Three Weeks on a Ketogenic Diet Reduces Free Testosterone and Free Estradiol in Middle-Aged Obese Men and Women

Background Beta-hydroxybuturate (β-OHB) supplements are commonly utilized in sports by both recreational and professional athletes. In a recent study, we observed a drop in testosterone levels following the oral ingestion of racemic sodium-β-OHB. In this investigation, we aim to determine whether a single oral dose of ketone ester (study I) and prolonged endogenous ketosis (study II) also reduces testosterone levels. Design This investigation integrated samples from two distinct studies. Study I was a randomized, controlled, crossover trial with ten healthy, young male participants receiving either a weight-adjusted ketone ester or control (water, CTR) and vice versa following an overnight fast. Repeated blood sampling was used to monitor plasma β-OHB and testosterone levels. Study II, another randomized, controlled, crossover trial, included 11 middle-aged participants (five males). They followed either a ketogenic diet (KD) characterized by low carbohydrates and high fat content or a standard diet (SDD) for three weeks. After each study period, participants underwent examination following an overnight fast, with repeated measures employed to analyze concentrations of plasma β-OHB and sex hormone levels. Results Study I: Testosterone decreased from 23.8 ± 2.4 nmol/l to 22.3 ± 2.5 nmol/l 300 minutes after the ketone ester and increased from 20.9 ± 2.1 nmol/l to 22.2 ± 1.9 300 minutes after CTR. This difference was not significant, p = 0.06. Study II. Total testosterone was unaffected after the KD compared to the SDD in men (20.2 ± 1.23 nmol/l vs. 18.2 ± 1.23 nmol/l (p = 0.1)) and was lower after KD in women (0.87 ± 0.06 vs. 1.1 ± 0.06 nmol/l (p < 0.0001)). Sex hormone-binding globulin (SHBG) increased in men after KD compared with SDD (31.2 ± 2.6 nmol/l vs 25.0 ± 2.6 nmol/l, p < 0.0001) and women (26.5 ± 3.05 nmol/l vs 24.2 ± 3.05 nmol/l, p = 0.003). The free androgen index decreased after KD in men (ratio: 0.65 ± 0.05 vs. ratio: 0.74 ± 0.05, p = 0.04) and in women (ratio: 0.036 ± 0.006 vs. SDD 0.05 ± 0.006, p = 0.0001). Free estradiol index was also found lower after KD in men (ratio: 3.1 ± 0.8 vs. ratio: 4.8 ± 0.8, p = 0.0003) and in women (ratio: 1.2 ± 2.2 vs. 9.8 ± 2.2, p = 0.0001). Conclusion Our findings indicate that the acute ingestion of ketone ester may not reduce testosterone levels in healthy young males. However, a three-week exposure to KB from a KD results in an increase in SHBG in men and women with obesity as well as it lowers free testosterone and estradiol for men and women. We thus present evidence of crosstalk between alterations in a metabolite, β-OHB, and the regulation of the hypothalamic-pituitary-gonadal axis from a KD. The clinical impact of this reduction remains to be investigated. This trial is registered with NCT04156477 and NCT05012748.

Ketone supplements are mainly available as ketone salts or ketone esters, and both approaches result in an elevated circulating concentration of β-OHB ranging from 0.5 mmol/l to 5.5 mmol/l [6,7].Many of these approaches lead to weight loss which in turn raise testosterone [8].Exogenous ketosis is obtainable through various methods of which infusion of racemic sodium-β-OHB (salt) and ingestion ketone esters are frequently employed [9].Sustained intake of β-OHB does not consistently result in weight loss, making it a suitable option for maintaining ketosis.
Testosterone improves physical performance [10], maintains skeletal muscle mass and strength on top of its key functions on libido and sexual function, mood, cognition, and cardiovascular health [11,12].A potential testosterone decline may counteract some of the potential positive efects associated with ketosis [13].
Recently, we showed that an oral sodium-β-OHB dose decreased testosterone in healthy young males [14].To further investigate whether ketosis is associated with lower testosterone levels, we aimed to investigate if this fnding is compound specifc and if longer lasting endogenous ketosis also lowers testosterone.
Consequently, we examined the efects of a three-week ketogenic diet (KD) on testosterone and estradiol concentrations in samples from middle-aged, obese men and women, as well as the efects of a single oral ketone ester dose on testosterone concentrations in samples from healthy young males.We hypothesized that testosterone concentrations would decrease following acute ketosis induced by ketone ester intake and longer-lasting endogenous ketosis from KD.

