When it comes to exercise in a competitive or professional setting, research has focused on developing nutritional strategies that optimize fuel selection during training and competition, thereby allowing for optimal or even enhanced performance. The majority of these strategies revolve around the availability of carbohydrates (CHO), knowing that this is the primary energy source during high-intensity exercise and is required to sustain longer duration or endurance exercise of higher intensities (Cermak & van Loon, 2013; Romijn et al., 1993). Reductions in CHO availability (i.e., muscle and liver glycogen stores) are associated with fatigue and performance impairment. As such, an approach to spare endogenous CHO reserves is physiologically attractive (Cermak & van Loon, 2013; Jeukendrup & Jentjens, 2000). Examples of popular nutritional strategies include CHO loading and CHO feeding. However, given the limited storage and rate of absorption of CHO, one might ask whether another method may be a more efficient alternative (Coggan & Coyle, 1985; Coyle et al., 1983).
Ketone bodies (KB) have recently sparked interest as one such alternative to spare CHO. Produced predominantly in the liver through the process of ketogenesis, KBs (beta-hydroxybutyrate [ßHB], acetoacetate [AcAc], and acetone) are capable of acting as a substrate for most tissues including the brain, heart, and muscle when glucose is otherwise unavailable (Robinson & Williamson, 1980). While there are three different types of KBs, ßHB is the most commonly occurring form with the largest contribution as an energy substrate (Harvey et al., 2019; Laffel, 1999). Ketogenesis occurs more frequently under physiological conditions such as periods of fasting, prolonged exercise, or when adopting a very low-CHO (ketogenic) diet (Cox & Clarke, 2014; Robinson & Williamson, 1980). Under all of these circumstances, KBs become a pertinent energy source while glucose is restricted or depleted. Ketosis, or hyperketonaemia, is defined as plasma ketone concentrations that exceed 0.2 mM, according to Robinson and Williamson (1980). Other researchers have defined it as ßHB concentrations that reach greater than 1 mM (Laffel, 1999; Mitchell et al., 1995). While no universal definition appears to exist, a plasma concentration that exceeds normal circulating levels of approximately 0.1 mM is acceptable as hyperketonaemia. Initially, research in this field concentrated on endogenous KB production by implementing the ketogenic diet and subsequently assessing exercise performance (Harvey et al., 2019; Shaw, Merien, Braakhuis, Maunder, & Dulson, 2019; Zajac et al., 2014; Zinn et al., 2017). It was quickly determined that reduced or depleted CHO stores, often observed in those adopting the ketogenic diet, are not beneficial to performance, despite the presence of KBs as an alternative energy source (Burke et al., 2017). In fact, it has been shown to be detrimental to endurance exercise performance during which CHO availability becomes a limiting factor (Harvey et al., 2019).
Accordingly, the latest development of exogenous forms of KBs has given researchers the ability to investigate their effects without compromising CHO reserves (Clarke et al., 2012). In other words, they allow for the inclusion of glycogen replete individuals, which would otherwise not be possible with ketogenic diet-induced ketosis. There have since been claims that exogenous ketone metabolism is preferential over both CHO and fat by effectively suppressing glycolysis during exercise workloads that typically favor CHO oxidation (Cox et al., 2016). However, it is unclear whether this effect allows for a preservation of CHO stores to be utilized at a later time or instead creates an impairment of CHO oxidation during exercise. Given the increase in research on exogenous KBs, a systematic review would provide a comprehensive overview of observed effects compared to nutritive or nonnutritive control options.
It is important to note that these exogenous ketones come in various forms: ketone monoesters (KME), ketone diesters, ketone salts (KS), or ketone precursors. Ketone esters are predominantly formulated as ßHB monoesters, given that ßHB is the most commonly occurring KB (Harvey et al., 2019; Laffel, 1999). This monoester combines the acid ßHB, with the alcohol hydroxybutyl, to produce a compound that is easily broken down and absorbed in the small intestine and then further metabolized in the liver to provide two molecules of ßHB (Sivva et al., 2016). It is, therefore, very effective at elevating ketone concentrations in the blood. Research has shown that participants consuming this KME can reach a 50% higher circulating concentration than when consuming KS that contain equivalent amounts of ßHB (Stubbs et al., 2017). These salts are composed of KBs, typically ßHB, bound to either sodium, potassium, or calcium salts (Evans et al., 2016; Stubbs et al., 2017). They often contain both d-ßHB and l-ßHB isoforms, the latter of which are less readily oxidized and therefore may not be as effective at increasing ketone concentrations (Stubbs et al., 2018). Given their composition, they are also more likely to cause gastrointestinal (GI) distress in large quantities (Evans et al., 2016; Stubbs et al., 2017). Additionally, a lesser known ketone diester combines a different KB, AcAc, with butanediol to deliver two AcAc molecules and one molecule of racemic ßHB once ingested (Stubbs et al., 2018). AcAc diesters have not been used in many studies, likely because they are currently far less palatable and have lower GI tolerability than both the ßHB monoesters and KS (Leckey et al., 2017). There are further a small number of studies that utilize a ßHB precursor known as 1,3-butanediol. This precursor produces one molecule of ßHB and is easily absorbed, similar to KME. Previous systematic reviews included studies using all forms of exogenous ketones and identified that the majority of studies using KS or ketone diesters had null or even ergolytic effects (Margolis & O’Fallon, 2020; Valenzuela et al., 2020). For the above-mentioned reasons, studies utilizing KS and diesters will be excluded from this review, making this the first systematic review of its kind to focus solely on the effects of ketone precursors and monoesters on endurance performance.
In theory, KBs are a practical alternative energy source with the potential to spare CHO reserves when consumed in an exogenous form. What is unknown is whether this novel development is truly worthwhile for use in endurance exercise endeavors. As such, this systematic review will aim to determine whether exogenous KME or precursors have the capacity to enhance endurance exercise performance, specifically, does consumption result in faster time trials (TTs) or extended time to exhaustion (TTE). Secondary outcomes include assessing: (a) GI effects, (b) the relationship between ketone dose and resulting blood concentration, and (c) the relationship between ketone concentration and changes in performance.
Methods
Search Strategy
This systematic review was completed in accordance with PRISMA guidelines (Page et al., 2021) and was registered on PROSPERO (CRD42020218915). The literature search took place between November 2020 and May 2021 and included three different databases: PubMed, SPORTDiscus, and CINAHL. A search strategy was developed with the key terms “humans,” “ketosis,” and “exercise performance” (see Table 1). Retrieved studies were first screened for inclusion by evaluating titles and abstracts by two independent reviewers. A third reviewer was consulted if there were any unresolved discrepancies. Next, studies were assessed in their full text by two independent reviewers for final inclusion in the review. There were no restrictions on publication date or language. Reference lists from the relevant publications were manually searched for any articles missed by the database searches.
