Original Article

Serum testosterone and urinary excretion of steroid hormone metabolites after administration of a high-dose zinc supplement

  • European Journal of Clinical Nutrition volume 63, pages 6570 (2009)
  • doi:10.1038/sj.ejcn.1602899
  • Download Citation

Contributors: KK managed the realization of the study, sample and statistical analysis and interpretation of the results and led the writing of the paper. MKP initiated the trial, was in charge of the design and the implementation and contributed to the writing. HG, JM and WS instigated the study and assisted with the interpretation of the data and the preparation of the manuscript.



To investigate whether the administration of the zinc-containing nutritional supplement ZMA causes an increase of serum testosterone levels, which is an often claimed effect in advertising for such products; to monitor the urinary excretion of testosterone and selected steroid hormone metabolites to detect potential changes in the excretion patterns of ZMA users.


Fourteen healthy, regularly exercising men aged 22–33 years with a baseline zinc intake between 11.9 and 23.2 mg day−1 prior to the study.


Supplementation of ZMA significantly increased serum zinc (P=0.031) and urinary zinc excretion (P=0.035). Urinary pH (P=0.011) and urine flow (P=0.045) were also elevated in the subjects using ZMA. No significant changes in serum total and serum free testosterone were observed in response to ZMA use. Also, the urinary excretion pattern of testosterone metabolites was not significantly altered in ZMA users.


The present data suggest that the use of ZMA has no significant effects regarding serum testosterone levels and the metabolism of testosterone in subjects who consume a zinc-sufficient diet.


The ergogenic effect of numerous nutritional supplements has been investigated to a great extent in the field of sports nutrition. Countless different supplements can be found on the market, which are advertised to be valuable for athletes. Yet, only a few substances available have been scientifically proven to improve athletic performance (Burke et al., 2000).

Even though supplementation may reverse negative effects of nutritional deficiencies (and consequently improve athletic performance), this cannot be transferred directly to non-deficient athletes. Therefore, the advertised effects of nutritional supplements have to be evaluated critically based upon independent trials.

The interrelations of the trace element zinc and the male sexual hormone testosterone (T) have been known for many years. Dietary zinc deficiency has been found to cause hypogonadism and growth retardation for the first time in the 1960s (Prasad et al., 1963). Reduced testosterone synthesis due to impaired action of superordinate hormones such as gonadotropin-releasing hormone, luteinizing hormone and follicle-stimulating hormone (McClain et al., 1984), and altered enzymatic conversion of testosterone (Om and Chung, 1996) have been identified as the main reasons for lower testosterone levels in zinc deficiency.

In 2000, a placebo-controlled, double-blind study showed that the use of the nutritional supplement ZMA by semiprofessional athletes resulted in an increase of plasma testosterone levels of approximately 30% and significantly improved muscle strength when compared to control athletes (Brilla and Conte, 2000).

Except for this trial, supplementation of zinc has only been reported to increase testosterone in pathological conditions linked to a low zinc status (Favier, 1992) and in elderly men (Haboubi et al., 1988; Prasad et al., 1996).

Therefore, the aim of the present study was to reinvestigate the claimed effect of the administration of ZMA on serum testosterone levels in young, physically active, healthy men in an independent placebo-controlled, double-blind trial.

A further objective was to monitor the urinary excretion of testosterone and selected steroid hormone metabolites to detect potential changes in the excretion patterns of ZMA users.

Materials and methods

The nutritional supplement ZMA

The product ZMA (manufacturer: SNAC System Inc., Burlingame, CA, USA) was purchased from an Internet distributor of sport nutrition (Bodybuilding.com, 2005).

According to the manufacturer's information, the supplement contained zinc (30 mg per recommended dose of three capsules, present as monomethionine and aspartate), magnesium (450 mg, as aspartate) and vitamin B6 (10.5 mg, as hydrochloride). Before the study, the zinc and magnesium content of the supplement was analysed by a commercial laboratory using Inductively Coupled Plasma–Mass Spectrometry after acid digestion. The concentrations of both elements were similar to those labelled.

