Association between subclinical right ventricular alterations and aerobic exercise capacity in type 2 diabetes

Background: Type 2 diabetes (T2D) leads to cardiovascular remodeling, and heart failure has emerged as a major complication of T2D. There is a limited understanding of the impact of T2D on the right heart. This study aimed to assess subclinical right heart alterations and their contribution to aerobic exercise capacity (peak oxygen consumption; peak VO2) in adults with T2D. Methods: Single center, prospective, case-control comparison of adults with and without T2D, and no prevalent cardiac disease. Comprehensive evaluation of the left and right heart was performed using transthoracic echocardiography and stress cardiovascular magnetic resonance. Cardiopulmonary exercise testing on a bicycle ergometer with expired gas analysis was performed to determine peak VO2. Between group comparison was adjusted for age, sex, race, and body mass index using analysis of covariance (ANCOVA). Multivariable linear regression, including key clinical and left heart variables, was undertaken in people with T2D to identify independent associations between measures of right ventricular (RV) structure and function with peak VO2. Results: Three hundred and forty people with T2D (median age 64 years, 62% (211) male, mean glycated hemoglobin (HbA1c) 7.3%) and 66 controls (median age 58 years, 58% (38) male, mean HbA1c 5.5%) were included. T2D participants had markedly lower peak VO2 (adjusted mean 20.3 (95% confidence interval (CI): 19.8-20.9) vs 23.3(22.2-24.5) mL/kg/min, P < 0.001) than controls and had smaller left ventricular (LV) volumes and LV concentric remodeling. Those with T2D had smaller RV volumes (indexed RV end-diastolic volume: 84 (82-86) vs 100 (96-104) mL/m, P < 0.001) with evidence of hyperdynamic RV systolic function (global longitudinal strain (GLS): 26.3 (25.8-26.8) vs 23.5 (22.5-24.5)%, P < 0.001) and impaired RV relaxation (longitudinal peak early diastolic strain rate (PEDSR): 0.77 (0.74-0.80) vs 0.92 (0.85-1.00) s-1, P < 0.001). Multivariable linear regression demonstrated that RV end-diastolic volume (j3 = -0.342, P = 0.004) and RV cardiac output (j3 = 0.296, P = 0.001), but not LV parameters, were independent determinants of peak VO2. Conclusion: In T2D, markers of RV remodeling are associated with aerobic exercise capacity, independent of left heart alterations.