Take a couple deep breaths: New keto paper published

For whatever reason, ketogenic diet studies seem to trigger visceral reactions these days, whether positive or negative. The latest edition is no exception, and the internet is champing at the bit to tear it apart.

When people start critically reading research, they tend to ride a U-shaped curve, as follows:

u shaved curve

The natural progression of critically reading research


When you first start, you’re most likely going to accept whatever you read as fact. As you get a little more experienced, you start looking for holes to poke in every study, ultimately becoming hypercritical. Eventually you’ve been around the block enough times to realize that there are any number of alternative approaches, methods, and interpretations for just about any study you could read. At that point, you start reading papers differently; instead of asking, “How many holes can I poke in this?” you start asking, “What information can this paper provide me?” The threshold of a “publishable paper” is not perfection, it’s a meaningful contribution with sound and repeatable methodology.

Hypocrisy disclaimer: I speak from experience on this. I once summited the peak of that U-shaped curve, and was dreadfully hypercritical of every paper I read, mostly for the gleeful ego-stroking that comes with feeling more sciencey than scientists. Then I started doing research and found out this stuff is really, really, really hard.

The only reason I bring this up is because the whole internet has descended on this paper like a ravenous pack of wolves. Everybody wants to poke holes, and there are certainly aspects that are quite pokable. But much of the chatter I’ve seen about this paper has been way overblown, complete with personal attacks, vague accusations of impropriety, and even calls for retraction. None of these are warranted. Frankly, if reviewers were as harsh as Facebook fitness group posters, our field would have about 3 suitable papers published per year. And this is coming from a person who serves as a reviewer for multiple journals, and often seems to come off as “the harsh reviewer.”

There are several things about this paper that limit the findings, as reviewed by Sci-Fit.net (to be clear, I think Sci-Fit wrote an excellent, fair review that was not excessively harsh or nitpicky). I agree that the lack of randomization and existence of meaningful baseline differences between groups are problematic. I also disagree with the decision to report performance outcomes relative to body weight, and to adjust the testing loads based on current body weight, instead of keeping them consistent from pre-testing to post-testing (at least that’s what the methods appear to indicate). The body composition changes also appear to be at odds with the reported caloric intakes, but this is neither rare nor unique in literature with self-reported diets.

Having said that, this study has some strengths. It’s a relatively long intervention for this type of work, has a fairly large sample size, used sound methodology to confirm that ketosis was achieved, and used reasonably well-trained participants. And, although the data are reported relative to body weight, the study used multiple indices of exercise performance, and it’s not difficult to convert those data back to raw values. All in all, this paper gives us an indication of what can happen when you implement a fairly “real-world” keto approach in trained individuals, in terms of body composition and a variety of exercise outcomes. While there are aspects of the study that have justifiably been questioned by readers, there’s no reason to call for retraction or accuse the authors of unethical behavior. It’s not a perfect study, but it makes a contribution to the literature.

The results tell us a few things: application of ketogenic diets may induce weight loss in this population, which makes sense given the satiating nature of the diet. This effect is most likely dictated by reduced caloric intake (rather than bioenergetic magic of ketones), whether or not the diet logs reflect it. We also see that absolute performance changes (i.e., not scaled to body weight), were pretty similar for both groups. Taken together with the rest of the keto literature (which is how we should interpret any study, by the way- in the context of the literature preceding it), there are some things we know about keto:

It reduces appetite

It helps people lose weight in ad libitum conditions, mostly due to appetite reduction

It’s suitable for low-intensity endurance athletes; the more glycolytic the activity becomes, the more unfavorable keto seems to be

It’s great for certain clinical populations with medical issues pertaining to carbohydrate metabolism

It’s pretty inconvenient, and a bit difficult to make a “well-rounded” keto diet that provides sufficient micronutrients (not impossible; just requires a bit more effort)


So let’s call back the dogs of war and set down our pitchforks. I’m not saying research shouldn’t be critically analyzed, reviewed, and discussed in public forums, we just need to retain a reasonable level of respect and civility when doing so. Half the time you’re reading a research paper, several of the authors on the paper are people who have a master’s degree, work 70-80 hours per week doing highly skilled labor, and earn less than the minimum wage. The only thing fueling them is a desire to make an impact on the field, investigate exciting questions, and share their results with the public. It’s important that we critique the work, without bashing the worker. Depending on contextual factors, keto diets may be better, worse, or no different than other diets. From a pragmatic standpoint, the study in question adds a little more to what we know about keto diets, and when they may be justifiable.