With respect to recent reports regarding contaminations (Geyer et al., 2004b) and adulterations (Geyer et al., 2004a) of nutritional supplements, prior to the trial the supplement was confirmed to contain none of the following anabolic steroids: testosterone, 19-nortestosterone, prohormones of testosterone and 19-nortestosterone, tetrahydrogestrinone, stanozolol, metandienone and trenbolone. The analysis of the supplement was performed according to a previously described method (Parr et al., 2004). For the detection of tetrahydrogestrinone, stanozolol, metandienone and trenbolone, this method was adapted to liquid chromatographic-tandem mass spectrometric (LC/MS-MS) conditions, which are used for the detection of synthetic steroids in urine samples (Thevis et al., 2005). The limits of detection (signal/noise ratio 3) of the listed steroids were calculated to be in the range of 0.2–5 ng g−1.


The placebo-controlled, double-blind trial was approved by the Local Ethics Committee of the German Sport University Cologne. All subjects gave their written consent prior to the trial. The study included 14 healthy male volunteers, who reported to exercise regularly on a recreational or semi-competitive basis (2.5–10 h week−1). After stratification by weight (<80 and 80 kg), the subjects were randomly assigned to the study groups (Table 1). Subjects and investigators were unaware of the group assignments until the end of the study.

Table 1: Anthropometric data and baseline dietary zinc intake of the participants of the trial

In the course of the trial, the daily intake of zinc was determined on the base of diet history interviews, which were conducted with EBISpro software. This method has been validated and compared to other computerized methods (Landig et al., 1998). The interviews were performed by an experienced nutritionist, and were designed to reflect the habitual dietary intake of the subjects in the month before and during the trial period.

The daily zinc intakes of all subjects (range: 11.9–23.2mg day−1) were higher than the recommended daily allowance of 11 mg day−1 (Institute of Medicine, 2001), so all subjects were considered to be not zinc deficient.

Supplementation trial

According to their group assignment, all subjects ingested either three capsules per day of ZMA or placebo for 56 days. Placebo capsules containing D-lactose monohydrate were indistinguishable from ZMA in size and colour. The participants were asked to swallow the capsules with water between dinner and bedtime at minimum 1 h after the last food intake. At least 2 weeks before and during the trial, all subjects had to refrain from using any nutritional supplement containing zinc.

Blood and spot urine samples were collected the morning before the start of the supplementation (week 0). For the investigation of the time course of potential changes caused by the administration of ZMA, blood and urine samples were taken weekly within and at the end of the supplementation period of 8 weeks (weeks 1–8). Samples were collected between 0900 and 1100 hours. To account for circadian rhythms of all measured parameters, the subjects reported to the laboratory at the same time of day every week. After centrifugation, serum was stored at –20°C until analysis. Urine samples were also kept frozen until analysis.

Analytical methods

Serum and urinary zinc concentrations were assessed on a Perkin Elmer 2380 atomic absorption spectrometer (AAS). After dilution with 0.01 N nitric acid (serum: 1:10; urine: 1:2), the samples were directly aspirated into the AAS system. The urinary zinc excretion rate, which is the most commonly used measure in the evaluation of urinary zinc, was calculated by multiplying the urinary zinc concentration with urine flow.

Serum total testosterone (total T) and serum sexual hormone-binding globulin (SHBG) were both measured with commercially available immunoassays on a Modular Analytics E170 (Roche, Mannheim, Germany). Serum albumin was determined with a ready-to-use kit using the bromcresol green method for the Modular/P-Modul system (Roche). Serum free testosterone was calculated from total T, SHBG and albumin according to a previously described method (Vermeulen et al., 1999).

Urinary creatinine was measured with an automatic enzymatic assay for the Modular/P-Modul system. Urinary-specific gravity was determined on a Paar DMA 38 Density Meter and for pH analysis, a Consort C831 analyser was used. Urinary flow was calculated as the volume of the spot urine over the time difference between the spot sample and the preceding urination, which the subjects were instructed to record to the minute.

