Abstract Body (Do not enter title and authors here):
Introduction: Sex-based differences in cardiac remodeling have been identified in patients with non-ischemic cardiomyopathy (NICM). There is conflicting data regarding sex differences in outcomes in these patients, which may be related to heterogenous risk profiles. We sought to assess sex specific modulators of risk, based on assessment with cardiac magnetic resonance (CMR) imaging.
Methods: We conducted a retrospective analysis of patients with NICM who underwent cardiac MRI between 2001 and 2017. The primary endpoint was a composite outcome of death, heart transplant, heart failure admission and LVAD. The study cohort was divided into male and female subgroups, and univariate cox regression was performed to evaluate differential risk factor profiles.
Results: We evaluated 634 patients: 255 females (40.3%) and 377 males (59.7%), see fig 1(A) for baseline demographic characteristics and CMR parameters. On Kaplan-Meier analysis, outcomes were similar in males and females (p=0.662). On Cox regression, hypertension (HR: 1.91 [CI 1.22-3.0]; p=0.005), diabetes (HR: 3.17 [CI 1.94-5.18]; p<0.001) and LGE (HR: 1.11 [CI 1.01-1.21]; p=0.030) were associated with adverse outcomes in females only. While LA ejection fraction (LAEF) (HR 0.87 [CI 0.82 - 0.93]; p<0.001) and eGFR (HR 0.88 [CI 0.80-0.95]; p=0.002) were associated with adverse outcomes in males only, see fig 2, left atrial volume index (LAVi) and LA reservoir strain were associated with the composite outcome in both males and females.
Conclusion: Despite similar outcomes in men and women with NICM, we found significant differences in the modulators of risk and cardiac structure and function based on sex. Notably, LGE and diabetes were unique predictors of risk in females, despite similar prevalence. On the other hand, LAEF and renal function were uniquely predictive of adverse outcomes in males. Further work is needed to elucidate sex specific biologic mechanisms of risk to better characterize sex specific phenotypes.