Does Protein Supplementation Support Adaptations to Arduous Concurrent Exercise Training? A Systematic Review and Meta-Analysis with Military Based Applications

We evaluated the impact of protein supplementation on adaptations to arduous concurrent training in healthy adults with potential applications to individuals undergoing military training. Peer-reviewed papers published in English meeting the population, intervention, comparison and outcome criteria were included. Database searches were completed in PubMed, Web of science and SPORTDiscus. Study quality was evaluated using the COnsensus based standards for the selection of health status measurement instruments checklist. Of 11 studies included, nine focused on performance, six on body composition and four on muscle recovery. Cohen’s d effect sizes showed that protein supplementation improved performance outcomes in response to concurrent training (ES = 0.89, 95% CI = 0.08–1.70). When analysed separately, improvements in muscle strength (SMD = +4.92 kg, 95% CI = −2.70–12.54 kg) were found, but not in aerobic endurance. Gains in fat-free mass (SMD = +0.75 kg, 95% CI = 0.44–1.06 kg) and reductions in fat-mass (SMD = −0.99, 95% CI = −1.43–0.23 kg) were greater with protein supplementation. Most studies did not report protein turnover, nitrogen balance and/or total daily protein intake. Therefore, further research is warranted. However, our findings infer that protein supplementation may support lean-mass accretion and strength gains during arduous concurrent training in physical active populations, including military recruits.


Introduction
Concurrent training is defined as the combination of resistance and endurance training as part of a periodised physical training model [1]. The simultaneous development of strength, power and endurance is required by many athletic and exercising populations to meet the physical demands of their chosen sporting discipline (e.g., soccer, rugby, hockey) or exercise activity (e.g., circuits, cross-fit training) [2][3][4][5][6]. Similarly, recruits undergoing arduous military training routinely engage in concurrent training so as to meet the training and operational demands of military life [7][8][9][10][11][12][13]. Military recruit training programmes are designed to transform civilians into trained soldiers, therefore, physical training is necessarily arduous, involving a combination of aerobic training, strength and conditioning, obstacle courses, swimming, circuit training and loaded marching [14,15]. Despite the requirement of concurrent training in athletic and military recruit populations, and the positive effects protein supplementation may have on training outcomes, the majority of systematic reviews and meta-analyses have focused mainly on the effects of protein the aim of the findings of this review being applied to military recruits undertaking military training. Military recruits are typically aged between 16-35 years [47], and are required to meet aerobic fitness and muscular strength test standards such as maximal strength and muscular endurance tests [8]. Studies that did not meet all the PICO were excluded from this review.
Population: (a) stated as healthy active male or females; (b) aged between 16-35 years. Intervention: (c) include both endurance/aerobic training and resistance/weight training, circuit training, cross-fit training, military training but not high-intensity interval training (HIIT); (d) daily protein supplementation included but not with vitamins and/or antioxidants or through an increased intake of whole-food protein sources in the diet; (e) studies assessing body composition and/or performance were required to have ≥two sessions per week and be ≥four weeks in duration; (f) studies assessing muscle recovery were required to be <one week in duration; (g) training sessions performed at moderate or vigorous intensity (e.g., jogging, running, cycling, weight training) [17].
Comparison: (h) changes in outcome measures across repeated timepoints; (i) participants grouped by supplement condition.

Search Strategy and Study Selection
The final electronic database searches were completed in February 2021 in three databases (PubMed, Web of science and SPORTDiscus) using the terms "protein" or "protein supplementation", "training", and "concurrent training"" either alone or concurrently. The reference lists of all papers that met the inclusion criteria were interrogated to identify additional studies not found in the electronic search, until no further studies could be identified [48]. First, the study title and abstract were screened by one reviewer (SC) followed by the full text by the same reviewer. The characteristics that were extracted from each study included: author, participant sample, total protein intake (g·kg −1 ·day −1 ), training intervention, protein timing and dose. The study selection process is outlined in Figure 1.

Risk of Bias Assessment
Studies were evaluated for methodological quality according to the COnsensus based standards for the selection of health status measurement instruments (COSMIN) checklist by two separate reviewers (SC and HC) ( Table 1). This review used the recommended "worst score counts" method to obtain a total score for study quality. This was done by obtaining a quality score per measurement by taking the lowest rating of any item in a criteria box [49]. Each COSMIN item for all categories were scored from 4-1, where 4 was low risk and 1 was high risk. Each study needed a mean score of ≥ 3 to be included in this review [49].

