Effect of exogenous and endogenous ketones on respiratory exchange ratio and glucose metabolism in healthy subjects

Abstract

This study examined the effect of exogenous ketone bodies (KB) on oxygen consumption (V̇o₂), carbon dioxide production (V̇co₂), and glucose metabolism. The data were compared with the effects of endogenous ketonemia during both, a ketogenic diet or fasting. Eight healthy individuals [24.1 ± 2.5 yr, body mass index (BMI) 24.3 ± 3.1 kg/m²] participated in a crossover intervention study and were studied in a whole-room indirect calorimeter (WRIC) to assess macronutrient oxidation following four 24-h interventions: isocaloric controlled mixed diet (ISO), ISO supplemented with ketone salts (38.7 g of β-hydroxybutyrate/day, EXO), isocaloric ketogenic diet (KETO), and total fasting (FAST). A physical activity level of 1.65 was obtained. In addition to plasma KB, 24-h C-peptide and KB excretion rates in the urine and postprandial glucose and insulin levels were measured. Although 24-h KB excretion increased in response to KETO and FAST, there was a modest increase in response to EXO only (P < 0.05). When compared with ISO, V̇o₂ significantly increased in KETO (P < 0.01) and EXO (P < 0.001), whereas there was no difference in FAST. V̇co₂ increased in EXO but decreased in KETO (both P < 0.01) and FAST (P < 0.001), resulting in 24-h respiratory exchange ratios (RER) of 0.828 ± 0.024 (ISO) and 0.811 ± 0.024 (EXO) (P < 0.05). In response to EXO there were no differences in basal and postprandial glucose and insulin levels, as well as in insulin sensitivity. When compared with ISO, EXO, and KETO, FAST increased homeostatic model assessment β-cell function (HOMA-B) (all P < 0.05). In conclusion, at energy balance exogenous ketone salts decreased respiratory exchange ratio without affecting glucose tolerance.

NEW & NOTEWORTHY Our findings revealed that during isocaloric nutrition, additional exogenous ketone salts increased V̇o₂ and V̇co₂ while lowering the respiratory exchange ratio (RER). Ketone salts had no effect on postprandial glucose metabolism.

INTRODUCTION

Ketone bodies (KBs) serve as an energy substrate during fasting, prolonged exercise, or a low carbohydrate (CHO) diet (1, 2). In addition to its substrate effect, KBs exhibit cellular signaling capacities in many tissues (3). Although fasting or a low CHO diet increases endogenous ketone production, ketone supplements such as salts or esters are a popular approach to achieving ketosis without caloric or dietary restriction. Ketone supplements have been studied as potential therapeutics suggesting potential effects concerning enhanced physical performance and improved cardiovascular and metabolic health, among others (for reviews, see Refs. 3 and 4).

In adults with normal (5) or impaired glucose tolerance, exogenous ketones improved oral glucose tolerance (6–8). A glucose-lowering effect had been also observed during exercise with prior administration of exogenous ketone salts (9, 10). The underlying mechanisms are only partially known, e.g., there is evidence that KBs have an insulinotropic effect (11–13). Alternatively, KBs were found to reduce the rate of gluconeogenesis and lipolysis under postabsorptive conditions (14).

Since, however, exogenous ketosis does not decrease nonesterified fatty acids (NEFA) concentrations (15, 16), it does not mirror endogenous ketosis as observed during fasting or in response to a ketogenic diet. Therefore, the effect of oral ketone supplements on fasting and postprandial glycemia needs to be investigated and compared with endogenous ketonemia.

