Reply
Myburgh, Kathryn H PhD; van der Merwe, Johann MBChB; Brooks, Naomi E PhD
Author Information
Department of Physiological Sciences, Stellenbosch University, Stellenbosch, South Africa.
Reply:
We welcome the comments of Prof Green on our article, which focuses on testing the hypothesis that one mechanism for an effect of creatine supplementation on muscle is a conversion of testosterone to its more bioactive metabolite dihydrotestosterone (DHT).1 In addition to questions related specifically to our study, Prof Green has used this forum to open a discussion on the serious topic of potential contamination of nutritional supplements that are used for research purposes.2 In short, he suggests that researchers in this field should be responsible for having the components of the supplements to be used in their studies verified in depth and independently. He further suggests that if this is not done, this journal should not accept the manuscript for publication.
We will present details of the supplement and supplement testing and answer the specific questions before contributing to the broader discussion. We will debate whether the responsibility lies with the researchers or with the accredited laboratories and the additional structures that are in place to accredit and monitor those laboratories.
There was no conflict of interest for any of the authors (no personal financial gain at the time of the study and before or after the study was completed; no appointment in any advisory capacity). The study was financed independently from any industry contract or donation, whether financial or in kind. If this was not the case, such acknowledgment would have been provided at the end of the discussion section. Prof Green makes a good point that such information should be stated in print with the article.
However, for the purposes of discussion, it should be added here that interaction with industry and use of their commercial product in research does not necessarily reflect conflict of interest. Indeed, relevance of academic endeavor requires bidirectional communication between the producers of knowledge and those who commercialize the knowledge. Contract research does not imply that industry will have an influence on the research results. Typically, academic institutions require legal documentation to be in place before research of contractual nature, and typically, such a document will include a short specified period (eg, 6-12 months) during which results remain unpublished if they do not reflect the desired result from the point of view of the commercial entity, with complete academic freedom thereafter.
For our study, sufficient supplement was purchased at the outset to ensure that the same batch would be used for the entire study. The batch number and the expiry date were provided. The batch was analyzed by Quotient Bioresearch Ltd's subsidiary company, HFL Sports Science Inc (Fordham, Cambridgeshire, United Kingdom). It owns 2 dedicated testing laboratories with extensive testing capability far beyond the testing of sports supplements. This relationship was clearly stated. Quotient Bioresearch Testing Laboratory is certified by the United Kingdom Accreditation Service (UKAS).3 Samples from early, mid, and end of the batch were analyzed, and results reported no difference within the batch. The testing included gas chromatography mass spectrometry and liquid chromatography mass spectrometry screens for 25 different substances, either androgenic steroids or related compounds, or stimulants. None were detected with a lower limit of detection of 10 ng/g (compared with the publication reporting 50 ng/g 19-norandrostenedione contamination, as mentioned by Prof Green). Such information is available to researchers but not always with retail purchase of supplements. The information can be requested from source.
Prof Green asserts that only inert substances should be referred to as a placebo. However, a placebo may be “an inert or innocuous substance used especially in controlled experiments testing the efficacy of another substance.”4 Because creatine is frequently ingested with carbohydrate, it would be appropriate to test its efficacy over and above any possible effect of the other components of the typical supplementation regimen, in which case, these components would be considered innocuous relative to the substance being tested. Other studies investigating the effects of creatine supplementation have also used carbohydrate as their control for creatine and considered, as we have, that it was proper to call it the placebo.5-8 The use of carbohydrate as a placebo is not limited to studies testing the effect of creatine.9 In these studies, the outcome measurements have varied according to the specific study aims and, in some cases, included intramuscular effects. In the case of our study, the primary aim was to test the efficacy of a creatine supplement on testosterone conversion to DHT, both measured in serum samples. The effect of exercise was not considered. The quantity of placebo we provided was primarily practical. The equivalent number of capsules needed to be taken when on the creatine supplement or on the placebo. The quantity of carbohydrate we used during loading did not differ dramatically from other studies in which the quantity of carbohydrate was also not equivalent but rather isocaloric.10
We used the phrase “a popular ergogenic and androgenic [sic] aid” on one occasion in the introduction of our article.1 We acknowledge that the phrase could be presented as “a popular ergogenic aid with some purported androgenic properties.” The aim of our study was to determine if there was a particular effect of creatine supplementation on a reaction in a steroid hormone pathway that would result in greater presence of a more potent androgen, DHT. Several studies refer to DHT as androgenic while focusing on effects other than secondary sexual characteristics.11,12 Lee et al 11 reported the ability of DHT to reduce H2O2-induced reactive oxygen species production. The authors first introduced the hormone to the readers, and we quote: “Among androgenic hormones, dihydrotestosterone (DHT) is generated by a 5[alpha]-reduction of testosterone. Unlike testosterone, DHT cannot be aromatized to estradiol, therefore DHT is considered a pure androgenic steroid.”11 Substances that block or reduce the function of 5-alpha-reductase are considered “antiandrogenic” (eg, Flores et al 13; Liu et al 14). In one of these examples, fatty acids (albeit in vitro) were considered antiandrogenic due to their ability to reduce the activity of this enzyme.14 Substances that stimulate 5-alpha-reductase can be considered “androgenic.”
