediated, immediate-type allergic reactions can appear with the first known encounter with a drug, even though these types of reactions require pre-existing sensitization (ie, formation of IgE antibodies) to the drug. This most often occurs with drugs that are administered intravenously. The explanation for this phenomenon is that sensitization has occurred as a result of exposure to a different but cross-reacting compound, typically a different medication or a cosmetic. The best studied example is that of muscle relaxants (ie, neuromuscular-blocking agents), which can trigger anaphylaxis upon first exposure because patients can become sensitized to over-the-counter cough remedies (pholcodine in Europe) or cosmetics (in the United States) [41]. (See "Perioperative anaphylaxis: Clinical manifestations, etiology, and management", section on 'Neuromuscular-blocking agents'.)
Another example of individuals with preformed IgE molecules reacting upon first exposure to a drug was identified in patients with anaphylaxis to cetuximab [42]. Some of these patients had antibodies against the oligosaccharide galactose alpha-1,3-galactose found in the drug and developed anaphylaxis upon intravenous administration of cetuximab. This oligosaccharide is also found in beef, pork, and lamb, and a subset of patients had experienced allergic reactions when eating these meats. Because most of the affected patients lived in the southeastern United States, it has been postulated that patients become sensitized to the oligosaccharide as a result of bites from ticks or other ectoparasites [43].
Drugs with high allergenic potential — The different types of immunologic drug reactions and the categories of drugs that are most commonly implicated in each are summarized in the table (table 3).
OBJECTIVE TESTING — Objective testing may be appropriate for some types of allergic drug reactions. Skin testing for type I reactions is supported by extensive clinical experience but is not standardized or validated for most drugs. The majority of in vitro tests for drug allergy are investigational. In addition, immunodiagnostic tests for drug allergy can require experience to interpret properly.
Testing for immediate reactions — Type I, IgE-mediated reactions typically begin within one hour of the first administered dose, as discussed previously. (See 'Definition and classification' above.)
Markers of anaphylaxis — Elevations in tryptase (serum or plasma) may be detectable in blood samples collected within several hours of an acute allergic event because this mediator is released by mast cells and basophils (table 4). Histamine may also be detectable, but assays are less reliable, and elevations are seen only during a very short time interval after the reaction.
Any elevation of tryptase is consistent with anaphylaxis, although normal levels do not exclude the diagnosis. Tryptase levels should be tested again after the reaction has resolved to detect a relative increase as well as to exclude indolent mastocytosis, an important risk factor for severe anaphylaxis. Elevations are most likely to be detectable following anaphylaxis with hemodynamic changes and if blood is collected one to three hours from the onset of symptoms [44]. The interpretation of tryptase in the diagnosis of anaphylaxis is not always straightforward. As an example, a tryptase level of 9 ng/mL during anaphylaxis and 3 ng/mL the following day is suggestive of anaphylaxis, even though both levels are below the upper limit of normal (ie, 11.4 ng/mL). In this example, the serum level tripled, which may be meaningful, since personal tryptase levels are usually quite constant. The interpretation of tryptase levels is discussed in greater detail separately. (See "Laboratory tests to support the clinical diagnosis of anaphylaxis".)
Skin testing for drug-specific IgE — A variety of medications can be used to perform prick and/or intradermal skin testing for the purpose of determining if the drug interacts with drug-specific IgE bound to cutaneous mast cells. This type of testing is only used to evaluate suspected type I allergic reactions. A positive wheal-and-flare response appearing within 15 to 20 minutes indicates the presence of drug-specific IgE on the patient's mast cells and supports the diagnosis of a type I reaction (table 1).
The clinician must first ascertain that the medication does not cause direct mast cell degranulation, as described with opioids, quinolones, and vancomycin. Medications that activate mast cells directly cannot be studied with this type of skin testing. In addition, the drug solution used for testing should not be nonspecifically irritating to the skin. Nonirritating concentrations of several antibiotics have been determined (table 5) [45].
For most drugs, the full range of metabolites and intermediate forms of the drug to which patients may become allergic has not been defined, and testing reagents are not available. Thus, just the native (unmetabolized) form of the drug is used in testing, and this may only detect a fraction of allergic patients. Penicillin is the one exception to this, and the important metabolites and metabolite/protein complexes that are required to detect allergic patients have been characterized. Penicillin skin testing is discussed in detail separately. (See "Penicillin skin testing".)