In addition to a “glucose-lowering effect,” exogenous ketones may exert a “glucose-sparing effect” by reducing CHO oxidation as reflected by alterations in the respiratory exchange ratio (RER). We have conducted a highly controlled crossover study in a whole-room indirect calorimeter (WRIC) and compared the effect of ketone salt supplementation with an isocaloric mixed diet versus a ketogenic diet, complete fasting, and an isocaloric control diet. We have analyzed fasting, postprandial glucose and insulin levels, RER, insulin secretion by C-peptide excretion, and insulin sensitivity by a mixed meal model. Each intervention period lasted for 24 h at a normal level of physical activity (PAL 1.65). Three meals a day were given to capture the circadian effects of ketone supplements on glucose metabolism and RER.

METHODS

Study Participants

Young adults were recruited at Kiel University (Germany) from September 2020 to July 2021 to take part in a study investigating the impact of one-day fasting, ketogenic diet, or exogenous ketones on control of energy balance in healthy participants. Results on energy balance were previously published (17). The trial was registered at clinicaltrials.gov as NCT04490226. Exclusion criteria were chronic diseases, regular use of medication, alternative eating habits (vegan, vegetarian, etc.), food allergies or intolerances, claustrophobia, smoking, high habitual physical activity (≥1 h/day), recent weight loss diet or weight loss of >5 kg in the last three months, pregnancy, or lactation. Women were only included when using hormonal contraceptives continuously to avoid the influence of the menstrual cycle on energy expenditure (18). The study protocol was approved by the ethics committee of the Medical Faculty at Kiel University, Germany (D522/20) in accordance with the Declaration of Helsinki. All subjects provided written informed consent before participation.

Participants were invited to attend an in-person screening conducted within 2 wk before the start of the interventions. Screening examinations took place after an overnight fast. Height was measured with a stadiometer (seca 274; seca GmbH & Co. KG, Hamburg, Germany). Body weight was measured on a calibrated scale, and fat mass was assessed using air-displacement plethysmography (BodPod, COSMED, Rome, Italy), both in underwear. Fat mass index (FMI) was calculated as fat mass divided by height squared (kg/m²).

Study Protocol

Each participant went through four different 24-h interventions in the WRIC within 3 wk. Each intervention week was initiated by a 3-day run-in period with a controlled diet and identical macronutrient composition to adapt macronutrient oxidation rates to macronutrient intake (19) followed by two 24-h intervention days in the WRIC. An outline of the study protocol is given in Fig. 1. Participants were advised to maintain their habitual physical activity levels (<1 h/day of exercise) before entering the WRIC to ensure equal baseline conditions. Energy requirement was measured with ad libitum energy intake over 24 h inside the WRIC preceding the fasting intervention. Participants entered the WRIC in the evening before each intervention period to adapt to the environment. A highly standardized diurnal rhythm (from 6:30 AM to 6:30 AM) was maintained throughout the study (wake-up at 6:30 AM; meals at 7:00 AM, 1:00 PM, 7:00 PM; and bedtime at 10:30 PM).

Figure 1.

Outline of the study protocol with 24-h interventions in a whole-room indirect calorimeter (WRIC) under four different conditions: ISO, isocaloric control diet; EXO, exogenous ketone salts plus isocaloric control diet; KETO, isocaloric ketogenic diet; and FAST, total fasting. Twenty-four-hour urine and blood samples were collected throughout the intervention days. Prescribed physical activity level was achieved by cycling on an exercise bike. A 3-day run-in period with a controlled diet preceded each intervention phase to ensure equal baseline conditions. The order of KETO and EXO interventions was randomized. CHO, carbohydrate.

Physical activity level (PAL) was 1.65 on all days in the WRIC and was performed on a bicycle ergometer (opticare basic and ergoselect 4, ergoline GmbH, Bitz, Germany) for 3×20 min three times a day (at 8:00 AM, 2:00 PM, and 8:00 PM, total: 180 min/day). Women were requested to cycle at 50 W and men at 75 W with a constant cadence (55–65 rpm). Physical activity was continuously monitored using step counts and acceleration volume per minute by a triaxial accelerometer (activPAL 4, Paltechnologies Ltd., Glasgow, UK) fixed on the thigh to ensure adherence to the protocol.

