Residency · Residency · Allergy Immunology

Drug Allergy and Hypersensitivity Reactions

Epidemiology and Classification

Drug allergy is reported by approximately 10 percent of the general population, yet rigorous evaluation confirms true drug hypersensitivity in only 10 to 20 percent of those carrying such a label. This enormous discrepancy between reported and confirmed drug allergy underscores the critical importance of systematic evaluation and de-labeling. True drug hypersensitivity reactions account for approximately 5 to 10 percent of all adverse drug reactions (ADRs).

Adverse drug reactions are broadly classified into two categories. Type A (augmented) reactions are dose-dependent, predictable, and related to the known pharmacologic action of the drug, comprising approximately 80 percent of all ADRs. Type B (bizarre) reactions are dose-independent, unpredictable, and unrelated to the drug's pharmacologic mechanism; this category encompasses true drug hypersensitivity reactions.

Drug hypersensitivity is further classified according to the Gell and Coombs system, which has been updated to incorporate modern immunologic understanding. Type I (IgE-mediated) reactions develop within minutes to one hour and manifest as urticaria, angioedema, or anaphylaxis. Type II (cytotoxic) reactions develop over hours to days and include entities such as drug-induced hemolytic anemia (as seen with penicillin or methyldopa) and drug-induced thrombocytopenia (as seen with heparin or quinine). Type III (immune complex-mediated) reactions develop over 1 to 3 weeks and include serum sickness, vasculitis, and drug fever. Type IV (T cell-mediated) reactions have been further refined by Pichler's subclassification: Type IVa reactions involve Th1 cells and macrophages, producing contact dermatitis; Type IVb reactions involve Th2 cells and eosinophils, producing DRESS and maculopapular exanthems; Type IVc reactions involve cytotoxic T lymphocytes causing keratinocyte death, as seen in Stevens-Johnson syndrome and toxic epidermal necrolysis; and Type IVd reactions involve neutrophilic infiltration, as seen in acute generalized exanthematous pustulosis.

Immediate (Type I) Drug Reactions

Clinical Features

Immediate drug hypersensitivity reactions develop within one hour of drug administration, though onset may be delayed up to 6 hours in some cases. Clinical manifestations range from urticaria and angioedema to bronchospasm, hypotension, and full anaphylaxis. The most commonly implicated drug classes include beta-lactam antibiotics, nonsteroidal anti-inflammatory drugs, neuromuscular blocking agents, platinum-based chemotherapy agents, and biologic therapies. IgE-dependent reactions require prior sensitization or the presence of cross-reactive IgE antibodies. Non-IgE-mediated immediate reactions, which do not require prior sensitization, include reactions to radiocontrast media, vancomycin (mediated through the MRGPRX2 receptor), and opioids (via direct mast cell activation).

Diagnosis

The diagnostic workup for immediate drug hypersensitivity begins with skin testing, employing a sequential approach of prick/puncture testing followed by intradermal testing with the suspected drug or its metabolites. Validated skin testing protocols exist for penicillin (discussed in detail in the penicillin allergy lecture), local anesthetics, and platinum-based agents. Establishing the non-irritating concentration (NIC) for each drug tested is essential to avoid false-positive irritant reactions. The sensitivity of skin testing is inherently limited for drugs whose allergenic determinants have not been fully characterized.

Serum-specific IgE assays are available for a limited number of drugs, including penicilloyl V and G, amoxicilloyl, ampicilloyl, cefaclor, insulin, chlorhexidine, suxamethonium, gelatin, and latex. The basophil activation test (BAT), which measures upregulation of CD63 and CD203c on basophil surfaces following drug exposure, is a promising research tool with particular utility for neuromuscular blocking agents, beta-lactams, and quinolones, though it is not yet widely available in clinical practice. The drug provocation test (DPT), a graded challenge in which the drug is administered in incrementally increasing doses under observation, remains the gold standard for confirming or excluding immediate drug hypersensitivity when skin testing is negative.

