Central vs peripheral venoarterial ECMO in postcardiotomy cardiogenic shock

Objectives Central or peripheral venoarterial extracorporeal membrane oxygenation (va ECMO) is widely used in postcardiotomy cardiogenic shock (PCS). Available data suggest controversial results for both types. Our aim was to investigate PCS patients treated with either peripheral (pECMO) or central ECMO (cECMO) concerning their outcome. Methods Between April 2006 and October 2016, 156 consecutive patients with va ECMO therapy due to PCS were identified and included in this retrospective analysis. Patients were divided into cECMO and pECMO groups. Statistical analysis of risk factors concerning 30-day mortality of the mentioned patient cohort was performed using IBM SPSS Statistics. Results Fifty-six patients received cECMO and 100 patients were treated with pECMO due to PCS. In the group of cECMO peripheral vascular disease was significantly more present (cECMO 19 [34%] vs pECMO 14 [14%]; P < .01). On-site ECMO complications occurred significantly more frequent in patients treated with cECMO (cECMO 44 [79%] vs pECMO 54 [54%] g/dL; P < 0.01). More often cECMO patients required a second look operation due to mediastinal bleeding (cECMO 52 [93%] vs pECMO 61 [61%] g/dL; P < .01). Thirty-day mortality was comparable with nearly 70% in both cohorts (cECMO 39 [70%] vs pECMO 69 [69%]; P = .93). Conclusion Patients supported by cECMO or pECMO due to refractory PCS did not show significant differences in 30-day mortality, despite a lower incidence of on-site ECMO complications and re-exploration in pECMO patients. PCS itself is associated with high mortality and peripheral cannulation might help to save resources compared with central cannulation.