The MARPELLA risk score for prediction of mortality in all-cause refractory cardiogenic shock treated with microaxial transvalvular pump: the MARPELLA study

Abstract

Abstract Background Although Impella may serve as an alternative strategy in order to stabilize the heart function without the detrimental effects of catecholamines or to bridge the patients in definite therapies in the setting of cardiogenic shock (CS), leading to a widespread use of this device, data concerning reliable prediction or predefining which patients would benefit from the implantation of such a device are completely lacking. Purpose To evaluate and to compare the prognostic accuracy of the Acute-Physiology-And-Chronic-Health II (APACHE II), the Simplified-Acute-Physiology-Score II (SAPS II), the Sepsis-related-organ-failure-assessment (SOFA), the intra-aortic-balloon-pump (IABP), the CardShock Score in refractory CS treated with Impella 2.5/CP. Moreover, we aimed to assess the prediction-of-Cardiogenic-shock-Outcome-foR-AMI-patients-salvaGed by VA-ECMO (ENCOURAGE) and the Survival-after -Veno-Arterial-extracorporeal-membranoxygenation (VA-ECMO) (SAVE) score, though initially for VA-ECMO patients created, in patients with CS as well as to develop a new prognostic score in this setting. Methods Single center study of consecutive Impella patients with CS admitted to Cardiology department of our university, from February 2013 until December 2020. Results A total of 401 patients were included in the analysis. 31% of the patients were supported with Impella CP, whereas 153 (38,1%) patients were resuscitated prior admission. Causes of CS was acute myocardial infarction in 311 (77,5%) patients, followed by dilatative cardiomyopathy/myocarditis (11,1%) and aortic stenosis (6%). The expected mortality according to scores was: SOFA 50%, SAPS II 70%, IABP Shock 55%, CardShock 60%, APACHE II 65%, ENCOURAGE 75% and SAVE score 70%. We observed a survival of 48,9% on hospital discharge and 45,6% after 12 months follow-up. Among the traditional scores estimated, the ENCOURAGE score was the most effective predictive model of mortality outcome presenting a moderate area under curve (AUC) of 0.72, followed by the CardShock, which derived an AUC of 0,7. The SAPS II, SOFA, IABP and the SAVE scores failed to predict outcome in this setting. According to the main predictors of outcome in our population derived from univariate analysis (vasoactive score >31, lactate>4,7 mmol/l, pH<7,31, Creatinine>1,33 mg/dl, Horowitz Index<238, age>71 years and prior resuscitation) as well as the odds ratio derived from binary regression analysis on mortality, a new score, the MARPELLA score, was created. This score reached an AUC of 0.83 (Figure 1). A mortality of 37%, 62% and 83,2% was observed in the low-, intermediate and high-risk group of the MARPELLA score, respectively (Figure 2). Conclusion MARPELLA Score is a new more potent score in the setting of all-cause CS that may guide clinicians to optimize the therapy in this group of patients, outweighing the traditional CS and intensive care unit scores. Funding Acknowledgement Type of funding sources: None.