ECMO was associated with significantly higher in-hospital mortality (aOR 4.40) and adverse events compared to Impella in patients with Takotsubo cardiomyopathy complicated by refractory cardiogenic shock.
Cohort (n=2,025)
Yes
Does ECMO compared to Impella improve in-hospital mortality in patients with Takotsubo cardiomyopathy and cardiogenic shock?
In Takotsubo cardiomyopathy with cardiogenic shock, ECMO without LV unloading is associated with higher in-hospital mortality and bleeding compared to Impella, though mortality differences resolve when ECMO is combined with LV unloading.
Effect estimate: aOR 4.40 (95% CI 1.76-6.54)
Absolute Event Rate: 37.7% vs 25%
p-value: p=<0.001
Background: The use of mechanical circulatory support (MCS) devices in cardiogenic shock is growing. We aim to study trends and compare different MCS modalities in this population. Methods: The National Readmission Database (2016-2020) was queried to identify TTC-CS requiring MCS. Cohorts were stratified as ECMO (extracorporeal membrane oxygenation) compared to other short-term percutaneous left ventricular assist devices (Impella). The propensity score matching (PSM) was used to remove confounders. Pearson's x2 test was applied to PSM-matched cohorts to compare outcomes. Additionally, we used multivariate regression and reported predictive margins for adjusted trend analysis. Results: Among 2,025 TTC-CS hospitalizations requiring MCS, 1,790 required Impella vs. 235 on ECMO. ECMO was more common in metropolitan teaching hospitals (72.2 % vs 56.1 %, p < 0.05). On PSM cohorts (N = 131), ECMO had higher in-hospital mortality (38.9 % vs. 20.6 %, p < 0.001), major bleeding (15.3 % vs. 2.3 %, p < 0.001), acute blood loss anemia (48.9 % vs. 19.1 %, p < 0.001) among others. Our subgroup analysis comparing ECMO when Left ventricular (LV) unloading was provided by either IABP or Impella, and Impella alone showed no difference in the short-term mortality (42.2 % vs. 33.3 %, p: 0.384). However, the rates of major bleeding (17.8 % vs. 0.0 %, p: 0.003) and acute blood loss anemia (55.6 % vs. 22.2 %, p: 0.001) were higher for ECMO cohort. Conclusion: In the absence of LV unloading, the ECMO utilization in TTC-CS had higher mortality and adverse events than Impella. The mortality difference was nonsignificant when concomitant LV unloading was provided with Impella or IABP in these patients.
Ali et al. (Mon,) conducted a cohort in Takotsubo cardiomyopathy complicated by refractory cardiogenic shock (n=2,025). ECMO (Extracorporeal Membrane Oxygenation) vs. Impella was evaluated on In-hospital mortality (aOR 4.40, 95% CI 1.76-6.54, p=<0.001). ECMO was associated with significantly higher in-hospital mortality (aOR 4.40) and adverse events compared to Impella in patients with Takotsubo cardiomyopathy complicated by refractory cardiogenic shock.