1e:T4f2,Little work has been carried out on the epidemiology of the two serious skin reactions--toxic epidermal necrolysis (TEN) and Stevens-Johnson syndrome (SJS). We collected details of all the hospitalized cases of TEN and SJS in the Federal Republic of Germany for the years 1981 through 1985 inclusive. Inquiries by telephone, letter, and personal visits produced an overall response of 91%; 259 cases of TEN and 315 cases of SJS were identified. From these data, we were able to calculate an overall annual risk of 0.93 and 1.1 per million for TEN and SJS, respectively. The average age group was higher for TEN (63 years) than for SJS (25 years). Women are markedly more at risk for TEN in the ratio of 2:1, these figures being reversed for SJS. The mortality was 34% (87/259) for TEN and only 1% (2/315) for SJS. An association with previous medication defined as "definite, probable, possible" could be established for 89% of cases of TEN and 54% of cases of SJS. The drugs most commonly involved were antibiotics (TEN, 40%; SJS, 34%), followed by the analgesics (TEN, 23%; SJS, 33%). As with the drug groups, the incidences being based on the defined daily doses, were high for sulfonamides, beta-lactam antibiotics, and some nonsteroidal anti-inflammatory drugs.1f:T453,Eighty-seven patients with toxic epidermal necrolysis were observed at Hôpital Henri Mondor in Créteil, France, over the last 12 years. The mean percentage of body surface area involved was 39%. Erosive mucous membrane lesions, identical to those of Stevens-Johnson syndrome, were present in all but three cases. Necrolysis was sometimes generalized within 24 hours but usually spread progressively after a Stevens-Johnson syndrome-like aspect at the onset. Mortality was 25%. Infection, mainly with Staphylococus aureus and Pseudomonas aeruginosa, was the first cause of death, clearly responsible in ten of 20 cases. Age, extension of necrolysis, idiopathic nature of toxic epidermal necrolysis, ingestion of many drugs, elevation of urea, creatinine, and glucose levels, neutropenia, lymphopenia, and thrombocytopenia were statistically linked to a bad prognosis. A multivariant analysis showed that three of these prognosis factors are of paramount importance, namely: age, area of necrolysis, and serum urea level. Pigmentary changes and sicca syndrome were frequently observed sequelae in survivors.20:T51d,• We looked retrospectively for all cases of toxic epidermal necrolysis that occurred in France over a 5-year period to appreciate the incidence of this disorder and its drug etiology. Of the 399 cases identified, we obtained detailed information on 344 cases and validated 253 cases. From response rates (66% to 98%), we estimated the actual total number of cases to be 333, and the incidence of toxic epidermal necrolysis in France to 1.2 cases per million per year. An independent estimation derived from death certificates gave a figure of 1.3 cases per million per year. When the number of cases with a given drug present were related to the defined daily doses of that drug sold over the 5-year period, the highest ratios were for sulfadiazin (230 · 10^-8), isoxicam (41 · 10^-8), oxyphenbutazone (18 · 10^-8), phenytoin (14 · 10^-8), fenbufen (13 · 10^-8), and cotrimoxazole (12 · 10^-8). This first nationwide study confirmed the rarity of toxic epidermal necrolysis. Within the two main classes of responsible drugs (antibacterial sulfonamides and nonsteroidal anti-inflammatory agents) the risks linked to different drugs appeared quite different, even for closely chemically related products. (Arch Dermatol.1990;126:37-42)21:T89c,

	Year	Citations

References