Abstract DP127: The Problem of Equipoise: Did we really enroll typical MeVOs in ESCAPE-MeVO?

Introduction: Recent trials for EVT in patients with medium vessel occlusion did not demonstrate benefit of EVT. Some have proposed that lack of equipoise for certain MeVO populations such as younger patients, more proximal occlusions (M2) and those with severe deficits lead to biased enrollment. To assess this, we directly compared these variables in the ESCAPE-MeVO cohort to a population of MeVO treated with EVT as usual care in a Multi-State US stroke registry during the same time of enrollment for ESCAPE-MEVO Methods: We used data from the ESCAPE-MeVO trial and a multi-state US health system registry as a counterfactual enrollment population for the ESCAPE-MeVO trial. The deidentified registry cohort was restricted to patients with MeVO location, treated with endovascular thrombectomy (EVT) that fit the ESCAPE-MeVO trial enrollment interval and inclusion and exclusion criteria. We compared demographic and clinical characteristics between the cohorts. Categorical data were presented as frequencies and percentages. Pearson’s chi-squared test was used for comparisons. Continuous data were presented as the median and interquartile range (IQR). A Wilcoxon rank sum test was used for comparison. A p-value ≤0.05 considered as statistically significant. Results: Compared with the ESCAPE-MeVO study population the registry cohort was significantly younger, median age, 72 (63 – 81) vs. 75 (64 – 82), p=.02. The median NIHSS score 12 (8 – 19) and M2 location of occlusion 333 (80%) was higher in the registry cohort than in the ESCAPE-MeVO trial 8 (5 – 11), p=<0.001 and 226 (43%), p=<0.001 respectively. Figure 1 and Table 1 Conclusions: ESCAPE-MeVO enrolled patients were older (75 vs 72), had lower NIHSS scores (8 vs 12) and had less M2 occlusions (43% vs 80%) than what was seen in a typical EVT treated MeVO population from our registry. A lack of equipoise in younger patients with more proximal MeVOs and higher baseline stroke scales who would have been treated with EVT rather than randomized may account for the selection bias in ESCAPE-MeVO. This selection bias was likely also present in other trials of MeVO EVT (DISTALS and DISCOUNT) which had similar published baseline criteria to ESCAPE-MeVO. Therefore, the results of ESCAPE-MeVO and other MeVO trials may apply more directly to a somewhat narrow patient population rather than the broad range of MeVO patients seen in clinical practice.