Exact Search Terms Used for Electronic Searches of the Following Databases: PubMed, SPORTDiscus, and CINAHL
Database | Search strategy |
---|---|
PubMed | (Humans) OR (Men)) OR (Women)) OR (Athletes)) (Ketosis) OR (Ketone bodies)) OR (Ketone supplement)) OR (Ketone ester)) OR (Exogenous ketones)) OR (Ketone*) (Exercise performance) OR (Physical performance)) OR (Endurance) |
SPORTDiscus | (Humans) OR (People) (Exogenous ketones) OR (Ketone*)) OR (Ketosis) (Endurance performance) OR (Exercise)) OR (Training)) OR (Running)) OR (Cycling)) OR (Time trial)) OR (Physical performance) |
CINAHL | (Exogenous ketones) OR (Ketone bodies)) OR (Ketone esters)) OR (Ketone supplement) (Physical performance) OR (Endurance)) OR (Time trial)) OR (Running)) OR (Cycling)) OR (Training)) OR (Exercise) |
Note. Within each database, search strategies were combined with AND.
Inclusion Criteria
Studies were eligible for inclusion in the review if they were randomized crossover or parallel controlled trials evaluating the effects of exogenous ketones on endurance exercise performance in humans. They also required a nutritive or nonnutritive control containing no ketones. Only ketones in the form of either KME or ketone precursors that were ingested prior to and/or during the exercise session rather than postexercise (for recovery) or chronically (over the course of more than 1 day) were included. There were no restrictions on the ketone dose, study duration, sample size, or participant age, sex, body mass, or training status.
Exclusion Criteria
Studies were excluded from the analysis if they used animal models or ketogenic diets rather than exogenous ketones as the intervention. Studies using ketone diesters or KS were also excluded for reasons previously covered. Other reviews, commentaries, and editorial articles were not included.
Data Extraction
Data were manually extracted from the included studies onto a standardized excel spreadsheet. Sex, age, body weight, training status, and VO2max were obtained from each study to provide descriptive characteristics of the participants. Performance test outcomes from the experimental and control groups were isolated for the meta-analyses. The type and dose of ketone supplement along with the peak blood concentration during the exercise session were recorded for regression analysis. The remaining data were extracted for descriptive purposes and included the mode of exercise, the exercise test and duration, the time of administration of the supplement, the control intervention, GI effects, and whether the participants were in a fed or fasted state for the exercise session.
Quality Assessment
Cochrane’s tool for assessing risk of bias was used to analyze the quality of the studies included in the review (Higgins et al., 2011). Ratings of low, high, or unclear risk of bias were assigned to each study for the following criteria: random sequence of generation and allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), and selective reporting (reporting bias).
Statistical Analysis
A meta-analysis was performed to assess the effects of exogenous ketone precursors or monoesters versus a control on endurance performance. This consisted of both an overall analysis of all studies as well as separate subanalyses consisting of ≥3 studies that measured performance using the same type of test. The standardized mean difference (Hedge’s g) between experimental and control trials and 95% CI were computed using a random-effects model. A random-effects model and standardized effect sizes were chosen because the included studies had independent operating researchers with differing units of measurement reported, making it unlikely that they were functionally equivalent. The absolute difference in means of the experimental and control trials, the standard deviation (SD) of the difference of the paired measurements (sd), and the standard error (SE) were used to compute the standardized mean difference. The sd was calculated in one of four ways: (a) using Cohen’s d, the sd is equal to the mean of the paired measurements divided by Cohen’s d; (b) using the length of the confidence interval (CI), the sd is equal to the square root of the sample size multiplied by the length of confidence interval, which is then divided by two times the t0.025 quantile from the t(n−1) distribution; (c) using the p value, the t-test statistic can first be determined, after which the sd is equal to the mean of the paired measurements divided by the value of the t-test statistic, which is then multiplied by the square root of the sample size; and (d) using the correlation between the paired measurements, the sd is equal to the square root of the following: the SD of one paired measurement (s1) squared, plus the SD of the other paired measurement (s2) squared, minus two times the correlation multiplied by s1s2 (Hogg & Tanis, 2010). There was one study for which we were not able to compute the sd, so we imputed the correlation of the paired measurements by using the median correlation from the remaining studies. The sd was then computed the same way as above (d). The SE was calculated by dividing the sd by the square root of the sample size. Tau and I2 statistics, which estimate the SD of underlying effects across studies and describe the percentage of variation across studies that is due to heterogeneity rather than sampling error (chance), respectively, were computed to assess heterogeneity between studies (Higgins & Thompson, 2002). Forest plots were generated to visualize the data. Meta-analyses were performed using the software Comprehensive Meta-Analysis (version 3.0; Biostat Inc., Englewood, NJ) with a significance level of .05. SPSS software (Version 28.0.1.0; IBM Corp, Armonk, NY) was used for all correlations. Spearman correlations were run for the comparisons between ketone dose and ßHB concentration and between percent change in ßHB concentration and percent change in performance with a significance level of .05.
Results
Study and Participant Characteristics
The literature search resulted in 569 retrieved articles, of which eight studies met the criteria to be included in the systematic review (Cox et al., 2016; Evans & Egan, 2018; Evans et al., 2019; Poffé et al., 2020a, 2020b, 2020c; Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019). A flow diagram of study retrieval and selection is presented in Figure 1. All participant and study characteristics are presented in Tables 2 and 3. A total of 80 individuals (77 men and three women) participated in these studies, all of which followed a crossover design, and were published between 2016 and 2020 and included representation from the United Kingdom, Ireland, Belgium, and New Zealand. Sample sizes ranged from eight to 12 participants. All individuals were trained athletes whose average age ranged from 25 to 38 years. Ketone supplements were provided in the form of KME or precursors only. The ketone dose ranged from approximately 500 to 922 mg/kg of body weight and the total exercise duration ranged between approximately 80 and 195 min of activity either in the form of cycling or running. Of the eight studies, three required the participants to consume the ketone supplement and undergo the performance test in a fasted state (Cox et al., 2016; Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019). All but one study (Shaw, Merien, Braakhuis, Plews, et al., 2019) provided sufficient CHO to equate to approximately 1–1.2 g/min alongside ketone supplementation. Three different outcomes under the scope of endurance performance were identified across the eight studies: four assessing TTE (Evans & Egan, 2018; Poffé et al., 2020a, 2020b, 2020c), three assessing TT completion for time (Evans et al., 2019; Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019), and one assessing TT completion for distance (Cox et al., 2016). For this reason, separate subanalyses were completed for the TTE outcome and the TT for time outcome.
