Increases in Serum Growth Hormone Concentrations Associated with GHB Administration (2024)

Search for other works by this author on:

Article history

Abstract

Introduction

Gamma-hydroxybutyrate (GHB) is an endogenous short-chain fatty acid that initially gained attention because of its ability to cross the blood–brain barrier, and was later used medicinally as a general anesthetic (1, 2). Due to an unpredictable duration of action, its use in anesthesia has now ceased, but GHB (as Xyrem^®) has since been licensed for the treatment of narcolepsy (associated with cataplexy) (3–5) and for alcohol withdrawal (6).

It is however the non-medicinal use of GHB which has been the subject of greater attention, from boththe media and scientific community. In the last 15 years, GHB (and its related analogs such as gamma-butyrolactone (GBL)) has been linked to cases of drug-facilitated sexual assault (DFSA), more commonly termed “date-rape” (7–10). In such cases toxicological proof of GHB administration is difficult due to its endogenous nature (0.2–1 mg/L in urine), and rapid elimination from the body (half-life 20–45 min), which results in a return to baseline concentrations in urine after only 8 h post-administration (11–18).

The recreational use of GHB has also become popular in association with the “clubbing scene”, particularly among hom*osexual men (19, 20), as a consequence of the disinhibitory effects associated with low doses (21).

The non-medicinal use however began when it was reported that GHB administration produced an increase in serum growth hormone (GH) concentrations (22). In this initial study, six healthy male volunteers were given a 2.5-g dose (~35 mg/kg, assuming 70 kg body weight) of GHB by intravenous (i.v.) injection, with five volunteers falling asleep due to the associated hypnotic (soporific) effects at this dose. The mean reported GH C_max occurred at 60 min post-injection (32.1 ng/mL), and demonstrated a significant increase of ~20-fold from mean basal concentrations. Perhaps unsurprisingly this observation led to the use of GHB as a muscle-building supplement among body builders (23).

A later study used the relationship between GHB and GH to investigate the central mechanism of action for GHB in the brain through the co-administration of several receptor agonists (24). A 1.5-g oral dose (~21.4 mg/kg, assuming 70 kg body weight) to 10 male volunteers produced rises of ~8–9 ng/mL in serum GH, peaking 45 min after administration. This equated to roughly three times the reported control values in the study. No major side effect including sedation was observed following the administration of GHB in this instance.

Van Cauter et al. (25) focused on the ability of GHB to augment the natural increase in GH serum concentrations associated with the onset of slow-wave sleep. Eight healthy male volunteers were given doses of 2.5 g, 3.0 g or 3.5 g at 22:45 h prior to a “bedtime” of 23:00 h. Basal serum concentrations were typically <10 ng/mL prior to sleep, while GH C_max associated with the onset of sleep in the control group was ~20 ng/mL. All three doses of GHB produced augmented GH concentrations with the larger doses producing a doubling in the GH C_max compared with GH pulses observed in the absence of GHB administration. Neither IGF-I nor IGFBP-3 concentrations were shown to increase following any dose. The authors suggested that GHB would only act as a GH secretagogue if sleep is induced since no increase in GH concentration was seen prior to the onset of sleep.

Several circ*mstances have been elucidated, which are capable of decreasing or negating the increase in GH associated with GHB administration. The administration of GHB is unable to produce an increase in GH in cocaine addicts, and if co-administered with benzodiazepine or serotonin receptor antagonists (24, 26, 27).

The study reported here seeks to clarify whether the production of GH post-GHB administration is solely associated with the onset of sleep. Serum collected from volunteers (six men and six women), who remained awake following a low therapeutic oral dose (25 mg/kg) of GHB in the morning, was analyzed to determine GH and GHB concentrations. Notably, the study reports data in female volunteers, which has so far been under investigated. GH serum concentrations were compared with GHB concentrations and the effectiveness of GHB as a GH secretagogue is considered.

