Ketogenesis is a process whereby the liver utilizes fatty acids to synthesize ketones: beta-hydroxybutyrate (BHB), acetoacetate, and, to a lesser degree, acetone. Ketones are produced in low amounts when dietary carbohydrates are limited (e.g., prolonged fasting or starvation) or during prolonged exercise, but they can also be produced in response to diets that are ketogenic. Under normal physiological conditions, blood BHB levels are <0.5 mmol/L (reviewed in Laffel, 1999) and, after weeks of fasting, blood BHB levels of 4–7 mmol/L have been reported (reviewed in Cahill, 2006). Previously reported ranges for blood BHB levels after exogenous ketone administration are in the range of 0.5–3.0 mmol/L (Kolb et al., 2021) or up to 7.0 mmol/L (Cox et al., 2016).

For over a century, ketogenic diets have been used as effective treatments for childhood seizures and epilepsy (Thiele, 2003). Prior to the discovery of insulin, a ketogenic diet was used as a therapeutic approach to control type 1 and 2 diabetes (Ludwig, 2020). Recently, the role of ketone bodies (synonymous with ketones) in producing energy (i.e., adenosine triphosphate) has gained traction for its potential application in the area of sports nutrition -- to improve endurance performance and recovery from exercise. Benefits in exercise performance with the consumption of ketones have been reported in some studies, while in other studies, impairments in exercise performance were reported (Cox et al., 2016; Wroble et al., 2019; Mansor and Woo, 2021; Whitfield et el., 2021; Evans et al., 2022). In a recent systematic review and meta-analysis, neither ketone esters nor ketone salts were found to improve physical performance or metabolic, respiratory, or cardiovascular responses to exercise (Valenzuela et al., 2020). Additional studies are needed to better understand the impacts of various factors (e.g., training level of athletes; type of exercise; nature, dose, and duration of ketone supplementation; and need for metabolic adaptation) on physical performance.

Ketone drinks and supplements contain exogenous sources of ketones that, following consumption, have been used to increase blood ketone concentrations to levels of ketosis (blood BHB >0.5 mmol/L) without the dietary restrictions imposed by a ketogenic diet (Stubbs et al., 2017). Examples of these exogenous ketone sources include ketone salts (KS) and BHB ketone monoesters (KME), the latter which are broken down to butanediol and BHB (Soto-Mota et al., 2020). BHB provided by ketone drinks and supplements is thought to be an energy source that could improve sporting performance and recovery, and may have implications for other endpoints such as glucose control (Falkenhain et al., 2022), immune modulation (Neudorf et al., 2019), cardiac function (Cuenoud et al., 2020; Yurista et al., 2021) and cognitive function (Poff et al., 2021). To maximize translational potential of exogenous ketone interventions, they must be well tolerated under the intended conditions of use. Studies to date of various exogenous ketone supplements have highlighted the potential for gastrointestinal (GI) or systemic symptoms to occur following their consumption. It has previously been reported that the consumption of KME can lead to mild flatulence, nausea, and dizziness when consumed at rest, and upper abdominal discomfort when consumed during exercise (Clarke et al., 2012; Egan and D’Agostino, 2016; Vandoorne et al., 2017); as well, consumption of KS drinks pre-exercise has been associated with diarrhea and vomiting (Waldman et al., 2018). Each of these studies assessing GI symptomology utilized a simplistic questionnaire (symptom/no symptom). Severity of symptoms was assessed by Stubbs et al. (2019), wherein the GI tolerability of KME and KS was reported. Severe symptoms included nausea, abdominal cramps and diarrhea following consumption of a high-dose of KS (3.2 mmol/kg body weight) in a fasted state. Symptoms resolved completely by the end of the study. KS may be associated with more severe GI symptoms than KME due to the high intakes of sodium and potassium, which can result in water retention and diarrhea secondary to a hyperosmolar gut lumen (reviewed in Stubbs et al., 2019).