Management

Acute management of immediate drug reactions follows standard anaphylaxis protocols, with epinephrine as the first-line intervention supplemented by antihistamines and corticosteroids as appropriate. The offending drug and known cross-reactive agents should be avoided. When no acceptable therapeutic alternative exists, drug desensitization may be undertaken. Desensitization involves the gradual administration of incrementally increasing doses of the drug over a period of hours, inducing a state of temporary tolerance. The underlying mechanism involves progressive depletion of mast cell and basophil mediators without triggering full degranulation. Established indications for drug desensitization include aspirin in aspirin-exacerbated respiratory disease, platinum-based and taxane chemotherapy agents, penicillin and sulfonamide antibiotics, and biologic therapies. Crucially, the tolerance state is maintained only through continuous dosing and is lost within 24 to 48 hours of drug discontinuation. Desensitization must be performed in a monitored setting such as an intensive care unit or procedure suite.

<image>A Gell and Coombs classification diagram adapted for drug hypersensitivity reactions. Four main panels arranged in a grid: Type I (top left): mast cell with IgE crosslinked by drug-hapten, showing urticaria and anaphylaxis timeline (<1 hour). Type II (top right): drug bound to cell membrane with IgG/complement-mediated cytolysis, showing examples of drug-induced hemolytic anemia and thrombocytopenia. Type III (bottom left): immune complex deposition in vessel walls, showing serum sickness and vasculitis timeline (1-3 weeks). Type IV (bottom right): subdivided into IVa (macrophage granuloma), IVb (eosinophilic infiltrate/DRESS), IVc (cytotoxic T cell killing keratinocyte/SJS-TEN), IVd (neutrophilic pustules/AGEP). Each panel includes: mechanism, timeline, key effector cell, clinical example, and diagnostic test.</image>

Delayed (Type IV) Drug Reactions

Maculopapular Exanthem (MPE)

Maculopapular exanthem is the most common drug eruption, accounting for approximately 90 percent of drug-related rashes. It typically develops 5 to 14 days after initiation of the offending drug, though the latency may be shorter in previously sensitized individuals. The rash follows a morbilliform pattern, usually beginning on the trunk and spreading centrifugally to the extremities, and is characteristically pruritic. The eruption is generally self-limited, resolving upon drug discontinuation. The most commonly implicated agents include aminopenicillins (with the notable caveat that the association between ampicillin/amoxicillin and rash during acute Epstein-Barr virus infection represents a transient pharmacologic interaction rather than true drug allergy), sulfonamides, allopurinol, and anticonvulsants. An important clinical imperative is to differentiate simple maculopapular exanthem from early DRESS syndrome by checking the complete blood count with differential for eosinophilia, liver function tests, and renal function at the time of presentation.

Drug Reaction with Eosinophilia and Systemic Symptoms (DRESS/DiHS)

DRESS syndrome is distinguished by the longest latency period of any drug eruption, with onset typically 2 to 8 weeks after drug initiation. The clinical presentation includes high fever, a diffuse maculopapular rash that may become edematous or purpuric, prominent facial edema (a hallmark feature), and generalized lymphadenopathy. Laboratory abnormalities include eosinophilia (which may be absent in the initial stages), atypical lymphocytosis, and evidence of visceral organ involvement, with hepatic involvement being the most common (manifesting as elevated liver enzymes), followed by renal dysfunction and thyroiditis (which characteristically appears late). Human herpesvirus 6 (HHV-6) reactivation is characteristic of DRESS and is detectable by PCR in approximately 60 percent of cases. The RegiSCAR scoring system provides a standardized diagnostic framework, with a score of 5 or more points establishing a definite diagnosis of DRESS. Mortality ranges from 5 to 10 percent.

The most commonly implicated drugs include allopurinol (the most common cause worldwide), the aromatic anticonvulsants carbamazepine, phenytoin, and lamotrigine, and antimicrobials such as trimethoprim-sulfamethoxazole, dapsone, minocycline, and vancomycin. Important pharmacogenomic associations have been identified linking specific HLA alleles to susceptibility. HLA-B58:01 is associated with allopurinol-induced DRESS and SJS/TEN, and pre-prescription screening is recommended in Southeast Asian, African American, and Korean populations, with the American College of Rheumatology now recommending screening in all patients. HLA-A31:01 is associated with carbamazepine-induced DRESS in European and Japanese populations, while HLA-B*15:02 is associated with carbamazepine-induced SJS/TEN in Southeast Asian populations.

Treatment requires immediate drug discontinuation and systemic corticosteroids, typically prednisone at 1 mg/kg with a very slow taper over 3 to 6 months to prevent relapse. Patients must be monitored for autoimmune sequelae, particularly autoimmune thyroiditis and type 1 diabetes mellitus, for at least 1 year following the acute episode.

