Split routines are pretty much synonymous with bodybuilding. A recent survey of 127 competitive bodybuilders found that every respondent trained with a split routine. Every one! Moreover, 2/3 of respondents trained each muscle only once per week (what is popularly known as a â€śbro-splitâ...
Recent Blog Posts
July 17, 2015
Split routines are pretty much synonymous with bodybuilding. A recent survey of 127 competitive bodybuilders found that every respondent trained with a split routine. Every one! Moreover, 2/3 of respondents trained each muscle only once per week (what is popularly known as a â€śbro-splitâ€ť) and none worked a muscle more than twice weekly. The theory behind such routines is that growth is maximized by blasting a muscle with multiple exercises from multiple angles and then allowing long periods of recovery.
Things werenâ€™t always this way, though.
Old-school bodybuilders such as Steve Reeves and Reg Park swore by total-body routines, working all the major muscles each and every session over three non-consecutive days-per-week. Proponents thought that the greater training frequency was beneficial to packing on lean mass.
Thing is, the choice to use one type of routine or another has been almost exclusively based on anecdote and tradition. Surprisingly little research has been carried out on the topic, and no study had directly compared muscle growth in a total-body routine versus a bro-split.
My lab carried out a controlled experiment to investigate the effect of training frequency on muscular adaptations. The study was recently published in the Journal of Strength and Conditioning Research. Hereâ€™s the scoop.
What We Did
Nineteen young men with an average of more than 4 years lifting experience were randomly assigned to a resistance training program using either a total-body (all muscles worked in a session) or split-body routine (2-3 muscle groups worked per session). The program consisted of 21 different exercises spread out over a 3 day-per-week training cycle. The volume of the routines were matched so that both groups performed an equal number of sets and reps over the course of each week. All subjects performed 3 sets of 8-12RM per exercise. Training was carried out for 8 weeks. The table below shows the program design for both routines.
Subjects were tested pre- and post-study. We used B-mode ultrasound to measure the thickness of the biceps, triceps, and quads, and assessed maximal strength via 1RM for the back squat and bench press. Subjects were advised to consume their normal diets and we monitored food intake by analysis of a self-reported diary.
What We Found
Subjects in both groups significantly increased hypertrophy in the arm and leg muscles. That said, muscle mass increased significantly more in the biceps/brachialis for the group performing total body training compared with those in the split routine group. There was a trend for greater increases in the quads (i.e. vastus lateralis) and the effect size â€“ a measure of the â€śmeaningfulnessâ€ť of results â€“ markedly favored the total body group. Although no significant between-group differences were found in triceps thickness, the effect size again showed an advantage to total body training.
With respect to strength, both groups significantly increased 1RM performance in the bench press and squat from baseline. There were no significant between-group differences in either of these measures, although the effect size for the bench press did seem to favor the total body group.
How Can You Use This Info?
On the surface it would seem that a total-body routine is superior to a one-muscle-per-week bro-split for building muscle. All of the muscles we investigated showed greater growth from a higher training frequency. For the biceps, these results were â€śstatistically significant,â€ť meaning that that there was a greater than 95% probability that results did not occur by chance. While results in the quads and triceps did not reach â€śsignificance,â€ť other statistical measures indicate a pretty clear advantage for the higher frequency routine. These results would seem to be consistent with the time-course of protein synthesis, which lasts approximately 48 hours (there is even some evidence that the time course is truncated as one gains lifting experience). Theoretically, repeated spiking of protein synthesis after it ebbs would result in greater muscular gains over time.
Before you jump the gun and ditch your split, a few things need to be considered when extrapolating results into practice.
First and foremost, itâ€™s important to remember that the study equated volume between conditions. This was done to isolate the effects of frequency on muscular adaptations – an essential strategy for determining causality. However, a primary benefit of a split routine is the ability to increase per-workout volume while affording ample recovery between sessions. Since there is a clear dose-response relationship between volume and hypertrophy, total weekly volume needs to be factored into the equation. Certainly itâ€™s possible that a split routine with a higher weekly volume would have performed as well or even better than the total body routine. Or perhaps not. We simply donâ€™t know based on the current literature.
In addition, the vast majority of subjects in the study reported using a split routine as the basis of their usual workout programs, with muscles worked just once per week. This raises the possibility that the novelty factor of the total body routine influenced results. There is in fact some research showing that muscular adaptations are enhanced when program variables are altered outside of traditional norms. Itâ€™s therefore conceivable that participants in the total body group benefited from the unaccustomed stimulus of training more frequently.
Drawing Evidence-Based Conclusions
Given the available info, hereâ€™s my take on how the findings can be applied to your training program. There does seem to be a benefit to more frequent training sessions if max muscle is the goal. In this regard, itâ€™s best to directly work each muscle at least twice a week; any less and you’re probably not stimulating protein synthesis frequently enough to optimize hypertrophy. Training each muscle three times a week, at least for periods of time, may provide additional benefits for spurring further gains.
Given the novelty factor, itâ€™s reasonable to speculate that periodizing frequency over the course of a long-term training cycle might be the ideal option. Progressing from periods of working muscles twice to three times per week (and perhaps more) and then cycling back again will conceivably provide a novel stimulus that elicits continued gains. But remember: any discussion of training frequency must take total weekly volume into account. Greater training frequencies (from the standpoint of total training sessions per week) using a split routine can be employed to maximize total weekly volume and thus potentially drive greater hypertrophy over time.
May 9, 2015
Itâ€™s a commonly accepted tenet that resistance training adaptations follow a â€śstrength-endurance continuumâ€ť whereby lifting heavy loads maximizes strength increases while light load training leads to optimal improvements in local muscle endurance. Conventional wisdom also postulates that at least moderately heavy loads are required for building muscle. General training guidelines proclaim that loads lighter than about 65% 1RM are insufficient to stimulate fast-twitch muscle fibers necessary for growth. The so-called â€śhypertrophy rangeâ€ť is generally considered to be 6-12 reps/set.
