Evaluation of Plasmapheresis vs Immunoglobulin as First Treatment After Ineffective Systemic Corticosteroid Therapy for Patients With Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis.

IMPORTANCE

Stevens-Johnson syndrome and toxic epidermal necrolysis (SJS/TEN) are severe cutaneous adverse reactions, and patients with SJS/TEN frequently require intensive care. However, there is limited evidence on the clinical outcomes of immunomodulating therapy, including plasmapheresis and intravenous immunoglobulin (IVIG) in patients with SJS/TEN.

OBJECTIVE

To compare clinical outcomes of patients with SJS/TEN who were treated with plasmapheresis first vs IVIG first after ineffective systemic corticosteroid therapy.

DESIGN, SETTING, AND

PARTICIPANTS

This retrospective cohort study used data from a national administrative claims database in Japan that included more than 1200 hospitals and was conducted from July 2010 to March 2019. Inpatients with SJS/TEN who received plasmapheresis and/or IVIG therapy after initiation of at least 1000 mg/d of methylprednisolone equivalent systemic corticosteroid therapy within 3 days of hospitalization were included. Data were analyzed from October 2020 to May 2021.

EXPOSURES: Patients who received IVIG or plasmapheresis therapy within 5 days after initiation of systemic corticosteroid therapy were included in the IVIG- and plasmapheresis-first groups, respectively.

MAIN OUTCOMES AND MEASURES: In-hospital mortality, length of hospital stay, and medical costs.

RESULTS

Of 1215 patients with SJS/TEN who had received at least 1000 mg/d of methylprednisolone equivalent within 3 days of hospitalization, 53 and 213 patients (mean [SD] age, 56.7 [20.2] years; 152 [57.1%] women) were included in the plasmapheresis- and IVIG-first groups, respectively. Propensity-score overlap weighting showed no significant difference in inpatient mortality rates between the plasmapheresis- and IVIG-first groups (18.3% vs 19.5%; odds ratio, 0.93; 95% CI, 0.38-2.23; P = .86). Compared with the IVIG-first group, the plasmapheresis-first group had a longer hospital stay (45.3 vs 32.8 days; difference, 12.5 days; 95% CI, 0.4-24.5 d; P = .04) and higher medical costs (US $34 262 vs $23 054; difference, US $11 207; 95% CI, $2789-$19 626; P = .009).

CONCLUSIONS AND RELEVANCE: This nationwide retrospective cohort study found no significant benefit to administering plasmapheresis therapy first instead of IVIG first after ineffective systemic corticosteroid treatment in patients with SJS/TEN. However, medical costs and length of hospital stay were greater for the plasmapheresis-first group.