Stevens-Johnson syndrome after sertraline.

Sir, Sertraline (Zoloft1) is a new selective inhibitor of serotonin re-uptake. This family of antidepressant drugs is considered to be safe, and cutaneous adverse reactions have rarely been reported (1). We report here a patient who developed StevensJohnson syndrome (SJS) after starting treatment with sertraline (Zoloft1). A 96-year-old woman was admitted with a cutaneous and mucosal eruption. Sertraline and arginine chlorhydrate (Arginine Veyron1) treatment had been initiated 3 weeks before the eruption. Sertraline had replaced paroxetine (Deroxat1), initiated 7 weeks before the eruption for depression. Her general condition was good, and her temperature was 373C. The cutaneous lesions were found on the face, trunk and proximal parts of the limbs, and were erythematous or purpuric, with an atypical ̄at target appearance, without signi®cant epidermal detachment. Nikolsky's sign was negative. She had painful oral erosions and conjunctivitis. Histological examination of a skin biopsy showed total necrosis of the epidermis, direct immuno ̄uorescence was negative, and the blood cell count was normal. Serology for herpes virus I and II showed previous immunization and serology for Mycoplasma pneumoniae was negative. Sertraline and arginine chlorydrate treatment were withdrawn and the cutaneous eruption disappeared in 7 days. Control of serology for Mycoplasma pneumoniae was negative. The diagnosis of SJS was retained because of the association of macular and purpuric lesions with an atypical ̄at target appearance involving the trunk and proximal limbs, 2 different mucosal sites and total necrolysis of the epidermis (2). This syndrome is usually related to drug intake (3), and serology for herpes and Mycoplasma pneumoniae were negative. According to the standards of the French drug surveillance system, sertraline and arginine chlorhydrate were the only culprit drugs (4). Imputability of arginine chlorhydrate is nil, because it is an amino acid, and no side effect has been reported with this drug. One case of erythema multiforme with involvement of the trunk and face has been reported with sertraline (5) but, to our knowledge, this is the ®rst case of SJS observed with sertraline treatment. Nevertheless, SJS and toxic epidermal necrolysis (TEN) have been reported with 2 other members of the family of selective inhibitors of serotonin re-uptake ( ̄uvoxamine and ̄uoxetine) (6, 7). Cross-reactivity between several members of this family has been observed in 2 cases (8). In the case reported here, paroxetine was not imputable because it had been initiated 7 weeks before the eruption. If a severe cutaneous side effect (SJS or TEN) occurs, it would be better to change the drug family (8).