Recent research has challenged these established tenets. It has been proposed that if light loads are lifted to muscular failure, near-maximal recruitment of fast-twitch fibers will occur resulting in muscular adaptations similar to those obtained from training heavy.
A meta-analysis from my lab published last year in the European Journal of Sports Science found substantial increases in muscle strength and hypertrophy following low-load training. However, the magnitude of increases were not as great as that associated with using heavier loads, and a trend for superior gains was in fact shown when lifting weights >65% 1RM. I covered the specifics of this meta-analysis in a previous post.
The caveat: All previous studies employed untrained subjects, raising the possibility that results were attributed to the â€śnewbie effectâ€ť that states those new to training build muscle from pretty much any activity — even cardio!
To achieve clarity on the topic, my lab carried out a well-controlled study on the effects of high- versus low-load training using resistance-trained individuals, which was just published in the Journal of Strength and Conditioning Research. Hereâ€™s what you need to know.
What We Did
Eighteen young men with an average of more than 3 years lifting experience were randomly assigned to a resistance training program using either moderately heavy loads (8-12RM) or light loads (25-35RM). All other aspects of the program were held constant between groups to isolate the effects of load on muscular adaptations. The program consisted of 3 sets of 7 different exercises targeting the major muscle groups (bench press, shoulder press, lat pulldown, seated pulley row, back squat, leg press, and leg extension). Training was carried out on 3 non-consecutive days-per-week (M, W, F) for 8 weeks.
Testing was conducted pre- and post-study. We used b-mode ultrasound to measure the thickness of the biceps, triceps, and quads. We assessed maximal strength via 1RM for the back squat and bench press. Finally, we measured changes in muscle endurance by having subjects perform the bench press at 50% of their 1RM to volitional failure.
What We Found
Both groups significantly increased lean mass in their biceps, triceps, and quads, but no statistically significant between-group differences were noted in any of these muscles (i.e. both groups had similar muscle growth over the course of the study). On the other hand, the heavy load group showed significantly greater strength increases in the back squat and a trend for greater increases in the bench press compared to the light load condition. Conversely, local muscle endurance was markedly greater for the low-load group.
Reconciling the Data
The primary take-home points from the study are as follows:
â€˘ Gains in muscle mass are about the same regardless of repetition range provided training is carried out to muscle failure
â€˘ Maximal strength requires the use of heavy loading
â€˘ Muscle endurance is best obtained from the use of light loads
To really understand the practical implications of the study, however, we need to look a bit deeper at the results.
The superior strength gains for heavy load training are consistent with the principle of specificity, which effectively states that training adaptations are specific to the imposed demands. No surprise here. From a mechanistic standpoint, the ability to exert maximal force has a high neural component, and the associated neural adaptations appear to be optimized through the use of heavy loads. Previous work from my lab showed that these adaptations exist even at the far left aspect of the strength-endurance continuum, as a powerlifting-type routine (3RM) was found to produce greater strength increases compared to a bodybuilding-style workout (10RM). It also makes intuitive sense that you need to train heavy to â€śget a feelâ€ť for using the maximal loads required to perform a 1RM.
The greater improvements seen in local muscle endurance from light-load training were expected as well. Although the topic hasnâ€™t been well-studied, it stands to reason that low-load training is associated with adaptations specific to enhancing buffering capacity, thereby allowing for the performance of a greater number of submaximal repetitions. Again, a basic application of the principle of specifity.
On the other hand, I readily admit to being surprised by the fact that muscle growth was similar between conditions. While a number of previous studies had shown no differences in gains between light- and heavy-load training, I figured this was due to the â€śnewbie effect.â€ť No way could you build appreciable muscle using 30 reps per set.
Or so I thought.
Iâ€™m now a believer.
Whatâ€™s particularly interesting, though, are the potential implications for how muscle growth actually manifests when training in different loading zones. A previous study from my lab showed that muscle activation was markedly greater when performing reps at 75% 1RM versus 30% 1RM. A follow up study (currently in review) found that the heavy-load superiority for activation held true when training at 80% 1RM versus 50% 1RM as well. Combined, these findings suggest that the recruitment and/or firing frequency in the high-threshold motor units associated with the largest type II fibers is suboptimal when training at low-loads. It therefore can be hypothesized that if muscle growth is indeed similar across loading zones â€” as found in the current study â€” hypertrophy from light-load training necessarily must be greater in the type I fibers. Indeed, emerging research out of Russia indicates that this is in fact that case with multiple studies showing that light loads promote greater gains in type I fibers while heavy loads increase type II fiber hypertrophy to a greater extent (Netreva et al 2007; Netreba et al 2009; Netreba et al 2013; Vinogradova et al 2013).
Bottom line: If your goal is to build as much muscle as possible, it seems appropriate to train across the spectrum of loading zones; use lighter loads to target type I fibers and heavier loads to target type IIs. In this way, you ensure maximal development of all fiber types.
An interesting point to keep in mind is that none of the subjects in my study trained with more than 15 reps/set during the course of their usual lifting routines and the majority never went above 10 reps. This raises the possibility that their endurance-oriented type I fibers were underdeveloped in relation to the strength-oriented type II fibers. If so, itâ€™s possible that their type I fibers had a greater capacity for growth, which was realized in those who trained using light loads.
The study had some notable limitations. For one, the training period lasted only 8 weeks; whether results would have diverged over a longer time-frame is undetermined. For another, muscle thickness was measured only at the approximate mid-point of each muscle. Research has shown that muscles often hypertrophy in a non-uniform manner. Thus, it is possible that other aspects (i.e. distal or proximal) of the muscles studied might have differed in their growth response.
A final and important point to consider. While people often dismiss light-loads as being for wimps, nothing could be further from truth. Training to failure with high reps is highly demanding and the associated acidosis extremely uncomfortable. To this end, approximately half the subjects in the low-load group puked during the first week of training and several others experienced nausea and/or light-headedness. Although these issues tended to dissipate as time went by, they nevertheless can negatively affect adherence to the program. If you choose to incorporate light-loads into your program, be prepared for a grueling workout!
