Early puberty: adulthood metabolic consequences and childhood nutritional determinants

Abstract

University of Minnesota Ph.D. dissertation. July 2013. Major: Epidemiology. Advisor: Mark A. Pereira. 1 computer file (PDF); x, 198 pages.

Background: Accumulating evidence suggests puberty is occurring earlier today than ever before. Pubertal timing may be a harbinger for abnormal metabolic health. Thus, identifying its upstream determinants and downstream metabolic health consequences may provide an avenue to primordial prevention. The objective of this dissertation was to investigate pubertal timing in relation to type 2 diabetes (T2D), adipose depots, non-alcoholic fatty liver disease (NAFLD), and early-life nutrition.
 Methods: This dissertation includes three separate but related research manuscripts. The first aimed to investigate pubertal timing and T2D in a Brazilian adult population (35-74y) who were born and came of age before the rise of overweight and obesity in Brazil. The second examined pubertal timing in relation NAFLD and fat depots in a biracial cohort of young adults (18-30 years at entry, 42-56 years at measurement of NAFLD) from the Coronary Artery Risk Development In Young Adults (CARDIA) study. Finally, the third paper determined the role of diet quality and risk of early pubertal timing in a biracial sample of girls (9-10 years) from the National Heart, Lung, and Blood Institute Growth and Health Study (NGHS). Multivariable linear regression was used for continuous outcomes (glucose homeostasis measures, adipose depots, liver attenuation, and age at menarche) and Poisson regression to generate prevalence ratios (PR) and relative risks (RR) for dichotomous outcomes (T2D, NAFLD, and early menarche). In the last project, a diet quality score, derived by summing standard deviation scores for the densities of nutrients/chemicals indicative of prudent or Western dietary patterns, was analyzed in relation to incident early menarche.
 Results: In the first paper: early menarche [<11 years vs. 13-14 years (reference)] was associated with 22% (95% confidence interval: 3-42%) higher prevalence of T2D in Brazilian adults, adjusted for confounders and BMI at age 20 years. In the same model, a 1-standard deviation higher relative leg length (RLL), a marker of early life growth and maturation, was associated with a 10% (5-15%) and 11% (7-15%) lower prevalence of T2D in women and men, respectively. The association between RLL and T2D was stronger among females with earlier menarche (p for interaction on multiplicative scale = 0.02), and among adults who were overweight or obese at age 20 years (p for interaction on multiplicative scale = 0.02) or 35-74 years (p for interaction on multiplicative scale = 0.03). In the second paper: in CARDIA, a 1-year increment in age at menarche was associated with 10% (2-18%) lower prevalence of NAFLD, and inversely associated with visceral, subcutaneous, and intra-abdominal muscle fat, after adjusting for confounders and BMI measured at 18-30 years. RLL was also associated with these measures, but was attenuated after control for BMI. Lastly, in the third paper: a better diet quality was associated with lower risk of early menarche (<11 years) after adjustment for confounders and % body fat from skinfolds (1-SD increment in diet quality score, RR=0.77; 95% CI: 0.66-0.90).
 Conclusions: These results indicate that early menarche is associated with higher prevalence of diabetes in an adult Brazilian population that underwent nutritional transition after coming of age, and non-alcoholic fatty liver disease in a biracial U.S. population, after adjusting for potential early-life confounders and early adulthood BMI. Relative leg length, while only loosely associate with age at menarche, was inversely associated with T2D but not NAFLD. Age at menarche and relative leg length did not appear to be preferentially related to specific adipose depots. These markers likely represent unique aspects of early-life growth and maturation, and depend on the context of the population under study. The results of the final research paper suggest that risk for early menarche may be reduced through dietary modification, independent of fatness. In light of the secular trend toward earlier age at puberty in the U.S. and abroad, it is important that future research continues in this arena under the realm of primordial prevention. Continued research is needed to determine the extent to which early-life modifiable lifestyle factors, such as dietary patterns and physical activity, can influence long-term health and disease through effects on pubertal timing.