SWITCH Location Analysis
(2025)Objective
To determine whether the treatment effect of decompressive craniectomy differs according to the precise location of deep intracerebral hemorrhage (basal ganglia alone vs with PLIC involvement vs with PLIC and thalamic involvement)
Study Summary
• Adjusted marginal risk reduction with DC: 15.6% for BG alone, 11.4% for BG+PLIC, and 9% for BG+PLIC+thalamus (P interaction=0.95)
• No evidence of treatment effect modification by ICH location, even in prognostically unfavorable locations involving PLIC and thalamus
Intervention
Decompressive craniectomy (≥12 cm diameter, without hematoma evacuation) plus best medical treatment vs best medical treatment alone
Inclusion Criteria
Post-hoc analysis of SWITCH trial participants with supratentorial severe deep ICH from intention-to-treat population, excluding those with missing central baseline imaging, primary thalamic ICH, or lobar ICH after central review
Study Design
Arms: BG alone (n=26): DC+BMT (n=13) vs BMT alone (n=13); BG+PLIC (n=94): DC+BMT (n=50) vs BMT alone (n=44); BG+PLIC+thalamus (n=64): DC+BMT (n=28) vs BMT alone (n=36)
Patients per Arm: 184 total (91 DC+BMT, 93 BMT alone) across 3 location subgroups
Outcome
• Death at 180 days: Point estimates favored DC across locations (P interaction=0.40 adjusted)
• Full mRS range: Lower odds for shift to worse outcomes with DC (P interaction=0.24 adjusted)