gms | German Medical Science

Background: Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are severe skin reactions mainly caused by drugs. They are characterized by specific skin lesions with epidermal detachment and hemorrhagic erosions of the mucous membranes. Morbidity and mortality are high with frequent long-lasting sequelae in survivors.

Materials and Methods: The German registry on severe skin reactions (dZh) started to ascertain all hospitalized cases of SJS/TEN in Germany almost 25 years ago. The dZh performs a regular survey including all hospitals and departments which are capable to treat SJS/TEN. It operates on a nationwide level with systematic case ascertainment and independent case validation. It participated in international case-control studies evaluating the risk of medications to induce SJS/TEN. Since 2003 data of the German registry are included in the international RegiSCAR-project. Thus, validated cases of SJS/TEN from Germany included in the RegiSCAR-study over a period of 10 years were analyzed in terms of demographic data and risk factors.

Results: In total, 681 cases of SJS and TEN as well as the overlap were validated as “probable” and “definite” in terms of the clinical diagnosis between 01/2003 and 12/2012. 52% meet the criteria of SJS with < 10% of skin detachment, 14% those of TEN >30% and 34% of SJS/TEN-overlap with 10 - 30% of skin detachment. Related to the percentage of women in the general population of Germany, they are 1.5 x more frequently affected than men. For persons of 65 years and older the risk is three times higher than for younger people. The overall mortality is 27%, but increases with the severity of the disease to e.g. 48% in TEN. In addition, older age has a substantial impact on the risk to die.

Within two weeks before onset of SJS/TEN approx. 50% of the patients were exposed to a high risk drug which was started within 8 weeks before. Such so-called “highly suspected” or “strongly associated” medications were found to have a high relative risk to induce SJS/TEN in two case-control studies. They included allopurinol, antibacterial sulfonamides, certain antiepileptics, nevirapine and non-steroidal inflammatory drugs (NSAIDs) of the oxicam-type. So called “suspected” or “associated” drugs were also identified in these studies and showed a lower but still significant relative risk. They mainly included antibiotics and NSAIDs of the phenylacetic acid type. In the German cohort investigated here, allopurinol was taken by 22% of SJS/TEN-patients in the relevant time period before disease onset, which is thus still the most important risk factor. It is followed by cotrimoxazole (sulfamethoxazole / trimethoprim) with 8.4% of exposed patients, lamotrigine with 7%, carbamazepine with 5.8% and phenytoin with 4.6% of exposed patients. Of course, these exposure rates have to be seen in prospective with the overall exposure of the general population to these drugs. In doing so, the risk of allopurinol is still the highest followed by antibacterial sulfonamides and certain antiepileptics.

Conclusion: In comparison to analyses of earlier years, the average age of SJS/TEN-patients as well as the mortality rate have increased. Concerning the inducing medications, a high risk could be confirmed for drugs known to induce SJS/TEN. The indication to prescribe such drugs, e.g. allopurinol, and the broadening of the indication, e.g. for lamotrigine, have to be checked more thoroughly, in order to avoid or at least reduce the occurrence of SJS/TEN.