Abstract Background Prediction of events is critical in decision-making for transcatheter edge-to-edge mitral valve repair (TEER). A new score has been reported for this aim, the MitraScore, that showed adequate performance for predicting death and heart failure rehospitalization in an external validation cohort described in the original publication. Purpose Our aim was to perform an external validation of this prognostic score in an independent sample. Methods The MitraScore assigned 1 point to each independent predictor of mortality found in the validation cohort: age ≥75 years, anemia, glomerular filtrate rate <60 mL/min/1.73 m2, left ventricular ejection fraction <40%, peripheral artery disease, chronic obstructive pulmonary disease, high diuretic dose, and no therapy with renin-angiotensin inhibitors. We calculated the MitraScore in every subject of a prospective cohort of patients treated with TEER in a single institution from November 2011 to January 2024. Association of the score with mortality and the combined end-point of heart failure admission or death was assessed by the Cox proportional hazard method and Kaplan Meier curves. Discrimination was studied by receiver operating characteristics (ROC) curve. Results A total of 191 consecutive patients (median age [p25-75] 74 [64-80] years, 33% female) were included in the study. Our sample differed significantly from the original derivation and validation cohorts in several baseline variables (figure 1). After a median follow up of 24 [9-50] months, 92 patients (48.4%) had 160 admissions for heart failure and 73 patients died (38.2%). MitraScore, as a continuous variable, had a significant association with mortality and the combined end-point in follow-up, either by univariate Cox analysis (HR 1.36 [1.11-1.66] per each score point, p=0.003 for all-cause mortality, and HR 1.35 [1.11-1.65] per each score point, p=0.003 for the combined end-point), and Kaplan Meier analysis (figure 2). Discrimination by ROC analysis was nearly significant for both end-points (C statistics 0.58 [0.50-0.67], p=0.052 for all-cause mortality, and HR 0.58 [0.49-0.66], p=0.068 for the combined end-point). Conclusion The MitraScore, considered as a continuous variable, was significantly associated with mortality and the combined end-point of heart failure admission or death in follow-up in the present external validation study. These data support the use of this prognostic index in risk stratification for patients treated with TEER.
Read full abstract