Effect of Acute Caffeine Intake on Fat Oxidation Rate during Fed-State Exercise: A Systematic Review and Meta-Analysis

Abstract

Pre-exercise intake of caffeine (from ~3 to 9 mg/kg) has been demonstrated as an effective supplementation strategy to increase fat oxidation during fasted exercise. However, a pre-exercise meal can alter the potential effect of caffeine on fat oxidation during exercise as caffeine modifies postprandial glycaemic and insulinemic responses. Hypothetically, the effect of caffeine on fat oxidation may be reduced or even withdrawn during fed-state exercise. The present systematic review aimed to meta-analyse investigations on the effect of acute caffeine intake on the rate of fat oxidation during submaximal aerobic exercise performed in the fed state (last meal < 5 h before exercise). A total of 18 crossover trials with randomised and placebo-controlled protocols and published between 1982 and 2021 were included, with a total of 228 participants (185 males and 43 females). Data were extracted to compare rates of fat oxidation during exercise with placebo and caffeine at the same exercise intensity, which reported 20 placebo-caffeine pairwise comparisons. A meta-analysis of the studies was performed, using the standardised mean difference (SMD) estimated from Hedges' g, with 95% confidence intervals (CI). In comparison with the placebo, caffeine increased the rate of fat oxidation during fed-state exercise (number of comparisons (n) = 20; p = 0.020, SMD = 0.65, 95% CI = 0.20 to 1.20). Only studies with a dose < 6 mg/kg of caffeine (n = 13) increased the rate of fat oxidation during fed-state exercise (p = 0.004, SMD = 0.86, 95% CI = 0.27 to 1.45), while no such effect was observed in studies with doses ≥6 mg/kg (n = 7; p = 0.97, SMD = -0.03, 95% CI = -1.40 to 1.35). The effect of caffeine on fat oxidation during fed-state exercise was observed in active untrained individuals (n = 13; p < 0.001, SMD = 0.84, 95% CI = 0.39 to 1.30) but not in aerobically trained participants (n = 7; p = 0.27, SMD = 0.50, 95% CI = -0.39 to 1.39). Likewise, the effect of caffeine on fat oxidation was observed in caffeine-naïve participants (n = 9; p < 0.001, SMD = 0.82, 95% CI = 0.45 to 1.19) but not in caffeine consumers (n = 3; p = 0.54, SMD = 0.57, 95% CI = -1.23 to 2.37). In conclusion, acute caffeine intake in combination with a meal ingested within 5 h before the onset of exercise increased the rate of fat oxidation during submaximal aerobic exercise. The magnitude of the effect of caffeine on fat oxidation during fed-state exercise may be modulated by the dose of caffeine administered (higher with <6 mg/kg than with ≥6 mg/kg), participants' aerobic fitness level (higher in active than in aerobically trained individuals), and habituation to caffeine (higher in caffeine-naïve than in caffeine consumers).

Keywords: aerobic exercise; breakfast; fat metabolism; stimulant; supplementation.

Figure 1

Study selection.

Figure 2

Effect of caffeine ingestion as compared to placebo on fat oxidation rate during fed-state exercise. The forest plot shows standardised mean differences with 95% confidence intervals (CI) for 18 studies with 20 comparisons. The diamond at the bottom of the graph represents the pooled standardised mean difference with 95% CI for all trials following random effects meta-analyses. The size of the plotted squares reflects the relative statistical weight for each study. (1) first placebo-caffeine comparison within the same study. (2) second placebo-caffeine comparison within the same study.

Figure 3

Effects of acute caffeine intake compared with placebo on the rate of fat oxidation during fed-state exercise in interventions with doses: (A) <6 mg/kg; (B) ≥6 mg/kg. The graph shows standardised mean differences with 95% confidence intervals (CI). The diamond at the bottom of the graph represents the pooled standardised mean difference with 95% CI for all trials after random-effects meta-analyses. The size of the squares reflects the relative statistical weight of each study. (1) first placebo-caffeine comparison within the same study. (2) second placebo-caffeine comparison within the same study.

Figure 4

Effects of acute caffeine intake compared to placebo on the rate of fat oxidation during fed-state exercise in (A) aerobically trained individuals; (B) active/untrained. The graph shows standardised mean differences with 95% confidence intervals (CI). The diamond at the bottom of the graph represents the pooled standardised mean difference with 95% CI for all trials after random-effects meta-analyses. The size of the plotted squares reflects the relative statistical weight of each study. (1) first placebo-caffeine comparison within the same study. (2) second placebo-caffeine comparison within the same study.

Figure 5

Effects of acute caffeine intake compared with placebo on the rate of fat oxidation during fed-state exercise: (A) in caffeine-habituated individuals; (B) in non-caffeine-habituated individuals. The graph shows standardised mean differences with 95% confidence intervals (CI). The diamond at the bottom of the graph represents the pooled standardised mean difference with 95% CI for all trials after random-effects meta-analyses. The size of the plotted squares reflects the relative statistical weight of each study. (1) first placebo-caffeine comparison within the same study. (2) second placebo-caffeine comparison within the same study.

Figure 6

Effects of acute caffeine intake compared to placebo on the rate of fat oxidation during fed-state exercise when the time between the last meal and the onset of exercise was (A) ≤2 h; (B) >2 h; and (C) not specified. The graph shows standardised mean differences with 95% confidence intervals (CI). The diamond at the bottom of the graph represents the pooled standardised mean difference with 95% CI for all trials after random-effects meta-analyses. The size of the plotted squares reflects the relative statistical weight of each study. (1) first placebo-caffeine comparison within the same study. (2) second placebo-caffeine comparison within the same study.