Sex Differences in Left Ventricular Remodeling for Risk Stratification of Patients With Aortic Regurgitation

Importance: Left ventricular (LV) dilatation is an established prognosticator in aortic regurgitation (AR). Current guidelines recommend aortic valve surgery (AVS) using LV end-systolic diameter index (LVESDi) with a uniform threshold, irrespective of sex. While LV end-systolic volume index (LVESVi) may better characterize LV remodeling, it was only recently included in European guideline recommendations, with a threshold of 45 mL/m2 for both men and women. Objective: To assess sex differences in LV remodeling using linear and volumetric dimensions and their association with outcomes in AR. Design, setting, and participants: This was a multicenter cohort study of patients with moderate-severe AR and preserved LV ejection fraction (LVEF) between December 2003 and December 2022, with a median (IQR) follow-up of 7 (4-11) years. The study took place at 5 centers in the Netherlands, Singapore, Hong Kong, Canada, and Romania. Patients with at least moderate-severe AR and preserved LVEF (≥50%) were included. Those with symptoms, acute AR, significant other valvular disease, or prior valve surgery were excluded. Data were analyzed from January to November 2024. Exposure: LV dilatation assessed by LVESDi and LVESVi. Main outcomes and measures: All-cause mortality during medical management and following AVS. Results: A total of 808 patients (mean [SD] age, 56 [19] years; 488 men and 320 women) were included, 323 of whom underwent AVS. Mean (SD) baseline LVESDi did not differ between sexes (women: 20 [5] mm/m2 vs men: 20 [4] mm/m2; P = .77), whereas men had larger mean (SD) LVESVi (39 [16] mL/m2 vs 31 [15] mL/m2; P < .001). During follow-up under medical management, 74 patients died. Adjusted 6-year survival was lower in women (80% vs 89%; P = .001). Receiver operating characteristic curve analysis identified LVESDi 20 mm/m2 or greater for both sexes, LVESVi 40 mL/m2 or greater for women, and LVESVi 45 mL/m2 or greater for men as thresholds associated with mortality. These cutoffs were validated using age-adjusted cubic splines and remained associated with outcomes after multivariable adjustment, with a differential effect by sex for LVESVi but not for LVESDi. After AVS, survival did not differ by sex (85% women vs 89% men; P = .31). Only preoperative LVESVi was associated with mortality, with a significant sex interaction (HR, 1.03; 95% CI, 1.00-1.06; P = .04). Conclusions and relevance: In this study among individuals with moderate-severe AR, similar LVESDi thresholds (20 mm/m2) for both sexes, but lower than currently recommended by guidelines, were independently associated with mortality. In turn, LVESVi thresholds were 40 mL/m2 for women and 45 mL/m2 for men, suggesting the need for sex-specific cutoffs to improve risk stratification.