New study: An inconvenient truth for volume addicts like me

Well, dammit. Like most people who recreationally lift objects repeatedly, I was brought up on the muscle magazines that pushed 25-set chest workouts and the more is always better narrative. While I rarely do the traditional 5-day bodybuilding split these days, there’s still a great deal of nostalgia whenever I switch things up with a marathon arm or chest workout.

I get it. It can be satisfying to hit the gym and beat up a muscle group or two for an hour. Between the fatigue, the pump, and the eventual soreness, these high-volume workouts alleviate all concerns that we may not have sufficiently stimulated growth, and scratch the obsessive compulsive itch that drives many in the bodybuilding/physique world. Nonetheless, there’s an inconvenient truth that we volume addicts must face: Research tells us that we’re probably spinning our wheels.

As a bodybuilder, the primary goal of hypertrophy-oriented training is to increase the ratio of muscle protein synthesis (MPS) to muscle protein breakdown (MPB), resulting in the accumulation of muscle mass. A recent study subjected rats to 1, 3, 5, 10, or 20 sets of resistance exercise and observed that MPS was essentially maxed out around 3-5 sets.

Now, a single study never tells the whole story, especially in non-human models. But this particular study seems to corroborate previous findings in humans. For example, a study on German Volume Training found that 5 sets per exercise was as good, if not better, than 10 sets. A previous meta-analysis of several studies suggested that performing multiple per exercise was clearly more favorable than 1, but that differences between 2-3 sets and 4-6 sets per exercise were fairly modest. Moreover, it’s understood that the MPS response to resistance exercise doesn’t last forever; it tends to peak around 24 hours after exercise, and remains elevated for about 48-72 hours.

Altogether, the data seem to suggest that the shotgun approach to training (i.e., 25 sets of chest over a 90-minute period) is probably overkill. And if you train that way, you’re probably putting in major effort for diminishing muscle returns, exhausting yourself in the process, and training with less-than-ideal training frequency to accommodate your split and recovery.

So whether you’re coming from a “leave no doubt” mindset, you get a cathartic release from training a muscle group to its absolute limits, or you get a nostalgic kick out of these high-volume, low-frequency bodybuilding splits, you’re going to have to come to terms with what the currently available data tell us: When it comes to efficacy and efficiency, this approach is no better, and likely worse, than programs featuring lower volume per session and higher frequency. That doesn’t mean you can’t or shouldn’t do them- if lifting weights is your hobby, you better enjoy the way you train. But understand that you’re spending a whole lot of effort for diminishing returns, and you have relinquished your right to complain if your results aren’t everything you dreamed they’d be.

I’m back! And here’s a cool new study about metabolic adaptation

prep face pic

The dark days of contest prep, where metabolic adaptation reigns and cheeks are a thing of the past


It’s been a while since I posted anything. I’m running a huge, time-consuming study at the moment, and had to take care of my comprehensive exam, which is basically a cumulative exam over the last eight years of my life. But now the exam is done, and I have metamorphosed from a PhD student to a PhD candidate.

I’m not really “known” for much of anything at this point in my career, but those who do know me most often associate me with my papers on metabolic adaptation and contest preparation in physique athletes. As a bodybuilder, I’ve long been interested in how metabolic rate fluctuates in response to under (and over) feeding, and how these fluctuations influence weight loss and weight maintenance success. A recent paper sheds more light on the topic of metabolic adaptation, or the exaggerated drop in metabolic rate that occurs during weight loss.

The researchers had two weight loss groups: One group did a weight loss diet for 16 weeks straight. The other group did the same weight loss diet, but they alternated between two weeks of dieting and two weeks of weight maintenance. So, instead of 16 straight weeks of weight loss, they would lose weight for two weeks, then take a two-week “break” in which they ate enough to keep their weight stable. In total, they did 16 weeks of dieting for weight loss, but the intervention took 32 weeks total for this group. The overall caloric deficit was equal for both groups.