All subjects were resident at the same Rugby Institute and were tested between 7.00 and 8.00 hours on each occasion, before any training. There was no deviation from this protocol for any subject.
We retrospectively had a sample of the remaining supplement (the same batch; in storage since the completion of the study) analyzed for content of creatine. The analysis was undertaken by J Muller Laboratories (Pty) Ltd, accredited by the South African National Accreditation System.15 Analysis (high-performance liquid chromatography with variable wavelength detector, combined with analysis of nitrogen, ash, and water content of the sample) indicated that the sample contained 89.55% creatine monohydrate with an additional moisture content of 9.3%. Therefore, the total of the powder accounted for by creatine monohydrate and water was 98.85% with an additional 0.1% accounted for by creatinine. An ultraviolet scan between the wavelengths of 200 and 360 nm revealed no interferences except for creatinine. Considering that the supplement had been stored for a considerable time since the study was done, one can understand that the moisture content may not have been at that level originally.
Getting back to the more wide-reaching debate on the responsibility for testing the supplements used in research studies for potential contaminants, Prof Green mentions that there are organizations that will certify a company after testing their supplements independently. As stated above, the supplement we used was tested in a UKAS certified laboratory. UKAS is the only national body recognized by the UK government for the purposes of accrediting testing laboratories and calibrating laboratories, as well as accrediting the bodies entrusted with certification and inspection. The UK Accreditation System is a signatory to the European cooperation for Accreditation and uses international standards. In this system, testing laboratories are accredited, and to standardize across different testing laboratories, calibrating laboratories are also certified. Further checks are in place because certification is dependent on both test results and regular inspection. A list of accredited laboratories is accessible to the public and also contains the names of laboratories removed from the list, presumably due to failure in one aspect or another of the rigorous procedure.
What level of systematic consideration and action should be expected from the researchers regardless of any requirements laid down by the possible publishing journal? We argue that if the supplement is purchased from a company that is able to supply verifiable component analysis and proof of regular inspection and certification by an internationally recognized system that has multiple levels of control and no seals have been broken, their trust is justified. There would be no further need for the researcher to submit samples for analysis.
It will require a wider academic debate to discuss how supplements should be tested, for example, if they are imported in bulk format and repackaged outside of the original controlled environment. A laboratory that operates under the ethical code required for academic research could also be considered a credible analysis facility if independent of the laboratory performing the research. Such a laboratory often does not submit to national accreditation but still follows good scientific practice, achieving results worthy of publication in the best journals in their field.
Good scientific practice, scientific integrity, and integrity of the peer-review process contribute to methodical consideration of results before publication. Results are not simply assumed to be correct and acceptable. Interpretation and discussion within the framework of current knowledge and theories or establishment of new theories are equally subjected to review. All published results are subject to verification by other researchers and to advances in understanding, which may render them interpretable in a different way. No declarations can influence this process.
What guidelines should a journal present to potential authors in this regard? If guidelines are presented, is it necessary to require a declaration? Is a declaration relevant if no guidelines have been presented? Who should be tasked to contribute to the development of such guidelines or declaration statements? How extensively should this be debated? Will there be a requirement for oversight of the declaration at the institutional level? These questions cannot be answered by the authors of a single publication.
Prof Green raises a point to which we do not wish to respond: that the proposed declaration would be analogous to the declaration of ethical approval that is required by medical journals.
Kathryn H. Myburgh, PhD
Johann van der Merwe, MBChB
Naomi E. Brooks, PhD
Department of Physiological Sciences, Stellenbosch University, Stellenbosch, South Africa