—PRISMA flow diagram for study selection.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 32, 3; 10.1123/ijsnem.2021-0280
Characteristics of Studies and Participants Included in the Review
Study | Reference | Sample size | Population | Age (year) | Weight (kg) | VO2max (mL·kg−1·min−1) |
---|---|---|---|---|---|---|
1 | Cox et al. (2016) | 8 (6 M, 2 F) | Elite endurance athletes | 29.4 ± 1.0 | 84.9 ± 5.2 | M, 5.37 ± 0.3 L/min F, 3.3 ± 0.1 L/min |
2 | Evans and Egan (2018) | 11 (11 M, 0 F) | Team-sport athletes | 25.4 ± 4.6 | 78.6 ± 5.3 | 53.9 ± 2.2 |
3 | Scott et al. (2018) | 11 (11 M, 0 F) | Trained runners | 38.0 ± 12.0 | 67.3 ± 6.5 | VO2peak = 64.2 ± 5.0 |
4 | Evans et al. (2019) | 8 (7 M, 1 F) | Trained runners | 33.5 ± 7.3 | 68.8 ± 9.7 | 62.0 ± 5.6 |
5 | Shaw, Merien, Braakhuis, Plews, et al., (2019) | 9 (9 M, 0 F) | Trained cyclists | 26.7 ± 5.2 | 69.6 ± 8.4 | VO2peak = 63.9 ± 2.5 |
6 | Poffé et al. (2020b) | 12 (12 M, 0 F) | Cyclists and triathletes | 25.0 ± 6.0 | 72.0 ± 8.0 | 62.4 ± 6.6 |
7 | Poffé et al. (2020c) | 12 (12 M, 0 F) | Trained cyclists | 26.0 ± 6.0 | 70.0 ± 7.0 | 62.5 ± 5.5 |
8 | Poffé et al. (2020a) | 9 (9 M, 0 F) | Trained cyclists | 29.0 ± 5.0 | 71.0 ± 7.0 | 61.0 ± 2.9 |
Note. Values are represented as mean ± SD. M = males; F = females.
Ketone Supplement, Control Intervention, and Exercise Parameters for Each Study Included in the Review
Reference | Type of supplement | Dose (mg/kg) | Peak [ßHB] (mM) | Control/placebo | Mode of exercise | Exercise test | Exercise duration (min) | Fed vs. fasted | CHO provision in ketone trial |
---|---|---|---|---|---|---|---|---|---|
Cox et al. (2016) | (R)−3-hydroxybutyl (R)−3-hydroxybutyrate KE | 573 | 2.5 | Isocaloric, taste-matched CHO drink | Stationary cycling | 1 hr 75% Wmax + 30 min TT for distance | 90 | Fasted overnight | 60% cals from CHO (1.2 g/min) |
Evans and Egan (2018) | ß-Hydroxybutyrate (R) 1,3-butanediol KE | 750 | 2.61 | Taste-matched 6.4% CHO-electrolyte solution | Running indoors | Loughborough intermittent shuttle test | ∼80 | Fed (3 hr before exercise) | 6.4% CHO-electrolyte solution (1.2 g/min) |
Scott et al. (2018) | 1,3-butanediol ketone precursor | 500 | 1.0 | Isocaloric CHO drink | Treadmill running | 1 hr 75% VO2peak + 5 km TT | ∼80 | Fasted overnight | 60 g CHO |
Evans et al. (2019) | (R)−3-hydroxybutyl (R)−3-hydroxybutyrate KE | 573 | 1.33 | Taste-matched 8% CHO-electrolyte solution | Treadmill running | 1 hr 65% VO2max + 10 km TT | ∼90 | Fed (2 hr before exercise) | 8% CHO-electrolyte solution (1 g/min) |
Shaw, Merien, Braakhuis, Plews, et al., (2019) | 1,3-butanediol ketone precursor | 700 | 0.75 | Orange-flavored drink | Stationary cycling | 85 min ∼73% VO2peak + TT equivalent to 7 kJ/kg | ∼115 | Fasted overnight | None |
Poffé et al. (2020b) | (R)−3-hydroxybutyl (R)−3-hydroxybutyrate KE | 918 ± 102 | 3.3 | Taste-matched collagen peptan and water | Stationary cycling | 3 hr submaximal intermittent + 15 min TT + sprint to exhaustion | ∼195 | Fed (2 hr before exercise) | 500 mL CHO drink + energy bar (60 g/hr) |
Poffé et al. (2020c) | (R)−3-hydroxybutyl (R)−3-hydroxybutyrate KE | 726 ± 75 | 3.5 | Taste-matched collagen peptan and water | Stationary cycling | 1 hr warm-up + 30 min TT + sprint to exhaustion | ∼90 | Fed (2 hr before exercise) | Energy gel + CHO solution (60 g/hr) |
Poffé et al. (2020a) | (R)−3-hydroxybutyl (R)−3-hydroxybutyrate KE | 922 ± 85 | 3.0 | Taste-matched collagen peptan and water | Stationary cycling | 3 hr submaximal intermittent + 15 min TT + sprint to exhaustion | ∼195 | Fed (2 hr before exercise) | Energy gel + CHO solution (60 g/hr) |
Note. Peak [ßHB] signifies the highest concentration reached during exercise. KE = ketone ester; TT = time trial; CHO = carbohydrate.
Endurance Performance
All of the eight studies assessed running or cycling performance. When comparing endurance performance among all included studies, no significant differences were found between the ketone and control trials (Figure 2; Hedge’s g = 0.136; 95% CI [−0.195, 0.467]; p = .419) with evidence of moderate to substantial interstudy heterogeneity (I2 = 56.888%; p = .023; T = 0.354). We further subanalyzed results based on the type of endurance test. Four of the studies evaluated endurance by means of TTE (Evans & Egan, 2018; Poffé et al., 2020a, 2020b, 2020c), three by means of time to completion of a TT (Evans et al., 2019; Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019), and one by maximum distance achieved in a TT (Cox et al., 2016). TTE sessions consisted of sprints to exhaustion (sprints lasted between approximately 50 s and 4 min), that immediately followed endurance bouts lasting between 80 and 195 min. No significant differences were found in endurance performance between the ketone and control trials based on TTE values (Hedge’s g = −0.002; 95% CI [−0.312, 0.308]; p = .989) with some evidence of heterogeneity (I2 = 19.144%; p = .294; T = 0.139). TTs also followed preload endurance bouts with total exercise time lasting between 80 and 115 min. TT distances ranged from 5 to 10 km. No significant differences were found in endurance performance between the ketone and control trials based on TT values (Hedge’s g = 0.057; 95% CI [−0.282, 0.395]; p = .744) with no significant evidence of heterogeneity (I2 = 0%; p = .664; T = 0). One study was not included in the subanalyses as there were no other studies assessing maximum distance in a TT as the endurance test (Cox et al., 2016). In this study, participants on average cycled 2% (411 ± 162 m) further in a 30-min TT that followed 60 min of exercise at 75% Wmax, after consuming a KME supplement compared to the control trial (Cox et al., 2016).
