Integrating Echocardiography Parameters With Explainable Artificial Intelligence for Data-Driven Clustering of Primary Mitral Regurgitation Phenotypes.

BACKGROUND: Primary mitral regurgitation (MR) is a heterogeneous clinical disease requiring integration of echocardiographic parameters using guideline-driven recommendations to identify severe disease. OBJECTIVES: The purpose of this preliminary study was to explore novel data-driven approaches to delineate phenotypes of MR severity that benefit from surgery. METHODS: The authors used unsupervised and supervised machine learning and explainable artificial intelligence (AI) to integrate 24 echocardiographic parameters in 400 primary MR subjects from France (n = 243; development cohort) and Canada (n = 157; validation cohort) followed up during a median time of 3.2 years (IQR: 1.3-5.3 years) and 6.8 (IQR: 4.0-8.5 years), respectively. The authors compared the phenogroups' incremental prognostic value over conventional MR profiles and for the primary endpoint of all-cause mortality incorporating time-to-mitral valve repair/replacement surgery as a covariate for survival analysis (time-dependent exposure). RESULTS: High-severity (HS) phenogroups from the French cohort (HS: n = 117; low-severity [LS]: n = 126) and the Canadian cohort (HS: n = 87; LS: n = 70) showed improved event-free survival in surgical HS subjects over nonsurgical subjects (P = 0.047 and P = 0.020, respectively). A similar benefit of surgery was not seen in the LS phenogroup in both cohorts (P = 0.70 and P = 0.50, respectively). Phenogrouping showed incremental prognostic value in conventionally severe or moderate-severe MR subjects (Harrell C statistic improvement; P = 0.480; and categorical net reclassification improvement; P = 0.002). Explainable AI specified how each echocardiographic parameter contributed to phenogroup distribution. CONCLUSIONS: Novel data-driven phenogrouping and explainable AI aided in improved integration of echocardiographic data to identify patients with primary MR and improved event-free survival after mitral valve repair/replacement surgery.

Clinical significance of myocardial contraction fraction in significant primary mitral regurgitation.

BACKGROUND: The optimal timing for mitral valve (MV) surgery in asymptomatic patients with primary mitral regurgitation (MR) remains a matter of debate. Myocardial contraction fraction (MCF) - the ratio of the left ventricular (LV) stroke volume to that of the myocardial volume - is a volumetric measure of LV myocardial shortening independent of size or geometry. AIM: To assess the relationship between MCF and outcome in patients with significant chronic primary MR due to prolapse managed in contemporary practice. METHODS: Clinical, Doppler-echocardiographic and outcome data prospectively collected in 174 patients (mean age 62 years, 27% women) with significant primary MR and no or mild symptoms were analysed. The impact of MCF< or ≥30% on cardiac events (cardiovascular death, acute heart failure or MV surgery) was studied. RESULTS: During an estimated median follow-up of 49 (22-77) months, cardiac events occurred in 115 (66%) patients. The 4-year estimates of survival free from cardiac events were 21±5% for patients with MCF <30% and 40±6% for those with ≥30% (P<0.001). MCF <30% was associated with a considerable increased risk of cardiac events after adjustment for established clinical risk factors, MR severity and current recommended class I triggers for MV surgery (adjusted hazard ratio: 2.33, 95% confidence interval: 1.51-3.58; P<0.001). Moreover, MCF<30% improved the predictive performance of models, with better global fit, reclassification and discrimination. CONCLUSIONS: MCF<30% is strongly associated with occurrence of cardiac events in patients with significant primary MR due to prolapse. Further studies are needed to assess the direct impact of MCF on patient management and outcomes.

Significance of Left Ventricular Ejection Time in Primary Mitral Regurgitation.

