Background
The protein requirements of active individuals are the subject of much debate. According to the International Society of Sports Nutrition, their Position Stand on Protein specifically states that “protein intakes of 1.4 – 2.0 g/kg/day for physically active individuals is not only safe, but may improve the training adaptations to exercise training.” [
1] Other investigators have suggested a 1.2 – 1.4 g/kg/day and 1.6 – 1.7 g/kg/day for endurance and strength-trained athletes, respectively [
2]. Thus, a ceiling of 2.0 g/kg/day is ostensibly the consensus in relation to the protein needs of athletic individuals. Data on high protein diets are somewhat misleading in that investigators operationally define ‘high’ incorrectly. The definitions of a high protein diet include: intakes greater than 15–16 % of total energy, as high as 35 % of total calories or intakes that merely exceed the RDA [
3]. One study examined the effects of progressive resistance training combined with a protein-enriched diet (i.e., via consuming more lean red meat) on various indices of body composition, physical performance and health in 100 women aged 60–90 years. The investigators discovered that a protein-enriched diet equivalent to approximately 1.3 g/kg/day achieved through lean red meat was a safe and effective way for enhancing the effects of weight training on lean body mass and muscle strength in elderly women [
4]. This aforementioned study would fit the published definition of a high protein diet. Another investigation ascertained the effects of an isoenergetic high-protein diet on strength and fatigue in ten young active women during high-intensity resistance exercise. They followed a control diet (55, 15 and 30 % of energy from carbohydrate, protein and fat, respectively) and a high-protein diet (respective values, 30, 40 and 30) for 7 days each in a random counterbalanced design. Subjects on the isoenergetic high-protein, moderate-fat diet maintained muscular strength and endurance during high-intensity resistance exercise without experiencing fatigue earlier compared with a control diet [
5]. That study suffers from two profound drawbacks. First, seven days hardly constitutes a ‘diet.’ And secondly, basing a diet on percentages can be quite misleading. Instead, one should operationally define what a ‘high’ protein diet is via the amount consumed daily per kilogram of body weight. Inasmuch as 2.0 g/kg/d seems to be the upper limit of what active individuals purportedly need, it is our contention that for a diet to truly be considered high in protein, daily consumption should necessarily exceed 2.0 g/kg/d. Previous work from our laboratory examined a true high protein diet (4.4 g/kg/d). In essence, consuming over five times the recommended daily allowance of protein had no effect on body composition in resistance-trained individuals who otherwise maintain the same training regimen. That investigation was the first interventional study to demonstrate that consuming a hypercaloric, high protein diet does not result in any significant body composition alterations [
6]. Thus, the purpose of the present investigation was to determine the effects of a high protein diet (>3 g/kg/d) combined with a traditional heavy resistance training program on indices of body composition, performance and health in resistance-trained individuals.
Discussion
This is the second investigation from our laboratory that has examined the effects of a true high protein diet (i.e., > 2 grams per kg body weight daily). Previously published work has shown that the mere addition of extra protein does not lead to substantive changes in body composition (i.e., no statistically significant change in FFM, fat mass or % body fat) in trained individuals who otherwise do not alter their exercise regimen [
6]. This is the first investigation in which a high protein diet in conjunction with a periodized heavy resistance training program was performed; moreover, subjects did not perform any aerobic exercise during the treatment period. In brief, our data suggests that consuming protein well above the recommended dietary allowance (RDA) can favorably alter body composition as long as changes are also made in one’s exercise training regimen. This is in contrast with our original pilot study in which subjects consumed five times the RDA for protein (~4.4 g per kg daily) for eight weeks. In that investigation, there were no statistically significant changes in body composition in the high protein diet group. Thus, this follow-up study was undertaken to ascertain if changing the resistance-training regimen in conjunction with a high protein diet could indeed affect the adaptive response.
It should be noted that other investigators have suggested that trained individuals may have a lower requirement for protein due to increased efficiency of use of protein. Accordingly, “several studies have shown that strength training, consistent with the anabolic stimulus for protein synthesis it provides, actually increases the efficiency of use of protein, which reduces dietary protein requirements” [
10]. If indeed regular heavy resistance training enhances efficiency, there would be no effect of added protein vis a vis body composition alterations. In other words, the consumption of protein in amounts far above the RDA should have little to no effect on body composition. Our investigation demonstrates that protein intakes that are approximately 60 % greater than even the highest recommended intakes (i.e., 2 grams per kg body weight daily) produce favorable alterations in body composition when combined with a periodized heavy resistance training regimen.