Statistical analysis status: It’s complicated

When it comes to the statistical analysis of this paper, there’s a lot to chew on. But there’s reasons for the cumbersome nature of the analysis. When we study weight loss or weight maintenance, compliance and drop out are major issues. This leads to two huge questions: 1) Do we analyze the data for everyone who was assigned the diet, or everyone who properly adhered to it? 2) What do we do about data that are missing due to dropout or poor compliance?

The paper reports two analyses: Intent to treat (ITT), and per protocol. The intent to treat analysis analyzes the data for everyone who began the study, even if they dropped out or had poor compliance. The per protocol analysis only uses people who finished the study with sufficient compliance (i.e., they did the whole thing and followed directions reasonably well).

There are pros and cons to each approach. From an application standpoint, the ITT analysis ensures that the randomization of groups holds true, and incorporates the likelihood of successful compliance as part of the intervention’s effect. For example, if you gave a miserable diet that no one could adhere to, the ITT would show a low level of effectiveness. You could pick out the 10% of people crazy enough to follow it and find a great deal of weight loss for those lunatics, but overall the success of the intervention as a whole is questionable based on an ITT analysis.

Per protocol analysis takes the other approach. It aims to determine how effective the treatment was, if you followed your directions and stuck to the plan reasonably well. If you’re a coach who knows your athlete will comply with whatever you give them, this might be a more meaningful analysis for you to interpret, because you aren’t concerned about your athlete’s willingness to comply.

Finally, there’s the issue of how to account for missing data. There’s plenty of methods out there. The simplest is to just remove subjects who don’t have all of their data present. The next simplest is called “last observation carried forward” (LOCF). Let’s say we have two visits: pre-test, and post-test. If you were there for the pre-test but dropped the study before the post-test, the researchers would just assume your value stayed the same- they would “carry forward” your value from pre-testing to post-testing. More advanced methods, such as single regression imputation or multiple imputation, use the trends in the existing data to “predict” the value for missing data.

In this paper, the researchers used LOCF. I typically don’t like LOCF for weight loss/weight maintenance research, because I tend to suspect that most individuals have missing data in follow-up testing because they fell off the weight loss wagon, and their value is likely worse than the previous observation available. I would’ve preferred to see single regression imputation at the very least, or better yet, multiple imputation or the use of linear mixed effect models. Either way, it’s just a preference thing- these researchers carried out a thorough and ambitious project, so no need to get worked up over a judgment call like this.

Finally, they used three different methods to calculate “adjusted” changes in energy expenditure over time. It’s important to adjust these values somehow, because we expect that metabolic rate will naturally decline due to the loss of mass. One method aimed to mathematically account for changes in lean mass and fat mass. Another method used a linear regression equation derived from this specific sample. The third method used a previously published prediction equation. Again, total judgment call here, and they decided to provide all three. Good on them.



Either way you slice it (ITT or per protocol), the group with diet breaks lost more weight. In absolute terms, metabolic rate dropped to a similar degree in both groups. However, when you account for the amount of mass that each group lost, the group with diet breaks had less of an “adaptive” drop in metabolic rate (i.e., the drop in metabolic rate that can’t be explained by the loss of tissue). What’s really interesting is that this was observed solely looking at resting metabolic rate, even though non-exercise activity thermogenesis (NEAT) is more markedly influenced by metabolic adaptation. One might presume that NEAT, which wasn’t measured, would show an even larger difference than resting metabolic rate, but this is purely speculative.

Another interesting observation is that the group with diet breaks lost more weight, but also maintained their weight loss advantage at the final follow-up visit. Some weight regain was observed, but the absolute amount regained was similar in both groups (with a slight advantage observed in the group with diet breaks). As a result, the weight loss gap between the groups at the end of the 16 weeks of weight loss was similar to the difference observed at week 48.



A single study never definitively proves anything, and there’s always plenty to nitpick in a big, intricate study like this one. However, these findings are reasonably promising, and as a practitioner/competitor, would lead me to believe that this intermittent approach to dieting may be just as effective, if not more so, than traditional, continuous, linear dieting. However, the authors suggest that what happens during the “diet break” is critical. They contend that it’s important to actually achieve energy balance (i.e., to get calories all the way up to a true maintenance level). However, if you return to ad libitum or uncontrolled eating habits, you’re very likely to overeat and regain weight during the diet breaks. So, the proper application of this study is not to actually take a complete “break” from dieting altogether, it’s to increase your calories to maintenance level for a couple weeks.

