Abstract 1017 Development and Validation of a Multivariable Risk Model for Prediction of Long-Term Mortality in Patients undergoing Percutaneous Edge-to-Edge-Repair for Severe Mitral Valve Regurgitation

Background: It remains difficult to predict outcome after transcatheter edge-to-edge repair (TEER) for severe mitral regurgitation and classical surgical risk scores are not suitable to predict patient's outcome.

Objective: The aim of this study is to develop a treatment specific multivariable risk model to predict all-cause mortality, considering hemodynamic variables.

Method: A lasso-penalized Cox-proportional hazard regression model was used to identify independent predictors of all-cause mortality during a median follow-up period of 18±13 months. Validation was performed using internal bootstrap resampling. Receiver operating characteristic (ROC) analysis was used to determine discriminative capacity of the developed risk model.

Results: 580 patients (median age 76 years [72-82 years], 59.7% male) at high operative risk (LogEuro-SCORE 17.0% [11.0%-27.0%]) were enrolled. All-cause mortality rate was 32.4% (n=188). Five independent predictors of all-cause mortality were identified: left ventricular stroke work index (LVSWi, hazard ratio (HR) 0.97, 95% confidence interval (CI) 0.95-0.99; p=0.048), chronic obstructive lung disease (HR 1.62, 95% CI 1.10-2.39; p=0.015) , tricuspid annular plane systolic excursion (TAPSE, HR 0.96, 95% CI 0.93-0.99; p=0.037), New York Heart association (NYHA) functional class IV (HR 1.55, 95% CI 1.06-2.26; p=0.023) and NT-proBNP levels (HR 1.47, 95% CI 1.25-1.72; p<0.001). At internal bootstrap resampling validation, the developed risk-score showed good discrimination (area under the ROC curve 0.71 [0.61-0.78]; p=0.001). Conversely, the discriminative ability of the LogEuro-SCORE II was fairly modest (AUC 0.59 [0.54-0.65]; p=0.001).

Conclusions: A treatment specific risk model that incorporates hemodynamic indices may be useful for risk assessment of patients undergoing TEER. External validation of study results for broad clinical application is needed.

Clin Res Cardiol (2022). https://doi.org/10.1007/s00392-022-02002-5
Development and Validation of a Multivariable Risk Model for Prediction of Long-Term Mortality in Patients undergoing Percutaneous Edge-to-Edge-Repair for Severe Mitral Valve Regurgitation
R. Osteresch¹, K. Diehl¹, P. Dierks¹, A. Ben Ammar¹, A. Fach¹, J. Schmucker¹, L. A. Mata Marín¹, C. Frerker², I. Eitel², H. Wienbergen¹, R. Hambrecht¹
¹Bremer Institut für Herz- und Kreislaufforschung (BIHKF), Bremen; ²Medizinische Klinik II / Kardiologie, Angiologie, Intensivmedizin, Universitätsklinikum Schleswig-Holstein, Lübeck;
Background: It remains difficult to predict outcome after transcatheter edge-to-edge repair (TEER) for severe mitral regurgitation and classical surgical risk scores are not suitable to predict patient's outcome. Objective: The aim of this study is to develop a treatment specific multivariable risk model to predict all-cause mortality, considering hemodynamic variables. Method: A lasso-penalized Cox-proportional hazard regression model was used to identify independent predictors of all-cause mortality during a median follow-up period of 18±13 months. Validation was performed using internal bootstrap resampling. Receiver operating characteristic (ROC) analysis was used to determine discriminative capacity of the developed risk model. Results: 580 patients (median age 76 years [72-82 years], 59.7% male) at high operative risk (LogEuro-SCORE 17.0% [11.0%-27.0%]) were enrolled. All-cause mortality rate was 32.4% (n=188). Five independent predictors of all-cause mortality were identified: left ventricular stroke work index (LVSWi, hazard ratio (HR) 0.97, 95% confidence interval (CI) 0.95-0.99; p=0.048), chronic obstructive lung disease (HR 1.62, 95% CI 1.10-2.39; p=0.015) , tricuspid annular plane systolic excursion (TAPSE, HR 0.96, 95% CI 0.93-0.99; p=0.037), New York Heart association (NYHA) functional class IV (HR 1.55, 95% CI 1.06-2.26; p=0.023) and NT-proBNP levels (HR 1.47, 95% CI 1.25-1.72; p<0.001). At internal bootstrap resampling validation, the developed risk-score showed good discrimination (area under the ROC curve 0.71 [0.61-0.78]; p=0.001). Conversely, the discriminative ability of the LogEuro-SCORE II was fairly modest (AUC 0.59 [0.54-0.65]; p=0.001). Conclusions: A treatment specific risk model that incorporates hemodynamic indices may be useful for risk assessment of patients undergoing TEER. External validation of study results for broad clinical application is needed.
https://dgk.org/kongress_programme/jt2022/aP1611.html