Background: Cardiogenic shock (CS) remains as a high mortality condition, despite advances. The use of mechanical circulatory support (MCS) devices such as VA-ECMO, Impella, ECPELLA, and TandemHeart has increased in recent years; however, comparative evidence about their clinical efficacy, safety profile and functional recovery remains controversial due to the limited evidence.
Methods: A narrative review was conducted including clinical trials and meta-analyses identified through PubMed. The outcomes included 30-day mortality, odds ratios (OR), complications, and changes in left ventricular ejection fraction (LVEF) for VA-ECMO, Impella, ECPELLA, TandemHeart, and intra-aortic balloon pump (IABP) in patients with cardiogenic shock. Also, Bayesian network meta-analyses were used to obtain adjusted odds ratios when available.
Outcome: Outcomes showed that 30-day mortality remained high for all MCS devices (43-78%), based on shock severity and device type. VA-ECMO in refractory shock (RS) showed 78% mortality, Impella and ECPELLA demonstrated mortality of 49% in severe shock (SS) and 57% in RS respectively and, TandemHeart 43% in SS. Adjusted comparisons revealed no significant differences in 30-day mortality between MCS devices: Impella versus IABP (OR 1.06, 95% CrI 0.71–1.49), TandemHeart versus IABP (OR 1.30, 95% CrI 0.51–3.70), ECPELLA versus VA-ECMO (OR 0.77, 95% CrI 0.50–1.14), and VA-ECMO versus Impella (OR 1.23, 95% CrI 0.77–1.92). In terms of complications, review also revealed that TandemHeart had the highest rate of severe bleeding (90.4%) and limb ischemia (33.3%). ECMELLA presented high rates of renal replacement therapy (58.5%) and hemolysis (33.6%). VA-ECMO had a significant risk of peripheral vascular complications (11%). VA-ECMO demonstrated an improvement in LVEF of 19% (p < 0.01), Impella: 10.3 ± 1.2% (p < 0.001), ECPELLA: 19.1 ± 2.4% (p < 0.001) and TandemHeart showed an absolute improvement of 30 percentage points in baseline LVEF.
Conclusion: None of the devices demonstrated a significant survival advantage over the others. Differences between MCS interventions appear to be determined by patient selection, shock severity, and study heterogeneity. Future studies should report additional post-shock outcomes such as NYHA functional class, MLWHFQ, KCCQ-12, SF-36, 6-minute walk test and readmission rates.