Experimental

Reagents

GHB (sodium salt), GBL and trifluoroacetic acid were all purchased from Sigma-Aldrich Company Ltd, Poole, UK. Sodium hydroxide pellets, chloroform, anhydrous potassium dihydrogen phosphate and anhydrous disodium hydrogen phosphate were purchased from Fisher Scientific, Loughbourgh, UK. Deuterated(d6)-GHB (4-hydroxy-2,2,3,3,4,4-hexadeuterobutyric acid sodium salt) was supplied as 1 mg/mL in methanol by LGC Standards, Teddington, UK. All water was purified to 18 MΩ · cm, using an Elga Maxima coupled to an Elga Purelab Option - R15, Waters, UK. Xyrem^® (500 mg/mL) was obtained from UCB Pharma Ltd, Berkshire, UK. Serum tubes (Becton, Dickinson Vacutainers™, 10 mL, red top, conventional closure, no additive) were purchased from MidMeds, Essex, UK.

GH assay

GH concentrations were measured using a commercially available immunoassay (Immulite 1000, growth hormone assay (LKGRH1), Siemens, Llanberis, UK). The assay is a solid phase, two-site chemiluminescent immunometric assay. As reported by the manufacturer: the lower limit of analytical sensitivity is 0.01 ng/mL and the reportable range is 0.05–40 ng/mL. The specified intra-assay precision is between 5.3 and 6.5 % over the range 1.7–31 ng/mL. The specified inter-assay precision is between 5.5 and 6.2% over the range 3.0–18 ng/mL. Assay measurements were calibrated against WHO NIBSC 2^nd IS 98/574.

Volunteer recruitment

Ethical approval for the GHB administration study was obtained from our institutional research ethics committee (approval number CREC/06/07-30). Written informed consent was obtained from the volunteers (six men and six women). Males had a mean age of 25 years (range 21–36 years) and a mean body mass index (BMI) of 23.7 kg/m² (21.7–27.1 kg/m²). Females had a mean age of 26 years (22–32 years) and BMI of 23.0 kg/m² (19.5–25.9 kg/m²). Prior to the study, volunteers were assessed to be in good health. Exclusion criteria included a history of liver disease, succinic semi-aldehyde dehydrogenase deficiency and those breast feeding. All volunteers were negative for the current use of sedatives, recreational drugs and pregnancy (females only), by analysis of a urine sample collected 1 week before GHB adminstration and by self-reporting.

Study design

On the day of the study, volunteers (in groups of three) were asked to finish a light breakfast (avoiding fried foods) by 7.30 am before arriving at the secure study suite. At 10.00 am, a single dose (25 mg per kg body weight) of GHB was administered (time = 0 h) in the form of the pharmaceutical preparation Xyrem^® (sodium oxybate, 500 mg/L). The mean dose was 1.8 g (as the sodium salt) and ranged from 1.4 to 2.6 g (equivalent to 1.2–2.1 g GHB). The preparation was diluted with 60 mL water prior to administration as directed by the manufacturer.

Sample treatment

Serum samples: 10 mL of whole blood was allowed to clot for 30 min in the Vacutainer™ in which it was collected. Samples were centrifuged at 1,000 g for 10 min. The supernatant (serum) was then transferred into polypropylene tubes, which were stored at –20°C until analysis.

GHB analysis

The developed method for the analysis of GHB in urine was validated with respect to range, linearity, accuracy, repeatability and reproducibility. Sample extraction, instrumental and data interpretation criteria are outlined in our previously published papers (14, 18).

Statistics

Statistical analysis was performed using SPSS statistics version 22. GH concentrations were checked for normality using the Shapiro–Wilks statistic, and shown to be considered normal within each time point for the range of 25–300 min for males and −10 to 240 min for females. A subsequent repeated measures ANOVA was performed with the within-subject factor being time (relative to GHB administration), and used to identify any significant increase in GH concentrations. Following the identification of a significant rise in GH concentration, a pairwise (least significant difference, LSD) analysis was performed to identify at what time point significance occurred. Statistical analysis was performed separately on males and females, and when testing for normality GH concentrations below the limit of detection (LOD) of the assay were excluded using a pairwise approach. For the repeated measures ANOVA, GH concentrations below the LOD were assumed to be half the LOD of the assay. Statistical significance is reported at a 95% confidence level in all cases. A Spearman's Rank Order Correlation (rho) was used to calculate the strength of the relationship between GHB and GH concentrations.