—Effects of ketone supplementation on overall endurance performance.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 32, 3; 10.1123/ijsnem.2021-0280
Ketone Dose and ßHB Concentrations
All studies reported peak blood or plasma ßHB concentrations during exercise of above 0.2 mM, indicating that participants reached a state of ketosis after supplementation. Figure 3b shows the correlation (rs = .68, p = .06) between average mg/kg ketone dose and peak concentration. Overall, higher ketone doses tended to result in higher peak concentrations. Figure 3a shows the two studies utilizing ketone precursors have lower peaks than those using KME, despite one implementing a relatively high dose (Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019). All but one of the studies using KME surpassed the hypothesized ergogenic threshold, reaching peak concentrations well over 2 mM, whereas neither study using ketone precursors met this threshold.
—(a) Peak ßHB concentration in plasma or blood for each included study. Ketone dose is shown in brackets below each study (mg/kg). (b) Average ketone supplement dose (mg/kg) as a function of peak ßHB concentration (mM). Dotted lines represent the hypothesized threshold that must be reached in order to see ergogenic effects (2 mM) (Clarke et al., 2012; Cox et al., 2016; Hashim & VanItallie, 2014; Stubbs et al., 2018) rs = .68, p = .06. (c) Percent increase from preketone to peak blood ßHB concentration as a function of percent change in performance of ketone trial compared to control for each study rs = .048, p = .91.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 32, 3; 10.1123/ijsnem.2021-0280
In order to further investigate the relationship between ketones and performance, we compared the change in ßHB concentrations to the change in performance among the ketone trials in all studies. Figure 3c shows the correlation (rs = .048, p = .91) between percent increase in performance of the ketone trial over control and the percent increase in blood ßHB concentrations.
GI Effects
Seven of the eight studies assessed incidence of symptoms of GI distress following ingestion of the ketone supplements (Table 4). Six of the seven studies reported higher levels of GI distress in the ketone trial compared to the control (Evans & Egan, 2018; Evans et al., 2019; Poffé et al., 2020a, 2020b, 2020c; Shaw, Merien, Braakhuis, Plews, et al., 2019), and the other indicated no significant difference in symptoms between trials (Scott et al., 2018). Two of the studies (Evans & Egan, 2018; Evans et al., 2019) used interviews to assess the presence or lack of presence of symptoms following the trials, which ranged from belching, cramps, flatulence, reflux, urge to defecate, and nausea, all of which were seen in both the ketone and control trials, to stitches, heartburn, and vomiting, which were seen only in the ketone trials. Three of the studies (Poffé et al., 2020a, 2020b, 2020c) used questionnaires which separated symptoms into systemic, lower abdominal, and upper abdominal categories and had participants rate each symptom on a scale of 0–8. One study (Scott et al., 2018) also used a Likert scale questionnaire, though participants simply rated their GI comfort on a scale of 0–10 at various timepoints throughout the trial, with 0 being very comfortable and 10 being extremely uncomfortable. Finally, one study (Shaw, Merien, Braakhuis, Plews, et al., 2019) used a 27-item questionnaire, which had participants rate the presence of symptoms from low to high. All but one study (Scott et al., 2018) recorded incidences of GI disturbances within 10 min of completion of each trial. One of the eight studies (Cox et al., 2016) did not record any information regarding GI discomfort within the trials.
Method of Measurement Used and Incidence of GI Symptoms in Participants in Ketone Trials Compared to the Control Trials
Study reference | Method of measurement | Results | Incidence of GI symptoms in ketone trial (compared to CON) |
---|---|---|---|
Cox et al. (2016) | N/A | N/A | N/A |
Evans and Egan (2018) | Interview | Ketone—9 of 11 participants reported symptoms CON—4 of 11 | Increased |
Scott et al. (2018) | Likert scale questionnaire | GI comfort = 2 ± 2 out of 10 for Ketone and CON | No difference |
Evans et al. (2019) | Interview | Ketone—5 of 8 participants reported symptoms CON—4 of 8 | Increased |
Shaw, Merien, Braakhuis, Plews, et al. (2019) | Questionnaire | Ketone—5 of 9 reported low/moderate belching, one reported severe abdominal pain CON—no similar symptoms reported | Increased |
Poffé et al. (2020b) | Likert scale questionnaire | Total GI discomfort (out of 96): Ketone = 13 ± 10 CON = 12 ± 12 | Increased |
Poffé et al. (2020c) | Likert scale questionnaire | Total GI discomfort (out of 96): Ketone = 12 ± 12 CON = 7 ± 12 | Increased |
Poffé et al. (2020a) | Likert scale questionnaire | Total GI discomfort (out of 96): Ketone = 16 ± 11 CON = 14 ± 13 | Increased |
Note. N/A = not applicable; CON = control; GI = gastrointestinal.
Quality Assessment
Cochrane’s tool for assessing risk of bias revealed that the majority of the studies included in the review presented with low risks of performance, detection, attrition, and reporting bias, while selection bias was largely unclear across the studies (Table 5).
Risk of Bias for Publications Included in the Review
Reference | Selection bias | Performance bias | Detection bias | Attrition bias | Reporting bias | |
---|---|---|---|---|---|---|
Random sequence generation | Allocation concealment | Blinding of participants/personnel | Blinding of outcome assessment | Incomplete outcome data | Selective reporting | |
Cox et al. (2016) | L | L | L | H | L | U |
Evans and Egan (2018) | U | U | L | L | L | L |
Scott et al. (2018) | U | U | L | L | U | L |
Evans et al. (2019) | U | U | L | L | L | L |
Shaw, Merien, Braakhuis, Plews, et al. (2019) | L | U | H | H | L | L |
Poffé et al. (2020b) | H | U | L | L | L | L |
Poffé et al. (2020c) | L | U | L | L | L | L |
Poffé et al. (2020a) | L | U | L | L | L | L |
Note. H = high; L = low; U = unclear.
Discussion
The present meta-analysis results suggest that acute ingestion of exogenous ketones, both in the form of precursors and monoesters, do not significantly impact endurance exercise performance. This lack of effect exists when evaluating all studies together as well as through subanalysis based on type of endurance test. Furthermore, correlational analyses did show a trend between ketone dose ingested and the resulting peak in plasma concentrations (p = .06), though this was not present between percent increase in ßHB concentrations and resulting percent changes in performance (p = .82). Records of GI incidences indicate an increased number of GI symptoms reported in ketone trials compared to control in three studies and higher ratings of discomfort on a Likert scale in ketone trials among three of the four remaining studies. This suggests that GI symptoms or discomfort may be greater with ingestion of exogenous ketones compared to a control.