Data Synthesis and Analyses
The analysis was conducted by first extracting the relevant information from all study groups at baseline and at the end of the intervention. This included the number of participants (n), p values, mean, standard deviation (SD) and 95% confidence intervals (if available). To compare the effects of protein supplementation against placebo conditions, pooling was used of the continuous data as standardized mean difference (SMD) represented as Cohen's d effect sizes (ES), standard error (SE) and 95% confidence intervals calculated for each main outcome using the reported mean change differences (delta scores), n and corresponding SDs [17,19]. If the mean change difference was not reported this was calculated based on the reported pre-and-post mean and SDs in each study. If a study used multiple protein-supplemented groups, we combined the data into an overall protein-supplemented group for subsequent analyses [17]. When a study used multiple performance outcome measures, the relative . VO 2peak and lower body 1RM were Nutrients 2021, 13, 1416 4 of 17 prioritised for muscle strength and aerobic endurance [17]. Effect sizes were classed as small (0.2), medium (0.5), and large effects (0.8) [50]. Effect size was calculated using the following equations: E1: Cohen's d = (M2 − M1)/SD pooled E2: SD pooled = √ ((SD 1 2 + SD 2 2 )/2) Nutrients 2021, 13, x FOR PEER REVIEW 4 of 18 Figure 1. Flow chart of study retrieval process.

Risk of Bias Assessment
Studies were evaluated for methodological quality according to the COnsensus based standards for the selection of health status measurement instruments (COSMIN) checklist by two separate reviewers (SC and HC) ( Table 1). This review used the recommended "worst score counts" method to obtain a total score for study quality. This was done by obtaining a quality score per measurement by taking the lowest rating of any item in a criteria box [49]. Each COSMIN item for all categories were scored from 4-1, where 4 was low risk and 1 was high risk. Each study needed a mean score of ≥ 3 to be included in this review [49].

Data Synthesis and Analyses
The analysis was conducted by first extracting the relevant information from all study groups at baseline and at the end of the intervention. This included the number of A random-effects model was applied with heterogeneity across studies tested using I 2 test. I 2 values of 25%, 50%, and 75% were considered low, moderate and high, respectively [17]. Each study was weighted (%) based on its inverse within study variance and between study variance using the Meta-Essentials spreadsheet 1.4 (Microsoft Excel 2016, Washington, DC, USA). Meta-Essentials was used for the meta-analysis, creation of forest and Egger's funnel plots (including the trim and fill method) and running statistical analysis, with alpha set at p ≤ 0.05.  Table 1 outlines the quality assessment scores for each study. All 11 studies were considered eligible for this review based on their COSMIN quality assessments scores. Egger's regression analysis found asymmetries in the funnel plot ( Figure 2) suggesting that results might be influenced by biasing factors such as publication bias. One study was particularly responsible for this asymmetry [57] as they showed the strongest beneficial effects of the treatment group when compared with the placebo group. When this outlier was removed the funnel plot was symmetric (p > 0.05). The funnel plot with this outlier included can be seen in Supplementary Figure S1.

Participant Characteristics and Study Interventions
Details of the studies' characteristics are provided in Tables 2 and 3. The sample sizes ranged from 10 to 387 with a total participant sample size of 681 (645 men, 61 women) for all studies. The reported mean age of participants ranged between 18 and 31 years. Intervention durations ranged from one day to six months with seven studies using a standardised concurrent endurance and resistance training programme [53][54][55][56][57][58], two studies using a military training programme [51,58] and two studies using an acute loaded march protocol [59][60][61].

Participant Characteristics and Study Interventions
Details of the studies' characteristics are provided in Tables 2 and 3. The sample sizes ranged from 10 to 387 with a total participant sample size of 681 (645 men, 61 women) for all studies. The reported mean age of participants ranged between 18 and 31 years. Intervention durations ranged from one day to six months with seven studies using a standardised concurrent endurance and resistance training programme [53][54][55][56][57][58], two studies using a military training programme [51,58] and two studies using an acute loaded march protocol [59][60][61]. Data reported as mean ± standard deviation where possible. U.S. = United states, g·kg −1 ·day −1 = grams per kilogram of body mass per day, PLA = placebo, PRO = protein, WP = whey protein, CHO = carbohydrate, CON = control, AM = morning. PLA = 0 g CHO, 0 g PRO, 0 g fat; CON = 0 g PRO, 8 g CHO and 3 g fat; PRO = 10 g PRO, 8 g CHO and 3 g fat.
Participants who weighed <81.8 kg received one portion and those weighing >81.8 kg received two portions post-exercise.