Results

Basal GH concentrations

As expected, basal serum GH concentrations were larger in females (mean = 5.4 ng/mL, range = 1.08–11.2 ng/mL) compared with males (mean = 0.26 ng/mL, range 0.03–1.01 ng/mL). GH concentrations in females first decreased from the time of GHB administration (Figure 1a), but no such trend could be observed in males (Figure 1b), possibly due to the associated smaller concentrations.

GH concentrations following GHB administration

Eleven of the twelve individuals in the study showed a marked increase in GH concentrations following GHB administration with the exception being a singular female volunteer. Figures 1a and b demonstrate the substantial variation that was observed in C_max both between the sexes and within both males and females. For males the mean serum GH C_max = 3.67 ng/mL (range 7.21– 0.12 ng/mL) and for females the mean serum GH C_max = 9.11 ng/mL (range 20.6–4.34 ng/mL). The difference between mean basal (−10 min) concentrations and mean C_max in males and females was broadly similar, being 3.41 ng/mL for males and 3.75 ng/mL for females.

Mean GHB concentrations for all 12 volunteers are plotted in Figure 3 and are reported in more detail elsewhere (18). GHB serum elimination profiles were similar in male (mean C_max = 59.7 µg/mL) and female (mean C_max = 59.2 µg/mL) volunteers. The inter-individual variation was also relatively small for all 12 volunteers (mean C_max = 59.5 µg/mL, SD = 10.2).

Statistical analysis

GH concentrations were significantly different between time points within each sex (repeated measures ANOVA, males (F(14,70) = 8.32, P < 0.05), females (F(14,70) = 6.33, P < 0.05)).

Significant increases (LSD, P = 0.05) in GH concentration occurred in both male and female volunteers at 45 and 60 min in agreement with visual data inspection.

Correlation of GHB and GH data

The use of a linear model resulted in a statistically significant correlation (P < 0.05, N = 180) with an r-value of 0.390 demonstrating a positive correlation between GHB and GH concentrations of medium strength following GHB administration. These data however result in a low shared variance of 15%, therefore predicting serum GH concentrations based on the serum GHB concentrations post-GHB administration has little value.

Normalized GH concentrations

In order to compensate for the large difference in the absolute GH concentrations among the volunteers, GH concentrations were normalized relative to each volunteer's C_max (100%). As expected, this significantly reduced the spread of data (Figure 3).

The normalized data are able to demonstrate that, while absolute increases in GH concentrations in males and females had a similar mean value (3.41 ng/mL for males and 3.75 ng/mL for females), the relative increase in males is much more significant due to the lower basal GH concentrations. Female concentrations increased on average 2.0 times pre-administration basal values (range = 1.2–4.0 times) while in men the mean increase was 29 times (range = 4.2–68 times).

Due to the greater starting female GH concentrations, taking the pre-administration values as the sole basal concentration was likely to result in an underestimate in the relative GH increase. Therefore, an alternative approach whereby the average GH concentration of the pre-administration sample and the 150-min sample (which was taken to represent the baseline GH concentration post the GHB induced pulse) was used to give an improved representation of the increase in relative female serum GH concentrations. While it may be considered that GH secretion would be suppressed post-stimulation due to negative feedback, only minor (if any) increases in serum GH concentrations were observed between 150 and 480 min (Figures 1a and b), suggesting that any inhibitory effect on endogenous GH production was minor. Using this approach female concentrations increased on average 3.0 times pre-administration basal values (2.3–5.0 times) while in men the mean increase was 16 times (3.0–25 times).

