Reply to PL Valenzuela et al.

Dear Editor:

We appreciate the opportunity to respond to the Letter to the Editor in which Valenzuela et al. (1) questioned the statement from our systematic review that based on heterogeneity across studies there is currently insufficient evidence to warrant recommendations of using ketone supplements to enhance physical performance (2). The authors suggest that the high level of heterogeneity observed in our work was due to the inclusion of 1 chronic study (3) along with 9 (4–12) acute ketone supplement and physical performance studies. They go on to state that no heterogeneity was detected in a recent meta-analysis assessing the impact of acute ketone supplementation on physical performance by their group (13). Due to the limited number of published studies and the narrative nature of our systematic review (2), we believe that the inclusion of the work by Poffé et al. (3) was appropriate and that proper context with regards to discrepancies in chronic compared with acute ketone supplementation was provided. In addition, even if the Poffé et al. (3) study was excluded, a high level of variance in performance outcomes remains between acute ketone supplementation studies. Specifically, Cox et al. (4) reported a significant 2% increase, whereas Leckey et al. (7) reported a significant 2% decrease in physical performance after acute ketone supplementation. In an effort to assess the degree of heterogeneity between acute ketone supplement and performance studies, we have updated our model and present results below.

A total of 10 studies (4–12, 14) were included in this meta-analysis. Nine studies (4–12) captured in our original systematic review (2), and 1 additional study (14) which was included in the work by Valenzuela et al. (13), were analyzed. To match the Valenzuela et al. (13) study, only the main performance outcome variables were assessed in the current meta-analysis (Table 1). Although not included in this analysis, it should be corrected that the fatigue index which was reported by Waldman et al. (12) as W/s should have been reported as ΔW/s, resulting in a negative, not positive, effect of ketone supplementation on performance. Effect sizes (ESs) for differences in physical performance outcomes were determined as standard mean differences between ketone supplementation and controls divided by pooled SD. Meta-Essentials by Van Rhee et al. (15) with Microsoft Excel 2010 (Microsoft Corp.) was used to conduct the meta-analysis. Inputted data included sample sizes, performance outcome means and SDs, and a correlation coefficient for within-participant measurements. When individual data were available, the correlation coefficient (r) for the specific study was used. When a correlation coefficient could not be determined for a specific study, the mean average (r = 0.95) across all studies was used. To assess heterogeneity, both Q and I² statistics were determined (16). To account for high levels of heterogeneity and small sample bias, random effects were applied and ES was generated as Hedges’ g (16). Meta-analysis was also conducted by supplement type: ketone ester or ketone salt/precursors. Performance data are presented as ES and 95% CI.

TABLE 1.

Description of the ketone supplements, performance outcomes, and exercise modes included in the meta-analysis¹

Reference	Ketone supplement	Performance outcome	Mode
Cox et al. (4)2	R-3-hydroxybuty1-R 3-hydroxybutyrate monoester (576 mg/kg) + CHO (110 g) vs. CHO (isocaloric)	Distance traveled during 30-min time trial	Cycle ergometer
Dearlove et al. (14)2	R-3-hydroxybuty1-R 3-hydroxybutyrate monoester (330 mg/kg) vs. placebo	Lower-body power	Cycle ergometer
Evans et al. (6)2	R-3-hydroxybuty1 R-3-hydroxybutyrate monoester (573 mg/kg) + CHO (1 g/min exercise) vs. CHO (1 g/min exercise)	10-km time trial	Treadmill
Evans and Egan (5)2	R-β-hydroxybutyrate-R 1,3-butanediol monoester (750 mg/kg) + CHO (90 g) vs. CHO (90 g)	20-m shuttle run time to exhaustion	Outdoor track
Leckey et al. (7)2	1,3-butanediol acetoacetate diester (500 mg/kg) vs. placebo	31.2-km time trial	Cycle ergometer
O'Malley et al. (8)3	β-hydroxybutyrate salt (300 mg/kg) vs. placebo	10-km time trial	Cycle ergometer
Rodger et al. (9)3	β-hydroxybutyrate salt (11.7 g) vs. placebo	Lower-body power	Cycle ergometer
Scott et al. (10)3	1,3-butanediol (500 mg/kg) + CHO (60 g) vs. CHO (isocaloric)	5-km time trial	Treadmill
Shaw et al. (11)3	R,S-1,3-butanediol (350 mg/kg × 2) vs. placebo	Time trial	Cycle ergometer
Waldman et al. (12)3	β-hydroxybutyrate salt (11.38 g) vs. placebo	Lower-body power	Cycle ergometer

¹CHO, carbohydrate.

²Ketone ester supplement.

³Ketone salt or precursor.

Main findings indicate that there was no difference in overall performance when participants consumed a ketone supplement as compared with control (Figure 1). Similarly, there was no difference in performance outcomes when consuming a ketone ester or ketone salt/precursor as compared with control (Figure 1). High levels of heterogeneity (Q = 91.53, I² = 90%) were detected across these 10 studies. High levels of heterogeneity were detected in studies providing a ketone ester (Q = 78.78, I² = 95%) and to a lesser extent ketone salts/precursors (Q = 12.74, I² = 69%).

FIGURE 1.

Meta-analysis of impact of ketone supplements on physical performance. Values are ES, 95% CI. Data were derived from healthy participants consuming a ketone supplement or control to determine the impact on physical performance. Individual white circles represent the ESs of ketone esters. Individual black circles represent the ESs of ketone salts/precursors. Grey diamonds are the aggregated data from the 2 supplement types. The white diamond is the aggregated data from all studies. ES, effect size.

The lack of overall performance benefit with ketone supplementation in the current meta-analysis is in agreement with findings from Valenzuela et al. (13). Despite similar results in overall performance outcomes, discrepancies remain in the severity of heterogeneity. Discordant results in heterogeneity between the current and the past meta-analysis (13) can likely be attributed to the models used, and the data inputted into the models to generate ESs and 95% CIs. It is worth noting that when comparing ESs and 95% CIs, results from the current meta-analysis agree with findings from the primary studies from which these data were generated. Specifically, the CIs in the current analysis did not cross the 0-ES line, indicative of a significant finding, for studies reporting a significant increase (4) or decrease (7, 12) in physical performance when consuming a ketone supplement as compared with control. Conversely, Valenzuela et al. (13) reported that the CIs in all 10 studies of their meta-analysis crossed the 0-ES line, suggesting no significant increases or decreases in physical performance when consuming a ketone supplement compared with control in individual studies. Agreement in results from primary articles in the current meta-analysis suggests our model was appropriate.

In conclusion, overall data suggest there is no ergogenic effect of consuming ketone supplements as compared with control. However, high levels of heterogeneity across ketone supplement performance studies persist, despite removal of the Poffé et al. (3) study from our model. High levels of heterogeneity suggest that caution is warranted when attempting to interpret aggregated results from multiple studies (17). Outcomes from this meta-analysis do not change the overall conclusion from our systematic review (2). Currently there is insufficient evidence to warrant a recommendation of ketone supplement use to enhance physical performance.

Notes

Supported by the US Army Medical Research and Development Command.

Author disclosures: The authors report no conflicts of interest.

The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Army or the Department of Defense. Any citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement or approval of the products or services of these organizations.