Thermic effect of a meal and appetite in adults: an individual participant data meta-analysis of meal-test trials

BACKGROUND Thermic effect of a meal (TEF) has previously been suggested to influence appetite. OBJECTIVE The aim of this study was to assess whether there is an association between appetite and TEF. Second, to examine whether protein intake is associated with TEF or appetite. DESIGN Individual participant data (IPD) meta-analysis on studies were performed at the Department of Nutrition, Exercise and Sports, University of Copenhagen, Denmark. Five randomized meal-test studies, with 111 participants, were included. The included studies measured energy expenditure (EE) in respiration chambers and pre- and postprandial appetite sensations using Visual Analog Scales (VAS). The primary meta-analysis was based on a generic-inverse variance random-effects model, pooling individual study Spearman's correlation coefficients, resulting in a combined r-value with 95% confidence interval (95% CI). The I (2) value quantifies the proportion (%) of the variation in point estimates due to among-study differences. RESULTS The IPD meta-analysis found no association between satiety and TEF expressed as the incremental area under the curve (TEFiAUC) (r=0.06 [95% CI -0.16 to 0.28], P=0.58; I (2)=15.8%). Similarly, Composite Appetite Score (CAS) was not associated with TEFiAUC (r=0.08 [95% CI -0.12 to 0.28], P=0.45; I (2)=0%). Posthoc analyses showed no association between satiety or CAS and TEF expressed as a percentage of energy intake (EI) (P>0.49) or TEF expressed as a percentage of baseline EE (P>0.17). When adjusting for covariates, TEFiAUC was associated with protein intake (P=0.0085). CONCLUSIONS This IPD meta-analysis found no evidence supporting an association between satiety or CAS and TEF at protein intakes ∼15 E% (range 11-30 E%).