Previous reviews of the literature have examined all types of exogenous ketones and both endurance and power performance (Margolis & O’Fallon, 2020; Valenzuela et al., 2020). While Margolis and O’Fallon (2020) state the effects of exogenous ketones on physical performance to be inconclusive, Valenzuela et al. (2020) concluded that ketone supplementation has no effect on exercise performance. Still, both discuss the impact that different ketone types may have on performance, mainly through their ability to raise ßHB concentrations and their effects on GI function. Studies using KS and ketone diesters have shown consistently null or negative effects to date and were therefore excluded from the present review (Evans et al., 2018; James and Greer, 2018; Leckey et al., 2017; O’Malley et al., 2017; Rodger et al., 2017). Ketone precursors have lacked attention in the literature and were even grouped in with the ketone esters by Valenzuela et al. (2020). Our analysis suggests that the ketone precursor 1,3-butanediol is not as efficient as KME at raising ßHB concentrations. Scott et al. (2018) and Shaw, Merien, Braakhuis, Plews, et al. (2019) reached peak concentrations of only 1.0 and 0.75 mM, respectively, compared to the average of ∼2.7 mM among the studies using KME (Cox et al., 2016; Evans & Egan, 2018; Evans et al., 2019; Poffé et al., 2020a, 2020b, 2020c), which is far from the hypothesized ergogenic threshold of 2 mM (Clarke et al., 2012; Cox et al., 2016; Hashim & VanItallie, 2014; Stubbs et al., 2018). This smaller change in concentration is likely due to a difference in the delivery of ketone equivalents previously discussed by Stubbs et al. (2018). Ketone precursors deliver just one equivalent compared to the two provided by KME.
While no statistically significant relationship was found between dose and resulting peak concentration, this may be due to many factors, including the small number of studies, all with small sample sizes. Regardless, Figure 3b shows a clear trend in the data suggesting that higher doses lead to higher peak concentrations. Interestingly, this nonsignificant relationship persisted when comparing only studies using KME (p = .14). Also important to note, we found no significant relationship between percent increase in ketone concentration and percent increase in the ketone over control trial performance. This outcome suggests that increased levels of ketosis do not equate to larger improvements in performance, or more ketones in circulating concentration do not lead to a greater ergogenic effect. Though, this would be worthwhile to investigate in a larger sample of studies.
Another factor that affects resulting ßHB levels is whether the individual consuming the ketones is in a fasted or fed state. It is known that ingesting ketones immediately following a meal diminishes the peak blood ßHB concentration thereafter (Stubbs et al., 2015, 2017). With exercise, peak concentrations become further reduced due to the increase in metabolism (Cox et al., 2016). Among the studies included, Cox et al. (2016) achieved a peak ßHB concentration of about 2.5 mM in participants who were fasted overnight. Interestingly, the remaining two studies with participants in a fasted state achieved the lowest peak concentrations of ßHB (Scott et al., 2018; Shaw, Merien, Braakhuis, Plews, et al., 2019). These were also the only two to utilize ketone precursors, suggesting that ketone supplement type may have more of an impact on resulting concentration. Evans et al. (2019) performed the only study that utilized KME and did not achieve a peak ßHB concentration above the 2 mM threshold. The fed state of the participants likely played a role in these results, as Evans et al. (2019) and Cox et al. (2016) administered the same dose of the same KME; however, participants in the Cox et al. (2016) study were fasted (peak =∼2.5 mM), whereas those in the Evans et al. (2019) study were fed (peak = ∼1.3 mM). Regardless, consumption of ketone supplements under fasted conditions is not practical for real-life competition where individuals follow nutritional routines meant to optimize their fuel stores (Burke et al., 2015). All four of the remaining studies (Evans & Egan, 2018; Poffé et al., 2020a, 2020b, 2020c) managed to surpass the 2 mM threshold of ßHB concentration, with participants all fed 2–3 hr before the exercise session. Thus, it is possible to achieve target ßHB concentrations with appropriate timing and dosing even after meal ingestion. Based on recommendations for precompetition nutrition, strategies like this could be further investigated in future studies (Burke et al., 2015).
While records of GI effects showed that ingestion of KME or precursors caused a larger number or a greater severity of GI disturbances than reported in the control condition across the majority of studies, it should be recognized that seven of the eight studies included data on GI comfort. What is more, three reported incidences of symptoms and four reported ratings of discomfort on a scale, making it difficult to compare results across all seven studies. The data were not statistically analyzed and therefore any conclusions should be interpreted with caution. Fasted consumption of larger doses or poor palatability of the formulated drink may have contributed to an increase in symptoms. For example, Shaw, Merien, Braakhuis, Plews, et al. (2019) noted, all participants in their study disliked the taste of the ketone precursor. Overall uncertainty in effects on GI function is consistent with previous studies, which have found that it is primarily context-dependent (Stubbs et al., 2019). Dose, time of administration, and exercise duration and intensity are all factors that can affect the appearance and severity of GI symptoms, not unlike CHO consumption during exercise (Stubbs et al., 2019). Based on the ketone trials among the included studies, it is unclear whether higher doses equate to higher incidences of GI symptoms. More data are required to corroborate this statistically. Stubbs et al. (2019) found that symptom load and severity did not differ between drinks containing KME combined with CHO and isocaloric CHO when consumed during 195 min of exercise, suggesting that KE and CHO together may improve GI tolerability. Nonetheless, ketone ingestion will likely need to be adjusted based on individual preference similar to other substances consumed before or during exercise.
