Abstract
Introduction
Stevens-Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis Syndrome (TENS) are part of a spectrum of autoimmune conditions, which cause epidermal detachment and keratinocyte necrosis. Due to the rare incidence, a dramatic heterogeneity in treatment algorithms and prolonged discussion of controversial therapies has ensued. We queried a large national data network to better understand how these therapies are actually implemented and the relative impact on survival.
Methods
The TriNetX Global Health Research Network (41 healthcare organizations) was queried using ICD-10 codes (L51.1 - 51.3) to identify patients diagnosed with SJS or TENS from 2000 to 2020. A treatment frequency map was constructed to determine the most common treatment groupings, and cohorts were indexed based on the most frequent isolated and combined therapy algorithms: systemic steroids (SS), diphenhydramine (DH), cyclosporine (CS), intravenous immunoglobulin (IVIG) and tumor necrosis factor alpha inhibitors (TNFαi). Cohorts were case matched for demographics against a restricted control group (RC; patients who did not receive any of the above-mentioned therapies) and an unrestricted control group (UC, all patients diagnosed with SJS or TENS) to evaluate mortality risk, survival probability, and to uncover Type II error.
Results
Cohorts were UC (6,196), RC (2,248), SS+DH (3,459), SS (1,269), SS+CS (1,554), DH (479), CS (52), IVIG (10) and TNFi (10). The treatment map showed 36.3% of patients did not receive any of the listed therapies. Of those that did 48.2% initially got SS, 24.3% got DH, 15.4% got SS+DH and 3% got SS+CS. Patients who received SS had a 8.51% mortality risk and 2.84% risk reduction compared to UC (p = 0.017). However, the Hazard Ratio (HR) was 0.80 (95% CI: 0.57, 1.23) showing no survival advantage. Compared to RC risk reduction decreased to 0.47% (p = 0.667). SS+DH showed a risk reduction of 1.13% compared to UC (p = 0.113; HR 0.89, 95% CI: 0.69, 1.16), but this advantage resolved when compared to RC. Similarly, SS+CS had a risk reduction of 2.12% compared to UC (p = 0.039; HR 0.80, 95% CI: 0.58, 1.09), which decreased to 0.07% (p = 0.947) when compared to RC. DH and CS had no significant risk reduction (p = 0.25 - 1.00) or survival advantage. IVIG and TNFαi were underpowered for analysis.
Conclusions
The most common treatments for patients diagnosed with SJS or TENS are systemic steroids, diphenhydramine, or a combination of the two. Unfortunately, none of the above therapies confer a significant survival advantage. Furthermore, some therapies raised concern for Type II errors when the control group is not adjusted for alternative therapies.
Recurrent preseptal cellulitis in Boston keratoprosthesis type II implantation for ocular Stevens-Johnson syndrome