Objective: The aim of the present study was to identify predictors of weight loss in obese men participating in a 2-year behaviour modification programme.
Design: Longitudinal, clinical intervention study of a behaviour modifying weight loss program.
Setting: University Hospital, Stockholm, Sweden.
Subjects: Forty-four obese men (age, 42.7 +/- 1.1 years: BMI, 37.1 +/- 0.6 kg m(-2), mean +/- SEM) followed for 2 years.
Interventions: Behaviour modification weight loss programme.
Main outcome measures: Associations between plasma leptin and thyroid function tests, insulin resistance by homeostatic model assessment (HOMA), dietary recall and anthropometrically determined body composition.
Results: At baseline, there were significant correlations between plasma leptin and body mass index (BMI), fat-free mass (FFM) and insulin resistance. Median weight loss over 2 years was 4.9 kg (range, -27.2 to +11.9). Baseline serum leptin concentrations adjusted for BMI (leptin/BMI ratio) were significantly correlated with 2-year weight change (r = 0.34, P = 0.04). A subset of seven of the 44 men gained weight over the 2 years. These 'gainers' differed significantly in initial leptin/BMI ratio (0.62 +/- 0.07) compared with the 37 'losers' (0.42 +/- 0.03, P < 0.05). In a multiple regression model, baseline leptin, insulin and age predicted 22% of the variance in weight change with no additional significant contribution from BMI, FFM, waist:hip ratio, thyroid function tests or energy intake. There was a strong correlation between the change in leptin concentrations and the change in insulin resistance from baseline to 2-year follow-up (r = 0.54; P < 0.001).
Conclusion: Baseline plasma leptin concentrations predicted long-term weight loss. Inappropriate leptin secretion or disposal, corrected for BMI, was associated with failure to maintain weight loss in obese men in a behaviour modification weight loss programme.