Photodistributed Stevens–Johnson syndrome and toxic epidermal necrolysis: a systematic review and proposal for a new diagnostic classification

Abstract

Background Ultraviolet radiation (UVR) exposure is commonly reported as a risk factor for Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). However, minimal evaluation of photo-induced SJS/TEN has been conducted. Thus, this review identifies all cases of SJS/TEN that are linked to an acute exposure of UVR and outlines the unifying characteristics of these cases. Furthermore, the theoretical pathogenesis, differential diagnoses, and proposed diagnostic criteria are defined. Methods PubMed, Google Scholar, and other databases and websites were searched from inception to September 2021 to identify studies that met inclusion criteria. The following keywords were utilized: “Stevens-Johnson syndrome” and “toxic epidermal necrolysis” with “ultraviolet,” “photodistributed,” “photo-induced,” “photosensitivity,” and “photo.” One reviewer assessed study characteristics, with confirmation by a second. The risk of bias was assessed independently by another. Results Thirteen patient cases were identified, all reporting ultraviolet radiation prior to rash onset and an underlying causal drug. Case classifications included 7/13 SJS and 6/13 TEN. All cases described the rash as photodistributed with UVR exposure prior to rash onset (delay of 1–3 days) and a causal drug. 10 cases provided evidence that the photodistributed rash lacked linear demarcation (as in a sunburn) with satellite target-like lesions. No cases described a flu-like prodrome. Discussion Mucositis, palmar and plantar rash, a positive Nikolsky sign, and a prolonged disease course can help distinguish from photosensitive reactions, while a negative direct immunofluorescence test is important to distinguish from other photo-induced disorders. Conclusion Physicians should be aware that UVR may precipitate SJS/TEN in patients taking susceptible drugs. After a 24-h delay from UVR exposure, a non-distinct, photodistributed rash appears with no flu-like prodrome and progresses for at least 48 h to include vesiculobullous eruptions and mucous membrane involvement. Photodistributed SJS/TEN appears to be photo-drug-induced with a unique onset and rash presentation that should be recognized as a distinct diagnosis.

Comment on: Proposal for a new diagnostic classification of photodistributed Stevens–Johnson syndrome and toxic epidermal necrolysis

“…We read with interest the article by McKinley, Allen, and Michels titled "Photodistributed Stevens–Johnson Syndrome and Toxic Epidermal Necrolysis: A Systematic Review and Proposal for a New Diagnostic Classification". The authors reviewed a total of 13 SJS/TEN cases that reported preceding and temporally related ultraviolet radiation (UVR) exposure to support their hypothesis that UVR was responsible for the distinct distribution of epidermal necrosis in these cases [ 1 ]. We commend the authors for their thorough and detailed review that makes a valuable contribution to the SJS/TEN literature while stimulating a discussion.…”

“…McKinley et al specifically addressed this by excluding cases with a sharp cutoff typical of phototoxic reactions. However, they described SJS/TEN lesions beyond sun-exposed areas at first presentation, including mucosal and palmoplantar involvement in 12/12 and 5/6 cases, respectively [ 1 ]. Detailing the sequence in which the photoexposed and protected areas were initially affected would be helpful, otherwise this brings to question the veracity of attributing causality to UVR as an explanation for this clinical pattern.…”