Synthesis of Results
This review identified 11 individual studies; one focused only on performance [55], six on performance and body composition [51,53,[55][56][57][58] and four on muscle recovery adaptations only [52,[59][60][61]. Four studies found a benefit on muscle strength, five studies reported a benefit on body composition changes such as increasing FFM, reducing fat-mass or both, and one study reported a benefit on muscle recovery. The characteristics of these studies are outlined in Tables 2 and 3.

Performance Adaptations
McAdam et al. [51] observed a greater increase in muscle strength (push-up repetition performance) with protein supplementation (+6.8, 95%CI: 2.9-10.7) compared to a placebo (+2.6, −0.7-6.0 95% CI) during a two-minute maximal push-up test. There was no effect on run time performance (protein: −48.3 s, −63.0-33.6 s 95%; placebo: −74.2 s, −95.5-51.9 95% CI) during a two-minute maximal time trial. Similarly, Ormsbee et al. [55] found a greater increase in 1 RM bench press after six-months of concurrent training with protein supplementation (+27.4 ± 2.4 kg vs. +15.9 ± 2.8 kg, p = 0.003) but not 1 RM hip sled performance (protein: 72.3 ± 7.8 kg; placebo: 73.6 ± 9.0 kg). Both groups had a statistically significant increase in . VO 2peak at six months compared to baseline but no differences between groups were reported. Taylor et al. [56] reported a statistically significant change in 1 RM bench press performance with additional protein in female basketball players over an eight-week period (protein = +4.9 ± 2.1 kg vs. placebo = +2.3 ± 1.4 kg, p = 0.046). Walker et al. [58] reported a statistically significant increase in 1 RM bench press performance (protein: +3.5 ± 5.2 kg; placebo: +1.3 ± 4.4 kg, p < 0.05) and the number of push-ups (protein: +5.4 ± 6.8; placebo: +3.2 ± 6.8, p < 0.05) performed with protein supplementation over eight weeks of recruit military training. There was, however, no effect of protein on run time performance during a maximal three-mile time-trial (protein: −1.4 ± 0.4 s; placebo: −0.9 ± 3.3 s, p > 0.05). The remaining study also observed a greater increase in rowing time to exhaustion with leucine supplementation compared to a placebo (p = 0.008) [53]. The remaining three studies reported no statistical effects on exercise performance with protein supplementation compared to a placebo or control condition [54,57]. For instance, Longland et al. [57] reported no impact of protein supplementation on leg and bench press 1 RM or cycling time trial performance. Similarly, no differences in men or women were reported between groups for changes in . VO 2peak , 2000 m rowing time trial, leg and bench press 1 RM performance [54]. It was possible to include five and three studies in the meta-analyses for muscle strength [54][55][56][57][58] and aerobic endurance adaptations [54,55,57], respectively. One study was not included in the muscle strength analysis [51] whilst two studies were not included in the aerobic endurance adaptations [51,53] due to no SD being reported. Additionally, another study was removed from the . VO 2peak meta-analysis due to assessing time trial performance [58]. The results of the meta-analysis are reported as SMD and showed that protein supplementation improved performance outcomes when muscle strength and aerobic endurance parameters were analysed together (SMD = 0.89, 95% CI = 0.08-1.70). When performance outcomes were analysed independently, protein supplementation was found to enhance muscle strength adaptations during concurrent training compared to placebo (ES = 1.18, SMD = +4.92 kg, 95% CI = −2.70-12.54 kg) ( Figure 3). However, the meta-analysis found protein supplementation to not enhance aerobic endurance adaptations ( . VO 2peak ) with the analysis favoring placebo (ES = 0.79, SMD = −0.37 ml·kg −1 ·min −1 , 95% CI = −1.45-0.71) (Figure 4). The individual study effect sizes for muscle strength and aerobic endurance adaptations can be found in Supplementary  Figures S2 and S3. There was substantial heterogeneity between studies for muscle strength (I 2 = 94%) and aerobic endurance adaptations (I 2 = 95%).

