The MIDA Mortality Risk Score : Development and external validation of a prognostic model for early and late death in degenerative mitral regurgitation

Aims In degenerative mitral regurgitation (DMR), lack of mortality scores predicting death favours misperception of individual patients' risk and inappropriate decision-making. Methods and results The Mitral Regurgitation International Database (MIDA) registries include 3666 patients (age 66 ± 14 years; 70% males; follow-up 7.8 ± 5.0 years) with pure, isolated, DMR consecutively diagnosed by echocardiography at tertiary (European/North/South-American) centres. The MIDA Score was derived from the MIDA-Flail-Registry (2472 patients with DMR and flail leaflet-Derivation Cohort) by weighting all guideline-provided prognostic markers, and externally validated in the MIDA-BNP-Registry (1194 patients with DMR and flail leaflet/prolapse-Validation Cohort). The MIDA Score ranged from 0 to 12 depending on accumulating risk factors. In predicting total mortality post-diagnosis, the MIDA Score showed excellent concordance both in Derivation Cohort (c = 0.78) and Validation Cohort (c = 0.81). In the whole MIDA population (n = 3666 patients), 1-year mortality with Scores 0, 7-8, and 11-12 was 0.4, 17, and 48% under medical management and 1, 7, and 14% after surgery, respectively (P < 0.001). Five-year survival with Scores 0, 7-8, and 11-12 was 98 ± 1, 57 ± 4, and 21 ± 10% under medical management and 99 ± 1, 82 ± 2, and 57 ± 9% after surgery (P < 0.001). In models including all guideline-provided prognostic markers and the EuroScoreII, the MIDA Score provided incremental prognostic information (P < 0.002). Conclusion The MIDA Score may represent an innovative tool for DMR management, being able to position a given patient within a continuous spectrum of short- and long-term mortality risk, either under medical or surgical management. This innovative prognostic indicator may provide a specific framework for future clinical trials aiming to compare new technologies for DMR treatment in homogeneous risk categories of patients.