It has been shown that acute alterations in muscle protein synthesis (MPS) have little to no predictive value regarding chronic changes in lean body mass [
11]. Thus, studies that have examined acute changes in MPS are likely poor predictors of body composition in general. For instance, there was a fairly recent investigation of six healthy young men that performed an intense bout of leg-based resistance exercise (i.e., 4 sets of 8–10 repetitions to failure of the leg press, knee extension and leg curls) followed by the randomized consumption of drinks containing 0, 5, 10, 20, or 40 g whole egg protein. They found that 20 grams of egg protein maximally stimulated MPS and albumin protein synthesis after resistance exercise. Furthermore, “dietary protein consumed after exercise in excess of the rate at which it can be incorporated into tissue protein stimulates irreversible oxidation” [
12]. This study has been cited as the primary basis for a limitation in protein intake per meal. However, as previously mentioned, acute changes in MPS are a poor predictor of actual gains in fat free mass. Furthermore, this response was examined in egg protein; it is not known if a similar response would be found in whey, beef, casein, soy or any other whole protein source. Even if one were to assume the 20 gram per meal is sufficient to maximize the MPS response, that would translate into 60 grams of protein consumed daily (i.e., if one’s primary protein was consumed over three meals: breakfast, lunch, dinner). That amount is substantially less than what was consumed by the normal and high protein groups in the current study. We would posit that in order to establish causality regarding dietary protein consumption and changes in body composition, time course training studies are perhaps the best way to achieve that.
Hydration status is an important variable that has garnered increased attention because of its impact on body composition assessment. Work by Utter et al. has shown that several methods (e.g., air displacement plethysmography [Bod Pod], hydrostatic weighing and skinfolds) showed a significant decrease in FFM from the hydrated to the dehydrated state [
13]. We were unable to measure hydration status; however, we did have participants follow identical pre- and post-testing conditions. In this respect, it is possible that hydration status may have been a contributing factor to the variability we are reporting with our body composition data. Nevertheless, a prudent step for future studies is to determine hydration status prior to body composition assessment.
Our study discovered that consuming protein in amounts that are 3–4 times greater than the RDA result in a similar FFM increase for both the normal and high protein groups; however, the high protein group experienced a significantly greater loss of fat mass compared to the normal protein group in spite of the fact that they consumed on average ~400 kcals more per day over the treatment period. One could speculate the gains in FFM in both groups were the result of providing a different training stimulus than what each subject had previously used. Our prior study showed that the mere consumption of prodigiously high amounts of protein (>4 g per kg daily) had no effect on body composition if training was not altered even though there was a trend towards better body composition. In the current investigation, the high protein group demonstrated greater compliance vis a vis the training program than the normal protein group; perhaps this can explain in part why changes in fat mass were substantially greater in that group.
Alternative explanations for the decrease in fat mass in the high protein group include possible changes in resting and sleep energy expenditure. A recent study examined 25 participants who ate approximately 40 % excess energy for 56 days from 5 %, 15 %, or 25 % protein diets. If the extra calories consumed were from protein, both sleep and resting 24-h energy expenditure increased in relation to protein intake. However, this investigation found no relationship between changes in fat mass and changes in energy expenditure [
14]. It should be noted that there was not an exercise component to that study.
Another study examined 12 pairs of identical twins that overfed (~840 kcals extra daily) over 100 days [
15]. The average gain in fat mass and FFM were 5.4 kg and 2.7 kg. In comparison, our study showed an average loss of fat mass (0.3 kg normal group, 1.6 kg high protein group) and gain in FFM (1.5 kg for both the normal and high protein group). The authors of the identical twin overfeeding study noted that no single variable was a great predictor of body composition changes [
15]. The identical twin overfeeding study did not have an exercise component and did not use trained subjects. Thus, its relevance is questionable in athletic populations that purposefully engage in overfeeding.
In a similar study, participants were fed 140 % of energy needs, with 5, 15 or 25 % of energy from protein, for 56 days. They discovered that changes in vector magnitude (VM), a weight-independent measure of activity and activity-related energy expenditure (AEE) were positively correlated with weight gain; however, protein intake had no effect on changes in activity. Thus, “overfeeding produces an increase in physical activity and in energy expended in physical activity after adjusting for changes in body composition, suggesting that increased activity in response to weight gain might be one mechanism to support adaptive thermogenesis [
16]”. Again this study did not have an exercise training component.
It is possible that in the current investigation, changes in AEE and perhaps non-exercise activity thermogenesis (NEAT) might account in part for the greater changes in body composition in the high protein diet group [
17,
18]. According to Levine et al., NEAT can vary substantially between individuals by as much as 2000 kcals daily [
18]. Obese individuals tend to be seated more than lean individuals up to 2.5 h daily. That could represent an additional 350 kcals expended per day. Thus, it is possible that in the individuals in which NEAT increased the most during overfeeding, they were more likely to lose fat mass. And one could speculate that the more advanced training status of the high protein group might lend itself to greater NEAT.
