What price would you pay for muscles?

Not everyone works out for performance. I would count myself amongst the folks who work out basically for looks. There was a time when I lifted to get better at my sport, but the reality is that my career and most of my current hobbies don’t require me to perform at a much higher skill level than sewing two hollow tubes about 1mm in diameter together, which clearly doesn’t require heavy squats to improve.

This entry’s article came to my attention from Ryan Zielonka, who wrote,

“Quick question for you. I work out to look good, period. I’m also blessed with a full and thick head of hair. I recently started taking creatine (again, I’ve never been particularly consistent) and came across some anecdotal evidence as well as one study that shows the potential for hair loss while using creatine.

What are your thoughts on this? I’d hate to sacrifice my full head of hair for just a slight uptick in swoleness.”

This is a perfect example of trade-off decision making, which, for the most part, is the central issue in most medical decision making. There’s no question that every intervention, fitness or medical (which I think are one and the same, personally) carries risk and benefit. We choose to pursue an intervention based on whether we think the benefits outweigh the risks. Is it worth exposing your wrist to pain, a scar, the possibility of infection and the possibility of no improvement for the chance at getting rid of your carpal tunnel syndrome? Is the risk of injury, and the cost of time, discomfort and (to some extent) denial of “junk food” worth it to perform better or to look better?

Creatine has been shown to have many benefits, not the least of which are performance-related as well as aesthetic. We still don’t understand how creatine makes these benefits possible and there are several theories, none of which are truly dominant as to why it works. But what if one of these theories was true and what if the trade-off for using creatine for its benefits meant some form of hair loss? Would this change your decision to use it or not?

van der Merwe J, Brooks NE, Myburgh KH. Three weeks of creatine monohydrate supplementation affects dihydrotestosterone to testosterone ratio in college-aged rugby players. Clinical Journal of Sport Medicine, 19:399-404, 2009.


As much as I agree with the statement that the long-term safety of creatine supplementation has not yet been established, I don’t think that the question is going to be answered in my lifetime, but that doesn’t mean that we should stop studying it.

There are a few theories as to how creatine improves performance, but none have really emerged as dominant so far. One of these theories is that creatine somehow affects the production of testosterone. Testosterone can be converted to dihydrotestosterone (DHT) which is a more bioactive androgen (i.e. it takes less to have the same effect). It isn’t known whether creatine has an effect on DHT production and whether THIS might be one of the mechanisms by which creatine produces the observed effects that it has.

Although DHT is a more potent androgen, it is also linked to other conditions such as baldness (alopecia), male-pattern baldness as well as prostate hypertrophy. Understanding whether creatine alters DHT levels, therefore, has relevance not only in figuring out how creatine works, but also in determining whether long-term use may have other, possibly unwanted consequences.


The design of this study was a randomized controlled cross-over design. So, each of the subjects was randomly assigned to receive either a placebo or creatine first, and then after a 6-week washout period of taking neither placebo nor creatine, were given whatever they didn’t have in the first phase.

The method by which the subjects were randomized is not one that most trialists would condone: Subjects were given a number based on the order in which they arrived for initial testing. Odd numbers were assigned to one group, and even numbers to the other. Twenty subjects were recruited in total, but the study was only done on 16, due to 4 drop-outs.

The subjects were all 18-19 years old, male rugby players in their competitive season.

The total supplementation period was 21 days. A 7-day loading phase (25g per day) followed by the maintenance phase (5g per day) was used. Creatine was taken with glucose (25g). The placebo group got the glucose with no creatine. It’s not mentioned who was blinded in this experiment, only that it was “double blind”. The creatine and placebo were given in capsules.

In terms of training and diet, these athletes were all part of an institute and therefore trained together (for their positions) and ate together.

Body mass, and skin folds were measured for Day 0, 7 and 21 for both phases of the study. Blood samples were drawn on day 0, 7 and 21 for serum testosterone and DHT.

Comparisons were made using ANOVA’s for repeated measures with post hoc Tukey tests for significant ANOVAs. They did do baseline statistical comparisons, but I’ve bitched enough about how this is inappropriate, particularly for a cross-over trial.


None of the body-mass or skin-fold measurements changed substantially at the 7-day or 21-day points. This included percent body fat as well as fat-free mass.

Testosterone levels did not change substantially for either the placebo or the creatine phase at any time point. However, what did change was DHT levels. These levels were higher in the creatine phase at both day 7 and day 21, with the day 7 level being higher.

The baseline DHT level for the creatine phase was 0.98 (SD 0.37). At day 7, this rose to 1.53 (SD 0.50) and then fell at day 21 to 1.38 (SD 0.45). These increased levels were statistically significant. As a result, the DHT:T ratios were also higher and statistically significant.

What’s interesting is that the baseline DHT level for the placebo phase was 1.26 (SD 0.52). More on this in the discussion area.


Most of the discussion of this article centred around the potential health effects of increased serum DHT.

I think my only major grief with this study is that they reported that DHT levels were 56% higher after 7 days, and then 40% higher than baseline after 21 days on creatine. But the baseline level of DHT in the placebo phase (i.e. the mean level of DHT in the SAME subjects on day 0 of the placebo phase) was 1.26 (SD 0.52). When compared statistically, the DHT level at day 7 was only 21% higher than this baseline, and not statistically significant; which then begs the question as to whether the variability of DHT (either as normal variance in the assay itself, or as normal physiological variance in 18-19 year old rugby players) is high enough that the increased level of 1.53 is meaningful.

However, this issue is not only a good example of trade-off decision making, but it’s also a great example of how a single study isn’t always enough to put a particular issue into an appropriate context. Yes, the study shows an increase in DHT. Yes, increases in DHT are linked to male-pattern baldness, but I think it’s important to note that the study cited in this paper whose purpose it was to investigate DHT and baldness show that even at the highest levels of DHT measured in these rugby players, the association with male-pattern baldness is not very strong.

In the cited paper (Bang et al, Comparative studies on levels of androgens in hair and plasma with premature male-pattern baldness, Journal of Dermatological Science, 34:11-16, 2004), the median level of DHT in males (aged 26-43)with premature baldness was 2.8 with a range of 1.99-4.88. The range of DHT in males without premature baldness was from non-detectable to 2.74 with a median of 1.20 (aged 25-27). Mind you, these are DHT levels in the HAIR itself, not serum concentration levels (which were published in ng/ml, not nmol/L–the units used in the creatine study) so comparison is tricky.

It’s impossible to know whether taking creatine will make YOU go bald prematurely if your baseline DHT levels are hanging out in the grey zone of “just short of going bald” unless you know what your baseline DHT levels actually are. It’s also impossible to know whether these rugby players’ DHT levels at age 18-19 are compatible with the levels that would have been observed had the baldness researchers measured their prematurely bald subjects at age 18-19 (i.e. we don’t know which of these 16 rugby players is going to become prematurely bald).

However, it would appear that despite elevated levels of DHT in the serum of rugby players aged 18-19 after 7 and 21 days of creatine supplementation, that the elevation is not likely to correlate with those levels that one could deem necessary to produce male-pattern baldness. There’s also a fair amount of overlap in DHT levels between males who did have male-pattern baldness and those who didn’t in at least two studies examining the issue; so elevated DHT is probably not the ONLY causative factor in the case of whether you’ll go bald or not.

Bottom line: If you’re going to go bald, it’s probably inevitable. It seems unlikely that taking creatine is going to be the thing that tips you over the edge into Cueball-world (if you’re even hanging out at the edge that is). I, for one, will continue to take creatine for what that’s worth.

Click Here to view the Full Version of our Website