Stevens-Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis (TEN)

Stevens-Johnson syndrome and toxic epidermal necrolysis represent a continuum of severity defined by the degree of epidermal detachment: SJS involves less than 10 percent body surface area (BSA), SJS/TEN overlap involves 10 to 30 percent BSA, and TEN involves greater than 30 percent BSA. The onset is typically 1 to 3 weeks (4 to 28 days) after drug initiation, preceded by a prodrome of fever, malaise, and upper respiratory tract symptoms. Mucosal involvement affecting oral, ocular, and genital surfaces occurs in more than 90 percent of cases, and ocular involvement may result in permanent blindness. The Nikolsky sign, in which lateral pressure on the skin produces epidermal separation, is characteristically positive.

The pathogenesis of SJS/TEN is mediated by cytotoxic T lymphocytes and natural killer cells, which induce widespread keratinocyte apoptosis. Granulysin has been identified as the most important cytotoxic mediator, with Fas/FasL signaling and the perforin/granzyme pathway also contributing. Prognosis is assessed using the SCORTEN scoring system, which incorporates age, presence of malignancy, BSA involvement, heart rate, blood urea nitrogen, glucose, and serum bicarbonate; a score of 5 or more points carries approximately 90 percent mortality. Overall mortality rates are 1 to 5 percent for SJS and 25 to 35 percent for TEN.

The most commonly implicated drugs include allopurinol, carbamazepine, phenytoin, lamotrigine, sulfonamides, nevirapine, and oxicam NSAIDs. Treatment centers on immediate drug discontinuation (the single most important intervention), transfer to a burn unit or intensive care unit for supportive care and wound management, and consideration of specific immunomodulatory therapy. Cyclosporine A at 3 to 5 mg/kg/day has some retrospective evidence of benefit. Intravenous immunoglobulin (IVIG) has been proposed to block Fas/FasL-mediated apoptosis but has yielded mixed results. Etanercept, an anti-TNF biologic, demonstrated benefit in a single randomized controlled trial. The role of systemic corticosteroids remains controversial, as they may increase infection risk, though some centers employ a short course early in the disease course. Pharmacogenomic screening with HLA-B15:02 testing before prescribing carbamazepine in at-risk populations is mandated by an FDA boxed warning, and HLA-B58:01 testing before allopurinol is increasingly recommended.

Acute Generalized Exanthematous Pustulosis (AGEP)

Acute generalized exanthematous pustulosis is distinguished from other severe cutaneous adverse reactions by its characteristically rapid onset, typically within 48 hours of drug exposure. The eruption consists of hundreds of small, non-follicular, sterile pustules arising on an erythematous base, classically beginning in the flexural areas. Patients typically present with fever and neutrophilia, and mild organ involvement may occur. Mortality is less than 5 percent, and the condition is generally self-limited, resolving within 2 weeks of drug discontinuation. The most commonly implicated agents include aminopenicillins, macrolides, quinolones, hydroxychloroquine, and diltiazem. The underlying mechanism is classified as Type IVd, with drug-specific T cells producing interleukin-8 (CXCL8) that recruits a predominantly neutrophilic infiltrate.

<image>A clinical comparison chart of severe cutaneous adverse reactions (SCARs). Four columns for MPE, DRESS, SJS/TEN, and AGEP. Each column contains: representative clinical image description (morbilliform rash, diffuse edematous eruption with facial swelling, skin detachment with Nikolsky sign, pinpoint non-follicular pustules), onset timeline, key distinguishing features, characteristic lab findings (eosinophilia for DRESS, neutrophilia for AGEP), histopathology findings, mortality rate, and first-line treatment. A timeline bar at the bottom shows typical latency for each: MPE 5-14 days, DRESS 2-8 weeks, SJS/TEN 1-3 weeks, AGEP <48 hours. HLA associations listed where applicable.</image>

NSAID Hypersensitivity

Classification (ENDA/EAACI)

NSAID hypersensitivity reactions are classified into cross-reactive and selective subtypes based on their underlying mechanism. Cross-reactive reactions are driven by pharmacologic COX-1 inhibition rather than immunologic sensitization and occur across structurally unrelated NSAIDs. Three cross-reactive phenotypes are recognized: NSAID-exacerbated respiratory disease (NERD, also known as AERD), characterized by the triad of chronic rhinosinusitis with nasal polyps, asthma, and NSAID-induced respiratory reactions; NSAID-exacerbated cutaneous disease (NECD), in which NSAIDs exacerbate urticaria and angioedema in patients with underlying chronic spontaneous urticaria; and NSAID-induced urticaria/angioedema (NIUA), occurring in patients without pre-existing chronic urticaria.