March 29, 2015
These are the words of a noted fitness trainer in response to a bodybuilder who spoke of packing on some additional muscle. The trainer went on to say that you can only gain muscle for a couple of years; after that, youâ€™ve maxed out your genetic potential.
If the trainer is indeed correct in his claim, then everyone with a modicum of training experience is basically spinning his wheels in the gym; might as well just do a couple of 15 minute HIT workouts and maintain what youâ€™ve got. Fortunately for those of us who aspire to keep making gains, the comments made were both misguided and uninformed.
Donâ€™t get me wrong. There certainly are upper limits to how much muscle you can build, just as there are limits to muscular strength, aerobic endurance, and any other exercise-induced adaptation. This is commonly known as your â€śgenetic ceilingâ€ť; at a certain point, you hit your ceiling and further gains cease.
Thing is, how do you know if youâ€™ve reached your genetic ceiling?
Answer: You donâ€™t.
In fact, you canâ€™t.
All you can ascertain is whether or not your training regimen is producing positive changes in your physique. And if youâ€™re not in fact growing from your present routine, that doesnâ€™t mean you might not see results from an alternative strategy. The number of possible ways to vary program design is virtually unlimited. Unless you try each and every alternative, thereâ€™s no way to know if another approach might be the ticket to further gains.
Understand that the reason your muscles adapt to an exercise stimulus is a function of survival. Your body doesnâ€™t realize the reason you hit the gym is to look jacked in a tank-top; rather, it senses a high degree of physical stress that is deemed a threat to survival. In response, a coordinated series of intracellular events are initiated to strengthen the muscles and supporting tissues so that they are better prepared the next time you lift.
Problem is, the more you continue to provide similar stimuli, the less of a need for future adaptation. Further growth can only occur by subjecting your muscles to a novel overload stimulus.
The imprudent nature of the comments made by the aforementioned trainer is reflected in his own training practices. Namely, he is known to perform the same basic routine over and over each and every year. Why would the body respond to a stimulus that it perceives it can readily handle?
Answer: It wonâ€™t.
While a â€śceilingâ€ť may exist in theory, you never actually realize your full genetic potential; there is always the ability to further increase muscle mass. Indeed, muscular gains can be made even at very advanced levels, albeit at a much slower pace than when you first started training.
Numerous research studies â€“ including those from my own lab â€“ show that those with considerable training experience do in fact build appreciable muscle when a novel stimulus is applied. Thus, the claim that a couple of years hitting the weights maxes out your genetic potential is patently false. Because of the difficulties in carrying out studies on those near the limits of their hypertrophic ceiling, research on this population is scant. That said, I recently collaborated with a group in Brazil on a study involving off-season pro bodybuilders who werenâ€™t using performance enhancing drugs (the study is currently in journal review). Suffice to say, significant gains in fat-free mass (as measured by DXA) were noted after just 4 weeks of intense training. Anecdotally, Iâ€™ve worked with numerous competitive natural physique athletes whoâ€™ve added several pounds of lean body mass over the course of a regimented hypertrophy training phase.
Now the closer you get to your individual ceiling, the more essential it is to take a scientific approach to training and nutrition. From a training standpoint, this entails precise manipulation of resistance exercise variables. Here, the concept of â€śprogressive overloadâ€ť needs to be expanded beyond simply increasing load within a given rep range. Adaptation can and should be achieved by varying loading zones as well. If nothing else, changing up loading patterns provides a novel stimulus to your muscles that can spur new growth. Moreover, emerging evidence suggests that heavy, moderate, and light loads promote fiber type-specific increases in growth that can maximize whole muscle hypertrophy. Perhaps more importantly, volume of training should be progressively increased, culminating in a high-volume phase designed to promote functional overreaching. When properly executed, this results in a supercompensatory response that increases muscle in even the most advanced lifters. Many other advanced lifting strategies also can be employed to enhance results; you’re only limited by your determination and base of knowledge.
Bottom line; If someone tells you that youâ€™re done adding muscle, pay them no heed. Itâ€™s a self-limiting attitude that will keep you from achieving your full genetic potential.
January 25, 2015
Conventional wisdom states that eating small, frequent meals helps to optimize weight loss. In theory, eating frequently enhances a phenomenon called the thermic effect of food (TEF), which results in more energy expended after consumption of the meal. Whatâ€™s more, some postulate that multiple meals spaced throughout the day prevents the body from going into â€śstarvation mode,â€™ thereby keeping metabolism perpetually elevated.
There also is speculation that frequent feedings are beneficial for anabolism. This is based on the premise of a limit to how much protein can be used to maximize protein synthesis. It therefore follows that large boluses of protein result in extensive oxidation of amino acids, preventing their use in tissue building purposes.
Despite a seemingly logical rationale, the efficacy of consuming frequent meals to optimize body composition has not been well established in long-term studies. In an attempt to gain clarity on the topic, my lab recently carried out a meta-analysis where we pooled the data from all meal frequency studies. The analysis was a collaboration with my colleagues and frequent partners-in-science, James Krieger and Alan Aragon. Here’s the scoop…
What We Did
A thorough search of all English language journals was conducted for studies with the following inclusion criteria:
1. Randomized controlled trial
2. Compared unequal feeding frequencies of less than or equal to 3 meals a day with greater than 3 meals a day
3. Had a study duration of at least 2 weeks
4. Reported a pre- and post-intervention measure of body composition (body mass, body fat, lean mass)
5. Was carried out in human participants >18 years of age
A total of 15 studies were identified that met the criteria outlined and provided adequate data for analysis â€“ several of these studies went back as far as the early 1960â€™s! The studies were individually coded and a randomly selected number of them were subsequently recoded to ensure accuracy. The coded studies were then pooled and statistically analyzed to determine what, if any, body composition differences existed between feeding frequencies.