Study Design.
We used samples from two randomized, controlled trials, whereof other data have previously been published [15,16].

Acute Ketone Ester Ingestion.
In brief, the study included ten healthy young males investigated on three separate occasions of which we report data from the placebo (CTR) and ketone ester ingestion (KET) study days.In this study, the participants acutely ingested 714 mg/kg ketone monoester D-BHB-R-1,3-butanediol (HVMN V1 ketone ester) or taste-adjusted water and underwent blood sampling for fve hours [7,16].

Ketogenic Diet.
In brief, the study included 11 (fve males, six females) healthy, overweight participants.All participants received an individualized KD plan adjusted for calorie intake based on sex, weight, and approximate activity level.Each diet plan included options for every meal, ensuring a macronutrient distribution of 5 E% carbohydrates, 20 E% protein, and 75 E% fat.Tese diet plans were created by a dietitian using Vitakost Pro, a professional dietary tool that utilizes the Danish Food database.A low glycemic index source of carbohydrate was recommended during the KD.Fat intake ranged from 200 to 300 grams/day, with an even distribution between saturated and monounsaturated fatty acids.Participants were responsible for buying ingredients, preparing meals, and adhering to the diet in their usual environment.During the standard diet (SDD), participants were instructed to eat with a macronutrient distribution following the Nordic Nutritional Recommendations (45-60 E% carbohydrates, 10-20 E% protein, 25-40 E% fat).Te participants were randomly assigned in a 1 : 1 ratio to receive either KD or SDD in the frst study period.To ensure compliance, all participants were instructed to measure their blood β-OHB concentration using a point-of-care ketonometer (Freestyle Libre; Abbott Diabetes Care Ltd., England) every morning (7: AM) and evening (7: 00 PM) during both interventions.Results were reported daily.Participants measured β-OHB twice daily and those who failed to increase their plasma ketone levels to 0.3 mmol/l during the frst week of KD were excluded from the study.An experimental day was included at the end of each trial period [15].

Statistical Analysis.
A linear mixed model was used to compare the efects of ketone ester treatment compared with placebo and KD with SDD.Treatment, period, and 2 Journal of Nutrition and Metabolism treatment sequence were defned as fxed efects and patients as random efects.For the repeated measurements during the sampling period after either a single dose KE/placebo treatment or KD, a treatment-by-time interaction was added as fxed efects.Te reported p values were calculated using least-squares means analyses of the respective linear mixed models.Te residuals were tested for normality and homoscedasticity.Data were log transformed if needed.Te efect size of KE treatment compared with placebo treatment is presented as pairwise mean diference with SEM.All graphics and statistical analyses were performed using SigmaPlot 14 (San Diego, CA, USA) and R (Version 2022.12.0,RStudio, Posit, USA).

Study 1 (Oral 3OHB Ester Ingestion).
Ketosis was achieved and the participants reached a peak ketone concentration around 5.5 mmol/l.

Study 2 (Ketogenic Diet
). Eleven (fve males, six females, post-menopausal) healthy, overweight participants with body mass index (BMI): 28-40 kg/m 2 and age between 50 and 75 years were included.Tree weeks of KD induced a signifcant higher circulating β-OHB concentrations compared to a standard diet (SDD) (1.0 ± 0.1 mmol/l vs. 0.1 ± 0.01 mmol/l), as previously reported [15].We measured sex hormones at three time points after each of the three weeks intervention.Tere was no carry-over efect detected and no interaction of visit order.

Discussion
We investigated the efects of acute ketosis on human sex hormone regulation in men and women.We hypothesized that acute ketone ester ingestion would lower testosterone.However, this was not the case from drinking ketone ester contrary to our previous fndings after drinking sodiumβ-OHB [14].We further investigated the efect of endogenous prolonged KD-induced ketosis on sex hormones in middle-aged, obese men, and women.Here, we hypothesized that KD would lower testosterone.Tis was the case in men and women with increased SHBG and lower FAI for both sexes.We further found lower total estradiol as well as FEI in men and women after three weeks of KD.

