AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Determinants of a Prognostically Relevant Left-to-Right Ventricular Volume Ratio from Cardiac Magnetic Resonance: The Multi-Ethnic Study of Atherosclerosis
By Hoballah M, Chehab O, Abdollahi A et al.
Background: Increased left-to-right ventricular volume ratios (LRVR) derived from cardiac magnetic resonance (CMR) independently predict adverse cardiovascular outcomes including heart failure, atrial fibrillation, and all-cause mortality in asymptomatic individuals. The determinants and mechanistic basis of this prognostically relevant biventricular remodeling phenotype remain incompletely understood. Methods: We performed an analysis of 2,932 MESA participants free of clinical cardiovascular disease who underwent baseline CMR. The percentage LRVR was defined as the ratio of LV to RV end-diastolic volumes, expressed as a percentage. Multivariable linear regression assessed associations between LRVR and demographics, cardiovascular risk factors, biomarkers of myocardial stress (NT-proBNP), indices of cardiac remodeling (LV mass-to-volume ratio [LVMVR], LV end-diastolic volume [LVEDVi] indexed to body surface area, left atrial volume [LAVi] indexed to body surface area), and subclinical atherosclerosis (LDL, HDL, Agatston calcium score). Results: Among 2,932 participants (mean age 62 years; 51% female), the mean LRVR was 103.4 ± 15.5%. After adjustment for all covariates, older age was independently associated with a higher left-to-right ventricular volume ratio (LRVR) (β = 0.14; P < 0.001), whereas male sex was associated with a lower LRVR (β = –7.90; P < 0.001). Education level was also independently associated with LRVR; participants who completed high school, technical or associate degrees, or college and higher education had lower LRVR values (β = –1.84 and –2.56, respectively; both P < 0.05). Higher systolic blood pressure (β = 0.08; P < 0.001) and elevated NT-proBNP (β = 0.003; P < 0.001) were positively associated with LRVR. Additionally, LRVR was positively associated with larger left ventricular end-diastolic volume (β = 0.56; P < 0.001) and inversely associated with left ventricular mass-to-volume ratio (β = –5.06; P < 0.001) and left atrial volume (β = –0.08; P = 0.005), consistent with a pressure-mediated remodeling phenotype. Body mass index, diabetes, lipid levels, smoking status, and coronary artery calcium score were not independently associated with LRVR. Conclusion: In a multi-ethnic cohort without clinical cardiovascular disease, LRVR ratio, a previously validated prognostic marker, was linked to demographics and hemodynamic stress signatures, but not metabolic or atherosclerotic factors. These findings support the LRVR ratio as an integrative imaging biomarker of early, pressure-mediated biventricular remodeling, providing mechanistic insight into its association with adverse outcomes.