The main aim of this study was to investigate whether a high protein intake (2.9 g.kg−1.d−1) leading into and across repeated days of intensive training improved markers of recovery in resistance-trained individuals when both total energy intake and peri-exercise protein timing were controlled for. Overall, recovery was comparable between dietary strategies when markers of muscle damage or soreness were considered. Similarly, performance repetitions were not found to be significantly different between dietary conditions, further indicating that PROMOD (1.8 g.kg−1.d−1) may be sufficient for resistance trained individuals. Our data are in line with a multitude of other studies finding no beneficial effects for strength trainees of consuming more than 1.8 g.kg−1.d−1 of protein [12–17, 23]. It is suggested, therefore, that the benefits observed in the previous study  were likely due to nutrient timing and not absolute protein intake.
Peri-exercise protein intake was an important parameter that was controlled for in this study. The concept of nutrient timing has previously been shown to have significant impact on muscular hypertrophy (cross-sectional area and lean mass gains) and maximal strength (upper and lower body) performance when a mixed protein drink was consumed pre-post exercise across a 10 week resistance training programme . However, there is still some debate as to whether protein/energy intake post-exercise only is a more important contributing factor to net positive fractional synthetic rates (FSR; [24–27]) with or without carbohydrate intake. Previous research has indicated that consumption of whey protein (25 g) post exercise significantly augments myofibrillar FSR up to 5 h into recovery , suggesting that the consumption of protein during the acute recovery period is central to net protein synthesis.
Mechanistically, the inclusion of essential amino acids (EAAs) appear critical to potentiating a greater net protein synthesis over the 24 h recovery window . In the current study, participants consumed 0.32 g.kg−1 of net protein before and after exercise (which amounted to an average intake of 25.1 ± 7.9 g net protein per serve) in line with dosages used in previous research . Of the EAAs, L-leucine has been proposed to have significant influence on protein synthesis [30, 31] following resistance training (which in the current study was the dominant amino acid provided per serve). Current evidence infers that acute essential amino acid feeding may likely inactivate the tublerosclerosis complex, particularly tuberin (TSC2) leading to activation of mTOR and PDK1 pathways. This has bearing on key regulatory proteins during the initiation phase of myofibrillar resynthesis including: eukaryotic initiation factor 2 (eIF2), 4E binding proteins and the protein kinase S6 K1 . Additionally, the concept of nutrient ‘sensing’ has been proposed in which other proteins (Vps34) may be key to stimulating mTOR/PDK1 synthesis pathways . Minimising nutrient deprivation pre-exercise, and acute refeeding post exercise may therefore be required for maximal recovery gains (particularly when training frequency is considered). A possible reason why a recent meta-analysis  on this subject did not find any beneficial effect of nutrient timing is that the majority of included studies were performed on untrained individuals. The anabolic ‘window’ for untrained individuals may be prolonged for >2 days following resistance exercise in contrast to strength-trained individuals . Additionally, in the majority of studies included in the meta-analyses, protein intakes (as well as protein timing) were not matched between the treatment and control groups.
It has been previously described that exercise intensity may alter protein requirements for athletes . The exercise protocol applied in this study presented a realistic scenario of how strength athletes, especially powerlifters, train. Our study used a whole-body workout on three consecutive days in contrast to previous research  using a lower body protocol, in which an intense leg workout with 3 exercises was performed on the first day and then only the squat exercise on the following testing days. Additionally, we did not limit the repetition number to only 10 repetitions for each set, but encouraged the subjects to continue until volitional exhaustion which permitted a more intensive protocol over the testing days.
The increased difficulty level and muscle damaging potential of our exercise protocol was reflected in the CK values, which were ~4-times as high as previously reported  and exceeded the physiological range at T2 and T3. Elevated CK values 24 h or more after intense exercise have also been observed in previous research [4, 6, 34, 35]. Analogue to previous investigations , perceived muscle soreness was not significantly different between dietary conditions, despite earlier recorded onset of muscle soreness for PROHIGH at T2 and T3. This finding was unsurprising considering CK values were not significantly different between conditions, indicating that any myofibrillar damage due to the exercise protocol may have been comparable between dietary strategies.
Surprisingly, however, the exercise protocol did not influence TNF-α values. Previous studies implementing heavy lower body exercise protocols with resistance-trained individuals observed an increase in TNF-α immediately after exercise [5, 6]. In contrast, one research study measuring TNF-α response after an eccentric arm exercise protocol failed to observe significant changes in TNF-α . The reason for this discrepancy may be that strenuous training of a smaller muscle group was not sufficient to elicit the same level of inflammatory response compared to larger muscle groups. Although our exercise protocol utilised a challenging whole body workout, it is also feasible that a significant elevation of TNF-α occurred >1–5 h after exercise, as reported elsewhere , or that the inclusion of a post-exercise protein formula may have blunted the TNF-α, but not the CK, response.