What We Found
There was no effect of the number of daily meals on body mass (i.e. weight). Alternatively, initial analysis did show a positive association between feeding frequency and reductions in fat mass. Hereâ€™s the kicker: a sensitivity analysis showed that a single 2-week study by Iwao et al. highly affected results â€“ when this study was removed from analysis, the effect of meal frequency was no longer significant. Similarly, body fat percent was initially shown to correlate with greater decreases in body fat percentage, but the results were highly affected by a single study by Arciero et al. whose removal rendered the results insignificant. There was a trend for greater increases in fat free mass with higher meal frequencies, but again the results were primarly attributed to the Iwao et al. study.
The results of our analysis do not support a tangible benefit to eating small frequent meals on body composition as long as daily caloric intake and macronutrient content is similar. The theory that a greater feeding frequency increases post-prandial thermogenesis is fundamentally flawed. As shown in the accompanying table, a typical meal results in a TEF of approximately 10%. Since the TEF is dependent on the number of calories consumed in the meal, the net thermic effect is the same for 3 versus 6 meals on a calorie-equated basis. There also is no evidence that the body goes into â€śstarvation modeâ€ť when you go without food for more than a few hours as commonly claimed in fitness circles. I covered the research on this in a recent T-Nation article.
The studies in question lasted varying amounts of time and many used recall food diaries to assess caloric intake, which have been shown to lack accuracy in reporting. However, several studies were carried out in metabolic wards where every morsel of food and every step of activity was carefully monitored â€“ these studies showed no benefit to higher meal frequencies, providing further confidence in the validity of our findings.
A primary limitation of the analysis was that all studies to date were carried out in sedentary individuals. Thus, results cannot necessarily be generalized to those involved in regular exercise, particularly resistance training. There is compelling evidence that the muscles are sensitized to protein intake for at least 24 hours after a lifting session, suggesting a potential benefit to frequent feedings with protein rich foods in the post-exercise period. Whether this translates into greater long-term muscle growth remains to be determined.
It also isnâ€™t clear if our findings are applicable to diets that include higher daily protein intakes. All of the studies analyzed used low to moderate protein doses, with the exception of the study by Arciero et al. Interestingly, this study did show significant improvements in body composition when an energy-equated high-protein diet (approximately 34% of total calories) was consumed in 6 versus 3 daily meals.
Take Home Points
The number of daily meals consumed does not appear to have much if any impact on changes in body fat, at least across a wide spectrum of feeding frequencies. Thus, the decision on how many meals to eat from this standpoint should come down to personal preference: if you find a benefit to having the structure of multiple meals throughout the day, then go for it; on the other hand, if you prefer to eat less frequently, that’s fine as well. The most important factor in this regard is achieving a negative energy balance, as well as ensuring that adequate dietary protein is consumed.
Although our analysis did not show differences between meal frequencies with respect to lean body mass changes, there is a logical basis for a hypertrophic benefit to consuming several protein-rich meals in those involved in regular resistance exercise. The anabolic effects of a meal last a maximum of 6 hours or so. Thus, consumption of at least 3 meals spaced out every 5 to 6 hours would seem to be optimal for keeping protein synthesis continually elevated and thus maximizing muscle protein accretion. This hypothesis needs further investigation in a controlled long-term study.
December 21, 2014
It is often stated that heavy loads (>65% 1RM) are required to promote muscular adaptations; light loads are generally considered ineffective for enhancing these outcomes. Recently, this belief has been challenged by several researchers. It has been proposed that as long as training is carried out to muscular failure, light load training will recruit the full spectrum of motor units (and thus muscle fibers), allowing for gains similar to that of using heavy loads.
Last year, I published a review on the topic in the journal Sports Medicine titled, Is there a minimum intensity threshold for resistance training-induced hypertrophic adaptations?. After thoroughly scrutinizing the body of literature, I ultimately concluded: “Current research indicates that low-load exercise can indeed promote increases in muscle growth in untrained subjects, and that these gains may be functionally, metabolically, and/or aesthetically meaningful.”
However, a narrative review is limited to drawing inferences based on a general sense of the research evaluated; it cannot provide quantification of data. A big issue with resistance training studies is that they are very costly and time-consuming to carry out. This invariably leads to small sample sizes where studies lack statistical power to note a significant difference (a so-called a Type II error). I therefore decided to conduct a meta-analysis, where the data from all relevant studies are pooled to maximize statistical power. and thus provide greater clarity on the topic. I teamed up with my colleagues James Krieger, Jacob Wilson, and Ryan Lowery to carry out the analysis.
What We Did
A systematic search of the literature was conducted to identify studies that would potentially be relevant to the meta-analysis. We filtered through the studies and subjected them to rigid inclusion criteria. To meet eligibility, studies had to:
1. Be a randomized controlled trial involving both low (<60% 1RM)- and high-load (>65% 1RM) training
2. Span at least 6 weeks
3. Directly measure dynamic muscle strength and/or hypertrophy
4. Carried out training to momentary muscular failure in both protocols
A total of 13 studies were identified that met inclusion criteria. Three of these studies did not contain adequate data for computation of effect sizes, leaving a total of 10 studies for analysis. Studies were separately coded by two researchers, and we cross-checked our data for consistency. We then randomly chose 3 studies for recoding to ensure there was no “coder drift.” The results of these studies were converted into effect sizes for comparison between conditions.
What We Found:
No significant differences were seen between low- versus high-load training in either strength or hypertrophy, although a trend for greater increases was noted in both conditions.
What These Results Mean
Results of the meta-analysis support the findings of my narrative review on the topic, showing that substantial hypertrophy and even strength can be achieved by training with light loads. Based purely on statistical probability (i.e. the odds that results are due to chance), there was no difference between using heavy and light loads for gaining strength or muscle. However, several things need to be taken into account when drawing evidence-based conclusions.
First, there was a trend for greater results in both strength and hypertrophy. This is a topic that has not been extensively researched, thereby limiting the statistical power of the meta-analysis. The trends noted would suggest that there is actually a difference favoring the heavy load condition, but statistical power was not great enough to sufficiently detect such a difference. Looking beyond basic probability statistics, other analytic measures provide interesting insight into results. Of particular note was the fact that the effect size (a measure of the magnitude of the difference in results) for strength was was markedly higher in the heavy- vs. light-load condition (2.30 versus 1.23, respectively). The 95% confidence interval differential also favored using heavy loads (CI: -0.18â€“2.32). Moreover, all 9 studies that investigated strength as an outcome favored high-load training, and six of these studies showed a moderate to strong difference in magnitude of effect. In combination, this provides strong evidence that maximal strength gains require heavier loads.