The optimal timing for mitral valve (MV) surgery in asymptomatic patients with primary mitral regurgitation (MR) remains controversial. We aimed at evaluating the relation between left ventricular ejection time (LVET) and outcome in patients with moderate or severe chronic primary MR because of prolapse. Clinical, Doppler echocardiographic, and outcome data prospectively collected from 302 patients (median age 61 [54 to 74] years, 34% women) with moderate or severe primary MR were analyzed. Patients were retrospectively stratified by quartiles of LVET. The primary end point of the study was the composite of need for MV surgery or all-cause mortality. During a median follow-up time of 66 (25th to 75th percentile, 33 to 95) months, 178 patients reached the primary end point. Patients in the lowest quartile of LVET (<260 ms) were at high risk for adverse events compared with those in the other quartiles of LVET (global p = 0.005), whereas the rate of events was similar for the other quartiles (p = NS for all). After adjustment for clinical predictors of outcome, including age, gender, history of atrial fibrillation, MR severity, and current recommended triggers for MV surgery in asymptomatic primary MR, LVET <260 ms was associated with an increased risk of events (adjusted hazard ratio 1.49, 95% confidence interval 1.03 to 2.16, p = 0.033). In conclusion, we observed that shorter LVET is associated with increased risk of adverse events in patients with moderate or severe primary MR because of prolapse. Further studies are required to investigate whether shorter LVET has a direct effect on outcomes or is solely a risk marker in primary MR.

Cardiac Damage Staging Classification in Asymptomatic Moderate or Severe Primary Mitral Regurgitation.

Background: Optimal timing for intervention remains uncertain in asymptomatic patients with primary mitral regurgitation (MR). We aimed to assess the prognostic value of a new cardiac damage staging classification in patients with asymptomatic moderate or severe primary MR. Methods: Clinical, Doppler-echocardiographic, and outcome data prospectively collected in 338 asymptomatic patients (64 ± 15 years, 68% men) with at least moderate primary MR were retrospectively analyzed. Patients were hierarchically classified as per the following staging classification: no cardiac damage (stage 0), mild left ventricular or left atrial damage (stage 1), moderate or severe left ventricular or left atrial damage (stage 2), pulmonary vasculature or tricuspid valve damage (stage 3), or right ventricular damage (stage 4). Results: There was a stepwise increase in 10-year mortality rates as per cardiac damage stage: 20.0% in stage 0, 25.6% in stage 1, 31.5% in stage 2, and 61.3% in stage 3-4 (p < 0.001). The staging classification was significantly associated with increased risk of mortality (hazard ratio = 1.41 per one-stage increase, 95% confidence interval: 1.07-1.85, p = 0.015) and the composite of cardiovascular mortality or hospitalization (hazard ratio = 1.51 per one-stage increase, 95% confidence interval: 1.07-2.15, p = 0.020) in multivariable analysis adjusted for EuroSCORE II, mitral valve intervention as a time-dependent variable, and other risk factors. The proposed scheme showed incremental value over several clinical variables (net reclassification index = 0.40, p = 0.03). Conclusions: The new staging classification provides independent and incremental prognostic value in patients with asymptomatic moderate or severe MR.

Forward Left Ventricular Ejection Fraction: A Simple Risk Marker in Patients With Primary Mitral Regurgitation.

BACKGROUND: The timing of mitral valve surgery in asymptomatic patients with primary mitral regurgitation (MR) is controversial. We hypothesized that the forward left ventricular (LV) ejection fraction (LVEF; ie, LV outflow tract stroke volume divided by LV end-diastolic volume) is superior to the total LVEF to predict outcomes in MR. The objective of this study was to examine the association between echocardiographic parameters of MR severity and LV function and outcomes in patients with MR. METHODS AND RESULTS: The clinical and Doppler-echocardiographic data of 278 patients with ≥mild MR and no class I indication of mitral valve surgery at baseline were retrospectively analyzed. The primary study end point was the composite of mitral valve surgery or death. During a mean follow-up of 5.4±3.2 years, there were 147 (53%) events: 96 (35%) MV surgeries and 66 (24%) deaths. Total LVEF and global longitudinal strain were not associated with the occurrence of events, whereas forward LVEF (P<0.0001) and LV end-systolic diameter (P=0.0003) were. After adjustment for age, sex, MR severity, Charlson probability, coronary artery disease, and atrial fibrillation, forward LVEF remained independently associated with the occurrence of events (adjusted hazard ratio: 1.09, [95% confidence interval]: 1.02-1.17 per 5% decrease; P=0.01), whereas LV end-systolic diameter was not (P=0.48). CONCLUSIONS: The results of this study suggest that the forward LVEF may be superior to the total LVEF and LV end-systolic diameter to predict outcomes in patients with primary MR. This simple and easily measurable parameter may be useful to improve risk stratification and select the best timing for intervention in patients with primary MR.