Drug eruptions: approaching the diagnosis of drug-induced skin diseases

Journal of Drugs in Dermatology, June, 2003 by Simon Nigen, Sandra R. Knowles, Neil H. Shear

Abstract

Adverse drug reactions are a major problem in drug therapy, and cutaneous drug reactions account for a large proportion of all adverse drug reactions. Cutaneous drug reactions are also a challenging diagnostic problem since they can mimic a large variety of skin diseases, including viral exanthema, collagen vascular disease, neoplasia, bacterial infection, psoriasis, and autoimmune blistering disease, among others. Furthermore, determining that a particular medication caused an eruption is often difficult when the patient is taking multiple drugs.

In this review, we will describe and illustrate a thoughtful, comprehensive, and clinical approach to the diagnosis and management of adverse cutaneous drug reactions. A morphologic approach to drug eruption includes those that are classified as maculopapular, urticarial, blistering or pustular with or without systemic manifestations. Exanthematous drug eruptions, drug hypersensitivity syndrome, urticaria and angioedema, serum sickness-like reactions, fixed drug eruptions, drug-induced autoimmune blistering diseases, Stevens-Johnson syndrome, toxic epidermal necrolysis, drug-induced acne, acute generalized exanthematous pustulosis, lichenoid drug eruptions and photosensitivity eruptions will be discussed.

Introduction

Adverse drug reactions are a common complication in drug therapy. 3-8% of all hospital admissions are the result of adverse drug reactions (Black 1984, Lakshmanan 1986, Moore 1998), and these can cause significant disability to patients (Lazarou 1998). A recent study estimated that between 5 to 9% of hospital costs were related to adverse drug reactions (Moore 1998). Skin rashes represent an important proportion of adverse drug reactions. In a study examining the incidence of skin eruptions, approximately 45% of all adverse drug reactions were manifested in the skin (Naldi 1999). Cutaneous drug reactions occur in 2 to 3% of all hospitalized patients (Bigby 1986, Stewart 1979). Furthermore, the development of a skin eruption is frequently cited as a reason for discontinuation of treatment before the completion of a therapeutic course (Svensson 2000). Drug-induced diseases represent an important proportion of the disease burden in the general population and on the health care system, and cutaneous drug eruptions account for a large proportion of them.

Risk factors

Certain patient groups seem to be at increased risk of developing a cutaneous drug reaction. The incidence of developing a cutaneous reaction increases with the number of drugs taken (Bigby 2001). In addition, some drug interactions may contribute to the development of a skin eruption (Li 1996). The age of the patient also correlates with an increased risk of a cutaneous drug eruption; older patients (Nolan 1989, Naldi 1999), boys younger than 3 years old, and girls older than 9 years old (Ibia 2000) have all been found to be more prone to skin drug eruptions. Women are more likely than men to develop a skin drug eruption (Bigby 1986, Naldi 1999, Rademaker 2001).

Viral infections have also been shown to increase the risk of a drug rash; infectious mononucleosis (van der Linden 1998), cytomegalovirus (Klemola 1970), human immunodeficiency virus (Spira 1998), and human herpes virus-6 (Hashimoto 2001) have all been incriminated.

Moreover, some intrinsic factors influence the risk of a cutaneous drug eruption. Individual genetic variations in the metabolism of a drug and HLA association are part of these intrinsic factors. For example, the slow N-acetylation phenotype may predispose to sulfonamide reactions (Wolkenstein 1995), HLA-DR4 type may predispose to drug-induced pemphigus (Ruocco 1992), HLA-B7 to insulin allergy (Bertrams 1977), and HLA-B22 to fixed drug eruptions (Pellicano 1994). Finally, intercurrent diseases like systemic connective tissue disease may lead to immune perturbance and enhance the risk of a cutaneous drug eruption (Crowson 2001), as impaired renal and liver function increase risk of cutaneous rashes.

Cutaneous drug eruptions are more commonly associated with antimicrobials (Bigby 1986, 2001, van der Linden 1998, Ibia 2000). In the Boston Collaborative Drug Surveillance Program, amoxicillin ghad a reaction rate of 5.1%, ampicillin of 4.5%, sulfamethoxazole-trimethoprim of 3.7%, semisynthetic penicillin of 2.9%, penicillin G of 1.6%, and cephalosporin of 1.5% (Bigby 2001). Fluoroquinolones also are often involved (Naldi 1999, van der Linden 1998). Other compounds frequently implicated in drug eruptions are red blood cells, NSA1Ds, analgesics, and gold compounds (Bigby 2001, Oberholzer 1993, Hunziker 1997, Swanbeck 1992).

Approach

When a cutaneous eruption occurs in a patient taking drugs, making the right diagnosis is one of the key elements in the assessment of a patient before instituting any treatment or recommendations. This diagnosis can sometimes be very difficult to establish insofar as a cutaneous drug eruption can closely mimic other diseases. Moreover, if a patient is taking several drugs, as is often the case, identification of the causative drug becomes much more complex.