Body Composition Adaptations
Fat-free mass (FFM) was shown to increase to a greater extent with protein supplementation over an eight-week training period in military recruits (protein: +0.7 ± 1.2 kg; placebo 0.0 ± 0.9 kg, p < 0.05) [58]. Similarly, FFM was also shown to increase to a greater extent in female basketball players over an eight-week period (protein: +1.4 kg; placebo: +0.4 kg, p = 0.025) [56]. McAdam et al. [51] also observed greater reductions in fat-mass over an eight-week period in military recruits with protein supplementation compared to a placebo (protein: −4.5 kg; placebo: −2.7 kg, p = 0.04) after controlling for initial fat-mass [51]. A trend for greater reductions in fat-mass (protein: −1.0 ± 0.3 kg; placebo: −0.3 ± 0.4 kg, p > 0.05) and gains in FFM (protein: +2.4 ± 0.3 kg; placebo: +1.9 ± 0.3 kg, p > 0.05) were reported by Ormsbee et al. [55]. However, significant differences were observed only at three months into the six-month concurrent training intervention for gains in FFM (protein:+2.6 ± 0.2 kg; placebo: 1.7 ± 0.3 kg, p = 0.02) in sedentary men and women. Longland et al. [57] reported greater reductions in fat-mass (protein: −4.8 ± 1.6 kg; placebo: −3.5 ± 1.4 kg, p < 0.05) and gains in FFM (+1.2 ± 1.0 vs. +0.1 ± 1.0 kg, p < 0.05) with protein supplementation compared to a placebo over a four-week period. Conversely, no effect of leucine supplementation was reported after six-weeks on fat mass (body fat percentage) changes [53]. In total, five out of six studies reported a beneficial impact of protein supplementation on body composition adaptations. Four studies were included in the meta-analysis for changes in FFM [51,55,[57][58][59] whereas three studies were included in the meta-analysis for fat-mass [51,57,58]. One study was excluded from the FFM adaptations analysis due to no SD being reported [56] and one study was excluded from the fat-mass adaptations analysis due to body fat percentage being reported [53]. The meta-analysis found that protein supplementation enhanced gains in FFM (ES = 6.29, SMD = +0.75 kg, 95% CI = 0.44-1.06 kg) ( Figure 5). The meta-analysis also found protein supplementation to enhance reductions in fat-mass compared to placebo (ES = −0.99, SMD = 0.60 kg, 95% CI = −1.20-0.45 kg) ( Figure 6). The individual study effect sizes for FFM and fat-mass adaptations can be found in supplementary Figures 4 and 5, respectively. There was considerable heterogeneity between studies for FFM (I 2 = 98%) and fat-mass adaptations (I 2 = 91%).

Body Composition Adaptations
Fat-free mass (FFM) was shown to increase to a greater extent with protein supplementation over an eight-week training period in military recruits (protein: +0.7 ± 1.2 kg; placebo 0.0 ± 0.9 kg, p < 0.05) [58]. Similarly, FFM was also shown to increase to a greater extent in female basketball players over an eight-week period (protein: +1.4 kg; placebo: +0.4 kg, p = 0.025) [56]. McAdam et al. [51] also observed greater reductions in fat-mass over an eight-week period in military recruits with protein supplementation compared to a placebo (protein: −4.5 kg; placebo: −2.7 kg, p = 0.04) after controlling for initial fat-mass [51]. A trend for greater reductions in fat-mass (protein: −1.0 ± 0.3 kg; placebo: −0.3 ± 0.4 kg, p > 0.05) and gains in FFM (protein: +2.4 ± 0.3 kg; placebo: +1.9 ± 0.3 kg, p > 0.05) were reported by Ormsbee et al. [55]. However, significant differences were observed only at three months into the six-month concurrent training intervention for gains in FFM (protein:+2.6 ± 0.2 kg; placebo: 1.7 ± 0.3 kg, p = 0.02) in sedentary men and women. Longland et al. [57] reported greater reductions in fat-mass (protein: −4.8 ± 1.6 kg; placebo: −3.5 ± 1.4 kg, p < 0.05) and gains in FFM (+1.2 ± 1.0 vs. +0.1 ± 1.0 kg, p < 0.05) with protein supplementation compared to a placebo over a four-week period. Conversely, no effect of leucine supplementation was reported after six-weeks on fat mass (body fat percentage) changes [53]. In total, five out of six studies reported a beneficial impact of protein supplementation on body composition adaptations. Four studies were included in the meta-analysis for changes in FFM [51,55,[57][58][59] whereas three studies were included in the meta-analysis for fat-mass [51,57,58]. One study was excluded from the FFM adaptations analysis due to no SD being reported [56] and one study was excluded from the fat-mass adaptations analysis due to body fat percentage being reported [53]. The meta-analysis found that protein supplementation enhanced gains in FFM (ES = 6.29, SMD = +0.75 kg, 95% CI = 0.44-1.06 kg) ( Figure 5). The meta-analysis also found protein supplementation to enhance reductions in fat-mass compared to placebo (ES = −0.99, SMD = 0.60 kg, 95% CI = −1.20-0.45 kg) ( Figure 6). The individual study effect sizes for FFM and fat-mass adaptations can be found in Supplementary Figures S4 and S5, respectively. There was considerable heterogeneity between studies for FFM (I 2 = 98%) and fat-mass adaptations (I 2 = 91%).

