Appetite regulating hormones may influence compensatory increases in energy intake with exercise, although this causal relationship has been difficult to prove in a longitudinal trial.


38 subjects (29 female) aged 18 to 40 years performed aerobic exercise 6 days (6d), 2 days (2d), or 0 days per week for 12 weeks. Concentrations of ghrelin, leptin, glucagon like peptide 1 (GLP-1), and insulin were assessed before (fasting, minute 0) and after a standardized meal at minute 15, 30, 45, 60, 90, 120, 150, and 180. Linear mixed effect models were used to model the relationships between time point (12 weeks vs. baseline) and group over time (minutes 0 to 180) for each hormone. For 2d and 6d, total area under the curve (AUC) was calculated and used to predict % body fat lost and energy compensation, defined as the difference between expected weight loss (based on exercise energy expenditure, ExEE) and changes in bodily energy stores. Energy compensation was expressed as % energy compensated (compensation index, CI).


The 2d and 6d expended 1,490.7 + 122.1 and 2,750.5 + 145.1 kcal while exercising 188.8 + 4.12 and 320.5 + 3.7 min/week respectively (means + SE, P<0.01). CI did not differ between 2d and 6d (P=0.81), averaging 52%. Only 6d lost significant body fat (-7.29% + 2.13 vs -1.86% + 4.12, P=0.03). For the mixed effects model, ghrelin (P=0.03) and leptin (P<0.01) had significant group by time interactions, decreasing to a greater extent in 6d. Changes in AUC for ghrelin predicts percentage of fat loss controlling for CI and ExEE while changes in AUC for leptin predicts CI controlling for ExEE and fat loss.


12-week changes in ghrelin and leptin are influenced by exercise frequency in overweight to obese adults. Greater AUC change in ghrelin is an independent predictor of attenuated body fat loss while a greater AUC change in leptin is an independent predictor of CI. These findings represent a novel predictor of energy compensation and body fat loss with exercise.