Effect size data for hypertrophy also favored the high- versus low-load conditions (0.82 vs 0.39), although the differential was not nearly as compelling as for strength. Taken in combination with the trend for significance, this suggests a potential advantage for higher-load training when the goal is maximal hypertrophy.
When reconciling findings, the results of our analysis provide compelling evidence that the use of light loads can be effective for increasing muscle size as well as muscle strength. These findings have wide-ranging implications for many populations, particularly the elderly and those with medical conditions that might preclude the use of use of heavier loads (i.e. osteoarthritis, osteoporosis, etc). Alternatively, those seeking to maximize muscular adaptations would require at least some use of heavy loading. Despite an inability to detect significant differences between conditions, the findings indicate a clear advantage for the use of heavier loads to maximize strength gains. There is a suggestion that heavy loads promote greater hypertrophic increases as well, but this inference is not as convincing. With respect to hypertrophy, it can be hypothesized that combining high- and low-loads could optimize fiber-type specific growth across the spectrum of myofiber isoforms. This hypothesis warrants further study.
A primary limitation of the meta-analysis was that all of the studies analyzed were carried out in untrained individuals; no published study to date has evaluated the topic in well-trained individuals. The good news is that I have completed just such a study, where subjects were all experienced lifters. The study is currently in review. I hope to be able to share results and their implications soon. Stay tuned!
November 21, 2014
In the late 1990â€™s, Bill Phillips authored â€śBody for Life,â€ť which went on to become one of the biggest selling fitness books of all time. In the book, Phillips claimed that performing 20 minutes of high-intensity aerobic exercise (HIIT) after an overnight fast has a greater effect on fat loss than an hour of cardio performed following consumption of a meal. The rationale for the hypothesis was based on research showing that low glycogen levels cause your body to shift substrate utilization away from carbohydrates, thereby allowing greater mobilization of stored fat for energy.
While the theory that fasted cardio is superior for fat loss is certainly intriguing, it is based on an extrapolation of findings that might not translate into practice. Several years ago I authored a review of literature that discussed the contradictions of the research on the topic. While my review highlighted a number of inconsistencies that suggested fasted cardio might not work as claimed, one little issue continued to nag at me: The entire debate was based on acute data; no study had actually investigated the effects of fasted cardio on body fat when subjects were in an energy-deficit sufficient to produce weight loss.
My lab recently carried out a controlled longitudinal trial designed to achieve clarity on the topic. The paper titled, Body composition changes associated with fasted versus non-fasted aerobic exercise was just published in the Journal of the International Society of Sports Nutrition. Here is a rundown of what we found along with a discussion of relevant practical implications.
What We Did
Twenty young non-obese (BMI < 30) were recruited to participate in the study. Prior to the intervention, subjects were tested for body composition (weight, body fat percent, fat mass, fat-free mass, and waist circumference) using a Bod Pod (i.e. air displacement plethysmography), then pair-matched based on initial body mass measurements and randomly assigned to 1 of 2 groups: a fasted training (FASTED) group that performed exercise after an overnight fast (n =10) or a non-fasted training (FED) group that consumed a meal prior to exercise (n =10). These meals were provided in the form of a shake (Pursuit Recovery by Dymatize Nutrition) that contained 250 calories consisting of 40 g carbohydrate, 20 g protein, and 0.5 g fat.
Training was carried out 3 days a week for 1 hour per session on a treadmill. Subjects performed a 5 minute warm-up followed by 50 minutes of walking/jogging at 70% max heart rate. A 5-minute cool down was then provided to end the session. We chose this protocol because evidence shows that lipid oxidation during fasted aerobic exercise is maximized during low-to moderate-intensity steady-state cardio â€“ at higher intensities, the fasted condition allows for an acutely greater lipolysis but the overall oxidation rate is similar to the fed condition because more free fatty acids are available than can be oxidized (in fairness to Bill Phillips, this research came out after publication of his book).
Subjects were provided with customized meal plans intended to bring about a 500-calorie deficit. The meal plans were flexible so that subjects had wide choices of their preferred foods. Protein was kept high to help ensure preservation of lean body mass. Subjects recorded their food in an online web-based program (myfitnesspal.com) on a daily basis so that dietary intake could be continually monitored. Ongoing nutritional counseling was provided to subjects throughout the study period to promote compliance and adherence.
After 4 weeks, the subjects were retested on all body composition measures.
What We Found
Both groups lost a statistically significant amount of weight (1.6 kg vs. 1.0 kg in the FASTED vs. FED groups, respectively) and fat mass (1.1 kg vs. 0.7 kg in the FASTED vs. FED groups, respectively). However, no significant differences were noted between groups in any of the body composition outcomes.
Reconciling Findings with Practical Implications
On the surface, it might seem that the fasted cardio group had a slight advantage in terms of weight loss and fat loss. Iâ€™ve seen comments on social media to this effect, claiming a hidden â€śtrendâ€ť for a benefit to fasted cardio that our study simply was underpowered to detect.
Fact is, the claim is unsubstantiated.
The p-values (a determinant of the probability that results were due to random chance) were *highly* insignificant, averaging 0.8-0.9 for the various body comp outcomes. Moreover, differences in effect sizes (a measure of the magnitude of the effect) between groups were negligible, further indicating a lack of differences. On top of all that, the FASTED group had somewhat higher (non-significant) baseline body fat percentage, providing a potential advantage for slightly greater fat loss. There are certainly times that I have seen and reported trends in studies I’ve carried out where it was apparent that the small sample size obscured significant differences. That was most definitely NOT the case here. Based on the findings, any differences noted would be attributable to chance — I’d say with a high level of confidence that the sample size was not an issue in this regard.
