The effect of extremity vascular complications on the outcomes of cardiac support device recipients

Abstract

To assess the effect of extremity vascular complications (EVCs, including ischemia or vessel trauma) on the outcomes of patients receiving cardiac support devices (CSDs, including ventricular assist device [VAD] and extracorporeal membrane oxygenation [ECMO]). Institutional Review Board-approved, retrospective review of a prospectively maintained database of all temporary and permanent CSD recipients from 7/1/10 to 6/30/12. Patient demographics, procedural data, and outcomes were analyzed. The primary endpoint was all-cause mortality at 30-days post-CSD initiation. Of 208 patients who received CSDs, 31 (14.9%) experienced EVC: 13 (8.9%) of the 146 permanent VADs, 10 (26.3%) of the 38 temporary VADs, and 8 (33.3%) of the 24 ECMO patients. The 30-day mortality for CSD-EVC patients was not significantly higher than that of the CSD patients who did not experience EVC for permanent VAD (15.4% vs 4.5%; P = .15) and ECMO patients (50.0% vs 68.75%; P = 1.00), but was significantly higher for temporary VAD patients (80.0% vs 35.7%; P = .03). Within the CSD-EVC cohort, patients who received a temporary VAD had a significantly higher 30-day mortality and decision to withdraw care after EVC compared with those who received a permanent VAD (P = .01 and P < .01, respectively). Looking beyond the 30-day window, EVC was associated with higher mortality rates in the permanent VAD population (53.8% vs 25.6%; P = .025) but not the temporary VAD or ECMO groups. In temporary VAD recipients, EVCs result in higher 30-day mortality, more frequent withdrawal of care, and shortened survival time relative to the global temporary VAD group. EVC in permanent VAD recipients did not affect early (30-day) mortality rates, but strongly predicted a higher cumulative mortality risk for the 2-year study period. Overall ECMO mortality rates were high, and not significantly impacted by the occurrence of EVC. The nature of the EVC (cannulation site complication vs embolic injury) did not impact mortality. This data provides quality improvement targets for VAD programs.