The Epidemiology of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis in the UK

Abstract

Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are rare but life-threatening mucocutaneous diseases. SJS/TEN mostly manifest as a reaction to new drug use, but little is known about their incidence and epidemiology. We conducted a large observational study on the epidemiology of SJS/TEN using data from the UK-based Clinical Practice Research Datalink. Among 551 validated SJS/TEN patients, we calculated an incidence rate of 5.76 SJS/TEN cases per million person-years between 1995 and 2013, which was consistent throughout the study period and was highest in patients aged 1-10 years and 80 years or older. Within a 1:4 matched case-control analysis, black and Asian patients were at a 2-fold risk of SJS/TEN when compared with white patients. Among patients with epilepsy and gout, odds ratios for SJS/TEN were significantly increased only in the presence of recent new drug treatment with antiepileptics or allopurinol, respectively. We observed statistically significant associations between SJS/TEN and pre-existing depression, lupus erythematosus, recent pneumonia, chronic kidney disease, and active cancer, but confounding by drug use needs to be followed up. This large and longitudinal observational study on the epidemiology of SJS/TEN contributes to the understanding of this still underinvestigated severe skin disease in a European and largely white study population.

Incidence of and mortality from epidermal necrolysis (Stevens–Johnson syndrome/toxic epidermal necrolysis) in France during 2003–16: a four‐source capture–recapture estimate

“…< 20 years old) and middle‐aged people (i.e. 20–64 years old), which agrees with previous studies . The five‐fold increase in mortality in individuals ≥ 65 years of age in comparison with individuals 20–64 years old is higher than the two‐ to three‐fold increase reported in other studies …”

“…estimation in the U.K. (i.e. 5·76 cases per million inhabitants per year) in which incidence rates were adjusted on a previously calculated positive‐predictive value for the codes used in order to minimize the bias of misclassification and a possible overestimation of incidence . Our results are also close to those published by Velasco‐Tirado et al .…”

The risk of Stevens‐Johnson syndrome and toxic epidermal necrolysis in new users of antiepileptic drugs

“…In this database study, the well‐known risk of SJS/TEN risk in association with AED may lead physician to overdiagnose SJS/TEN in AED users, which may have led to the inclusion of some false‐positive cases, and thus to slightly overestimated ORs. However, our results were similar to prior results, which provides confidence in the validity of our cases . Second, the proportion of cases who were recently exposed to AEDs was lower in our study (7.5%) than in the two previous studies (both 25%), which may partially be explained by the use of different AEDs over time and by the decreasing use of phenobarbital.…”

“…In addition to underestimated ORs, this limitation could also have affected the assessment of concomitant use of other high‐risk drugs for ALDEN in some cases. However, in a previous study based in the same study population, we observed increased risk of SJS/TEN only among patients with a diagnosis of epilepsy as well as a recent new prescription (≤84 days) of an AED, but not in epilepsy patients in the absence of such a prescription, suggesting that bias due to unobserved AED treatment is low in our study popyulation ,. Third, because ALDEN automatically grades causality higher for drugs with previous evidence for an association with SJS/TEN, the assessed likelihood for a causal association could potentially be underestimated for newer AEDs that have not previously been associated with SJS/TEN.…”

Crizotinib-associated toxic epidermal necrolysis in an ALK-positive advanced NSCLC patient

“…Since the patient have not taken any other suspicious drugs beside crizotinib, both before and during the development of skin lesions, we consider TEN was most likely induced by crizotinib. It was reported that Blacks and Asian patients were at a higher risk than Caucasian (10). Some researchers found a strong association between HLA genotype and SJS/TEN in Han Chinese (11,12).…”