A 6-month prospective survey of cutaneous drug reactions in a hospital setting.

Abstract

Few prospective studies are available on the incidence and analysis of the characteristics of adverse cutaneous drug reactions in hospital settings. A 6-month prospective study was managed in our hospital among hospitalized patients to: (i) evaluate the incidence of cutaneous allergic reactions from systemic drugs; (ii) study characteristics of patients with cutaneous drug reactions; (iii) describe the adverse cutaneous reactions; and (iv) evaluate drug reaction imputability and preventability. All suspected allergic cutaneous reactions to systemic drugs were collected during a 6-month period (November 2000 to May 2001). Exhaustivity of recording was ensured by regular dissemination of information about this study to the practitioners; a simple method for inclusion by fax was established. Inclusion criteria were suspected cutaneous allergic reactions induced by systemic drugs responsible for hospitalization or developed during hospitalization. A physical examination was done by a dermatologist who completed a standardized questionnaire. Requested information included patient characteristics (associated disorders, severity score), drug intake (list and chronology of the drug intake during the 3 weeks preceding the adverse reaction) and characteristics of the skin reaction (type, course). A group comprising dermatologists and pharmacologists evaluated the drug imputability and preventability. Forty-eight cases were collected. A prevalence of 3.6/1000 among hospitalized patients was estimated. The prevalence rate was higher in patients hospitalized in medical departments (0.5%) than in surgical departments (0.01%) (P < 0.001). The cases were mostly recruited in departments of infectious diseases and dermatology. The most frequent associated disorders were: human immunodeficiency virus (HIV) infection (19%), connective tissue disease (10%) and viral or autoimmune hepatitis (12%). Of these patients, 31% had had a previous immunological drug reaction. Adverse cutaneous drug reactions were principally exanthematous (56%). Reactions were considered severe in 34% of cases because they were responsible for hospitalization (18%), increased the duration of hospitalization (14%) or were life threatening (2%). Principal imputable drugs were antibiotics, mainly penicillins. An imputability score was likely in 56% of cases, but it was only possible to conclude definitively in 44% of patients. In 15% of the cases, the side-effect was considered to be preventable. This study finds a lower incidence than other studies that reported an incidence of 2% of cutaneous drug reactions in hospitalized patients, but only allergic adverse cutaneous reactions induced by systemic drugs were collected in this study. The previous studies were principally done in selected patients hospitalized only in general internal medicine or medical divisions. Our results confirm some data already known about skin drug reactions: HIV infection as a risk factor (P < 0.0001), high prevalence of adverse cutaneous reactions due to antibiotics, and difficulty in ascertaining the imputability of a drug. A high proportion (34%) of these reactions was severe and 15% were avoidable; these two facts justify the development of an intensive programme of clinical pharmacology.