Earlier Versus Later Extracorporeal Cardiopulmonary Resuscitation in Refractory Out-of-Hospital Cardiac Arrest: A Nationwide Sequential Landmark Target Trial Emulation

BACKGROUND: Extracorporeal cardiopulmonary resuscitation (ECPR) may improve outcomes in refractory out-of-hospital cardiac arrest (OHCA), but the optimal timing of initiation remains uncertain, partly because prior comparisons did not restrict controls to patients still in refractory arrest at the same decision point. METHODS: We employed target trial emulation with a sequential landmark design in the Korean nationwide OHCA registry (2016-2023). Among patients with presumed cardiac aetiology who were still receiving emergency department (ED) CPR 15 minutes after ED arrival, landmarks were set at 15, 30, and 60 minutes, each comparing ECPR initiation within the subsequent window with no ECPR during that window among patients still in refractory arrest. Treatment effects were adjusted risk differences estimated by propensity score overlap weighting with 22 covariates. RESULTS: Of 58,668 patients with refractory OHCA, 909 received ECPR. Weighted 30-day survival (ECPR vs. control) was 31.0% vs. 16.0%, 21.6% vs. 7.2%, and 10.9% vs. 3.2% at the 15-, 30-, and 60-minute landmarks; adjusted risk differences were +15.0 percentage points (95% CI 8.0 to 22.0), +14.4 (10.4 to 18.4), and +7.7 (3.7 to 11.7) (all P<.001). Favourable neurological outcomes showed a similar pattern, and in the exploratory timing-gradient analysis, the adjusted risk difference decreased with longer door-to-ECMO times but remained statistically significant through 120 minutes. CONCLUSIONS: ECPR was associated with improved survival and neurological outcomes at every landmark, including initiation beyond 60 minutes, where weighted 30-day survival with continued conventional CPR was 3.2%. Benefit was largest with early initiation but did not disappear at the commonly applied 60-minute benchmark, which may be overly restrictive for selected patients in capable systems.