Regardless, it is the contribution of ketone utilization to overall fuel oxidation during exercise that is ultimately indicative of its potential influence on performance. Previous measurements of this involvement, such as those executed by Cox et al. (2016), utilized adjusted indirect calorimetry equations to estimate the contribution of exogenous ketones to overall oxidation. In their experiment consisting of 45 min of exercise at either 40% or 75% of Wmax following 573 mg/kg KME plus CHO ingestion, they found that ketone oxidation accounted for 16%–18% of total oxygen consumption during the exercise session. It is likely that these equations have led to overestimates of contribution, however, as a recent study by Dearlove et al. (2021) that used whole-body ketone tracers showed that ketone oxidation contributed minimally (∼4.46%) to overall energy expenditure during incremental intensity exercise lasting 60 min. Interestingly, they found that a higher dose of KME (752 mg/kg), while effectively doubling blood ßHB concentrations (∼2 vs. ∼4.4 mM), did not significantly increase ßHB oxidation rates compared to a lower dose (252 mg/kg). Additionally, ßHB oxidation peaked during the lowest exercise intensity (25% Wmax) after ingestion of both the high dose and low dose. Dearlove et al. (2021) attribute the consistent oxidation rates despite higher circulating ßHB concentration to the metabolic flexibility of skeletal muscle, proposing that reliance will be on the oxidation of lipids in order to preserve ketone oxidation for cerebral metabolism. Ketone oxidation in skeletal muscle may increase between ∼fivefold and ∼11-fold during exercise compared to at rest, suggesting that it is not a matter of limited ketone transport or ketolytic capacity, but rather a tight regulation on skeletal muscle ketone oxidation in order to reserve energy for peripheral tissues (Balasse et al., 1978; Mikkelsen et al., 2015). These speculations are in accordance with findings from Féry and Balasse (1986), who discovered that KB metabolism during exercise is a function of the initial degree of ketosis, with higher ketone levels leading to a greater reduction of the stimulatory effect of exercise on the Ra and MCR of KBs. These along with the recent findings by Dearlove et al. (2021) further suggest that at higher plasma levels, KBs may be preferentially utilized by nonmuscular tissues such as, and likely primarily, the brain. These results may also explain why there appears to be a lack of effect of acute exogenous KB consumption on performance as concluded in the present review.
Strengths and Limitations
This systematic review followed current PRISMA guidelines and included both a meta-analysis and multiple supplementary regression analyses. We performed a search of numerous databases as well as hand-searched for additional studies to ensure an exhaustive group of studies was included. This review is the first of its kind to assess only studies using KME and precursors. Although meta-analyses were performed, the small number of included studies and their small samples is a limitation and speaks to further research directions. Previous reviews encountered the same concerns. Margolis and O’Fallon (2020) included a total of 10 studies and Valenzuela et al. (2020), just 13. Given that this review intended to narrow the inclusion criteria based on type of ketone supplement and that this is still a relatively new area of research, the search resulted in an even smaller number of trials. We recognize that the inclusion of only eight studies creates less confidence in our conclusions and may have also influenced certain aspects of our study like the heterogeneity testing, as it has been shown that chi-squared tests have low power in situations with studies that have small sample sizes or are few in number (Higgins & Thompson, 2002). Therefore, while our test shows a statistically significant result of substantial heterogeneity (p = .023), this should be interpreted with caution. In fact, for this reason a p value of .10 is often used, in which case our results become nonsignificant (Higgins & Thompson, 2002). Regardless, the included studies varied widely in aspects of their methodologies, making it challenging to compare or detect commonalities between trials. For example, an important element that we wanted to assess between studies was the timing of ketone dose administration because timing can potentially affect ßHB concentration, GI function, and ultimately performance. The Appendix shows how greatly this aspect varied across the studies, with each one implementing a different method of administration. Not only did the dose vary, but so did the number of boluses that the overall dose was divided into, the percentage ratio of these boluses, and their time of administration related to the exercise session. Similarly, discrepancies in exercise duration and intensity, along with consistency of co-ingestion with CHO made it difficult to effectively compare the studies as a whole and therefore became a limiting factor to this review.
Future Directions
Future studies on the use of exogenous ketone supplements for endurance exercise performance enhancement should take into account the efficiency of KME at raising blood ßHB concentrations over other ketone supplement types such as precursors, diesters, or salts. It has also been shown that sufficient levels can be reached with prior meal ingestion and/or co-ingestion with CHO. These aspects should be considered along with attention to palatability and time of administration to effectively standardize and assess the effects of ketone supplementation while maintaining principles that we know to be beneficial to performance. More consistency in methodologies across studies is required to make definitive conclusions. Additionally, studies could aim to include more female participants as there is currently a large sample size discrepancy in the sex of individuals being studied. Of 80 participants that took part in the eight studies included in this review, only three participants were female. This lack of sex diversity poses more research questions regarding whether there are potential differences in the effects of exogenous ketones on endurance performance between males and females. To fully understand exogenous ketone metabolism, we will need an equal distribution of male and female athletes in studies.
Conclusions
Based on the results from this meta-analysis, we conclude that acute ingestion of exogenous ketone precursors and monoesters do not significantly improve endurance exercise performance. The overwhelming majority of studies show null effects when compared to both nutritive and nonnutritive controls. Future studies may consider making trials more applicable to real-life competition by feeding participants beforehand and coordinating the time of administration to minimize GI disturbances. More research is warranted to strengthen conclusions and better understand exogenous ketone metabolism specifics during endurance exercise.
Acknowledgments
The authors wish to thank the authors of the papers included in this systematic review, as well as the participants who volunteered their time to their respective research projects. The authors’ responsibilities were as follows: É. Doucet, P. Imbeault, K. Adamo, and E. Brooks: conceptualization; E. Brooks and J. Kara: literature search; É. Doucet: supervision and finalization of paper inclusion; E. Brooks and J. Kara: extraction of data; T.S. Nagpal: software; E. Brooks, G. Lamothe, and T.S. Nagpal: formal analysis and interpretation of results; E. Brooks: preparation of tables and figures and writing—original draft; É. Doucet, G. Lamothe, T.S. Nagpal, P. Imbeault, and K. Adamo: writing—review and editing; and all authors: read and approved the final manuscript.
