Michelena, H

Abstract Aims Prognosis in tricuspid regurgitation (TR) is shaped by complex clinical and haemodynamic interactions, complicating risk stratification. We aimed to use explainable artificial intelligence (EAI) to model these relationships and identify key predictors of outcomes. Methods and results We analysed long-term all-cause mortality in a large prospective registry of patients with native TR across the full severity spectrum. Patients underwent detailed clinical evaluation and quantitative Doppler-echocardiography. Supervised EAI models were applied to identify and rank predictors of mortality under medical therapy, capturing non-linear associations and interactions among baseline characteristics. Among 9761 patients [mean age 73 ± 15 years; 56.9% women; effective regurgitant orifice area (EROA) 40 (30–56) mm2; regurgitant volume 39 (29–53) mL/beat; 42% severe TR], 2676 deaths occurred during 3.8 ± 4.6 years of follow-up. EAI identified and ranked 20 prognostic determinants of survival. Haemodynamic measures, led by elevated pulmonary-artery systolic and right-atrial pressures, were strongest, followed by age and comorbidity. TR-related variables ranked next, led by right ventricular dysfunction and EROA, which outperformed integrative TR grade and TRIScore. Left-ventricular parameters were less influential. Non-linear effects, including those involving ejection fraction, and interactions such as higher EROA-related risk with elevated pulmonary pressures or lower TRIScore (both P < 0.03) were observed. Proportional hazards models confirmed EROA as the strongest TR severity marker (P < 0.0001), with excess mortality beginning at 40 mm2 and increasing further at 60 mm2. Conclusion EAI ranked mortality predictors in TR, highlighting patient factors, haemodynamics, TR-specific measures, and left-ventricular function. Among TR-specific measures, EROA and TRIScore were most prognostic. Recognizing their impact can refine risk stratification, guide intervention timing, and inform long-term management.

Abstract Background Left ventricular (LV) remodeling in patients with valvular heart disease may vary according to the type of overload (pressure vs volume) as well as patients’ characteristics and comorbidities. Different LV remodeling patterns can be defined based on either an increase in LV mass index (LVMI), relative wall thickness (RWT), or both, and have been largely studied in patients with aortic valve disease but not in patients with primary mitral regurgitation (PMR). Purpose To describe the prevalence and baseline characteristics of the different LV remodeling patterns in PMR patients undergoing mitral valve (MV) surgery, and their association with outcome. Methods Four LV remodeling patterns were identified before MV surgery according to LVMI and RWT (Fig1-a). A cutoff of >115 g/m2 in males or >95 g/m2 in females was used to define high LVMI, and >0.42 was used to define high RWT. Patients with more than mild concomitant aortic stenosis were excluded. The study outcome was all-cause mortality. Results A total of 1538 patients were included, of which 362 (23.5%) had normal LV geometry (NG), 774 (50.3%) had eccentric hypertrophy (EH), 286 (18.6%) had concentric hypertrophy (CH), and 116 (7.5%) had concentric remodeling (Fig 1-a). Compared to the NG group (Fig 2), the EH group had more females, with larger LV size and left atrial volume index (LAVI). Compared to the NG group, patients in the CH group were significantly older and had more proportion of females. Moreover, they were more symptomatic, had higher prevalence of arterial hypertension, atrial fibrillation, and worse renal function, with smaller LV end-diastolic diameters and thicker LV walls.. Compared to the NG group, the CR group included patients with significantly smaller LV cavities and thicker LV walls, but interestingly similar LAVI and systolic pulmonary pressure, suggesting that this group might represent patients with a rather acute onset MR. Of note, there was no difference among the groups as regards LV systolic dysfunction (ejection fraction ≤ 60%, P=0.314). However, patients with CH and CR had significantly worse e’ and E/e’ ratio. After a median follow-up of 6.2 years, 211 (13.7%) patients died. Kaplan-Meier analysis showed similar survival between the NG and EH groups (91.4 vs 91.6%, P = 0.964), with significantly lower survival in the CH (76.6%, P<0.001) and CR (81.9%, P=0.028) groups, compared to NG (Fig 1-b). On multivariable analysis, after correcting for all the significant variables (Fig 1-c), CH and CR patterns carried higher risk of mortality (HR 1.7, P=0.024 and 2.0, P=0.017) compared to NG, while EH carried no additional risk (P=0.768). Conclusion LV remodeling patterns help in phenotyping patients with PMR undergoing MV surgery, and are independently associated with different risks of post-operative mortality.Figure 1:Prevalence and outcome Figure 2:Baseline characteristics