Plasma exchange for severe alveolar hemorrhage in ANCA-associated vasculitis: emulation of a target trial.

Abstract

Patients with severe antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV)-related diffuse alveolar hemorrhage (DAH) face high mortality. In the PEXIVAS trial, plasma exchange (PLEX) did not reduce death or end-stage kidney disease, but only 9% had severe DAH. This study aimed to assess whether PLEX lowers mortality in patients with severe DAH. We emulated a target trial using retrospective data from a national multicenter cohort of patients with severe AAV-related DAH. The primary endpoint was 30-day mortality after ICU admission, analyzed using a Cox model adjusted for prespecified confounders. We included 184 patients (median age 66 [53-75] years; 51% female; 51% with granulomatosis with polyangiitis; 53% MPO-ANCA positive). Of these, 144 (78.3%) received PLEX and 40 (21.7%) did not. Baseline characteristics were similar, except for more severe renal impairment (creatinine 357 vs. 171 µmol/L, P = 0.01) and more frequent cyclophosphamide use (77% vs. 55%, P = 0.01) in the PLEX group. Severity at ICU admission (median SAPS II score: 42) and mechanical ventilation needs (54%) were comparable between groups. At 30 days, overall survival was 85%. No significant difference in mortality was observed between the PLEX and no-PLEX groups: 30-day survival was 85% (95% CI 81-90) with PLEX vs. 88% (95% CI 77-96) without (HR 1.23; 95% CI 0.57-3.89). Secondary outcomes were also similar. In this emulated target trial, PLEX did not reduce 30-day mortality in patients with severe AAV-related DAH.