References
Balasse, E.O., Fery, F., & Neef, M.A. (1978). Changes induced by exercise in rates of turnover and oxidation of ketone bodies in fasting man. Journal of Applied Physiology: Respiratory, Environmental, and Exercise Physiology, 44(1), 5–11. https://doi.org/10.1152/jappl.1978.44.1.5
Burke, L.M., Ross, M.L., Garvican-Lewis, L.A., Welvaert, M., Heikura, I.A., Forbes, S.G., Mirtschin, J.G., Cato, L.E., Strobel, N., Sharma, A.P., & Hawley, J.A. (2017). Low carbohydrate, high fat diet impairs exercise economy and negates the performance benefit from intensified training in elite race walkers. The Journal of Physiology, 595(9), 2785–2807. https://doi.org/10.1113/JP273230
Burke, L.M., Thomas, D.T., & Erdman, K.A. (2015). Nutrition and athletic performance—Joint position statement from the Academy of Nutrition and Dietetics (AND), Dieticians of Canada (DC), and the American College of Sports Medicine (ACSM). Medicine & Science in Sports & Exercise, 48(3), 543–568. https://doi.org/10.1249/MSS0000000000000852
Cermak, N.M., & van Loon, L.J. (2013). The use of carbohydrates during exercise as an ergogenic aid. Sports Medicine, 43(11), 1139–55. https://doi.org/10.1007/s40279-013-0079-0
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter, E., Todd King, M., Musa-Veloso, K., Ho, M., Roberts, A., Robertson, J., VanItallie, T.B., . . . Veech, R.L. (2012). Kinetics, safety, and tolerability of (R)-3-hydroxybutyl (R)-3-hydroxybutyrate in healthy adult subjects. Regulatory Toxicology and Pharmacology, 63(3), 401–408. https://doi.org/10.1016/j.yrtph.2012.04.008
Coggan, A.R., & Coyle, E.F. (1985). Reversal of fatigue during prolonged exercise by carbohydrate infusion or ingestion. Journal of Applied Physiology, 63(6), 2388–2395. https://doi.org/10.1152/jappl.1987.63.6.2388
Cox, P.J., & Clarke, K. (2014). Acute nutritional ketosis: Implications for exercise performance and metabolism. Extreme Physiology & Medicine, 3, 17. https://doi.org/10.1186/2046-7648-3-17
Cox, P.J., Kirk, T., Ashmore, T., Willerton, K., Evans, R., Smith, A., Murray, A.J., Stubbs, B., West, J., McLure, S.W., King, M.T., Dodd, M.S., Holloway, C., Neubauer, S., Drawer, S., Veech, R.L., Griffin, J.L., . . . Clarke, K. (2016). Nutritional ketosis alters fuel performance and thereby endurance performance in athletes. Cell Metabolism, 24, 256–268. https://dx.doi.org/10.1016/j.cmet.2016.07.010
Coyle, E.F., Hagberg, J.M., Hurley, B.F., Martin, W.H., Ehsani, A.A., & Holloszy, J.O. (1983). Carbohydrate feeding during prolonged strenuous exercise can delay fatigue. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology, 55(1), 230–235. https://doi.org/10.1152/jappl.1983.55.1.230
Dearlove, D.J., Harrison, O.K., Hodson, L., Jefferson, A., Clarke, K., & Cox, P.J. (2021). The effect of blood ketone concentration and exercise intensity on exogenous ketone oxidation rates in athletes. Medicine and Science in Sports and Exercise, 53(3), 505–516. https://doi.org/10.1249/MSS.0000000000002502
Evans, M., Cogan, K.E., & Egan, B. (2016). Metabolism of ketone bodies during exercise and training: Physiological basis for exogenous supplementation. The Journal of Physiology, 595(9), 2857–2871. https://doi.org/10.1113/jp273185
Evans, M., & Egan, B. (2018). Intermittent running and cognitive performance after ketone ester ingestion. Medicine & Science in Sports & Exercise, 50(11), 2330–2338. https://doi.org/10.1249/MSS.0000000000001700
Evans, M., McSwiney, F.T., Brady, A.J., & Egan, B. (2019). No benefit of ingestion of a ketone monoester supplement on 10-km running performance. Medicine & Science in Sports & Exercise, 51(12), 2506–2515. https://doi.org/10.1249/MSS.0000000000002065
Evans, M., Patchett, E., Nally, R., Kearns, R., Larney, M., & Egan, B. (2018). Effect of acute ingestion of ß-hydroxybutyrate salts on the response to graded exercise in trained cyclists. European Journal of Sport Science, 18(3), 376–386. https://doi.org/10.1080/17461391.2017.1421711
Féry, F., & Balasse, E.O. (1986). Response of ketone body metabolism to exercise during transition from postabsorptive to fasted state. The American Journal of Physiology, 250(5), E495–E501. https://doi.org/10.1152/ajpendo.1986.250.5.E495
Harvey, K.L., Holcomb, L.E., & Kolwicz, S.C., Jr. (2019). Ketogenic diets and exercise performance. Nutrients, 11(10), Article 2296. https://doi.org/10.3390/nu11102296
Hashim, S.A., & VanItallie, T.B. (2014). Ketone body therapy: From the ketogenic diet to oral administration of ketone ester. Journal of Lipid Research, 55(9), 1818–1826. https://doi.org/10.1194/jlr.R046599
Higgins, J.P., Altman, D.G., Gøtzsche, P.C., Jüni, P., Moher, D., Oxman, A.D., Savović, J., Schulz, K.F., Weeks, L., Sterne, J.A.C., & Cochrane Bias Methods Group. (2011). The Cochrane collaboration’s tool for assessing risk of bias in randomised trials. The British Medical Journal, 343, Article 5928. https://doi.org/10.1136/bmj.d5928
Higgins, J.P., & Thompson, S.G. (2002). Quantifying heterogeneity in a meta-analysis. Statistics in Medicine, 21, 1539–1558. https://doi.org/10.1002/sim.1186
Hogg, R.V., & Tanis, E.A. (2010). Probability and statistical inference (8th ed.). Pearson.
James, S., & Greer, B.K. (2018). Influence of exogenous ß-hydroxybutyrate on walking economy and rating of perceived exertion. Journal of Dietary Supplements, 16(4), 463–469. https://doi.org/10.1080/19390211.2018.1471562
Jeukendrup, A.E., & Jentjens, R. (2000). Oxidation of carbohydrate feedings during prolonged exercise: Current thoughts, guidelines, and directions for future research. Sports Medicine, 29(6), 407–24. https://doi.org/10.2165/00007256-200029060-00004
Laffel, L. (1999). Ketone bodies: A review of physiology, pathophysiology, and application of monitoring to diabetes. Diabetes/Metabolism Research and Reviews, 15(6), 412–426.
Leckey, J.J., Ross, M.L., Quod, M., Hawley, J.A., & Burke, L.M. (2017). Ketone diester ingestion impairs time-trial performance in professional cyclists. Frontiers in Physiology, 8, Article 806. https://doi.org/10.3389/fphys.2017.00806
Margolis, L.M., & O’Fallon, K.S. (2020). Utility of ketone supplementation to enhance physical performance: A systematic review. Advances in Nutrition, 11(2), 412–419. https://doi.org/10.1093/advances/nmz104
Mikkelsen, K.H., Seifert, T., Secher, N.H., Grondal, T., & Van Hall, G. (2015). Systemic, cerebral, and skeletal muscle ketone body and energy metabolism during acute hyper-D-ß-hydroxybutyratemia in post-absorptive healthy males. The Journal of Clinical Endocrinology and Metabolism, 100(2), 636–643. https://doi.org/10.1210/jc.2014-2608
Mitchell, G.A., Kassovska-Bratinova, S., Boukaftane, Y., Robert, M.F., Wang, S.P., Ashmarina, L., Lambert, M., Lapierre, P., & Potier, E. (1995). Medical aspects of ketone body metabolism. Clinical and Investigative Medicine, 18(3), 193–216.
