Procedural volume and outcomes in patients with cardiogenic shock who received Impella: insight from the J-PVAD Registry

Abstract Background/Introduction A percutaneous ventricular assist device (Impella) is increasingly used in patients with cardiogenic shock (CS). The association between the procedural volume and outcomes has been shown for extracorporeal membrane oxygenation; however, data are scarce for Impella. Purpose This study aimed to evaluate the association between Impella procedure volume and outcomes in patients with CS, using a nationwide prospective registry database. Methods Using the Japan Registry for Percutaneous Ventricular Assist Devices (J-PVAD) database from January 29, 2020, to December 31, 2021, we included 3,112 patients who received an Impella from 184 hospitals and calculated annual institutional volume that was defined as the number of all Impella procedures conducted at each institution during the year prior to the case. The institutional annual volume was classified into quartiles. We ultimately identified 1,390 patients who received an Impella for CS from 182 hospitals between February 1, 2021 and December 31, 2021. The primary outcome was in-hospital mortality and the secondary outcomes were 30-day mortality and major complications, including major bleeding, hemolysis, cerebrovascular events, access site-related infection, lower limb ischemia, acute kidney disease, thrombocytopenia, and vascular injury. Multivariable logistic regression and cubic spline analyses were performed to assess the nonlinear associations between Impella procedure volume and outcomes. Results The median annual institutional volume was 12 (interquartile range, 7–18) cases/year. Of 1,390 eligible patients, there were 382 patients in the first quartile (0–7 cases/year), 360 in the second quartile (8–12 cases/year), 331 in the third quartile (13–18 cases/year), and 317 in the fourth quartile (19–48 cases/year). The mean age was 66 years, 76% were men, 49% had acute coronary syndrome, 20% had out-of-hospital cardiac arrest, and there were no significant differences among the quartiles. The multivariable logistic regression analyses showed that patients in the fourth quartile compared to those with first quartile had a lower in-hospital mortality (adjusted odds ratio[aOR], 0.67; 95% confidence interval [CI], 0.47–0.96) (Figure 1) and 30-day mortality (aOR, 0.69; 95% CI, 0.48­–0.99). The Cubic spline analysis showed a decrease trend of in-hospital mortality with higher annual institutional volume between 12 and 48 cases/year, with little change between 0 and 12 cases/year (Figure 2). Regarding complications, the multivariable logistic regression analyses showed that patients in the fourth quartile compared to those with first quartile had a higher incidence of hemolysis (aOR, 3.05, 95% CI, 1.90–4.92) and thrombocytopenia (aOR, 2.87, 95% CI, 1.20–6.86). Conclusions This nationwide prospective study showed that higher annual institutional Impella volume was associated with a lower mortality but a higher incidence of hemolysis and thrombocytopenia.Figure 1Figure 2