Energy requirement under activity was measured using two identical WRICs (Promethion model GA-3m2/FG-250, Sable Systems International, Las Vegas, NV) at Kiel University. For details of WRIC setup see Ref. 20.

Diet Intervention

The highly controlled nutritional intervention comprised four 24-h interventions at a physical activity level (PAL) of 1.65: 1) isocaloric formula diet (ISO), 2) exogenous ketone salts provided with an isocaloric formula diet (EXO), 3) isocaloric ketogenic formula diet (KETO), and 4) endogenous ketone production due to total fasting (FAST). Commercially available formula diets were used to facilitate standardized meals for breakfast, lunch, and dinner (33% of energy content each). For ISO, formula diet was used with a macronutrient composition of 47.4 ± 1.6% CHO, 39.6 ± 1.1% fat, and 13.0 ± 0.8% protein (Next Level Meal Coconut, Strawberry or Cheese Cake, Runtime GmbH, Berlin, Germany). For EXO, the ISO meal was consumed together with 30 mL of ketone salts [KetoBlitz, KetoSports, Urbana IL, racemic blend of β-hydroxybutyric acid (BHB), potassium-BHB and sodium-BHB equivalent to 12.9 g of BHB] in 270 mL of water, resulting in a total daily intake of 38.7 g/day BHB (containing both d-BHB and l-BHB). For KETO, a balanced ketogenic formula diet was used with a macronutrient composition of 3.1% CHO, 88.8% fat, and 8.1% protein (KetoCal 4:1 Vanilla, Nutricia GmbH, Erlangen, Germany). Every food item was weighted to the nearest 0.1 g for each participant according to individual energy requirements. On isocaloric intervention days, participants were asked to eat their meals within 30 min without leftovers. Each intervention week was preceded by a 3-day run-in diet with a controlled macronutrient composition of 51.3 ± 0.6% CHO, 36.7 ± 0.4% fat, and 12.0 ± 0.2% protein. Leftovers were weighed and energy intake was calculated. Individual diet composition and actual energy and macronutrient intake were calculated using PRODI expert version 6.10 (Wissenschaftliche Verlagsgesellschaft Stuttgart; based on German Nutrient Data Base BLS 3.02). All food was provided and subjects were instructed to only consume the allocated foods and noncaloric beverages without caffeine.

Although participants were fed an isocaloric diet; energy balance was slightly negative on KETO (−192 ± 170 kcal/day, P < 0.05) compared with ISO, and EXO (for details, see Ref. 17). See also Hägele et al. (17) on correction of macronutrient oxidation rates for ketone intervention (KETO). Ketone salts led to an additional caloric intake of 183 kcal/day. Concomitantly, when compared with ISO EXO increased 24-h energy expenditure (+119 ± 40 kcal/day, P > 0.05, which is marginally above the minimal detectable change of the WRIC; 7). One participant had to be excluded from KETO analyses due to incorrect calculation resulting in a nonisocaloric energy intake.

Assessment of Blood and Urine Parameters

Plasma samples for the measurement of postprandial glucose, insulin, NEFA, and ketone bodies (acetoacetate, AcAc and d-BHB) were collected before (0) and 30, 60, 90, 120, 180 min after each meal and 240 min after breakfast and lunch. Additional fasting blood samples were taken the following mornings after each intervention day. Glucose concentration was determined via hexokinase method (OSR6121, Beckman Coulter, Brea, CA). Serum insulin was measured by chemiluminescent immunoassay (Alinity i Insulin Reagent Kit 04T75, Abbott, Wiesbaden, Germany). Assays for NEFA and KB were purchased from Sigma-Aldrich (MAK044 and MAK134, Sigma-Aldrich, Merck KGaA, Darmstadt, Germany). The ketone body concentrations of d-AcAc and d-BHB were both measured in serum and 24-h urine colorimetrically at 340 nm, their sum was regarded as the total ketone concentration for each time point. AcAc was not measurable in 28–35% of the samples. Insulin secretion was assessed by 24-h urinary C-peptide excretion (luminescence immunoassay). The 24-h urinary nitrogen excretion was measured photometrically from 24-h urine.