Certainly, one should not discount the role of protein’s thermic effect (i.e., TEF or thermic effect of feeding). Protein has a TEF of approximately 19–23 % in both obese and lean individuals whereas carbohydrate is approximately 12–14 % [
19]. In fact, a high protein meal (45 % total kcal) elicits a 30 % greater TEF than an isocaloric low protein meal (15 % total kcal) in active females [
20]. It should be noted that the TEF of fat is substantially less in the obese than in lean subjects [
19]. The subjects in our study did not alter fat intake; thus, that could not be an explanation for changes in body composition. One could speculate that subjects in the high protein diet group experienced a combination of enhanced TEF, AEE, NEAT and SEE; this might explain in part the decrease in fat mass. Furthermore, the high protein group was more compliant with the exercise training regimen.
The training regimen used in the current study was clearly effective in producing an adaptive response. Both groups experienced a significant increase in muscular strength, power and endurance. It should be emphasized that it is quite difficult for trained subjects to gain FFM. Thus, the fact that on average both the normal and high protein groups gained 1.5 kg of FFM is an important point. The high protein diet group was more ‘trained’ (i.e., years of training experience) than the normal protein group. This may explain in part why the HP group did not gain more FFM than the NP group. Thus, any gain in FFM and strength may be viewed as a possible result of the extra protein consumed. However, the HP group was also more compliant with the training program and that would certainly be another causative factor in promoting FFM gains.
On average, both groups experienced a gain in FFM and a loss of fat mass; nonetheless, our data demonstrate that there is a bit of individual variability in the response. At the high end, there were subjects in the normal and high protein group that gained up to 7 kg of FFM and lost up to 4 kg of fat mass. Conversely, there were subjects who lost FFM or gained fat mass. In general, our data suggest that vast majority of individuals (~70 %) that consume a high protein diet (>2 g per kg daily) do indeed get an improvement in body composition. A study by Hubal et al. showed that among 585 subjects that underwent a unilateral resistance training program of the elbow flexors, several subjects showed no gain in muscle size whereas others experienced a profound increase [
21]. This shows indeed that there is a fairly substantive genetic component to the exercise training response; it would also be reasonable that such a component exists with the addition of a dietary treatment (ex., increased protein intake).
This study also found no harmful effects of consuming a high protein diet on renal function. Thus, professionals who work with athletes (i.e., sports nutritionists, sports dietitians, clinical nutritionists, medical doctors, strength coaches, athletic trainers, etc.) should be aware that athletes can consume very high amounts of protein with no harmful effects over a period of several weeks. Whether side effects will occur over longer protein overfeeding periods has not yet been investigated.
Conclusion
In conclusion, this is the first randomized controlled trial that has examined the effects of consuming a high protein diet in conjunction with a periodized heavy resistance training regimen over the course of several weeks on markers of performance, health and body composition. This study as well as previous work from our lab suggests that gains in body fat are unlikely to occur with protein overfeeding [
6]. Furthermore, gains in FFM can indeed occur in trained subjects with no harmful side effects of high protein consumption. This investigation confutes the notion that trained subjects need only 1.5–2.0 grams of protein per kg body weight daily and that intakes above that are superflous. In fact, we would speculate that the minimal daily needs of dietary protein in trained individuals should be approximately 2 g/kg/d.
Limitations of this study include the fact that the NP group exceeded their baseline protein intake. This was contrary to the instructions given to them. Subjects in this group were instructed to maintain their regular protein intake of ≤ 2 grams of protein per kg daily. Nonetheless, this group ended up significantly increasing their protein intake. However, there was an increase in LBM with no overall changes in fat mass. It should be noted that the HP group had significantly more training experience than the NP group. This would likely result in less of an ability to gain LBM over a finite training period.
Dietary self-reports also present problems vis-á-vis its accuracy. However, much of the additional protein consumed by subjects was in the form of protein powder; it would seem reasonable to assume that subjects could provide an accurate dietary recall with such a simple dietary addition. Furthermore, from an entirely pragmatic perspective, the use of dietary recalls (i.e., our investigation examined intake daily) is the most logical option. Unless investigators literally measure every food and beverage consumed by a subject, it is pragmatically impossible to get 100 % accuracy particularly in free-living subjects. Future work should examine very highly trained athletes who undergo cycles of varying protein intake over a period of several months or years. This would at least provide information in terms of whether the highly trained respond more so to a change in training stimulus, diet or a combination of both.
Competing interests
All authors have declared no competing interests or financial interests concerning the outcome of this investigation.
Authors’ contributions
The study was designed by JA; data were collected by SO, AE, AG, MS, CP, TS and JA; data interpretation was undertaken by JA, CP and TS. Manuscript preparation was performed by JA, TS and CP. All authors approved the final version of this paper.