Based on my experiences as a researcher, competitor, and colleague of some great coaches, it appears that some people have metabolic rates that are more adaptive than others. For individuals who tend to experience relatively large drops in metabolic rate during contest prep, this might be a viable strategy for attenuating this unfavorable adaptation. But don’t get your hopes up too high- some degree of metabolic adaptation will still occur, and contest prep will still be rough. You’ll probably just be able to sneak a little more food into the prep diet, or get away with slightly less cardio.  But when it comes to making the most miserable days of contest prep just slightly more tolerable, a win is a win.

Navigating the post-competition period, part one: Physiological factors

Everybody worries about how they’ll get to the finish line, but we rarely consider what we’ll do beyond the finish line. For bodybuilding and physique athletes, contest prep and the post-competition period bring some pretty crazy changes, both physiologically and psychologically. To effectively navigate the weeks (and months) immediately following competition, knowing what to expect can help you plan for a successful transition. This article will discuss the physiological aspects of the post-competition recovery period; part two will discuss psychological factors.

I don’t want to rewrite everything I’ve written in the past three years, so I’ll summarize: Your body goes a little haywire during contest prep, assuming that you’ve effectively obtained a “stage-ready” level of leanness. A variety of hormonal changes and other physiological adaptations promote an environment of reduced anabolism, increased catabolism, suppressed metabolic rate, and absolutely insane hunger [1]. The side effects go even further, with common complaints of disrupted sleep, foggy thinking, extreme lethargy, and reproductive side effects including amenorrhea in females and low testosterone in males.

In my time at UNC, I’ve been unbelievably fortunate to be granted the freedom to pursue this research line, and to be connected with tremendous collaborators who share an interest in this topic. Our group wrote a review paper summarizing some of the unfavorable effects of contest prep [1], then followed up with a case study [2] and a pilot study [3] on post-competition recovery. I also competed in two competitions (including my pro debut!) recently, so I currently find myself directly in the middle of the post-competition recovery process. In this article, I aim to answer some common questions by drawing upon observations from the literature, the laboratory, and “the trenches,” which may help other athletes manage the post-competition period.


1) How quickly should you gain weight?

If you truly got into contest shape, you’re going to gain weight after the competition. This is normal, expected, and almost certainly a good thing. I’m hard-pressed to imagine a sum of money that could convince me to maintain a contest-ready physique for more than 6-8 weeks at most.

The good news is that, as our pilot study indicated [3], the initial weight gain you may experience in the first few days has very little to do with getting “fat.” We used a 2-compartment body composition model measured via A-mode ultrasound, and much of the initial weight gain fell in the “fat-free mass” compartment; bioelectrical impedance spectroscopy confirmed that the initial weight gain was largely water weight.

Simply put, you’re likely to see an abrupt increased in fluid, sodium, and carbohydrate intake following competition, which will increase total body water. Another frequently overlooked fact is that increasing food intake will likely increase the amount of food and waste within the gastrointestinal tract at any given time. So, there’s nothing wrong with just planing for a few quick pounds of weight gain in the first week or so after competition (depending on your body size), and I wouldn’t spend too much time fretting about it.

But what should you do after this initial increase?

I’m of the opinion that it entirely depends on the individual’s goals and circumstances. To make an informed decision on how to manage the “recovery diet,” one must consider the competitor’s weaknesses, the stage of their career, and the presence/severity of any lingering “side effects” of prep. I’ll illustrate with a few examples:

Competitor A is a long-time pro male bodybuilder, who competes with a fat-free mass index of 26 kg∙m-2. He hasn’t gained much lean mass in the last few years, is likely close to his genetic ceiling for muscle mass, and his primary weakness is his inability to achieve elite conditioning (leanness). This competitor has no significant lingering side effects from contest prep.

For Competitor A, a slower rate of weight gain might be advisable, which would consist of what many people consider a conservative “reverse dieting” approach. They might employ a strategy in which they immediately get to maintenance calories after the show, and slowly increase calories from there over time (note: The initial bump to maintenance calories probably won’t be very big, as the caloric deficit is often pretty marginal by the end of contest prep in many circumstances).

