Impact of different partition values on prevalences of left ventricular hypertrophy and concentric geometry in a large hypertensive population : the LIFE study.

Left ventricular (LV) hypertrophy and concentric remodeling have been defined by using a variety of indexation methods and partition values (PVs) for LV mass and relative wall thickness (RWT). The effects of these methods on the distribution of LV geometric patterns in hypertensive subjects remain unclear. Echocardiograms were obtained in 941 patients with stage I to III hypertension and LV hypertrophy by ECG. LV mass was calculated by using different methods of indexation for body size and different PVs to identify hypertrophy: LV mass/body surface area (g/m(2)) PV for men/women 116/104, 125/110, or 125/125; LV mass/height (g/m) PV 143/102 or 126/105; and LV mass/height(2.7) (g/m(2.7)) PV 51/51 or 49.2/46.7. RWT was calculated by either 2xend-diastolic posterior wall thickness (PWT)/end-diastolic LV internal dimension (LVID) or end-diastolic interventricular septum dimension+end-diastolic PWT/end-diastolic LVID. LV hypertrophy or remodeling was present in 63% to 86% of subjects, and LV hypertrophy was present in 42% to 77%. By any index, eccentric hypertrophy was the common LV geometric pattern. Use of interventricular septum dimension+PWT/LVID to calculate RWT slightly increased the prevalence of normal geometry and eccentric hypertrophy compared with the use of 2xPWT/LVID. Subjects with LV hypertrophy identified by only LV mass/height(2.7) PV 49.2/46.7 were more obese, whereas those identified by only LV mass/body surface area PV 116/104 were taller and thinner than those in the 2 concordant groups with or without LV hypertrophy by both criteria. By either criterion, there were no significant differences between different LV geometric patterns in clinical cardiovascular disease. Hypertensive patients with LV hypertrophy by ECG have a high prevalence of geometric abnormalities, especially eccentric hypertrophy, irrespective of method of indexation or PV. LV mass indexation by body surface area or height(2.7) identifies lean and obese subjects, respectively. We found no difference in prevalent cardiovascular disease in subjects identified by either criterion, suggesting a similar high risk.