For the determination of the urinary concentrations of unconjugated and glucuronidated steroid hormones, a gas chromatographic–mass spectrometric method routinely used in doping analysis was utilized (Geyer et al., 1998). The following urinary steroids were quantified:

T, epitestosterone (EpiT), 5α-dihydrotestosterone, 5α-androstane-3α,17β-diol (Adiol), 5β-androstane-3α,17β-diol (Bdiol), androsterone, etiocholanolone, DHEA, 5β-pregnane-3α,20α-diol (Pdiol), 11β-hydroxy-androsterone, 11β-hydroxy-etiocholanolone, tetrahydrocortisol and allo-tetrahydrocortisol. Urinary concentrations were corrected to a mean specific gravity of 1.022 g ml−1 according to Equation (1).

Additionally, the following ratios, which are usually recorded in routine doping analyses, were determined: T/EpiT, Adiol/Bdiol, androsterone/etiocholanolone, androsterone/T and androsterone/EpiT.

Statistical analysis

Statistical analysis was performed using SAS software, version 9.1. Serum and urinary hormone concentrations were normalized by applying natural log transformation before statistical analysis.

For determination of statistically significant changes of all measured parameters, a general mixed linear model for repeated measures was used (Littell et al., 1998). The assignment to each treatment group (‘group’) and the change within each treatment group over time (‘week × group’) were considered to be fixed effects. Subject variability was deemed to be a random effect. Variances were assumed to be different within each group.

Parameters were presumed to be affected by the administration of ZMA if the estimate of ‘week × group’ was significantly different from 0 (α=0.05). The mixed linear model also provided estimates for each point of time and each group as well as corresponding P-values, indicating significant differences from baseline values and/or from the control group.


Serum concentrations

Serum zinc was significantly raised following ZMA supplementation but remained unchanged in the placebo group (Figure 1, top). Even though serum zinc values differed from baseline concentrations at a significant level only after 5 and 6 weeks, the mixed linear model indicated a significantly positive trend of 0.29±0.13 nmol l−1 per week in the ZMA group (P=0.031).

Figure 1
Figure 1

Serum zinc concentrations and urinary zinc excretion rate. aSignificantly different from week 0, P<0.05; bsignificantly different from week 0, P<0.01; csignificantly different from placebo, P<0.05; and dsignificantly different from placebo, P<0.01.

The time course of serum total T and free T is shown in Figure 2. Within each group, there were no statistically significant trends in the concentrations of total T (ZMA: P=0.42; placebo: P=0.69) or free T (ZMA: P=0.33; placebo: P=0.56).

Figure 2
Figure 2

Serum total and serum free testosterone after use of ZMA (grey) or placebo (white).

Urinary parameters

As illustrated in Figure 1, bottom, the urinary zinc excretion rate was increased in the group supplementing ZMA. The mixed model showed a significant average increase of 13.6±6.4 pmol min−1 per week (P=0.035), whereas only at weeks 4 and 8, the differences from baseline levels reached statistical significance. In the placebo group, the urinary zinc excretion rate was not altered (P=0.75).

There were no significant differences in the urinary excretion of the monitored urinary steroid metabolites (Table 2). Consequently, none of the ratios recorded in routine doping analysis was significantly altered by the use of ZMA (data not shown).

Table 2: Mixed linear model estimates of the monitored urinary metabolites in each groupa

Supplementation with ZMA significantly elevated urinary pH by approximately 1 pH unit after 8 weeks (overall trend: P=0.011). Additionally, urine flow was almost doubled after 8 weeks of ZMA use and the mixed model revealed a significantly positive trend (Figure 3; Table 3).

Figure 3
Figure 3

Urine pH and urine flow in subjects using ZMA (grey) or placebo (white). aSignificantly different from week 0, P<0.05; csignificantly different from placebo, P<0.05; and dsignificantly different from placebo, P<0.01.

Table 3: General urinary parameters in each group

In spite of the increase in urinary flow, there was neither a significant change in urinary specific gravity nor in urinary creatinine concentration (Table 3).


The present supplementation trial could not confirm the results of a previous study conducted by Brilla and Conte, who reported an increase of plasma testosterone levels by about 30% after use of ZMA (Brilla and Conte, 2000). The reasons for this can only be speculated on. It has to be noted that the training level of the participants of the present trial seems somewhat lower than that of the previous study. However, this is unlikely to have caused such a discrepancy between the study results, since serum (respectively plasma) zinc levels before and after supplementation as well as baseline testosterone levels were similar in both trials.