Incremental area under the curve (iAUC) throughout the day was calculated for 15 h (7:00 AM to 10:00 PM) using trapezoidal rule. Homeostatic model assessment β-cell function (HOMA-B) was calculated using the following formula: HOMA-B = 20 × fasting insulin (μU/mL)/fasting glucose (mmol/mL) − 3.5. In addition, an oral mixed-meal model was used to simulate glucose-insulin dynamics and to estimate insulin sensitivity (S_I) using the SAAMII software (The Epsilon Group, Charlottesville, VA; v.2.3.2) (see Supplemental Material for model specification following: Refs. 21 and 22).

Statistical Analysis

Data were analyzed using the SPSS software package (IBM Corp. Released 2022. IBM SPSS Statistics for Windows, Armonk, NY: IBM Corp; v.29.0). Within-subject intervention effects were analyzed by two-way repeated-measures ANOVA followed by pairwise comparisons using the Bonferroni correction. Analyses only concerning the comparisons between ISO and EXO were performed using paired t tests. The associations were determined by Spearman correlation. Graphs are plotted using GraphPad Prism 10 (GraphPad Prism for Windows, GraphPad Software, La Jolla, CA). Data are presented as means ±SD, and a two-sided P of <0.05 was considered to be statistically significant.

RESULTS

Eight healthy participants (four women and four men) aged 24.1 ± 2.5 yr with a body mass index (BMI) of 24.3 ± 3.1 kg/m² and fat mass index (FMI) of 5.9 ± 2.4 kg/m² were included in this study. According to WHO criteria, five participants were normal weight, and three lived with overweight. Regarding body composition, two participants had an FMI above the age- and sex-adjusted 95th percentile (23).

The intervention increased blood ketone bodies throughout the day under FAST and KETO whereas EXO increased blood ketone levels after the meals only (Fig. 2A and Supplemental Data). With EXO, average metabolic KB availability was 10.7 ± 5.8 g/day suggesting a recovery rate of 55%. When compared with ISO and EXO, FAST and KETO increased lipolysis as reflected by increases in 15-h iAUCs of NEFA (Fig. 2B).

Figure 2.

Box plots showing 15-h incremental areas under the curve (iAUCs) of sum of ketone bodies (d-β-hydroxybutyrate and d-acetoacetate (A) and nonesterified fatty acids (NEFA; B), and urinary 24-h ketone excretion (C) in comparison between all interventions: isocaloric control diet (ISO), and with the addition of exogenous ketones (EXO), isocaloric ketogenic diet (KETO), and total fasting (FAST); n = 7. Data are presented as means ±SD. *P < 0.05; **P < 0.01; ***P < 0.001 (RM-ANOVA, Bonferroni correction).

RER and Macronutrient Balances

V̇o₂, V̇co₂, and RER values corresponding to the four interventions are shown in Table 1. When compared with ISO and FAST, KETO and EXO elevated V̇o₂ (both P < 0.05). Conversely, V̇co₂ values were variable across the different intervention periods, with lowest V̇co₂ values in FAST, followed by KETO and ISO whereas EXO records the highest V̇co₂ values (all P < 0.05). When compared with ISO, EXO significantly lowered RER (P < 0.05). However, the lowest RER values were observed in KETO and FAST.

Table 1.

