Myocardial structure and function by echocardiography in relation to glucometabolic status in elderly subjects from 2 population-based cohorts: A cross-sectional study

Abstract

Left ventricular (LV) diastolic dysfunction has been associated with impaired glucometabolic status. However, studies of older subjects are lacking. We examined associations between echocardiographic indices of LV diastolic function and LV mass index (LVMI) and glucometabolic status among middle-aged and elderly subjects free from heart disease, hypothesizing that the associations would be comparative to younger cohorts. We examined the Age Gene/Environment Susceptibility Reykjavik Study (Iceland; n = 607, 76 +/- 6 years) and the Malmö Preventive Project Re-Examination Study (MPP-RES) cohorts (Sweden; n = 1,519, 67 +/- 6 years), evaluating associations with multivariable regression analysis. In the Age Gene/Environment Susceptibility Reykjavik Study, LVMI was positively correlated with glycosylated hemoglobin (HbA1c) (P = .001). Otherwise, echocardiographic variables were not associated with glucometabolic status. In the MPP-RES, LVMI increased with increasing glucometabolic disturbance among both older (70-80 years) and middle-aged (57-69 years) subjects. Among older subjects, HbA1c was positively correlated with 2 variables reflecting LV diastolic function: late transmitral peak flow velocity (A) (P = .001) and early transmitral peak flow velocity (E)/early diastolic peak tissue velocity (Em) (P = .046). In middle-aged MPP-RES subjects, increasing glucometabolic disturbance was correlated with increasing late diastolic peak tissue velocity (Am) (P = .002) and, after age adjustment, with increasing A (P = .001) and decreasing Em/Am (P = .009). With age adjustment, Am and A were positively correlated with fasting glucose and HbA1c. Contrary to our hypothesis, in 2 independent cohorts of older individuals, associations between glucometabolic status and LV diastolic function were generally weak. These contrast with previous reports, as well as with observations among middle-aged subjects in the present study. Changes in LV diastolic function may be more age-related than associated with glucose metabolism in older subjects.