Supplementation of ZMA caused a small but significant increase in serum zinc and a more pronounced increase in urinary zinc. The fact that serum zinc was only mildly increased after ZMA use agrees with the general observation that serum zinc concentrations are well regulated (Krebs, 2000). The stronger increase of the urinary zinc excretion rate indicates that renal zinc excretion substantially contributed to zinc homeostasis in the subjects receiving ZMA. This gives additional support to the assumption that the basal zinc status of the participants of the study was good prior to the start of the supplementation (Capel et al., 1982; Verus and Samman, 1994).

The high zinc dose of additional 30 mg day−1 had no effects on serum testosterone concentrations in the subjects. Consequently, urine markers of testosterone and its metabolism were also not altered by the use of ZMA.

In contrast to the present study, previous trials which reported an increase of testosterone levels after zinc supplementation were conducted with zinc-deficient subjects: in pathological conditions, in which the zinc status is often poor, such as uraemia/haemodialysis (Antoniou et al., 1977; Mahajan et al., 1982), growth retardation (Ghavami-Maibodi et al., 1983), sickle-cell anaemia (Prasad et al., 1981) or infertility (Netter et al., 1981; Favier, 1992), it was shown that administration of the trace element caused a reversal of previously lowered testosterone concentrations. Kilic et al. (2006) reported that supplementation of zinc reversed reduced serum testosterone levels caused by exhaustion exercise.

In summary, it can be concluded that even if zinc supplementation may reverse lowered testosterone levels and restore disturbed testosterone metabolism in cases of mild or severe zinc deficiency, it is not capable of further increasing serum testosterone when sufficient zinc is provided by the regular diet.

However, it is noteworthy that the use of the supplement ZMA had significant effects on urinary pH and urine flow. There is only limited comparable data available on effects of supplementation of the major components of ZMA (zinc aspartate, zinc methionine and magnesium aspartate) on urinary pH and/or urine flow. Still, the increase in urinary pH after the use of ZMA stands in contrast to previous studies on the effects of magnesium-aspartate supplementation. Mühlbauer et al. (1991) reported that magnesium-aspartate supplementation caused a mild decrease in urinary pH, whereas Classen et al. (1987) concluded that magnesium oxide but not magnesium-aspartate affects the acid–base status as shown by an alkalization of the urine. So, further investigation is needed to clarify the effect of ZMA supplementation on urinary parameters of the acid–base balance and hydration status, possibly with more participants to detect changes over time more specifically.

In conclusion, the advertised testosterone-increasing effect of ZMA supplementation could not be confirmed in the present trial. Therefore, it seems unlikely that athletes with a balanced dietary zinc status will benefit from the use of ZMA with respect to the claimed effects on testosterone levels. Also, it has to be considered that the recommended dosage of such a supplement causes zinc intakes in the range of the tolerable upper intake level of 40 mg day−1 (Goldhaber, 2003). Further on, the present data indicate that the use of ZMA may interfere with the user's acid–base balance and hydration status.

From the present point of view, the use of ZMA or comparable high-dose zinc supplements is not generally advisable to athletes or other people with a sufficient dietary zinc intake.


  1. , , , (1977). Reversal of uraemic impotence by zinc. Lancet 29, 895–898.

  2. Bodybuilding.com (2005). SNAC ZMA. ,accessed May 2005.

  3. , (2000). Effects of a novel zinc–magnesium formulation on hormones and strength. JEPonline 3, 26–35.

  4. , , (2000). Dietary supplements and nutritional ergogenic aids in sport. In: Burke L, Deakin V (eds) Clinical Sports Nutrition. McGraw-Hill: Roseville. pp 455–553.

  5. , , , (1982). The assessment of zinc status by the zinc tolerance test in various groups of patients. Clin Biochem 15, 257–260.

  6. , , , , , (1987). Prevention of stress-induced damage in experimental animals and livestock by monomagnesium-L-aspartate hydrochloride. Magnesium 6, 34–39.

  7. (1992). The role of zinc in reproduction. Hormonal mechanisms. Biol Trace Elem Res 32, 363–382.