V̇o₂, V̇co₂, and RER in all four interventions

	ISO	EXO	KETO	FAST
V̇o2, L/day	605.4 ± 86.7a	632.7 ± 89.0b	638.1 ± 83.4b	610.1 ± 88.7a
V̇co2, L/day	499.7 ± 60.6a	511.9 ± 63.0b	471.5 ± 63.4c	447.4 ± 57.2d
RER	0.828 ± 0.024a	0.811 ± 0.024b	0.739 ± 0.009c	0.735 ± 0.019c

Values are means ± SD, n = 8. Means not sharing a common superscript letter are significantly different. EXO, control diet with the addition of exogenous ketones; FAST, total fasting; ISO, isocaloric control diet; KETO, isocaloric ketogenic diet; RER, respiratory exchange ratio; V̇co₂, carbon dioxide production; V̇o₂, oxygen consumption.

Macronutrient oxidation rates and their proportional utilization, normalized for the total energy expenditure are shown for ISO, KETO, and FAST in Fig. 3. When compared with KETO and FAST, 24-h carbohydrate oxidation rate was higher with ISO (P < 0.001). By contrast, the rate of fat oxidation was lower in ISO (compared with KETO and FAST; P < 0.001). There was no difference in the rate of protein oxidation between ISO, KETO, and FAST, whereas nitrogen balance was, however, negative with KETO and FAST (−3.1 ± 1.6 g/day and −11.0 ± 1.3 g/day; both P < 0.05).

Figure 3.

Macronutrient oxidation (in %, normalized for 24-h energy expenditure) compared between three interventions: isocaloric control diet (ISO), isocaloric ketogenic diet (KETO), and fasting (FAST); n = 7. CHO, carbohydrates; EE, energy expenditure; Ox, oxidation. Data are presented as means ± SD; ***P < 0.001 (RM-ANOVA, Bonferroni correction).

Regulation of Glucose Homeostasis

Comparing postprandial glucose and insulin levels as well as insulin sensitivity (S_I), no differences were observed neither between the three meals within an intervention day nor between ISO and EXO (Table 2).

Table 2.

Postprandial 3-h iAUCs of glucose and insulin and insulin sensitivity

	ISO	EXO
Glucose iAUC
Breakfast, mg/dL × 3 h	17.7 ± 40.9	33.6 ± 27.5
Lunch, mg/dL × 3 h	60.3 ± 48.3	69.2 ± 39.5
Dinner, mg/dL × 3 h	87.5 ± 42.1	68.5 ± 44.8
Insulin iAUC
Breakfast, mU/L × 3 h	177.3 ± 57.4	190.6 ± 72.6
Lunch, mU/L × 3 h	176.3 ± 59.6	153.1 ± 76.1
Dinner, mU/L × 3 h	189.9 ± 94.8	142.0 ± 53.1
Insulin sensitivity, SI
Breakfast, dL/kg/min per µU/mL	3.1 × 10−4 ± 3.2 × 10−4	2.1 × 10−4 ± 1.9 × 10−4
Lunch, dL/kg/min per µU/mL	1.3 × 10−4 ± 1.9 × 10−4	1.2 × 10−4 ± 2.2 × 10−4
Dinner, dL/kg/min per µU/mL	5.5 × 10−5 ± 1.2 × 10−4	2.0 × 10−4 ± 4.1 × 10−4

Parameters were assessed by oral mixed meal minimal model for three meals (breakfast, lunch, and dinner) in comparison between the isocaloric control diet (ISO) and with the addition of exogenous ketones (EXO); n = 8. iAUC, incremental areas under the curve; S_I, insulin sensitivity.

Insulin secretion assessed by C-peptide excretion tended to be higher with EXO compared with ISO (P = 0.103, Fig. 4A). The difference in C-peptide excretion showed high interindividual variation (ΔISO-EXO: +15.9 ± 22.4 µg/day; ΔISO-FAST: −43.46 ± 26.57 µg/day; ΔISO-KETO: −40.39 ± 23.78 µg/day) and was between ISO and EXO inversely associated with FMI (r = −0.77, P < 0.05; Fig. 4B) but not with parameters of insulin sensitivity (HOMA-B, S_I).

Figure 4.

