Transition from VA-ECMO to Durable VAD Systems. Do We Need Cardiopulmonary Bypass Machine? On Behalf of ECMO-VAD Study Group

Objectives: Many surgeons prefer to use cardiopulmonary bypass (CPB) machine when durable ventricular assist devices (VAD) is implanted in patients on venoarterial membrane oxygenation (VA-ECMO). The aim of this study was to investigate if omitting CPB machine at the time of surgery negatively influence the outcome.

Methods: Data of eleven high-volume VA-ECMO/VAD centers are collected and evaluated to identify patients who underwent VAD implantation after ECMO support between January 2010 and July 2018. Preoperative parameters and postoperative outcome are analyzed. Propensity score analysis was performed to identify two identical groups; one who were operated on without CPB (no CPB group) and the other group were operated on using CPB (CPB group).

Results: A total of 531 patients met the inclusion criteria. Nineteen patients were supported with CardioWest TAH were excluded from this analysis. The remaining 512 patients included 306 patients in the CPB group and 206 patients in no CPB group. After 1:1 propensity score matching 207 patients remained in each group and were compared. The total surgery time was significantly longer in the CBP group (288 ± 92 vs. 206 ± 64 minutes, p = < 0.001). The postoperative chest tube output on the day of surgery was comparable between the groups (1,580 ± 1271 vs. 1,321 ± 1047 mL, p = 0.09). However, resternotomy for bleeding was necessary in 41 vs. 30% of patients with or without CPB, respectively (p = 0.022), and a significantly higher number of PRBC were given to the patients who were operated with CPB (11 ± 8 vs. 9 ± 4 units, p = 0.009); the number of transufused FFP and platelets were comparable. A temporary postoperative right ventricular assist device implantation was necessary in 48 vs. 40% of the patients with or without CPB respectively (p = 0.082). Stroke rate of 20% comparable in both groups (p = 0.922). Postoperative renal failure, liver failure and respiratory failure were also comparable (p = 0.223–0.708). No significant difference in the postoperative survival between the two groups was observed (p = 0.096).

Conclusion: This study shows, omitting CPB at the time of durable VAD implantation for patients on VA-ECMO results in shorter surgery time, lower reexploration rate for bleeding and less blood product requirements. A trend toward lower RVAD use was also observed. Survival remains similar between both groups.