Derivation and validation of a novel right-sided heart failure model after implantation of continuous flow left ventricular assist devices

Abstract

BACKGROUND: The aim of the study was to derive and validate a novel risk score for early right-sided heart failure (RHF) after left ventricular assist device implantation. METHODS: The EUROMACS (European Registry for Patients with Mechanical Circulatory Support) was used to identify adult patients undergoing continuousflow left ventricular assist device implantation with mainstream devices. Eligible patients (n=2988) were randomly divided into derivation (n=2000) and validation (n=988) cohorts. The primary outcome was early (<30 days) severe postoperative RHF, defined as receiving short- or long-term right-sided circulatory support, continuous inotropic support for ≥14 days, or nitric oxide ventilation for ≥48 hours. The secondary outcome was all-cause mortality and length of stay in the intensive care unit. Covariates found to be associated with RHF (exploratory univariate P<0.10) were entered into a multivariable logistic regression model. A risk score was then generated using the relative magnitude of the exponential regression model coefficients of independent predictors at the last step after checking for collinearity, likelihood ratio test, c index, and clinical weight at each step. RESULTS: A 9.5-point risk score incorporating 5 variables (Interagency Registry for Mechanically Assisted Circulatory Support class, use of multiple inotropes, severe right ventricular dysfunction on echocardiography, ratio of right atrial/ pulmonary capillary wedge pressure, hemoglobin) was created. The mean scores in the derivation and validation cohorts were 2.7±1.9 and 2.6±2.0, respectively (P=0.32). RHF in the derivation cohort occurred in 433 patients (21.7%) after left ventricular assist device implantation and was associated with a lower 1-year (53% versus 71%; P<0.001) and 2-year (45% versus 58%; P<0.001) survival compared with patients without RHF. RHF risk ranged from 11% (low risk score 0-2) to 43.1% (high risk score >4; P<0.0001). Median intensive care unit stay was 7 days (interquartile range, 4-15 days) versus 24 days (interquartile range, 14-38 days) in patients without versus with RHF, respectively (P<0.001). The c index of the composite score was 0.70 in the derivation and 0.67 in the validation cohort. The EUROMACS-RHF risk score outperformed (P<0.0001) previously published scores and known individual echocardiographic and hemodynamic markers of RHF. CONCLUSIONS: This novel EUROMACS-RHF risk score outperformed currently known risk scores and clinical predictors of early postoperative RHF. This novel score may be useful for tailored risk-based clinical assessment and management of patients with advanced HF evaluated for ventricular assist device therapy.