Effect of growth on variability of left ventricular mass: assessment of allometric signals in adults and children and their capacity to predict cardiovascular risk

Abstract

Objectives: We sought to determine whether growth influences the relation between left ventricular mass and body size and whether use of different body size indexes affects the ability of ventricular mass to predict complications of hypertension.

Background: Allometric (or growth) signals between left ventricular mass and height have recently been reported to improve previous approaches for normalization of ventricular mass for body size.

Methods: Residuals of left ventricular mass-height2.7 relations were analyzed in a learning series of 611 normotensive, normal-weight subjects 4 months to 70 years old and, separately, in 383 children (< 17 years old) and 228 adults. Ten-year cardiovascular morbidity in a test series of 253 hypertensive adults was compared with groups with normal or high baseline left ventricular mass normalized for body weight, height, body surface area and allometric powers of height.

Results: The dispersion of residuals of ventricular mass versus height2.7 increased with increasing height or age in children but not in adults, suggesting that the effect of other variables on ventricular growth increases during body growth and stabilizes in adulthood. Therefore, we derived separate allometric signals for adults (predicted ventricular mass = 45.4 x height2.13, r = 0.48) and children (32.3 x height2.3, r = 0.85) (both p < 0.0001). Patients with left ventricular hypertrophy had 3.3 times higher cardiac risk with elevated left ventricular mass/height2.7 (p < 0.001), 2.6 to 2.7 times higher risk with left ventricular mass indexed for height, height2.13 and body surface area (all p < 0.01) and 1.7 times the risk with ventricular mass/weight (p > 0.1).

Conclusions: These results show the following: 1) Variability of left ventricular mass in relation to height increases during human growth; 2) allometric signals of left ventricular mass versus height are lower in adults and children than those obtained across the entire age spectrum; 3) height-based indexes of left ventricular mass at least maintain and may enhance prediction of cardiac risk by hypertensive left ventricular hypertrophy; and 4) the allometric signal derived across the entire spectrum of age appears to be more useful for prediction of cardiovascular risk than that computed in adults.