So does this mean the case is closed and that fasted cardio is worthless for fat loss?
As with every study, there were a number of limitations that must be taken into account when drawing evidence-based conclusions.
For one, the study spanned only four weeks in duration. Certainly this is a sufficient period of time to realize significant reductions in weight and fat mass (as was demonstrated here), but it remains possible that very slight differences between conditions *might* take longer to manifest. We limited the duration of the study in an effort to ensure that the subjects adhered to the diet and exercise protocol (longer term trials in young college students can be problematic with respect to adherence). Our study was not funded, so we could not offer remuneration as an incentive for sticking with the program. Ideally a 16 week protocol would better determine if any small effects would ultimately become significant.
Another potential confounding issue was the use of pre-menopausal women as subjects. Monthly menstrual cycles can influence body weight due to alterations in fluid balance. The fact that pre- and post-testing was conducted exactly one month apart would seem to control for any issues in this regard. However, some women have irregular menses and we cannot rule out the possibility that such fluctuations influenced results.
I will also note that subjects lost slightly less weight than anticipated. This seems to be due to consuming more calories than prescribed in the meal plans. Despite our best efforts to counsel subjects on what to eat and providing detailed instruction on how to to record foods in the online database, the subjects apparently under-reported their nutritional intake. That said, analysis of food diaries indicates that under-reporting was equally distributed between groups and thus this should not have affected overall results. Whether a larger caloric deficit would have provided an advantage to one condition versus the other is open to debate.
Finally, our findings are specific to young, non-obese women and cannot necessarily be generalized to other populations. It has been speculated that the true benefit of fasted cardio is specific to very lean individuals, such as pre-contest bodybuilders, who are trying to lose that last pound or two of stubborn fat. We cannot rule out such a possibility. I’ll note, however, that several of our subjects were off-season track athletes who were quite lean. In fact, four of the subjects (two in each group) had body fat levels that would be considered very low for women (13-16%). When analyzing the results of these four subjects, there was no evidence whatsoever that the fasting condition conferred any benefits. Admittedly this is a tiny sample and certainly cannot be taken as proof of anything. Nevertheless, it does provide a more controlled, objective perspective into potential benefits of fasted cardio than the usual “it worked for me” claims that lack any level of control or objectivity.
A single study is simply a piece in an evidentiary puzzle and can never considered the final word on a topic. What I do think is clear from our study, however, is that if there are any benefits from fasted cardio (still highly equivocal), they would be minor at best. So the best advice for those who are simply looking to get lean is to focus on total energy and macronutrient balance; whether you perform cardio fasted or fed should depend entirely on preference.
On the other hand, it remains possible that a small benefit could be seen by performing fasted cardio. If such an effect does exist, it would seem to be only meaningful to someone who is competing in a bodybuilding or physique competition, where minute differences in fat mass could make the difference between winning or losing a competition. I will point out, however, that it also is conceivable fasted cardio could have a negative effect in this regard. A recent study by Paoli et al showed that lipid utilization over 24 hours was actually higher when eating prior to cardio as opposed to remaining fasted. Thus, the best advice here would be to experiment and try to objectively determine what works best for you as an individual.
October 20, 2014
I recently collaborated with my colleague Menno Henselmans on a review paper that sought to provide clarity on the effects of rest intervals on muscle hypertrophy. Based on the current literature, we concluded that evidence was lacking to support the contention that rest interval length has an impact on growth. Problem is, there have been very few studies carried out to investigate the topic. Thus, it’s difficult to say with any degree of confidence as to whether there are or aren’t any benefits to varying how long you should rest between sets. For more on specifics of the review paper check out my blog post where it is discussed in detail.
Fast forward to today: A new study has just been published titled, Short rest interval lengths between sets optimally enhance body composition and performance with 8 weeks of strength resistance training in older men that sheds further light on how the duration of rest intervals may affect muscular adaptations. If you just read the abstract, you might think the answer is clear.
Not so fast…
Here’s my take:
22 older men (mean 68 yrs) were recruited for participation. Subjects were healthy but were not involved in resistance training. All subjects underwent a 4-week â€śbreak inâ€ť phase where they performed a â€śhypertrophy-typeâ€ť total body routine consisting of 2-4 sets of 8-15 reps per set. The subjects were then tested for various measures including strength and body composition, and then pair-matched based on 1RM bench press to perform an 8-week strength-type program with short rest (1 minute) or long rest (4 minutes) between sets. The strength-type routine consisted of 2-3 sets of 4-6 reps carried out 3 days a week. The exercises included leg press, flat bench machine chest press, lat pulldown, seated row, dumbbell step-ups, dumbbell Romanian deadlifts, knee extension, and knee flexion. Reps were performed with the intent to move the loads as fast possible while maintaining proper form. All sessions were supervised by trained personnel.
Results were determined over the final 8-week strength phase of the program. Significantly greater increases in fat free mass (FFM), 1RM bench press, and 1RM leg press were noted for the short-rest group compared to those who took long rest periods. The researchers investigated a wide array of additional outcomes including power measures, which also generally favored the short rest group as well.
Based on these results, it would appear that limiting rest between sets is beneficial to enhancing strength and hypertrophy. The increase in FFM for the short-rest group was 1 kg vs just a 0.3 kg increase for the long-rest group. The effect size — a measure of the meaningfulness of the results — was 0.37 indicating a fairly small effect. That said, a difference of 0.7 kg (equating to ~1.5 pounds) could certainly be meaningful for those seeking to maximize hypertrophy — particularly over a fairly short period (8 weeks). The effect sizes for strength were fairly large (0.65 and 0.76 for the 1RM bench and squat, respectively). Combined, these findings indicate that muscular adaptations are enhanced by taking short rest periods between sets.
But…a closer scrutiny of the study’s methodology gives reason for caution when drawing conclusions.