  8. , , , , (1998). Screening procedure for anabolic steroids—the control of the hydrolysis with deuterated androsterone glucuronide and studies with direct hydrolysis. In: Schänzer W, Geyer H, Gotzmann A, Hareck-Engelke U (eds) Recent Advances in Doping Analysis (5). Sport und Buch Strauss: Köln. pp 99–102.

  9. , , , , (2004a). High doses of the anabolic steroid metandienone found in dietary supplements. Eur J Sport Sci 3, 1–5.

  10. , , , , , (2004b). Analysis of non-hormonal nutritional supplements for anabolic–androgenic steroids—results of an international study. Int J Sports Med 25, 124–129.

  11. , , , , (1983). Effect of oral zinc supplementation on growth, hormonal levels, and zinc in healthy short children. Ann Nutr Metab 27, 214–219.

  12. (2003). Trace element risk assessment: essentiality vs toxicity. Regul Toxicol Pharmacol 38, 232–242.

  13. , , , , (1988). Zinc supplementation and erythropoiesis in the elderly. J Clin Pathol 41, 706.

  14. Institute of Medicine. Food and Nutrition Board (2001). Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academy Press: Washington, DC.

  15. , , , , , (2006). The effect of exhaustion exercise on thyroid hormones and testosterone levels of elite athletes receiving oral zinc. Neuro Endocrinol Lett 27, 247–252.

  16. (2000). Overview of zinc absorption and excretion in the human gastrointestinal tract. J Nutr 130 (Suppl), 1374S–1377S.

  17. , , , (1998). Validation and comparison of two computerized methods of obtaining a diet history. Clin Nutr 17, 113–117.

  18. , , (1998). Statistical analysis of repeated measures data using SAS procedures. J Anim Sci 76, 1216–1231.

  19. , , , , , (1982). Effect of oral zinc therapy on gonadal function in hemodialysis patients. A double-blind study. Ann Intern Med 97, 357–361.

  20. , , (1984). Hypogonadism in the zinc-deficient rat: localization of the functional abnormalities. J Lab Clin Med 104, 1007–1015.

  21. , , , , , et al. (1991). Magnesium-L-aspartate-HCl and magnesium-oxide: bioavailability in healthy volunteers. Eur J Clin Pharmacol 40, 437–438.

  22. , , (1981). Effect of zinc administration on plasma testosterone, dihydrotestosterone, and sperm count. Arch Androl 7, 69–73.

  23. , (1996). Dietary zinc deficiency alters 5 alpha-reduction and aromatization of testosterone and androgen and estrogen receptors in rat liver. J Nutr 126, 842–848.

  24. , , , (2004). Analytical strategies for the detection of non-labelled anabolic androgenic steroids in nutritional supplements. Food Addit Contam 21, 632–640.

  25. , , , (1981). Effect of zinc supplementation on serum testosterone level in adult male sickle cell anemia subjects. Am J Hematol 10, 119–127.

  26. , , , , (1996). Zinc status and serum testosterone levels of healthy adults. Nutrition 12, 344–348.

  27. , , , , (1963). Zinc metabolism in patients with the syndrome of iron deficiency anemia, hypogonadism and dwarfism. J Lab Clin Med 61, 537–549.

  28. , , , (2005). Screening for unknown synthetic steroids in human urine by liquid chromatography-tandem mass spectrometry. J Mass Spectrom 40, 955–962.

  29. , , (1999). A critical evaluation of simple methods for estimation of free testosterone in serum. J Clin Endocrinol Metab 84, 3666–3672.

  30. , (1994). Urinary zinc as a marker of zinc intake: results of a supplementation trial in free-living men. Eur J Clin Nutr 48, 219–221.

Download references


We acknowledge the support from the Manfred Donike Institute for Doping Analysis e.V.

Author information


  1. Institute of Biochemistry, German Research Centre of Elite Sport, German Sport University Cologne, Cologne, Germany

    • K Koehler
    • , M K Parr
    • , H Geyer
    •  & W Schänzer
  2. Institute of Training Science and Sport Informatics, German Research Centre of Elite Sport, German Sport University Cologne, Cologne, Germany

    • J Mester


  1. Search for K Koehler in:

  2. Search for M K Parr in:

  3. Search for H Geyer in:

  4. Search for J Mester in:

  5. Search for W Schänzer in:

Corresponding author

Correspondence to K Koehler.