1,3-butanediol, an organic compound that can be converted to BHB, has also been studied for its potential to increase ketone levels. 1,3-butanediol is converted to BHB, primarily in the liver, by alcohol and aldehyde dehydrogenases (Stubbs et al., 2021). There are two distinct stereoisomers of 1,3-butanediol: the R form and the S form; the R stereoisomer is naturally occurring and appears to be the configuration that can be converted to physiological ketone bodies most proficiently (Desrochers et al., 1992). To date, five clinical studies in which the efficacy and/or safety/tolerability of (R, S)-1,3-butanediol were assessed have been conducted (Kies et al., 1973; Tobin et al., 1975 [Studies 1 and 2]; Scott et al., 2019; Shaw et al., 2019). There were no significant adverse events (AEs) in the studies following the consumption of a beverage providing 15–40 g per day of (R,S)-1,3-butanediol. In the study by Shaw et al. (2019), a beverage providing 0.5 or 0.7 g/kg body weight (35–49 g for a 70-kg person) of (R,S)-1,3-butanediol resulted in nausea, euphoria, and dizziness; however, these adverse outcomes were mitigated with the administration of divided servings (as opposed to a single intake). Falkenhain et al. (2023) recently published the results of a clinical study in which the R form of 1,3-butanediol was administered to 12 healthy male and female adults. In this study, the administration of a single serving of 10 g of (R)-1,3-butanediol resulted in a peak blood BHB level of 1.2 mmol/L at 39 min, and an incremental area under the curve (iAUC) of 118 mmol/L x 240 min; furthermore, the serving was well-tolerated.

The primary objective of our study was to determine the level of ketones (measured as blood BHB) following the single-day consumption of 34.5 g of (R)-1,3-butanediol in a beverage. The secondary objectives were to assess GI symptomology, as well as any effects on sleepiness.

This study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of PearlIRB (protocol code #22-GMAT-101; Approved 12 May 2022). The study was registered at clinicaltrials.gov (NCT05384106).

In the study reported herein, three servings of a beverage containing Avela™ and each providing 11.5 g of (R)-1,3-butanediol were consumed in close succession at 0, 30, and 60 min by participants in a fasted state, and effects on BHB levels, GI symptomology, and sleepiness were assessed over a total period of 300 min.

Consumption of three servings of (R)-1,3-butanediol (11.5 g, each; 34.5 g, total) resulted in a mean blood BHB C_max of 2.10 ± 0.97 (SD) mmol/L, which was achieved roughly 1 h post consumption of the final beverage containing Avela™. In a recent publication, 12 healthy individuals who consumed a single serving of 10 g of (R)-1,3-butanediol following an 8-h fast achieved a mean peak blood BHB level of 1.2 ± 0.3 mmol/L (SD) approximately 40 min post consumption (Falkenhain et al., 2023). In the current study, the mean blood BHB iAUC_0–300 min was 376.73 ± 156.76 (SD) mmol/L*min, a value approximately triple that which was measured (iAUC_0–240 min of 118 ± 64 mmol/L*min) following the consumption of 10 g (R)-1,3-butanediol in the study by Falkenhain et al. (2023). This is consistent with total intakes administered in each study. Blood BHB levels appear to continue to rise with each subsequent consumption of a beverage prepared with Avela™ (Figure 2) to a peak (roughly 1 h following the last consumption of Avela™), followed by a slow decline to the end of the study, where the mean level was 1.11 ± 0.85 mmol/L. Consumption of a plant-based protein bar (providing 25 g carbohydrate, 14 g protein, and 9 g fat) at 240 min did not appear to change the kinetics of BHB disappearance in the 1 h post-consumption. In a study by Cuenoud et al. (2020) plasma BHB levels were elevated following the consumption of 12 g of D-BHB (in the form of salts), despite the consumption, 30 min after the D-BHB salts beverage, of a breakfast meal (consisting of 2 boiled eggs, 2 pieces of toast, 1 slice of cheese, and 1 portion of fruit jam, providing a total of 423 kcal, 20 g fat, 24 g protein, and 32 g carbohydrate). Thus, based on preliminary evidence, BHB levels achieved via exogenous ketone sources are not affected by the consumption of foods, but additional studies are needed to better understand this. Overall, maintenance of elevated levels of BHB in the blood for at least 5 h using exogenous administration appears quite feasible.