Study 1.
We found no diference in testosterone from drinking the ketone ester.Whether the efect is diferent from long term use remains uncertain.It is notable that by chance, the baseline values in the two groups at baseline were higher on the KET day.Te fnding is contrary to previous fndings where we observed with Na-D/L-β-OHB ingestion in a similar group of young healthy males [14].

Study 2.
During KD, β-OHB increased to approximately 1 mmol/l and remained at overnight fasting values around 0.1 mmol/l during the SDD.Notably, total testosterone declined in women and was unaltered in men.Among obese males we also observed a large (approximately 50%) increase in SHBG, and the calculated FAI decreased signifcantly, following KD.In women an SHBG signifcant increase of some 10% was also observed together with a signifcant decline in FAI.Obesity is a well-established risk factor for insulin resistance and diabetes [19].In postmenopausal women not using hormone replacement therapy, a lower SHBG and a high FAI is associated with cardiovascular events [20].It is also suggested that high levels of SHBG favors a healthy cardiometabolic profle [21] and testosterone in men is protective with regards to cardiometabolic health [22].
Hydroxycarboxylic acid receptor 2 (HCAR2) is present in hepatocytes [23] and hepatic macrophages promoting anti-infammation [24].Te age-related loss of HCAR2 in hepatic tissue in rodents is suggested to play an important role in hepatic lipid accumulation underlining the link between β-OHB and metabolic health [25].Hence, it is noteworthy that just three weeks of adopting the KD intervention resulted in elevated SHBG levels in obese men Journal of Nutrition and Metabolism and women, potentially improving their cardiometabolic risk profle.Metabolic associated fatty liver disease (MAFLD) correlates strongly with low levels of SHBG [26].MAFLD increase with age in men and women [27] and KD is shown to improve MAFLD, and this could link the KB increase to the prominent SHBG increase in our study [28].
A HCAR2 signaling efect cannot be ruled out either, as the pituitary gland and both sex gonadal glands express HCAR2 on their surface [23].Te unchanged LH and FSH after KD point towards a steady state hormone balance.Te signaling, if present, most likely takes place in the gonads as the pituitary hormones are completely stable in both studies.We found lower estradiol in men and women as well total as SHBG corrected FEI.In women, estradiol is a wellestablished protective factor against cardiovascular disease [29].On the other hand, low estradiols in women are a protective factor with regards to breast cancer [30].In men the absence or declining estradiol levels is correlated with sexual dysfunction [31].
One major limitation of our study was the lack of power in the KD study to investigate men and women separately, as it was designed to investigate cardiac metabolism with no obvious need to divide the participants in men and women.Investigating the β-OHB efect on sex hormones we needed to split the population into males and females, which left only fve men and six women in each group Both studies employed crossover design, with each participant serving as their own control, thereby enhancing the studies' power and the reliability of the results.Carry-over efects were ruled out as was the intervention order.
In conclusion, our fndings suggests that consuming ketone esters does not afect testosterone levels acutely in healthy young men, contrary our previous study on Na-D/Lb-OHB.Furthermore, our investigation revealed that a ketogenic diet for three weeks induced higher levels of SHBG in obese men and women.Tis fnding was accompanied by lower FAI and FEI in men and women after KD.Tis sheds light on a possible interplay between changes in β-OHB, a metabolite, and the control of the hypothalamic-pituitarygonadal axis, over time.

Figure 1 :Figure 2
Figure 1: Data from all ten participants in study 1.Panel (a) total testosterone, (b) sex hormone binding globulin (SHBG), (c) Follicle stimulating hormone (FSH), and (d) Luteinizing hormone (LH) concentrations are shown on the y-axis and time on the x-axis.Te black arrow at time � 0 minutes indicates the time of ingesting either ketone ester (KET) or placebo (CTR).Black circles illustrate the mean value (±SEM) on the KET day and white circles illustrate the mean value (±SEM) on the CTR day.Linear mixed model was used to test for between-treatment pairwise comparison (KET/CTR).

Figure 2 :Figure 3 :
Figure2: Hormone levels for fve male participants in study 2. Panel (a) Total testosterone, (b) sex hormone binding globulin (SHBG), (c) free androgen index (FAI) (total testosterone/SHBG), (d) total estradiol, and (e) free estradiol index (FEI) (total estradiol/SHBG).Black circles illustrate the mean value (±SEM) on the KET day and white circles illustrate the mean value (±SEM) on the CTR day.Linear mixed model was used to test for between-treatment pairwise comparison (KET/CTR).