Caffeine and Bicarbonate for Speed. A Meta-Analysis of Legal Supplements Potential for Improving Intense Endurance Exercise Performance

Abstract

A 1% change in average speed is enough to affect medal rankings in intense Olympic endurance events lasting ~45 s to 8 min which for example includes 100 m swimming and 400 m running (~1 min), 1,500 m running and 4000 m track cycling (~4 min) and 2,000 m rowing (~6-8 min). To maximize the likelihood of winning, athletes utilizes legal supplements with or without scientifically documented beneficial effects on performance. Therefore, a continued systematic evidence based evaluation of the possible ergogenic effects is of high importance. A meta-analysis was conducted with a strict focus on closed-end performance tests in humans in the time domain from 45 s to 8 min. These test include time-trials or total work done in a given time. This selection criterion results in a high relevance for athletic performance. Only peer-reviewed placebo controlled studies were included. The often applied and potentially ergogenic supplements beta-alanine, bicarbonate, caffeine and nitrate were selected for analysis. Following a systematic search in Pubmed and SportsDiscuss combined with evaluation of cross references a total of 7 (beta-alanine), 25 (bicarbonate), 9 (caffeine), and 5 (nitrate) studies was included in the meta-analysis. For each study, performance was converted to an average speed (km/h) from which an effect size (ES; Cohens d with 95% confidence intervals) was calculated. A small effect and significant performance improvement relative to placebo was observed for caffeine (ES: 0.41 [0.15-0.68], P = 0.002) and bicarbonate (ES: 0.40 [0.27-0.54], P < 0.001). Trivial and non-significant effects on performance was observed for nitrate (ES: 0.19 [-0.03-0.40], P = 0.09) and beta-alanine (ES: 0.17 [-0.12-0.46], P = 0.24). Thus, caffeine's and bicarbonate's ergogenic effect is clearly documented for intense endurance performance. Importantly, for all supplements an individualized approach may improve the ergogenic effect on performance.

Keywords: cycling; ergogenic aids; intense exercise; performance; rowing; running; supplements; swimming.

Figure 1

Flow chart displaying number of included and excluded studies for the evaluation of effects on intense endurance performance lasting 45 s to 8 min following supplementation with beta-alanine, bicarbonate, caffeine and nitrate using Medline (Pubmed) as search engine (see Methods for details regarding study criteria required for inclusion).

Figure 2

Quality assessment of studies used for meta-analysis studying the effects on intense endurance performance following supplementation with beta-alanine, bicarbonate, caffeine and nitrate. Bars denote the relative distribution of studies for each supplements with a high or low risk of bias. See Methods for details regarding criteria for high and low risk.

Figure 3

Standardized mean difference (effect size) with 95% confidence intervals as reported in studies of beta-alanine supplementations effect on performance (average speed) relative to pre supplementation in athletic endurance events lasting 45 s to 8 min. The combined effect size has been calculated as described in methods section. All other data is reported as mean ± SD. All performance tests were executed indoors on ergometers or treadmills unless stated differently. Regarding Dose: Dose in each study is displayed as days (d) of consumption with the daily dose as gram pr day (g/d). ^#Speed estimated from mean power reported.^£Group mean reported (n = 14); no info on placebo (n = 7) and beta-alanine (n = 7) group separately.

Figure 4

Standardized mean difference (effect size) with 95% confidence intervals as reported in studies of bicarbonate supplementations impact on performance (average speed) relative to placebo in athletic endurance events lasting 45 s to 8 min. The combined effect size has been calculated as described in methods section. All other data is reported as mean ± SD. All performance tests were executed indoors on ergometers or treadmills unless stated differently. Regarding Dose & Timing: Dose is listed as gram pr kilo bodyweight (g/kg) and “60 min” denotes time from ingestion to start of test. “30 + 60 min” denotes a 30 min period to consume capsules followed by additional 60 min before start of test and “240 & 120 min” denotes intake of capsules 240 and 120 min before start of test. For studies with chronic loading the amount of days (d) is listed as well as timing of last dose before performance test. ^#Speed estimated from mean power reported. ^***Significant effect of bicarbonate on performance (P < 0.001).

Figure 5

Standardized mean difference (effect size) with 95% confidence intervals as reported in studies of caffeine supplementations impact on performance (average speed) relative to placebo in athletic endurance events lasting 45 s to 8 min. The combined effect size has been calculated as described in methods section. All other data is reported as mean ± SD. All performance tests were executed indoors on ergometers or treadmills unless stated differently. Regarding Dose & Timing: Dose is listed as milligram pr. kilo bodyweight (mg/kg) and “60 min” denotes time for consumption before the performance test. ^#Speed estimated from mean power reported.^£Four of the 10 subjects were also subjects in part 1 of the study. ^**Significant effect of caffeine on performance (P < 0.01).

Figure 6

Standardized mean difference (effect size) with 95% confidence intervals as reported in studies of nitrate supplementation impact on performance (average speed) relative to placebo in athletic endurance events lasting 45 s to 8 min. The combined effect size has been calculated as described in methods section. All other data is reported as mean ± SD. All performance tests were executed indoors on ergometers unless stated differently. Regarding Dose & Timing: Dose is listed millimoles (mmol) and “140 min” denotes time for consumption before the performance test. ^#Speed estimated from mean power reported.

Figure 7

Association between performance test time and supplement induced change in performance (average speed in athletic events lasting 45 s to 8 min). Supplements investigated was beta-alanine (r² = 0.07), bicarbonate (r² = 0.1922, P < 0.05), caffeine (r² = 0.45, P < 0.05) and nitrate (r² = 0.0001).

Figure 8

Combined effect size with 95% confidence interval displaying the impact on performance (average speed) in athletic endurance events lasting 45 s to 8 min from supplementation with either beta-alanine (7 studies) relative to pre supplementation, bicarbonate (25 studies), caffeine (9 studies) or nitrate (5 studies) relative to placebo. Hatched lines display boundaries for effect sizes being trivial (<0.2), small (0.2–0.6) or large (>0.6). ^***Significant effect of bicarbonate on performance (P < 0.001). ^**Significant effect of caffeine on performance (P < 0.01).