Left ventricular mass-geometry and silent cerebrovascular disease: The Cardiovascular Abnormalities and Brain Lesions (CABL) study

Background Although abnormal left ventricular geometric patterns have prognostic value for morbidity and mortality, their possible association with silent cerebrovascular disease has not been extensively evaluated. Methods We examined 665 participants in the CABL study who underwent transthoracic echocardiography and brain magnetic resonance imaging. Participants were divided into 4 geometric patterns: normal geometry (n = 397), concentric remodeling (n = 89), eccentric hypertrophy (n = 126), and concentric hypertrophy (n = 53). Subclinical cerebrovascular disease was defined as silent brain infarcts (SBIs) and white matter hyperintensity volume (WMHV; expressed as log-transformed percentage of the total cranial volume). Results Silent brain infarcts were observed in 94 participants (14%). Mean log-WMHV was −0.97 ± 0.93. Concentric hypertrophy carried the greatest risk for both SBI (adjusted odds ratio [OR] 3.39, P < .001) and upper quartile of log-WMHV (adjusted OR 3.35, P < .001), followed by eccentric hypertrophy (adjusted ORs 2.52 [P = .001 for SBI] and 1.96 [P = .004] for log-WMHV). Concentric remodeling was not associated with subclinical brain disease. In subgroup analyses, concentric and eccentric hypertrophies were significantly associated with SBI and WMHV in both genders and nonobese participants, but differed for SBI by age (all ages for eccentric hypertrophy, only patients ≥70 years for concentric hypertrophy) and by race-ethnicity (Hispanics for eccentric hypertrophy, blacks for concentric hypertrophy; no association in whites). Conclusions Left ventricular hypertrophy, with both eccentric and concentric patterns, was significantly associated with subclinical cerebrovascular disease in a multiethnic stroke-free general population. Left ventricular geometric patterns may carry different risks for silent cerebrovascular disease in different sex, age, race-ethnic, and body size subgroups.