Competitor B is pretty new to the sport, 23 years old, and still in the early stages of his lifting career. He achieved decent conditioning for his competition, but simply didn’t have the muscle to compete at a high level yet. Based on these circumstances, it makes absolutely no sense to keep this kid on low calories for another 7 months. He’s better of jumping straight to a moderate surplus, and pushing calories up from there (at a more aggressive rate than Competitor A) to facilitate some lean mass gains.

Competitor C is the exact same as Competitor A, but he does have lingering side effects from contest prep that are clinically meaningful and disruptive. In my opinion, this individual’s approach should be much like Competitor B’s approach- when the caloric deficit caused all the issues, committing to a sustained, reasonably sized caloric surplus is often the most parsimonious route to alleviating these unfavorable effects. (note: see the website disclaimer; always discuss any such health issues with a physician or otherwise qualified medical professional, and follow their advice with regard to any necessary treatment or recovery strategies)


2) Are you “primed for growth” after a competition?

Sure you are, just not the good kind of growth. In the obesity literature, there’s a commonly observed phenomenon called post-starvation obesity; we commonly see that weight-reduced subjects will have a ridiculous appetite (hyperphagia), and will tend to preferentially regain fat instead of lean mass when they begin to regain weight [4, 5]. Our pilot study in physique athletes showed a similar pattern, unfortunately, with very little “true” lean mass gained within the first 4-6 weeks after competing [3].

Taken together, the evidence suggests that you should not be force-feeding yourself in hopes of capitalizing on some extra anabolic state after competition. If anything, this is likely to cause a whole lot of rapid fat gain. Research has also suggested that hyperphagia may continue, even after you’ve regained all of the fat you lost, if you have not fully restored the lean body mass you lost while dieting [4, 5]. So, even though the deck is stacked against you, diet and training strategies should be tailored to maximize your chances of regaining lean mass at a realistic rate during the post-competition period.


3) How long does it take to recover?

It depends on who you are, what you do, and how you define “recovered.” Case studies tell us that most factors, such as hormone levels, metabolic rate, and physical performance, generally return toward baseline after about 6-7 months [2, 6]. However, these subjects also tend to restore their weight and body fat levels to near-baseline levels by the end of this 6-7 month period. Had they attempted to maintain low caloric intakes and a contest-level body weight, it’s very safe to assume that recovery would be substantially delayed, or potentially halted altogether, depending on the magnitude of post-contest restriction.

However, there are some factors that may lag behind during recovery. For instance, some struggle with appetite and/or their psychological eating habits well after 7 months post-show. In fact, one case study [7] documented amenorrhea in a female competitor that persisted for 71 weeks following competition! As such, it’s almost impossible to give a single estimate for the time course of recovery- different components recovery at different rates, and the rate of recovery depends on the individual and the strategies they employ after competition.


4) Conclusions and practical applications (the stuff you can actually use)

When you get in contest shape, your body will hate you, and recovery will be required. In the post-competition period, how quickly you choose to increase caloric intake and/or gain weight should be determined based on the needs and goals of the individual- a “one size fits all” approach is not appropriate. Along the same lines, the rate at which these things occur will influence the timeline of recovery, so different individuals will recover at different rates.

To effectively navigate this time period, I think a few strategies come in handy. I’m going through this process as we speak, and I have made a point of emphasizing high protein intake, consumption of foods with plenty of fiber and low energy density, and training to maximize hypertrophy as much as possible.

The high protein consumption may help on several fronts- protein supports the process of regaining lean mass, has a high thermic effect of feeding, and promotes satiety. Collectively, these factors may impart favorable effects with regard to rebuilding muscle, fighting excessive hunger, and restoring metabolic rate. In fact, our pilot study found a correlation between post-competition protein intake and restoration of metabolic rate.

Consumption of high-fiber foods with low energy density is generally a good idea, even if you aren’t a physique athlete. The foods that meet this criteria often include micronutrient-dense fruits and vegetables, which carry a wide range of general health benefits. With respect to the post-contest period, these foods will also increase satiety and can help curb urges to overeat.

