Do leptin levels predict weight gain?--A 5-year follow-up study in Mauritius. Mauritius Non-communicable Disease Study Group.

Abstract

To investigate whether relative baseline leptin levels predict long-term changes in adiposity and/or its distribution. In a longitudinal study of 2888 nondiabetic Mauritians aged 25 years to 74 years who participated in population-based surveys in 1987 and 1992, changes in body mass index (BMI), waist/hip ratio (WHR), and waist circumference were compared between "hyperleptinemic," "normoleptinemic," and "hypoleptinemic" groups. "Relative leptin levels" were calculated as standardized residuals from the regression of log10 leptin on baseline BMI to provide a leptin measure independent of BMI. Analyses were performed within each sex. A linear regression model was used to assess the effect of standardized residuals on changes in BMI, WHR, and waist circumference, independent of baseline BMI, age, fasting insulin, and ethnicity. After adjusting for age and baseline BMI by analysis of covariance, there was no difference in changes in BMI, WHR, or waist circumference between men with low, normal, or high relative leptin levels. Among women, there was a significant difference in deltaWHR across leptin groups, such that the largest increase occurred in the "normal" leptin group. For both men and women, the linear regression models explained approximately 10% of variation in dependent variables, and the only significant independent variables were age, BMI, and being of Chinese origin, compared with Indian origin. These findings do not support a role for leptin concentration in predicting weight gain or changes in fat distribution in adults over a 5-year period.