C-peptide excretion (A) in comparison between isocaloric control diet (ISO), control diet with the addition of exogenous ketones (EXO), isocaloric ketogenic diet (KETO), and total fasting (FAST), and correlation between the difference of C-peptide excretion between ISO and EXO and fat mass index (FMI) (B) in women (●) and men (■); n = 8. Data are presented as means ± SD; *P < 0.05, **P < 0.01.

HOMA-B did not change in all interventions with caloric intake (ISO, EXO, and KETO) whereas FAST led to an increase in HOMA-B (Fig. 5; P < 0.05).

Figure 5.

Homeostatic model assessment β-cell function (HOMA-B) compared between pre- and postintervention shown for all diet conditions. ISO, control diet, and with the addition of exogenous ketones (EXO), isocaloric ketogenic diet (KETO), and total fasting (FAST); n = 7. Data are presented as means ± SD; *P < 0.05 (paired t test).

DISCUSSION

The results of the present study showed that exogenous ketones reduced RER with EXO suggesting a glucose-sparing effect (Table 1). Accordingly, O’Malley and coworkers (24, 25) found that exogenous ketone salts decreased CHO oxidation during exercise, which has been explained by a “Randle cycle-like” effect, i.e., the suppression of glycolysis by ketone bodies was attributed to inhibition of pyruvate dehydrogenase (26). By contrast, when ketone esters were administered together with oral glucose before exercise, no difference in exogenous and endogenous glucose oxidation or metabolic clearance rate (as determined by dual-stable isotope dilution technique combined with indirect calorimetry) was observed compared with CHO alone (27). The discrepant findings might be explained by a shorter duration of our study as well as by a single application of KB, whereas the intensity of exercise had been higher in the study by Howard et al. (27). Therefore, exercise may diminish glucose-sparing by increased glucose requirement.

Exogenous ketones represent a metabolic fuel that do not impact basal and postprandial glucose metabolism (see results). However, they may alter gluconeogenesis, increase glycogen storage, and decrease protein breakdown and protein oxidation (see results). Comparing the different increases in blood ketones between EXO, KETO, and FAST, we have to compare differences in the “input” of KB, which is 10.7 g/day of d-BHB in EXO (see results) whereas an endogenous KB production of ∼40–80 g results in blood ketones of ∼1–2 mM (28) as observed after 1 day in response to KETO and FAST (see Supplemental Information).

The KETO condition led to a slightly negative nitrogen balance despite isocaloric conditions which can be attributed to a lower protein content in the diet. Interestingly, HOMA-B was maintained with KETO compared with FAST (Fig. 5) despite similar increases in NEFA levels, fat oxidation, and ketone bodies in these interventions.

We found a lower RER with EXO whereas 24-h urinary nitrogen excretion rate was unchanged (see results). For comparison, a ketone ester drink led to increased postexercise muscle glycogen synthesis (assessed by muscle biopsies) together with higher glucose uptake and higher insulin levels during hyperglycemic clamp (29). However, in that study, muscle glycogen was depleted before the exercise protocol, whereas, in our study, only moderate-intensity exercise was performed, which should not have led to sizable glycogen depletion. The lower CHO oxidation by ketone salts during exercise was compensated by an increased fat oxidation (25). Unfortunately, we could not calculate fuel oxidation rates during exogenous ketone administration because of stoichiometry of AcAc, the final step in KB oxidation is 1.00 (analogous to carbohydrates), determining total volume of distribution and difficulties in accounting for total plasma ketones (d-BHB, l-BHB, AcAc, acetone), excretion (acetone via exhalation, urinary excretion of BHB and AcAc), and tissue. Also, changes in pH levels may have affected V̇co₂ leading to a higher RER when ketone esters, which lower pH levels, are administered (30). By contrast, ketone salts increase pH levels, which may result in an underestimation of RER and thus overestimation of fat oxidation (15).