Overall, performance repetition scores across each testing day were not significantly different between dietary conditions. However, it is noteworthy that within condition lower body performance was maintained with PROHIGH. In comparison, within condition only, squat performance significantly declined by T3 with PROMOD (despite no differences in overall number of repetitions performed throughout the assessment period between conditions: 64.5 ± 21.2 for PROMOD v 63.1 ± 19.4 for PROHIGH; p = 0.477). Aligned with this, a significant interaction effect was found for bioelectrical impedance PhA, with values increasing at T3 for PROHIGH in contrast to PROMOD. As PhA has been reported to be a proxy measure of muscle ‘quality’ [38–40], myofibrillar structure may have been maintained with PROHIGH which could have bearing for longer term performance gains during intensive periods of training. The results may indicate that a PROHIGH approach during repeated days of intensive exercise could support training maintenance pertinent to lower body exercise.
Previous research has shown that participation in a prolonged resistance training program is associated with an increase in PhA . The mean PhA for athletes training for strength and power has also been reported to be higher than endurance athletes (8.4 ± 0.8 v 8.0 ± 1.0; ) indicating that PhA may depend on muscle fiber composition. To our knowledge, this is the first report of short-term changes in PhA as a result of repeated days of intensive resistance exercise coupled with modified protein intake. However, such findings should be interpreted with caution in light of the lack of significant differences between dietary groups for performance repetition scores and biomarkers of muscle damage. Additionally, such findings may only be applicable to strength-trained athletes, and may not necessarily apply to other sporting disciplines in which athletes train multiple times a day including sport specific and resistance training.
A further explanation for the lack of significant differences between dietary strategies for repetition performance may have been individual variability, which appeared to be particularly pronounced between men and women as reported elsewhere [42, 43]. Whereas some of the participants in this study were not able to perform more than 8 repetitions per set on the squat exercises, others were able to exceed this number by performing more than 15 repetitions at 80% of individual 1 RM. The muscle fiber composition of vastus lateralis is a genetic trait, which could explain 45% of the proportion of muscle composition, whereas ~40% can be explained by environmental factors e.g. a specific training protocol. For this reason, slow-twitch (Type I muscle fiber) content varies considerably (14–86%) between individuals . Individuals with a higher slow-twitch muscle fiber content in the quadriceps have the genetic predisposition to perform more repetitions on the squat exercise, which likely influences the protocol intensiveness, overall post exercise muscle damage and potential net protein synthesis following both a PROMOD and/or PROHIGH diet.
Whilst the findings of this study indicate that a short term PROMOD approach may be sufficient to support markers of recovery in resistance-trained individuals undergoing repeated days of intensive exercise, the potential benefit of lower protein intakes (<1.8 g.kg−1.d−1) cannot be excluded. However, as it was noted that within group, lower body repetition performance significantly declined with PROMOD by the end of the assessment (along with reported differences in phase angle between dietary conditions), a lower protein intake may have resulted in further performance decrements. Future research on short-term lower protein intakes may be warranted to confirm this.
It is acknowledged that the acute nature of the dietary interventions and short-term cross over period may be study limitations. As participants in this study were experienced resistance-trained individuals who typically consumed protein intakes ~2.1 g.kg−1.d−1, a standardised approach to calorific intake in the week prior to the assessment period should have sufficed to evaluate whether total protein load influenced recovery across repeated training days. Whilst a longer wash-out period may have been beneficial, post-hoc assessment of potential order/carry-over effects revealed no overall significant differences for main variables or bodyweight between test periods. The dietary lead-in period prior to each assessment phase was therefore deemed satisfactory.
Participants were tested under the same conditions across assessment days, with peri-exercise protein intake and timing controlled for. Prior to each laboratory visit, participants were requested to maintain similar dietary patterns ensuring they were acutely fasted before arrival (3-4 h). However, individual variance in postprandial nutrient availability may have influenced study findings. Assessment in a longer term post-absorptive or overnight fasted state may have presented clearer findings. However, not only did our participants effectively act as their own controls by maintaining eating patterns prior to testing, but intensive training in an overnight fasted state may not have been realistic for such individuals.
Whilst the study design purposefully aimed to assess both male and female resistance-trained athletes, another limitation to the study was sample size (n = 14), which could result in the possibility of type II errors when interpreting the findings. Given that our sample size exceeded the a priori power analysis requirement of 10 subjects and that there was no significant effect between dietary conditions on any of the outcome measures (except phase angle), it is unlikely that the sample size masked a large effect of protein intake. Future research should consider evaluation of specific gender differences and overall training experience which may likely be confounding variables when assessing the impact of protein intake on recovery.