Selective reactions are immunologically mediated against a single specific NSAID and do not extend to structurally unrelated agents. These include single NSAID-induced urticaria/angioedema/anaphylaxis (SNIUAA), which is IgE-mediated to a specific compound (frequently pyrazolones or propionic acid derivatives), and single NSAID-induced delayed reactions (SNIDR), which are T cell-mediated and may manifest as SJS/TEN, DRESS, or fixed drug eruption.

Management of cross-reactive NSAID reactions requires avoidance of all COX-1 inhibitors, while COX-2 selective inhibitors such as celecoxib and etoricoxib are generally tolerated, though this should be confirmed through an oral provocation challenge. Aspirin desensitization is available for patients with AERD who require daily aspirin therapy.

Drug Provocation Testing (DPT) and Graded Challenges

Indications

Drug provocation testing is indicated when the pre-test probability of true drug allergy is low based on clinical history, when available skin testing is negative, or when definitive confirmation or exclusion of drug hypersensitivity is needed to guide future prescribing. It is important to distinguish drug provocation testing from drug desensitization: the former confirms the absence of allergy in a patient who is not truly allergic, while the latter induces temporary tolerance in a patient with confirmed allergy.

Protocols

A graded drug challenge typically involves a 2 to 3 step dose escalation, beginning with one-tenth of the therapeutic dose followed by the full dose, with 30 to 60 minutes between doses. For suspected delayed reactions, a multi-day provocation protocol may be employed, in which the drug is administered at full therapeutic doses over several days to unmask delayed-onset reactions. All provocation testing must be performed in a monitored setting with resuscitation equipment immediately available. Drug provocation testing is absolutely contraindicated in patients with a history of SJS/TEN, DRESS, AGEP, vasculitis, or organ-specific drug reactions such as drug-induced hepatitis or interstitial nephritis.

Key Clinical Pearls

• Only 10-20% of patients labeled "drug allergic" have a confirmed allergy upon evaluation - drug allergy de-labeling is a critical practice
• DRESS has the longest latency (2-8 weeks) and can cause autoimmune sequelae (thyroiditis, T1DM) months after resolution
• AGEP has the shortest latency (<48 hours) and is distinguished by sterile non-follicular pustules and neutrophilia
• SJS/TEN mortality is driven by BSA involvement; use SCORTEN for prognosis; granulysin is the key cytotoxic mediator
• HLA screening before prescribing: HLA-B15:02 for carbamazepine (SJS/TEN risk in Southeast Asians), HLA-B58:01 for allopurinol (all patients per ACR)
• NSAID cross-reactive reactions are driven by COX-1 inhibition, not immune mechanisms; COX-2 selective inhibitors are generally safe alternatives
• Drug desensitization induces temporary tolerance maintained only with continuous dosing; it is distinct from a graded challenge which confirms absence of allergy

References

1. Demoly P, et al. International consensus on drug allergy. Allergy. 2014;69(4):420-437.
2. Blumenthal KG, et al. Approaches to the diagnosis of drug hypersensitivity reactions. J Allergy Clin Immunol Pract. 2023;11(1):65-76.
3. Chung WH, et al. Granulysin is a key mediator for disseminated keratinocyte death in Stevens-Johnson syndrome and toxic epidermal necrolysis. Nat Med. 2008;14(12):1343-1350.
4. Kardaun SH, et al. Drug reaction with eosinophilia and systemic symptoms (DRESS): an original multisystem adverse drug reaction. Results from the prospective RegiSCAR study. Br J Dermatol. 2013;169(5):1071-1080.
5. Kowalski ML, et al. Classification and practical approach to the diagnosis and management of hypersensitivity to nonsteroidal anti-inflammatory drugs. Allergy. 2013;68(10):1219-1232.

Read this lecture as Markdown