Despite this limitation, medications for which skin testing for immediate reactions with the native (unmetabolized) form has proven useful in identifying a subset of allergic patients include the following:
●Other beta-lactam antibiotics (cephalosporins and imipenem). (See "Allergy evaluation for immediate penicillin allergy: Skin test-based diagnostic strategies and cross-reactivity with other beta-lactam antibiotics".)
●Neuromuscular blockers and dyes used to localize lymph nodes intraoperatively (ie, patent blue, isosulfan blue, methylene blue). (See "Perioperative anaphylaxis: Evaluation and prevention of recurrent reactions".)
●Carboplatin and other platin drugs. (See "Infusion reactions to systemic chemotherapy", section on 'Platinum drugs'.)
●Pyrazolones, such as metamizole [46].
●Local anesthetics. (See "Allergic reactions to local anesthetics".)
●Thiobarbiturates. (See "Perioperative anaphylaxis: Evaluation and prevention of recurrent reactions".)
●Therapeutic monoclonal antibodies, even if fully humanized, can still elicit immune reactions. The majority of reactions involve IgG antibodies, some neutralizing the efficacy of the therapeutic antibody. IgE-mediated reactions do occur but are rather rare. (See "Infusion-related reactions to therapeutic monoclonal antibodies used for cancer therapy".)
The results of skin testing with one of the above agents should be interpreted as follows:
●A positive result is indicative of allergy, provided nonirritating concentrations of the drug were used.
●A negative result does not exclude allergy, because the patient may be allergic to metabolites of the medication or metabolite/protein complexes, as explained previously.
Skin testing may be falsely negative if performed too soon after an anaphylactic reaction. For this reason, a period of two to four weeks should be allowed between a severe reaction and skin testing to the suspect drug. However, if skin testing is performed in this period, positive results are still valid. False-negative results can also occur in patients taking medications that blunt immediate skin responses, such as H1 and H2 antihistamines and tricyclic antidepressants. (See "Overview of skin testing for IgE-mediated allergic disease", section on 'Medications that should be discontinued'.)
In vitro tests — In vitro tests for immediate drug reactions are available, but these are largely considered investigational, as mentioned previously.
●Solid-phase immunoassays, such as the radioallergosorbent test and enzyme-linked immunosorbent assay, have been developed for a wide variety of drug-specific IgE, using sera from skin test-positive patients for standardization. These tests have been useful in research settings [47-49]. However, only for beta-lactam antibiotics and pyrazolone (analgesics not available in the United States) have in vitro test results been systematically compared with skin tests, with in vitro tests consistently proving less sensitive [46,50]. Many immunoassays for drug-specific IgE have been commercialized without published validation, and some of these are highly suspect since IgE-mediated reactions have never been documented.
●Flow cytometry assessing drug-induced basophil activation by means of increase in surface markers, such as CD63, has been studied in the diagnosis of immediate drug allergy [51-58]. Determination of cysteinyl leukotrienes released from blood leukocytes after drug incubation has been suggested to increase the diagnostic value of flow cytometry alone [59]. However, these tests are not standardized.
Testing for delayed reactions — A maculopapular rash appearing several days into a course of a systemic medication is the most common form of delayed-onset drug allergy [60,61]. Most of these reactions are mediated by T cells (type IV) (table 1). In contrast to IgE-mediated reactions, delayed drug reactions can recur after decades, even in the absence of interim reexposure [62].
Types II and II reactions — The diagnosis and evaluation of types II and III hypersensitivity reactions are discussed separately (table 1):
●Type II reactions (antibody [usually IgG]-mediated cell destruction) – (See "Drug-induced immune thrombocytopenia" and "Drug-induced neutropenia and agranulocytosis" and "Drug-induced hemolytic anemia".)
●Type III (immune complex deposition and complement activation) – (See "Serum sickness and serum sickness-like reactions" and "Overview of cutaneous small vessel vasculitis".)
Type IV reactions — The majority of drug reactions involve dermatitis and/or hepatitis, occasionally with systemic involvement (eg, drug-induced hypersensitivity syndrome/drug reaction with eosinophilia and systemic symptoms [DiHS/DRESS]). They are principally mediated by T cells and are classified as type IV reactions (table 1). Type IV reactions may be evaluated with various diagnostic tests, including patch testing, intradermal testing with delayed cutaneous readouts, or lymphocyte activation assays.