First and foremost, the researchers used DXA to measure body composition. The authors reported results for FFM, which as stated were higher for the short rest group. However, FFM encompasses all tissues in the body other than fat mass. This includes bone, connective tissue, and importantly water. You can probably rule out any differences associated with bone and connective tissue, which almost certainly would be minimal over an 8 week resistance training in terms of contribution to body mass. However, variances in water weight could easily have accounted for a large portion of the the reported 0.7 kg difference in FFM. It’s curious why the researchers did not choose to quantify the subject’s segmental muscle mass. There are equations that can be employed with DXA to obtain these values, which would have given a better sense as to true increases in muscle. Unfortunately, the reported data do not allow for a true understanding of changes in the lean component of body composition between groups.
Second, the subjects did not train to failure in either condition. The researchers stated that this was done to reduce neuromuscular fatigue and thus ensure that the subjects could tolerate the program over its duration. While I have no problem with that reasoning, it does raise a major issue: Since those in the short rest interval group had to lift again after only 60 seconds, they would have been taxed to a greater extent on each successive set. The long-rest group on the other hand would have sufficient time to recover prior to the next set, and thus would not have been substantially taxed at point during the workout. Now it is a bit difficult to determine how much the subjects were actually challenged on each set based on the study write up. Ideally the researchers should have quantified the level of effort exerted (perhaps by RPE or similar scale) to provide context. Without this info, I’m left wondering if the design was biased to produce a greater effect with shorter rest periods.
Finally and importantly, the study was carried out on elderly, untrained subjects. These individuals would no doubt have been sarcopenic, and their response to an exercise stimulus therefore would not necessarily mimic that of young, fit individuals. Thus, generalizability of results is limited to the population studied.
In conclusion, this is an interesting study that adds to the body of literature. However, caution must be exercised when attempting to draw conclusions as to the effects of rest interval length on muscular adaptations. The limitations of the study preclude extrapolation of results to those seeking maximal muscle mass.
The good news is that I am currently collaborating on a study on the topic that directly measures hypertrophy in well-trained subjects. Target completion for data collection is early next year. I will update when results are available.
October 2, 2014
I continue to hear the same claim. It appears in magazine and internet articles. It’s heard in fitness forums and social media outlets. It’s sometimes perpetrated by high-profile fitness pros.
“Just do compound lower body movements and your hamstrings will get all the work they need to grow.”
The statement has been made so many times a lot of folks simply take it as gospel. Problem is, it’s a claim that has no basis.
Let’s talk facts.
First, the idea that hammies would be maximally stimulated in a compound movement doesn’t make sense from an anatomical standpoint. The hamstrings complex as a whole is biarticular, crossing both the knee and hip joints. (Nerd note: the exception is the short head of the biceps femoris, which just crosses the knee and technically is not considered a true hamstrings muscle). At the knee, the hamstrings acts as a flexor; at the hip it acts as an extensor. What happens when you squat down? Try it. The hamstrings shorten at the knee and lengthen at the hip. Alternatively, the opposite occurs on the concentric portion of the movement with the hamstrings lengthening at the knee and shortening at the hip. Thus, the overall functional length of the muscle complex doesn’t change all that much throughout the movement — a phenomenon that is not ideal for maximizing force output.
Research supports the fact that hamstrings activity is low during compound exercise. A recent study from my lab showed that the biceps femoris was only ~25% as active as the vasti muscles and just a third as active as the the rectus femoris during the leg press. Safe to say that the hammies don’t get much work in the leg press.
Think that perhaps squatting shows substantially greater hamstrings activation?
Hamstrings activation during the squat has been shown to be only 27% of maximal voluntary isometric contraction. This led the author of the study to conclude, “Thus the squat is not an optimal exercise for training the hamstrings.”
Okay, so maybe you don’t put much stock in muscle activation and want to put forth the argument that hypertrophy is all that matters. Fair enough. A study by Weiss et al provides direct evidence that squats don’t do much for hamstrings muscle growth. Subjects performed four sets of squats to approximately parallel depth using either a low, medium, or high rep range. Training was carried out 3 days a week for 7 weeks. At the end of the study, results showed significant increases in hypertrophy of the quads for all conditions studied. The hammies: no changes from baseline seen in any of the conditions.
Bottom line: If you want to maximize hamstrings development, you need to incorporate single-joint exercises that directly target the musculature into your routine. Both exercises originating at the hip joint (SLDL, good morning, etc) and exercises originating at the knee (leg curls) are viable choices. Ideally, both types of movements should be combined to optimize growth of the muscle complex.
August 30, 2014
It’s commonly taken as gospel that you need to warm-up prior to lifting. The warm-up contains two basic components: a general warm-up to raise core temperature, and a specific warm-up to heighten neural activation. The combination of these procedures is purported to enhance exercise performance. However, while research does seem to support this contention during maximal or near-maximal efforts, studies are lacking as to the effects of warming up during submaximal lifting routines.
To help determine the impact of warming up on a typical bodybuilding-style workout, I recently collaborated with colleagues in Brazil to carry out a controlled study on the topic. Here is an overview of the methodology and findings of the study, as well as its practical implications.
What We Did
Fifteen young men were recruited to participate in the study. Subjects were “recreationally trained” meaning they had limited lifting experience (resistance training for less than a year on average). Each subject performed 4 exercise sessions on separate days (48-72 hours between sessions) using the following different warm-up strategies prior to each workout: a general warm-up; a specific warm-up; a combination general and specific warm-up, or; no warm-up. In the aerobic warm-up subjects performed 10 minutes of light cycling exercise at a speed of 40 km/hr. For the specific warm-up, subjects performed a light set (10 reps at 50% 1RM) of the specific exercise before performance of that exercise. The order of the warm-ups was counterbalanced between subjects as shown in the accompanying figure to ensure that this variable did not unduly influence results. Exercise sessions consisted of 4 sets of the bench press, squat and arm curl at 80% 1RM. All sets were carried out to the point of muscular failure.
What We Found
There were no significant differences between the number of repetitions performed in any of the warm-up conditions nor was their a difference in the fatigue index, which is a formula that assess the decline in number of repetitions across the first and last sets of each exercise. In combination, these findings indicate that the warm-up procedures analyzed in this study had no effect on performance.