Muscle Recovery Adaptations
Perceived muscle soreness after a six-mile hike was reduced by 7% with post-exercise protein supplementation compared to increases of 10% and 16% in the placebo and control conditions, respectively (p < 0.05) [61]. The remaining studies found no significant difference between protein and placebo conditions for the recovery of muscle function [60], muscle damage [59] or both [52]. Blacker et al. [60] reported no effect of protein compared to CHO on muscle function recovery. At 48 hours post-exercise, knee extensor isometric force was reduced by 10 ± 10% for the low caloric placebo condition (p = 0.008) but had returned to baseline in the CHO (p = 0.199) and protein condition (p = 0.099). At 72 hours post-exercise, participants in the placebo condition returned to baseline (p = 0.145), whereas both the CHO (p = 0.457) and protein conditions (p = 0.731) remained at baseline at 48 hours post-exercise. Only one study assessed the impact of protein supplementation on markers of exercise induced muscle damage and inflammation [59]. It was found that there were no differences between protein and isocaloric placebo conditions for changes in blood concentrations of cortisol (placebo: −0.79 ± 0.89; protein: 1.39 ± 1.08 µg·dL⁻ 1 , p = 0.160), C-reactive protein (placebo: 0.13 ± 0.77; protein: 0.99 ± 0.16 mg·L⁻ 1 , p = 0.305), creatine kinase (placebo: 278.65 ± 50.23; protein: 422.18 ± 149.87 U·L⁻ 1 , p = 0.722) or aldolase (placebo: 2.06 ± 0.46; protein: 1.98 ± 0.91 U·L⁻ 1 , p = 0.704). Based on the limited number of studies and available data, it was not possible to complete a metaanalysis of studies assessing the effect of protein supplementation on muscle recovery adaptations. These limitations include the SD not being reported [61] and different outcome measures, such as muscle damage [59], muscle function [52,60] and muscle soreness [61].

Muscle Recovery Adaptations
Perceived muscle soreness after a six-mile hike was reduced by 7% with post-exercise protein supplementation compared to increases of 10% and 16% in the placebo and control conditions, respectively (p < 0.05) [61]. The remaining studies found no significant difference between protein and placebo conditions for the recovery of muscle function [60], muscle damage [59] or both [52]. Blacker et al. [60] reported no effect of protein compared to CHO on muscle function recovery. At 48 hours post-exercise, knee extensor isometric force was reduced by 10 ± 10% for the low caloric placebo condition (p = 0.008) but had returned to baseline in the CHO (p = 0.199) and protein condition (p = 0.099). At 72 hours post-exercise, participants in the placebo condition returned to baseline (p = 0.145), whereas both the CHO (p = 0.457) and protein conditions (p = 0.731) remained at baseline at 48 hours post-exercise. Only one study assessed the impact of protein supplementation on markers of exercise induced muscle damage and inflammation [59]. It was found that there were no differences between protein and isocaloric placebo conditions for changes in blood concentrations of cortisol (placebo: −0.79 ± 0.89; protein: 1.39 ± 1.08 µg·dL⁻ 1 , p = 0.160), C-reactive protein (placebo: 0.13 ± 0.77; protein: 0.99 ± 0.16 mg·L⁻ 1 , p = 0.305), creatine kinase (placebo: 278.65 ± 50.23; protein: 422.18 ± 149.87 U·L⁻ 1 , p = 0.722) or aldolase (placebo: 2.06 ± 0.46; protein: 1.98 ± 0.91 U·L⁻ 1 , p = 0.704). Based on the limited number of studies and available data, it was not possible to complete a metaanalysis of studies assessing the effect of protein supplementation on muscle recovery adaptations. These limitations include the SD not being reported [61] and different outcome measures, such as muscle damage [59], muscle function [52,60] and muscle soreness [61].