Statistically significant inverse associations between BMI and fasting BHB and between body weight and each of fasting BHB, BHB iAUC_0–300 min, BHB C_max, and BHB T_max are interesting. Body weight may be a more sensitive predictor of ketogenic potential than BMI. Increased adiposity is associated with increased insulin, which potently inhibits lipolysis and ketogenesis (Horowitz et al., 1999; Horowitz et al., 2001). Indeed, in our study, of the 6 subjects with a BMI ≤23 kg/m², 5 had a fasting plasma BHB level greater than 0.2 mmol/L (of the remaining 19 subjects with a BMI >23 kg/m², 18 had a fasting plasma BHB level less than 0.2 mmol/L).

Additionally, the significantly greater BHB T_max (p = 0.046) and nearly significantly greater BHB iAUC_0–300 min (p = 0.06) determined for females versus males also are interesting, given that sex differences have been reported in metabolic changes following fasting. Increases in plasma fatty acid and ketone body concentrations (Merimee and Fineberg, 1973; Merimee et al., 1978; Haymond et al., 1982) with a concomitant decrease in blood glucose concentration (Merimee and Fineberg, 1973; Haymond et al., 1982) which normally occur during short-term fasting (60–86 h), are reportedly greater in women than in men (Mittendorfer et al., 2001). While effects of gender may be confounded by effects of adiposity (given that men typically have greater body weights than females), in a fasting study in which males and females were matched for body fat percentage, the mean plasma fatty acid and glycerol concentrations were higher in women than in men (following a 14-h fast), and continued to increase with duration of fasting (22-h fast); however, glucose concentrations were similar between sexes (Mittendorfer et al., 2001). Although the mechanisms for these differences are not known, Mittendorfer et al. (2001) hypothesized that they may be related to differences in major plasma hormones that regulate lipolysis.

To further evaluate the effects of body weight on fasting BHB, BHB C_max, BHB T_max, and BHB iAUC, we examined associations within each gender, since body weight was significantly lower in females than in males. Although all Spearman correlation coefficients were negative, only those for fasting BHB and BHB T_max were significant, and only for females (data not shown). Of course, the sample sizes are very small (14 males and 12 females) and so these results should be interpreted with caution. Additional studies are needed to better understand the effects of gender and body weight on ketone levels.

GI symptomology was assessed at multiple time points throughout the study to capture the presence or absence, as well as the severity, of 19 symptoms. Overall, (R)-1,3-butanediol was well-tolerated with no or mild GI symptoms reported by most participants. Post-baseline adjusted symptoms were rated as moderate by only one person, for each of urge to regurgitate, nausea and dizziness; this person tested positive for COVID-19 the following morning. The GI symptoms that were most frequently reported were mild belching, mild nausea, and mild dizziness, each of which was reported by eight, four, and nine persons, respectively, along with mild headache reported by 5 subjects. Past studies have reported diarrhea, decreased appetite, and headache with KS (Leckey et al., 2017; Waldman et al., 2018; Stubbs et al., 2019) or KME (Clarke et al., 2012). Stubbs et al. (2019) suggested that the salt bolus consumed with KS may affect fluid homeostasis in the gut leading to symptoms not observed in our study with (R)-1,3-butanediol. While symptoms of nausea and dizziness/light headedness have been reported following consumption of other ketone beverages (Clarke et al., 2012; Leckey et al., 2017; Vandoorne et al., 2017; Evans and Egan, 2018; Evans et al., 2018; Waldman et al., 2018; Stubbs et al., 2019), factors that may have contributed to these effects could be prolonged fasting (Torelli and Manzoni, 2010) and/or caffeine withdrawal (Sajadi-Ernazarov et al., 2022). Indeed, in our study, the subjects were fasted overnight and completed the study in a fasted state (until the 240 min mark, when the ALOHA bar was consumed). The maximum ratings of dizziness were at 120 and 240 min. Once the ALOHA bar was consumed (240 min), the dizziness ratings returned to baseline values in most subjects, thereby suggesting that the dizziness experienced may have been attributed to prolonged fasting, rather than the consumption of (R)-1,3-butanediol. In addition, it is noteworthy that the participants had an intense schedule of evaluations during the 5-h study, which entailed the completion of questionnaires, finger-pricking (which is nauseating for some people), and entering data electronically, repeatedly. Finally, (R)-1,3-butanediol has a bitter taste which may have contributed to the nausea. Of note, neither “other GI Symptoms” (which included nausea and dizziness) nor any of the other GI symptom categories was associated with BHB levels, even after adjusting for body weight (data not shown).

