Influence of fat-free mass on detection of appropriateness of left ventricular mass: The HyperGEN Study

Objectives: To evaluate the differences between using height^2.7 or fat-free mass for assessment of the appropriateness of left ventricular mass (LVM) in relation to hemodynamic load, and to evaluate the performance of Doppler as compared with M-mode-derived stroke volume for computation of predicted values of LVM. Design: Cross-sectional. Setting: Population-based. Participants: We studied 2299 participants from the Hypertension Genetic Epidemiology Network Study (prevalent cardiovascular disease in 342). Outcome measures: Individual predicted values of LVM were generated by equations using sex, stroke work (systolic blood pressure x stroke volume by either Doppler or M-mode) and either height^2.7 or fat-free mass, as measures of body build, in 228 normotensive, non-obese, non-diabetic participants. Observed LVM was divided by the predicted value and evaluated as 'excess of LVM'. Results: Among 1957 participants without prevalent cardiovascular disease, obese individuals (n = 1008) were slightly younger than non-obese individuals, whereas diabetic participants (n = 294) were slightly older. Excess of LVM was positively related to body mass index (BMI), independently of echocardiographic method and measure of body build, especially when height^2.7 and m-mode stroke work were used, and was greatest in the presence of concentric left ventricular hypertrophy (P < 0.0001). Excess LVM by height ^2.7 was progressively greater than that by fat-free mass, as BMI increased (P < 0.0001). In analyses of covariance of association of prevalent cardiovascular disease with age, sex, race, BMI, and excess of LVM (by each method), methods using height^2.7 were more associated with prevalent cardiovascular disease than were methods using fat-free mass (P<0.02). Conclusions: Deviation of LVM from values that compensate hemodynamic load can be similarly identified using different measures of body build and methods to generate stroke work. However, the use of height ^2.7 to compute LVM as a percentage of that predicted appears to identify deviations from compensatory values that are independently related to prevalent cardiovascular disease more effectively than does the use of fat-free mass.