BackgroundImpella™ is increasingly used in cardiogenic shock. However, thromboembolic and bleeding events are frequent during percutaneous mechanical circulatory support (pMCS). ObjectiveTherefore, we aimed to explore the optimal anticoagulation regime for pMCS to prevent thromboembolism and bleedings. MethodsThis hypothesis-generating multi-center cohort study investigated 170 patients with left-Impella™ support. We (A) compared bleeding/thrombotic events in two centers with therapeutic range (TR-aPTT) activated partial thromboplastin time (60–80s) and (B) compared events of these centers with one center with intermediate range aPTT (40–60s). ResultsAfter matching, there were no differences in patients' characteristics. In centers aiming at TR-aPTT, major bleeding was numerically lower with aPTT <60s within 48 h of left-Impella™ support, versus patients that achieved the aimed aPTT of ≥60s [aPTT ≥60s: 22 (37.3%) vs. aPTT<60s 14 (23.7%); Hazard ratio [HR], 0.62 (95%) CI, 0.28–1.38; p = 0.234]. Major cardiovascular and cerebrovascular adverse events (MACCE) did not differ between groups. In comparison of centers, TR-aPTT strategy showed higher major bleeding rates [TR: 8 (47.1%) vs. intermediate range: 1 (5.9%); HR, 0.06 (95%) CI, 0.01–0.45; p = 0.006]. MACCE were lower in the intermediate range aPTT group as well [TR 12 (70.6%) vs. intermediate range 5 (29.4%) HR, 0.32 (95%) CI, 0.11–0.92; p = 0.034]. ConclusionThis pilot analysis showed that lowering UFH-targets in left-Impella™ supported CS patients seems to be a safe and promising strategy for reducing major bleedings without increasing MACCE. This needs to be validated in larger, randomized clinical trials.