When compared with ISO the lower RER with EXO was not reflected by differences in plasma glucose levels (Table 2). By contrast, studying in healthy individuals or patients with impaired glucose tolerance other authors found lower glucose levels when ketone esters were administered together with an oral glucose tolerance test (OGTT) (5–8). Lower glycemia after an OGTT may, in particular, occur in subjects with hepatic insulin resistance since ketone bodies have been shown to suppress gluconeogenesis. The underlying mechanism of the suppression of gluconeogenesis by ketone bodies may be twofold. First, exogenous d-BHB was found to lower blood glucose in part by decreasing the availability of l-alanine for gluconeogenesis (12, 31, 32). Second, a small yet immediate increase in insulin secretion after ketone ester ingestion may be sufficient to inhibit hepatic glucose production (7, 33).

An insulinotropic effect of oral ketone supplements has been shown during an OGTT or hyperglycemic clamp (6, 8, 29). However, KB (either administered orally or intravenously) also increased insulin or C-peptide levels without simultaneous glucose administration (11–13, 33–35). A direct comparison of oral and intravenous ketone administration revealed a higher insulinotropic effect of oral ketone salts compared with intravenous administration, presumably because of an incretin effect via cholecystokinin (34). The lack of an insulinotropic effect in our study, despite comparable dosing, may be explained by the fact the ketone salt dose was divided into three portions administered together with the meals. Nevertheless, we found a tendency toward a higher insulin secretion rate that was associated with FMI of the subjects and would have reached significance with 14 subjects at a power of 80%.

Strengths and Limitations of the Study

The strengths of the study are the highly controlled and standardized setting of a WRIC and dietary intervention with an intraindividual crossover design for four 24-h conditions. However, there are at least three potential limitations concerning the results of this study. First, although we administered the maximal dosage of ketone salts as recommended by the manufacturer, plasma levels of ketones were low, which may be attributed to the fact that in our study, we have applied a racemic mixture of KB, thus, in EXO blood ketone levels reflect d-BHB only and therefore underestimate total KB. To address this issue, l-BHB values can only be determined using mass spectrum analysis, whereas our colorimetric assay did not discriminate between configurations (10). However, d-BHB rather than l-BHB is metabolized physiologically (15). To our knowledge, in humans, enzymes related to KB-metabolism are specific to the d-isomer whereas the metabolic fate of the l-isomer remains unclear. This issue may add to discrepancies between studies resulting in lower blood ketone increases with ketone salts versus esters (15). Concomitantly, ketone salts led to a higher urinary excretion of l-BHB (15). By contrast, ketone monoesters are generally characterized by a higher increase in plasma levels (15). Second, we could not calculate fuel oxidation during exogenous ketone administration because the assumptions of fuel oxidation are violated when ketones are administered. Third, the small sample size led to a limited power that does not allow to analyze sex effects.

Conclusions

In conclusion, even if carbohydrates and ketones usually do not coexist in a physiological state (since endogenous ketosis requires carbohydrate depletion), exogenous ketone salts lead to a lower RER (suggesting a lower glucose oxidation rate) without lowering glucose levels.

DATA AVAILABILITY

The data that support the findings of this study are not publicly available due to the data privacy statement in the subject information form.

SUPPLEMENTAL DATA

Supplemental Material: https://doi.org/10.6084/m9.figshare.24922659.

GRANTS

This study was funded by budgetary resources of the Christian-Albrechts-Universität Kiel, Germany.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.B.-W. conceived and designed research; R.D., F.A.H., U.S., and A.B.-W. performed experiments; R.D., F.A.H., and A.B.-W. analyzed data; R.D., F.A.H., M.J.M., and A.B.-W. interpreted results of experiments; R.D. prepared figures; R.D. and A.B.-W. drafted manuscript; R.D., F.A.H., M.J.M., U.S., G.R., and A.B.-W. edited and revised manuscript; R.D., F.A.H., M.J.M., U.S., G.R., and A.B.-W. approved final version of manuscript.