At face value it would appear that a warm-up is pretty much useless prior to submaximal resistance training. Despite the currently held belief that warming up enhances exercise performance, no benefits were seen between either a general warm-up, specific warm-up or combination of the two compared to no warm-up at all. Intuitively this seems to make sense in that the initial repetitions of a submaximal lifts are in effect their own specific warm-up and the need to increase core temperature might be superfluous from a performance standpoint when multiple reps are performed.
When applying these results to practice, however, several factors must be taken into account. First, the subjects in this study were recreationally trained; although they had some experience with resistance training they were in no way highly skilled lifters. It certainly is feasible that those with extensive lifting experience who have highly developed neuromuscular patterns might benefit from even slightly increased neural responses.
Second, you need to take into account the type of exercise performed. To this point, there did seem to be a slight advantage to performing a specific warm-up in the squat (although it did not rise to statistical significance) while there actually seemed to be somewhat of a detriment to the specific warm-up in the biceps curl. Thus, more complex movement patterns would seem to benefit from the “practice effect” of a specific warm-up while this would be of no value during performance of simple exercises.
Third, the absolute amount of weight lifted also must be considered. A good case can be made that a specific warm-up would have more utility for someone benching 400 pounds as opposed to 200 pounds. Even though the “heaviness” of the load would be similar on a relative basis, the neural benefits of doing a lighter set would seem to have greater transfer when lifting the heavier absolute load.
Finally, we did not investigate safety-related issues of warming up — only performance-aspects were assessed. Although no subjects in this study were injured during testing, the sample was too small and the duration of the protocol too short to draw conclusions on the topic. While resistance training with submaximal loads generally has very low risk of injury provided proper form is maintained, there nevertheless exists the possibility that warm-up procedures could reduce the risk even further. This seems especially pertinent when working with high absolute loads.
The take-home message is to consider your own situation when determining whether or not to warm up prior to a submaximal lifting session. Yes, a warm-up does take a bit of time and you might be able to skip the procedure if you are time-pressed without enduring any negative effects on performance. This is particularly valid if you are less experienced at training and/or lifting fairly light loads. On the other hand, if you are a highly experienced trainee lifting heavy absolute loads then there very well might be a benefit to warming-up — this study certainly cannot be used as evidence to the contrary. Also, understand that research only reports the means (i.e. averages) between groups. There were in fact differences between responses whereby some subjects did show a beneficial effects from warming up while others did not. Only through individual experimentation can you determine if a warm-up enhances your own performance. Finally, there are potentially safety issues that were not studied here; a warm-up certainly would not seem to hurt in this regard and possibly could be of some help.
Ribeiro AS, Romanzini M, Schoenfeld BJ, Souza MF, Avelar A, Cyrino ES. Effect of different warm-up procedures on the performance of resistance training exercises. Percept Mot Skills. 2014 Aug;119(1):133-45.
August 17, 2014
Wanted to update everyone on all that’s been happening; so much to share!
First, I’ve agreed to write a textbook on muscle hypertrophy, to be published by Human Kinetics — one of the leading publishers on the science of exercise and nutrition. The book will be geared towards fitness professionals and university programs. I’m totally stoked to provide an evidence-based resource on a subject that has long relied on gym lore and bro-science. Estimated pub date is April of 2016.
I’ve also agreed to write a monthly column for Flex Magazine. The column will discuss science-based application of hypertrophy and fat loss practices. It’s a real kick for me to be a regular columnist for a mag that I grew up reading. My first column is slated for the November issue.
Research-wise, I’m currently finishing up a study on body comp changes associated with fasted cardio and another on muscle activation during different loading intensities in the bench press. During the fall I have multiple studies set to get underway including a training frequency study investigating muscular adaptations in split vs. total body routines, another comparing functional transfer between the squat and leg press, and yet another that will evaluate the effects of protein timing pre- versus post-workout on muscle hypertrophy in well-trained subjects. I look forward to sharing the results of these and other studies currently in review when they become available.
Okay, that out of the way, here are some links that I thought you’d find informative. As always, I appreciate your continued support.
â€˘ I recently lectured at the CanFitPro Conference in Toronto. While there, I got a chance to record a few interview segments for Omar Isufâ€™s YouTube channel. In this segment we discuss repetition ranges for maximizing muscle hypertrophy. Give this a watch and youâ€™ll see why Omar lives up to his nickname, King of YouTube Fitness.
â€˘ I was interviewed along with my partner-in-science, Alan Aragon, on the We Do Science Podcast. Here Alan and I discuss the complexities of nutrient timing, delving into both the science and practical applications on the topic. Bonus discussion on a related topic: whether there is any fat loss benefit to doing fasted cardio. Click on Episode #8.
â€˘ Iâ€™ve appeared numerous times on Superhuman Radio; this segment might be my favorite yet. Here I discuss whether it’s possible to gain muscle simultaneously while simultaneously losing fat. Host Carl Lanore is consistently one of the best interviewers in the biz and he again shows why by asking all the right questions. l
â€˘ My friend and colleague Tom Venuto wrote an excellent post on delayed-onset muscle soreness and its relevance to muscular gains. The article covers the science in an understandable fashion, and provides solid take-home advice.
â€˘ Speaking of Tom Venuto, he wrote what I think is the most detailed review of my book, The M.A.X. Muscle Plan. Always an honor to receive praise from a true fitness pro such as Tom.
â€˘ In case you missed it, I recently published this study showing muscle activation during the leg press at 30% 1RM to failure produced significantly lower muscle activation compared to 75% 1RM. I also wrote an accompanying blog post where I break things down into consumer-friendly language and discuss the study’s implications.
â€˘ Finally, my good friend Bret Contreras wrote this terrific article that delved into the free-weights vs. machines debate. As mentioned earlier, I will be collaborating on a study examining this topic; Bret’s post provides excellent commentary on its complexities. While you’re at it, make sure to read through the references at the end of the article; it’s patently clear research doesn’t support the claims made by certain fitness pros.