●Patch testing – Drugs to be used in patch testing are mixed into petrolatum or 0.9 percent saline, applied to a small area of skin under occlusion for 48 hours, and then removed. The site is examined 48 to 96 hours after placement. Concentrations of various drugs for use in patch testing have been established [63].
Patch testing with drug preparations may be useful in evaluating patients with maculopapular exanthema, acute-generalized exanthematous pustulosis, DiHS/DRESS, and flexural exanthema. Mild systemic reactions (eg, transiently appearing generalized exanthema) can occur even after patch (or intradermal) tests in patients with very high sensitivity (eg, patients with DiHS/DRESS). It is only occasionally positive if the delayed-appearing skin reaction was mainly macular (no or only moderate cutaneous cell infiltration) or urticarial or isolated hepatitis or nephritis. Patch testing to the culprit drug is less likely to be positive (9 to 23 percent) in patients with past blistering reactions, such as Stevens-Johnson syndrome and toxic epidermal necrolysis. It is not dangerous to perform such testing, contrary to previous belief [64,65]. Patch testing to evaluate DiHS/DRESS is reviewed separately. (See "Drug reaction with eosinophilia and systemic symptoms (DRESS)", section on 'Diagnostic approach'.)
●Intradermal testing with delayed readout – This type of testing should only be performed if a commercially available injectable form of the drug is available. The concentration used should be known to be nonirritating, and a prick test should be performed initially to assure there is no immediate response [45]. A positive result consists of erythema and induration at the site, and the site is examined at 24 to 48 hours after placement [63].
Intradermal testing with delayed readout is slightly more sensitive than patch testing but somewhat less specific [61]. It has the same indications as patch testing and may be used in evaluating exanthems, acute-generalized exanthematous pustulosis, DiHS/DRESS, fixed drug eruption, and symmetrical drug-related intertriginous and flexural exanthema. It has low sensitivity for blistering skin reactions, as well as for isolated macular or urticarial skin eruptions. Some patients with DiHS/DRESS may show a slight reappearance of symptoms, even if the dose was minimal.
Only a few preparations of drugs for skin testing are commercially available, so preparations are often made by individual clinicians or centers. This makes it difficult to standardize the procedure or compare results. Overall, patch and intradermal testing with delayed readout appear to have good drug-related specificity, although they can be positive even when recent drug treatment did not cause a reaction. In addition, more work is needed to increase their sensitivity [60,66-72].
●In vitro tests for delayed reactions – In vitro tests for delayed reactions include lymphocyte transformation/activation tests, upregulation of activation markers on T cells (eg, CD69), cytokine production, and drug-induced cytotoxicity assays [73]. These tests are done in certain centers only, need to be done with a preparation of pure drug, and are considered research tools. If available, such tests may help in diagnosis, as they can be positive when skin tests are negative. Their specificity for most drug classes seems to be very good (mostly >95 percent), but the sensitivity needs to be improved [70].
Where available, in vitro testing may be useful in evaluating patients with maculopapular exanthema, acute-generalized exanthematous pustulosis, DiHS/DRESS, and flexural exanthema. The sensitivity of tests based on cell expansion, cell activation, or cytokine release is high in DiHS/DRESS but low in blistering reactions. The role of tests evaluating drug-induced cytotoxic reactions needs to be further evaluated and may be particularly interesting for bullous skin reactions [74], as well as for isolated drug-induced hepatitis or nephritis, where skin tests are often negative.
GRADED CHALLENGE — Graded challenge (also called drug provocation testing) involves administering a medication to a patient in a graduated manner under close observation. It is based on the principle that a certain amount of the medication is needed to elicit symptoms.
Indications — Graded challenge is used to exclude allergy to the medication in question, and is most appropriate for a patient who is unlikely to be allergic to that drug. Graded challenge should not be performed in a patient with a positive response in a prior drug allergy test (skin or in vitro).
Graded challenge does not modify the allergic response to the drug or prevent recurrent reactions. Therefore:
●Patients who tolerate a drug upon graded challenge prove that they are not allergic to the drug given at the dose used.
●A challenge procedure in a patient with a suspected IgE-mediated drug allergy could potentially induce anaphylaxis and should be performed by an allergy expert in a setting equipped to manage possible reactions.
