Severe cutaneous adverse reactions to drugs: a case series study

Abstract

Objective: to describe five cases of severe cutaneous adverse reactions to drugs (SCARD) that led to hospitalization and were investigated and diagnosed by the dermatology service of a university hospital. Methods: this is a descriptive observational study of the case series type in which were included patients aged 18 years old or older from both sexes, hospitalized between January 2015 and July 2019, in a university hospital in Rio de Janeiro, Brazil and whose reason for hospitalization was SCARD diagnosed by the dermatology service. For SCARD causality assessments, two internationally adopted criteria were used: the Naranjo algorithm and the WHO-UMC causality assessment system. Results: five cases of RCAG have been reported. Amongst these, three cases of erythroderma, one of Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN) and one of drug reaction with eosinophilia and systemic symptoms (DRESS). The cases of erythroderma have presented as suspected drugs substances such as hydrochlorothiazide, furosemide, captopril and carbamazepine, which is consistent with the literature reports. In the reported case of SJS/TEN, both the suspected drug, allopurinol, and the presence of an anti-HIV positive laboratory test are also referred to in the literature. The case of DRESS has shown allopurinol as a suspected drug, which is already related to this well-described skin reaction. The causalities were assessed as possible for the first two cases and probable for the following three. The results of the study supported the proposal for a model of pharmacovigilance of SCARD in a hospital environment, consisting of spontaneous reporting, periodic checking of medical records and for requests of opinion from the dermatology service registered in the hospital's computerized system. Conclusion: this study contributes to add knowledge and better understanding about SCARD, in regards to the nature of the type of reaction, identification of patients at risk, early detection of these and the responsible drugs. Furthermore, it proposes a model for monitoring SCARD to be implemented in the hospital, which combines passive and active pharmacovigilance strategies and encourage notification by health professionals.