Abstract
BACKGROUND: Left ventricular (LV) overload is a frequent complication during VA-ECMO associated with poor outcomes. Many strategies of LV unloading have been documented but lack of evidence shows which is better. We conducted a network meta-analysis to compare different LV unloading strategies. METHODS: We searched databases for all published studies on LV unloading strategies during VA-ECMO. The pre-defined primary outcome was all-cause mortality. RESULTS: 45 observational studies (34235 patients) were included. The Surface Under the Cumulative Ranking values (SUCRA) demonstrated that compared to no unloading strategy (15.4 %), IABP (73.8 %), pLVAD (60.8 %), atrial septostomy (51.2 %), catheter venting (48.8 %) were all associated with decreased all-cause mortality, in which IABP and pLVAD existed statistical significance. For secondary outcomes, no unloading group had the shortest VA-ECMO duration, ICU and hospital length of stay, and the lower risk of complications compared with unloading strategies. IABP was associated with reducing VA-ECMO duration, ICU and hospital length of stay, and the risk of complications (except for hemolysis as the second best) compared with other unloading strategies. CONCLUSIONS: LV unloading strategies during VA-ECMO were associated with improved survival compared to no unloading, but the tendency to increase the risk of various complications deserves more consideration.