LAUSR

Introduction: Extracorporeal cardiopulmonary resuscitation (ECPR) for out-of-hospital cardiac arrest (OHCA) is promising but unproven. We developed a concept for a randomized Bayesian adaptive clinical trial to define the interval after arrest during which patients derive benefit. Hypotheses: We hypothesized that the simulated design characteristics of the adaptive design would efficiently confirm or refute the existence of a therapeutic window for ECPR in a future trial. Methods: Through iterative simulation and design modification, we developed a Bayesian adaptive trial of ECPR for adults with OHCA. The goals of the trial design were to address the uncertainty surrounding the OHCA-to-ECPR interval within which clinical benefit might be preserved and the difference in prognosis between patients with non-shockable vs shockable rhythms. The treatment effect was defined as the mean 90-day utility-weighted Modified Rankin Scale (uwmRS) for each rhythm subgroup and estimated CA-to-ECPR interval. The trial adaptively lengthens or contracts the estimated CA-to-ECPR eligibility window depending on the probability of benefit from ECPR. We simulated trial performance under six potential ECPR efficacy scenarios. We compared the adaptive design to a fixed design. Results: The trial had 91-100% power to detect a benefit from ECPR for non-shockable rhythms and 69-98% power for shockable rhythms under the scenarios simulated (figure). The design had a high probability of identifying the maximum CA-to-ECPR interval within which ECPR produces a clinically significant benefit of 0.2 on the uwMRS. In null scenarios, the 97-99% of the simulated trials ended early declaring futility. Conclusions: The adaptive trial design helps to ensure the patient population most likely to benefit from treatment—as defined by rhythm subgroup and estimated CA-to-ECPR interval—are enrolled. The design also promotes early termination of the trial if continuation is likely to be futile.