Contraindications — Graded challenge to the suspect drug is contraindicated in patients with the following types of reactions:
●Blistering dermatitis (eg, Stevens-Johnson syndrome, toxic epidermal necrolysis)
●Sloughing of the skin
●Severe generalized hypersensitivity reactions involving internal organs (drug-induced hypersensitivity syndrome/drug reaction with eosinophilia and systemic symptoms)
●Milder dermatoses with mucous membrane lesions (eg, erythema multiforme)
Even minute amounts of the suspect drug can reactivate these reactions, and the response can then escalate even though no further drug is given [75]. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Management, prognosis, and long-term sequelae".)
Inclusion of placebo doses — Inclusion of a placebo is strongly recommended to exclude false-positive reactions in patients with anxiety or other subjective reactions. Some patients experience unintentional anticipatory reactions (classic conditioning), particularly following a true allergic reaction or a frightening experience in association with drug treatment. These can often be elicited by placebo doses, thus reassuring the clinician and patient alike that the response is not due to the drug in question.
Challenge protocols — The pace of the challenge and degree of caution exercised depends upon the likelihood that the patient may be allergic, the patient's medical stability, and the clinician's experience and level of comfort with the procedure.
Other considerations include:
●Patients should not be premedicated with antihistamines or glucocorticoids, because these agents may mask early signs of an allergic reaction.
●Beta-blockers taken for control of hypertension should be withheld for 24 hours before challenge, as these agents can interfere with treatment of anaphylaxis with epinephrine, should that be necessary. In contrast, beta-blockers that are taken to control arrhythmias should not be withheld without consulting a cardiologist.
●Patients with asthma, chronic obstructive lung disease, or other pulmonary diseases should have their pulmonary symptoms optimally controlled prior to undergoing challenge.
For immediate reactions — If the previous reaction occurred less than one hour after drug administration and there is concern about a possible IgE-dependent reaction, then the challenge should be performed by an allergy expert and in a setting equipped to treat anaphylaxis.
●The starting challenge dose should be low, typically between 1/10,000 and 1/1000 of the therapeutic dose, depending on the severity of the past reaction. Using this strategy, any symptoms that are elicited should be mild.
●The oral route is preferred when possible, since oral administration is generally associated with less severe symptoms compared with intravenous administration.
●Tenfold increasing doses are administered every 30 to 60 minutes until the full therapeutic dose is reached. If subjective symptoms appear, the clinician can proceed more slowly by using threefold increases for one or two steps. Once the full therapeutic dose has been achieved without incident, continuous therapy should be begun immediately with appropriate monitoring.
For delayed reactions — Challenge schemes for simple delayed exanthema to drugs are usually performed either in the context of research (to document sensitivity) or at a time that the patient actively needs the drug. In the latter situation, the drug may be slowly increased and the patient instructed to contact the clinician immediately if any adverse events are noted.
For delayed or nonimmediate types of reactions, challenge procedures are less standardized and are based upon the time course of the patient's reaction and the pharmacology of the drug involved [76]. The optimal approach is not known, and experts have differences in practice. In general, the time between doses should be long enough that delayed symptoms have time to develop before the next higher dose is administered. Some protocols may take days or even weeks.
An example of a challenge procedure for delayed reactions to cephalosporins involved administering doses at weekly intervals, starting with 1/100 of a usual dose on day 0, 1/10 of a dose on day 7, and a standard dose on day 14 [77]. However, other studies have documented that some delayed reactions only develop after several days at a full therapeutic dose or in the presence of a concomitant viral infection [78,79]. Thus, there may be certain drug reactions that cannot be easily elicited with any challenge protocol [80].
Other published examples of challenge protocols, when available, are discussed in specific topic reviews. (See "Sulfonamide allergy in HIV-uninfected patients", section on 'Desensitization'.)
Negative predictive value of challenge procedures — If a patient tolerates a drug challenge, it is highly unlikely but not impossible that that drug will cause symptoms if taken again in the future. A small number of studies have examined the negative predictive value of challenges for immediate and delayed reactions:
●A study examined the negative predictive value of drug challenges in 203 children [81]. Of these, 163 of the children underwent allergy evaluation for immediate or delayed reactions. All were labeled as "not allergic to the drug" based on 175 negative drug challenges, consisting of one full dose of the drug. Re-evaluation after at least a three-month interval showed that 91 (52 percent) of the 175 tested drugs were reported to have been used again: 65 children took the same antibiotic, 20 took the same nonsteroidal anti-inflammatory drug (NSAID), and 6 took other drugs. An immediate reaction was reported in 5 (5.5 percent) of these (four antibiotics, one NSAID). One patient developed an anaphylactic reaction and was skin test-positive upon repeat testing, suggesting either resensitization from the challenge or an error in original skin testing. The remaining false-negative challenges were delayed reactions.