Discussion

It is important to consider the pulsatile nature of GH release when considering whether any observable increases in GH concentrations seen in our data are associated with the administration of GHB or representative of typical pulsatile release. Previous research has reported the mean frequency of GH pulses in six healthy males to be 4.3 pulses (range = 2–8 pulses) over a 24-h period (28). Unlike the GH increases observed in this study natural GH pulses were shown to vary largely at their time of occurrence.

In the study presented, all 11 volunteers displaying an observable pulse of GH did so within 1 h of GHB administration which therefore related to ~20–35 min post the observed GHB C_max (Figures 1–3).

Our work clearly demonstrates that the administration of GHB is able to produce an observable and statistically significant increase in serum GH concentrations in both male and female volunteers, even when administered in the low therapeutic range (25 mg/kg). The variation in GH concentration was much larger than the variation in GHB concentration, suggesting a considerable inter-individual variation in the ability of GHB to stimulate GH release.

The six male volunteers produced a lower observed increase in GH concentration when compared with the data of Takahara et al. (22). Though a lower dose was given in our study (25 mg/kg) compared with this previous work (2.5 g, ~35 mg/kg), the increase was proportionally much lower with a mean C_max of 3.67 ng/mL rather than 32.1 ng/mL. It is however important to note that we used an oral administration, which represents the usual method of GHB administration, rather than i.v. route of administration used by Takahara. It should also be considered that the basal concentrations for the male volunteers were smaller in our study (mean, 0.26 ng/mL) compared with ~2 ng/mL which was reported previously. The relative increase in GH concentrations is therefore similar for both studies.

Van Cauter et al. gave a range of oral GHB doses (2.5, 3.0 and 3.5 g) to eight healthy male volunteers. These doses are equivalent to ~35, 42 and 50 mg/kg, respectively, and therefore are all in excess of the dose administered in this study (25). As their study focused on the augmentation of GH production associated with the onset of slow-wave sleep, study volunteers were given the GHB dose prior to bedtime (23:00 h) rather than in the morning, as in the study reported here. All three GHB doses were shown to augment the natural increase in GH associated with the onset of slow-wave sleep. The smaller dose produced an increase of ~10 ng/mL while the medium and larger doses produced a relative increase of ~20 ng/mL, doubling the increase seen without GHB administration. Though the authors suggest that GHB only acts as a GH secretagogue if sleep is induced, our work contradicts this statement as none of the participants in our study fell asleep (generally exhibiting mild sedation or mild euphoria).

It should be noted that no increase in GH concentration was seen in volunteer 1 (female). As volunteers were confirmed to be negative for the current use of sedatives and recreational drugs based on urinary analysis (1 week before GHB administration) and self-reporting, we can draw no firm conclusion to explain the non-response in this volunteer.

Despite the observed increases in absolute GH concentrations being small (up to ~10 ng/mL), the relative increase in GH concentrations particularly in males was significant, with GH concentrations on average 29 times greater than basal concentrations (Figure 3). Such increases may still be relevant from both a therapeutic and sports doping perspective. Although it is acknowledged a placebo control would have been beneficial, but was not part of the main study. The increases published here for this small GHB dose are ~15% of that seen from a 0.15-U/kg subcutaneous dose of somatropin (29). Given that the work of Van Cauter suggests that larger doses will be capable of producing an even greater response, the potential for GHB abuse in sport is noteworthy (25). Though current research in this area is limited, the long-term administration of GHB to alcoholics did not affect muscle mass or waist-to-hip ratio, but this may be a consequence of the GH release associated with GHB administration being suppressed in alcoholics (30). Further work on healthy individuals is therefore required to ascertain the potential effects of long-term GHB use in healthy individuals.

Conclusion

In conclusion, our work demonstrates that low oral doses of GHB are able to produce significant increases in GH concentrations in both male and female volunteers. Unlike previous work our data reveal that increases in serum growth hormone concentrations are observed in the absence of sleep (25).

References

© The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Issue Section:

Article

Email alerts

Citing articles via

More from Oxford Academic