O’Malley, T., Myette-Cote, E., Durrer, C., & Little, J.P. (2017). Nutritional ketone salts increase fat oxidation but impair high-intensity exercise performance in healthy adult males. Applied Physiology, Nutrition, and Metabolism, 42(10), 1031–1035. https://doi.org/10.1139/apnm-2016-0641
Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., Shamseer, L., Tetzlaff, J.M., Akl, E.A., Brennan, S.E., Chou, R., Glanville, J., Grimshaw, J.M., Hróbjartsson, A., Lalu, M.M., Li, T., Loder, E.W., Mayo-Wilson, E., McDonald, S., . . . Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews, The British Medical Journal, 372, 71. https://doi.org/10.1136/bmj.n71
Poffé, C., Ramaekers, M., Bogaerts, S., & Hespel, P. (2020a). Bicarbonate unlocks the ergogenic action of ketone monoester intake in endurance exercise. Medicine & Science in Sports & Exercise, 53(2), 431–441. https://doi.org/10.1249/MSS.0000000000002467
Poffé, C., Ramaekers, M., Bogaerts, S., & Hespel, P. (2020b). Exogenous ketosis impacts neither performance nor muscle glycogen breakdown in prolonged endurance exercise. Journal of Applied Physiology, 128(6), 1643–1653. https://doi.org/10.1152/japplphysiol.00092.2020
Poffé, C., Wyns, F., Ramaekers, M., & Hespel, P. (2020c). Exogenous ketosis impairs 30-min time-trial performance independent of bicarbonate supplementation. Medicine & Science in Sports & Exercise, 53(5), 1068–1078. https://doi.org/10.1249/MSS.0000000000002552
Robinson, A.M., & Williamson, D.H. (1980). Physiological roles of ketone bodies as substrates and signals in mammalian tissues. Physiological Reviews, 60(1), 143–187. https://doi.org/10.1152/physrev.1980.60.1.143
Rodger, S., Plews, D., Laursen, P., & Driller, M. (2017). Oral ß-hydroxybutyrate salt fails to improve 4-minute cycling performance following submaximal exercise. Journal of Science and Cycling, 6(1), 26–31. https://core.ac.uk/download/pdf/85165429.pdf
Romijn, J.A., Coyle, E.F., Sidossis, L.S., Gastaldelli, A., Horowitz, J.F., Endert, E., & Wolfe, R.R. (1993). Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. The American Journal of Physiology, 265(3), E380–E391. https://doi.org/10.1152/ajpendo.1993.265.3.E380
Scott, B.E., Laursen, P.B., James, L.J., Boxer, B., Chandler, Z., Lam, E., Gascoyne, T., Messenger, J., & Mears, S.A. (2018). The effect of 1,3-butanediol and carbohydrate supplementation on running performance. Journal of Science & Medicine in Sport, 22(6), 702–706. https://doi.org/10.1016/j.jsams.2018.11.027
Shaw, D.M., Merien, F., Braakhuis, A., Maunder, E.D., & Dulson, D.K. (2019). Effect of a ketogenic diet on submaximal exercise capacity and efficiency in runners. Medicine and Science in Sports and Exercise, 51(10), 2135–2146. https://doi.org/10.1249/MSS.0000000000002008
Shaw, D.M., Merien, F., Braakhuis, A., Plews, D., Laursen, P., & Dulson, D.K. (2019). The effect of 1,3-butanediol on cycling time-trial performance. International Journal of Sport Nutrition & Exercise Metabolism, 29(5), 466–473. https://doi.org/10.1123/ijsnem.2018-0284
Sivva, V., Cox, P.J., Clarke, K., Veech, R.L., Tucker, I.G., & Duffull, S.B. (2016). The population pharmacokinetics of d-ß-hydroxybutyrate following administration of (R)-3-hydroxybutyl (R)-3-hydroxybutyrate. The American Association of Pharmaceutical Scientists Journal, 18(3), 678–88. https://doi.org/10.1208/s12248-016-9879-0
Stubbs, B.J., Cox, P.J., Evans, R.D., Santer, P., Miller, J.J., Faull, O.K., Magor-Elliott, S., Hiyama, S., Stirling, M., . . . Clarke, K. (2017). On the metabolism of exogenous ketones in humans. Frontiers in Physiology, 8, Article 848. https://doi.org/10.3389/fphys.2017.00848
Stubbs, B.J., Cox, P.J., Kirk, T., Evans, R.D., & Clarke, K. (2019). Gastrointestinal effects of exogenous ketone drinks are infrequent, mild, and vary according to ketone compound and dose. International Journal of Sport Nutrition and Exercise Metabolism, 29(6), 596–603. https://doi.org/10.1123/ijsnem.2019-0014
Stubbs, B.J., Koutnik, A.P., Poff, A.M., Ford, K.M., & D’Agostino, D.P. (2018). Commentary: Ketone diester ingestion impairs time-trial performance in professional cyclists. Frontiers in Physiology, 9, Article 279. https://doi.org/10.3389/fphys.2018.00279
Stubbs, B.J., Willerton, K., Hiyama, S., Clarke, K., & Cox, P.J. (2015). Concomitant meal ingestion alters levels of circulating ketone bodies following a ketone ester drink. Proceedings of the Physiological Society, 34, PC235. https://www.physoc.org/abstracts/concomitant-meal-ingestion-alters-levels-of-circulating-ketone-bodies-following-a-ketone-ester-drink/
Valenzuela, P.L., Morales, J.S., Castillo-Garcia, A., & Lucia, A. (2020). Acute ketone supplementation and exercise performance: A systematic review and meta-analysis of randomized controlled trials. International Journal of Sports Physiology and Performance, 15, 298–308. https://doi.org/10.1123/ijspp.2019-0918
Zajac, A., Poprzecki, S., Maszczyk, A., Czuba, M., Michalczyk, M., & Zydek, G. (2014). The effects of a ketogenic diet on exercise metabolism and physical performance in off-road cyclists. Nutrients, 6(7), 2493–2508. https://doi.org/10.3390/nu6072493
Zinn, C., Wood, M., Williden, M., Chatterton, S., & Maunder, E. (2017). Ketogenic diet benefits body composition and well-being but not performance in a pilot case study of New Zealand endurance athletes. Journal of the International Society of Sports Nutrition, 14(1), 1–9. https://doi.org/10.1186/s12970-017-0180-0
Appendix
Timeline of administration of ketone supplement doses for each study. Six of the eight studies split the dose into three boluses, with four of the six administering them at a percentage ratio of 50:25:25 (a–d) and the other two at a ratio of 38:31:31 (f and h). The remaining two studies split the doses into two boluses at a ratio of 50:50 (e and g). Bottle symbol represents drink aliquots. TT = time trial; IMT = intermittent; TTE = time to exhaustion.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 32, 3; 10.1123/ijsnem.2021-0280