Finally, if you’re going to be in a caloric surplus, you might as well put those calories to use. Train to maximize hypertrophy, targeting each muscle group with sufficient volume and intensity at least twice per week. Personally, I tend to opt for an upper body/lower body split after competition, allowing me to train each muscle group every 48-72 hours to stimulate muscle growth. This will assist with nutrient partitioning, optimization of body composition, and recovery of lean mass, which may play a role in attenuating post-competition hyperphagia and overeating.

So plenty of protein, fruits, vegetables, and lifting are good. I know, I know- groundbreaking, controversial findings indeed.


Be sure to check out part two of this article, which discusses psychological aspects of the post-competition period!



  1. Trexler ET, Smith-Ryan AE, Norton LE: Metabolic adaptation to weight loss: implications for the athlete. J Int Soc Sports Nutr. 2014;11(1):7.
  2. Pardue A, Trexler ET, Sprod LK: Case Study: Unfavorable But Transient Physiological Changes During Contest Preparation in a Drug-Free Male Bodybuilder. Int J Sport Nutr Exerc Metab. 2017:1-24.
  3. Trexler ET, Hirsch KR, Campbell BI et al.: Physiological Changes Following Competition in Male and Female Physique Athletes: A Pilot Study. Int J Sport Nutr Exerc Metab. 2017:1-25.
  4. Dulloo AG, Jacquet J, Girardier L: Poststarvation hyperphagia and body fat overshooting in humans: a role for feedback signals from lean and fat tissues. Am J Clin Nutr. 1997;65(3):717-23.
  5. Dulloo AG, Jacquet J, Montani JP et al.: How dieting makes the lean fatter: from a perspective of body composition autoregulation through adipostats and proteinstats awaiting discovery. Obes Rev. 2015;16 Suppl 1:25-35.
  6. Rossow LM, Fukuda DH, Fahs CA et al.: Natural bodybuilding competition preparation and recovery: a 12-month case study. Int J Sports Physiol Perform. 2013;8(5):582-92.
  7. Halliday TM, Loenneke JP, Davy BM: Dietary Intake, Body Composition, and Menstrual Cycle Changes during Competition Preparation and Recovery in a Drug-Free Figure Competitor: A Case Study. Nutrients. 2016;8(11).



New article: Using fat-free mass index to guide your off-season

ffmi trexfit picture

My primary goal of this website, and my career in general, is to highlight and apply the symbiotic relationship between research and application. I find joy in using observations “from the trenches” to guide research questions, and in translating research findings back to applicable, real-world training and nutrition recommendations.

My newest article, “Using fat-free mass index to guide your off-season,” is an excellent example of this. The article, featured on BioLayne.com, translates my recent publication into some practical tips that help you use this valuable metric to guide your mass-gaining goals. Moreover, the BioLayne staff went above and beyond and put together a nice FFMI calculator to accompany the article and make life even easier.

As always, it’s an honor to be featured on Layne’s site, so I want to thank him and the BioLayne staff for their help and support in getting this article up. I’m about 4 days away from a nice, mass-filled off-season, so the timing couldn’t be better!

Check out the article, and please share if you like it.

Three new podcast appearances

mydnacoach podcast ep 96 Eric Trexler

Greetings from Phoenix!

The 2017 ISSN conference just wrapped up, and I’m at the airport to catch a red-eye back to Chapel Hill. After a fun and information-packed conference, I finally have a minute to post a quick update.

A couple weeks ago I had the pleasure of being Jonny Deacon’s guest on three episodes of the “My DNA Coach” podcast. We discussed a few ingredients that I have previously done some research on, including creatine, caffeine, and nitric oxide boosters.

Check out the links below and give them a listen!

Episode 96: Creatine

Episode 97: Caffeine

Episode 98: Nitric oxide boosters



ICYMI: Featured guest on “The Bodybuilding Summit”

bodybuilding summit image

A while back, I had the pleasure of sitting down for a podcast with Joshua Vogel. This was part of his “Bodybuilding Summit” event, which featured an incredible list of bodybuilding experts ranging from active researchers to world-class competitors.

I’ve been extremely lucky to have the opportunity to merge my personal interest in bodybuilding with my research activities (a couple examples here and here). In this podcast, we discuss all things bodybuilding, drawing upon experiences from both the gym and the laboratory. Check it out here!