●Other studies have described similar findings. Recurrent reactions in spite of negative prior challenges were reported in 7.6 percent of 118 adults re-exposed to the same beta-lactam antibiotic. All reactions were nonimmediate, and none were severe [82].
Such studies demonstrate that single dose challenge procedures have higher negative predictive values for immediate reactions than for delayed-onset reactions. Challenges may rarely result in resensitization, and thus, a prior negative challenge is not a guarantee that the drug will be tolerated, especially regarding delayed reactions. In addition, patients with a recent anaphylactic drug reaction, even if allergy testing and challenge are negative, should be retreated only under medical surveillance.
OPTIONS FOR FUTURE TREATMENT — There are three options for future treatment in patients with a confirmed drug allergy:
●Administration of an unrelated medication
●Careful administration of a related medication
●Desensitization to the culprit drug
Administration of an unrelated medication — The most straightforward option is administration of an unrelated medication that is safe and effective for the disorder in question. However, second-line therapies may confer their own risks, such as toxicities and higher costs.
Penicillin allergy illustrates the issues involved. Patients labeled penicillin-allergic often receive non-beta-lactam antibiotics, which can be more expensive, associated with side effects, and in some cases, less efficacious [83-86]. Specifically, patients carrying the diagnosis of penicillin allergy are more likely to be treated with vancomycin or quinolones [83,87-91]. The use of these broad-spectrum antibiotics contributes to the development and spread of drug-resistant bacteria [92-94]. One study assessed various risk factors for the development of vancomycin-resistant enterococcus (VRE) in a medical intensive care unit (MICU) [95]. Among pre-MICU treatment with various classes of antibiotics, quinolones had the strongest association with subsequent development of VRE (odds ratio = 8.6), whereas treatment with penicillins/beta-lactamase inhibitors was not associated with developing VRE [95]. Some broad-spectrum antibiotics are also associated with Clostridium difficile infection. Thus, use of alternative, unrelated medications is not without risk.
Administration of a related medication — The second alternative for drug-allergic patients is to receive a medication similar but not identical to the offending drug.
The likelihood of cross-reactivity among similar drugs is partly dependent on the type of allergic reaction in question. As an example, a patient with a T cell-mediated exanthema to amoxicillin is at low risk for reacting to a cephalosporin, whereas a patient with IgE-mediated anaphylaxis to amoxicillin is at some increased risk if given a cephalosporin. The related medication may be one for which skin testing is informative or it can be initially given via graded challenge. Cross-reactivity within classes of medications is discussed in specific topic reviews. (See "Allergy evaluation for immediate penicillin allergy: Skin test-based diagnostic strategies and cross-reactivity with other beta-lactam antibiotics" and "Infusion reactions to systemic chemotherapy".)
Desensitization to the culprit drug — The third option is desensitization to the culprit drug. Classical desensitization is predictably successful for type I, IgE-mediated allergy, although gradual dose escalation may be attempted with variable success to several other types of immunologic and nonimmunologic drug reactions. Drug desensitization is reviewed in detail separately. (See "Rapid drug desensitization for immediate hypersensitivity reactions".)
LONG-TERM MANAGEMENT OF PATIENTS PRONE TO DRUG ALLERGY — Once a drug allergy has been identified, the affected patient should be educated about avoidance and provided with a written list of the generic and brand names of the culprit drug, as well as possibly cross-reactive drugs. Patients with potentially severe reactions should carry wallet cards, wear identification jewelry, or register with a drug allergy information service.
As discussed previously, some individuals are vulnerable to allergic drug reactions as a result of genetic or metabolic abnormalities (eg, multiple drug allergy syndrome), frequent and recurrent drug exposure (eg, antibiotics in cystic fibrosis), or certain disease states related to immune dysfunction (eg, HIV infection). In a drug allergy-prone population, pharmacotherapy requires special management to avoid adverse events and to prevent (to the degree possible) sensitization or resensitization.
Limit antibiotic use — Efforts should be made to limit unnecessary exposure to antibiotics. These include the diagnosis and treatment of underlying conditions that predispose to infections. As examples, rhinitis and asthma should be aggressively treated with glucocorticoid nasal sprays and inhalers to minimize sinopulmonary infections, and diabetes mellitus should be tightly controlled. Less commonly, undiagnosed immunodeficiency is present.