No differences in the reporting of sleepiness post-baseline for any time points in the study were observed. Some scientists suggest that consumption of a ketogenic diet or exogenous ketones can improve sleep patterns, cognition and outcomes in neurologic conditions (reviewed in Hallböök et al., 2012; Murray et al., 2016; Evans and Egan, 2018). Additionally, others hypothesize a potential link between low-carbohydrate diets and feelings of euphoria (Brown, 2007). For this reason, an additional questionnaire was provided to subjects in the study herein to determine whether there was an effect on sleepiness following consumption of the study product. The SSS has the greatest reliability and validity in predicting performance on tasks related to alertness (e.g., reaction time, vigilance tests) following total sleep deprivation (Shahid et al., 2012). The SSS likely lacked the sensitivity to identify any changes in alertness in this study, but was selected given its simplicity of use (it is a 7-point Likert scale) and in consideration of the design of the study (at-home, conducted by the participants themselves). The study reported herein was an acute feeding study; longer-term feeding trials in patients with measurable disorders may be more appropriate for assessing the neurocognitive effects of ketones and effects on sleepiness (García-Rodríguez and Giménez-Cassina, 2021).

Several limitations were inherent to this study and warrant discussion. The study was conducted in the participants’ homes and the results of the study were contingent on the participants performing the trial, as intended. Subjects followed step-by-step prompts on a screen to mitigate issues that might arise with subject error in study conductance; they were prompted to advance to a subsequent step only after the prior step was completed. This helped to ensure that all study-related procedures were completed in the proper order at each time point. The Keto-Mojo monitor was utilized to measure blood BHB levels. This monitor is capable of measuring ketones from 0.1 mmol/L to 8.0 mmol/L; however, validity at detecting BHB levels greater than 3.0 mmol/L was not tested. In this study, there were 208 BHB measurements across all 26 participants and time points; of these, 8 were greater than 3 mmol/L. Of note, plasma butanediol levels were not measured and so, it is unknown at what level they circulate in the body; this could be measured in a future study. The requirement for a minimum 12-h fast and the administration of the study product in three servings in close succession (each separated only by a half-hour) on an empty stomach, without food, was considered to represent a best-case scenario in terms of increasing BHB levels, but a worst-case scenario regarding GI tolerability.

In this study, peak blood BHB levels of 2.10 ± 0.97 mmol/L were achieved roughly 1-h post consumption of the last (R)-1,3-butanediol beverage, when a total of 34.5 g (R)-1,3-butanediol would have been consumed. (R)-1,3-butanediol was well tolerated. Unlike reports following consumption of KME and KS, no subjects reported experiencing diarrhea following consumption of the study product. The most frequently reported symptoms of belching, nausea, headache and dizziness, were mild and transient in nature. No effect on subject sleepiness was reported.

This study demonstrates the feasibility of administering oral (R)-1,3-butanediol to achieve blood BHB levels typical of physiologic ketosis. There is a high likelihood of creating regimens for chronic administration which could potentially sustain a desired level throughout the day. Next steps require testing of such regimens in patients with measurable disease or performance endpoints.