Age-appropriate vaccinations should be administered and repeated at the recommended intervals. Avoidance of unnecessary exposures to contagious diseases may be helpful, especially in children and patients with immunodeficiency.
Proactive efforts to obtain culture data before initiating antibiotics can help to limit cumulative antibiotic exposure by avoiding treatment of viral infections and minimizing mid-treatment changes in antibiotic regimens. Enlisting the cooperation of the patient's other providers is important.
Consideration should be given to treating through mild cutaneous reactions, such as itching or mild rashes, especially early in therapy when an alternative antibiotic will usually be required. Treating through is possible in delayed-onset exanthema. This is typically considered when alternative treatments are problematic, and an allergy specialist considers continued exposure to the drug to be low risk. Ongoing vigilance for clinical and laboratory danger signs is necessary, and concomitant use of glucocorticoids (prednisone 0.5 mg per kg) may be required if the exanthem intensifies rather than improves.
Approach to initiating new drugs — In drug allergy prone patients, new drugs should be started at lower than normal doses when feasible. Administration under medical observation is sometimes necessary.
Patients with past Stevens-Johnson syndrome/toxic epidermal necrolysis and drug-induced hypersensitivity syndrome/drug reaction with eosinophilia and systemic symptoms (DiHS/DRESS) should be taught about the earliest symptoms of these reactions, such as fever or mucosal irritation, and know to stop medications immediately should these appear. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis", section on 'Prodrome' and 'Long-term management of patients prone to drug allergy' above.)
Many patients with prior DiHS/DRESS experience a period of drug intolerance during the weeks after the initial reaction, meaning that any new drug may exacerbate symptoms (exanthems, hepatitis), often without proof of a new sensitization. The symptoms may lead to the suspicion of a new drug allergy and further limit treatment. However, when the immune system is less activated (documented by circulating activated lymphocytes), the new drugs are again tolerated. This phenomenon has been termed a "flare up" reaction [96]. However, some patients with prior DiHS/DRESS do develop genuine sensitizations to new compounds, and some develop multiple drug hypersensitivity [96].
Patients with past IgE-mediated reactions to drugs should not be given premedications before gradual dose escalation of new drugs. Neither antihistamines nor glucocorticoids can prevent IgE-mediated anaphylaxis, and premedications may mask early symptoms and allow dosing to proceed more rapidly than advisable.
In contrast to the above scenarios, patients who have reacted with uncomplicated exanthemas, including mild urticaria, to a variety of different drugs may benefit from administration of an antihistamine when new drugs are initiated.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Drug allergy and hypersensitivity".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Drug allergy (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Risk factors for developing drug allergies include past drug allergy, genetic factors, recurrent drug exposure, and certain diseases (eg, HIV/acquired immunodeficiency syndrome). Individuals with two or more immunologic drug reactions to chemically unrelated medications are said to have "multiple drug allergy syndrome." (See 'Risk factors for drug allergy' above.)
●Patients who develop signs/symptoms of drug allergy while receiving multiple medications simultaneously are best approached in a systematic manner. Evaluation involves a meticulous history of past and present drug reactions, additional information gathering from the medical record, and analysis of temporal patterns between drug administration and onset of symptoms. Once this information has been assembled, it is combined with knowledge about the types of allergic reactions most often caused by various classes of drugs to identify potential culprit agents. (See 'Systematic approach to the patient' above and 'Clinical history' above and 'Review the medical record' above.)
●Overall, objective testing for the diagnosis of drug allergy is limited. Skin testing is helpful for some medications that cause type I, immediate hypersensitivity reactions, although skin testing has been fully validated only for penicillin. In some cases, types II, III, and IV reactions can be evaluated with in vitro studies, although the majority of these tests are still investigational. (See 'Objective testing' above.)
●Allergy to a drug can be excluded by performing a graded challenge. Graded challenge does not modify the allergic response to the drug or prevent recurrent reactions. Thus, challenges must be performed with caution and in a monitored setting, and patients who tolerate drug challenges prove that they are not allergic to that medication. (See 'Graded challenge' above.)
●There are three options for future treatment in patients with a confirmed drug allergy: administration of an unrelated medication, careful administration of a related medication, and desensitization to the culprit drug. (See 'Options for future treatment' above.)