Muscle Recovery Adaptations
Perceived muscle soreness after a six-mile hike was reduced by 7% with post-exercise protein supplementation compared to increases of 10% and 16% in the placebo and control conditions, respectively (p < 0.05) [61]. The remaining studies found no significant difference between protein and placebo conditions for the recovery of muscle function [60], muscle damage [59] or both [52]. Blacker et al. [60] reported no effect of protein compared to CHO on muscle function recovery. At 48 hours post-exercise, knee extensor isometric force was reduced by 10 ± 10% for the low caloric placebo condition (p = 0.008) but had returned to baseline in the CHO (p = 0.199) and protein condition (p = 0.099). At 72 h post-exercise, participants in the placebo condition returned to baseline (p = 0.145), whereas both the CHO (p = 0.457) and protein conditions (p = 0.731) remained at baseline at 48 h post-exercise. Only one study assessed the impact of protein supplementation on markers of exercise induced muscle damage and inflammation [59]. It was found that there were no differences between protein and isocaloric placebo conditions for changes in blood concentrations of cortisol (placebo: −0.79 ± 0.89; protein: 1.39 ± 1.08 µg·dL −1 , p = 0.160), C-reactive protein (placebo: 0.13 ± 0.77; protein: 0.99 ± 0.16 mg·L −1 , p = 0.305), creatine kinase (placebo: 278.65 ± 50.23; protein: 422.18 ± 149.87 U·L −1 , p = 0.722) or aldolase (placebo: 2.06 ± 0.46; protein: 1.98 ± 0.91 U·L −1 , p = 0.704). Based on the limited number of studies and available data, it was not possible to complete a meta-analysis of studies assessing the effect of protein supplementation on muscle recovery adaptations. These limitations include the SD not being reported [61] and different outcome measures, such as muscle damage [59], muscle function [52,60] and muscle soreness [61].

Discussion
This review identified 11 studies which investigated the effects of protein supplementation on exercise performance, body composition and muscle recovery adaptations to concurrent exercise training compared to a placebo in healthy adults, confirming the need for more work in this area. The key findings from the literature that met our inclusion criteria demonstrated that protein supplementation had a large effect on muscle strength and FFM adaptations to concurrent exercise training. There was limited evidence to suggest that protein supplementation can support aerobic endurance and muscle recovery adaptations.

Muscle Strength and Body Composition Adaptations
Longland et al. [57] reported no impact of protein supplementation on muscle strength adaptations despite a greater increase in FFM compared to a placebo condition. This was the only study to purposely induce a negative energy balance while participants consumed a total protein intake of 2.4 g·kg −1 ·day −1 . The study duration (four weeks) may have been too short for differences in strength development to be detected, particularly as protein supplementation is suggested to promote gains in FFM and muscle strength as the duration of training increases [62]. Forbes and Bell [54] also reported no effect of an additional 2.0-2.4 g·kg −1 ·day −1 of protein on muscle strength and body composition adaptations over a six-week period ( Table 2). It may be that the findings were also confounded by the intervention duration, given that it was shorter than each study reporting a positive effect of protein supplementation. The results were analysed by sex and the small sample size (15 women and 16 men) may have also limited the statistical findings as acknowledged by the study authors. Furthermore, the participants in the control condition consumed 1.2-1.4 g·kg −1 ·day −1 of protein, which may have been adequate to meet the demands of training, however, with no measure of nitrogen balance or protein turnover, this cannot be confirmed. The final study which observed no effect of protein supplementation on body composition adaptations may have had low total daily protein intakes (0.85 g·kg −1 ·day −1 ) [53].
The studies that reported a positive effect of protein supplementation on muscle strength and body composition adaptations provided protein to participants immediately post-exercise [51,[55][56][57][58]. This likely maximised myofibrillar protein synthesis in response to concurrent training [33,35,38,39] and modulated muscle strength and FFM adaptations [63,64]. Promoting MPS post-exercise is an important factor at enhancing skeletal muscle remodelling and adaptation [39,42,64,65]. Subsequently, this could have attenuated the interference effect of endurance training on strength adaptations [35,66] by promoting the activation of mTOR and inhibiting the activation of the tuberous sclerosis complex [27,28]. Skeletal muscle is sensitive to protein feeding for 24 hours postexercise and thus, consuming protein in 20-40 g doses evenly throughout the day is recommended [64,67,68]. More recently, it has been shown that consuming protein prior to sleep also augments MPS throughout the night [69]. Consuming protein prior to sleep, and subsequently increasing total daily protein intake may be advantageous at optimising MPS responses and supporting muscle strength and body composition adaptations when undertaking concurrent training [70].Nevertheless, it should be acknowledged that acute changes in MPS does not necessarily predict changes in muscle strength and FFM [71]. Instead it is likely the chronic and repetitive changes in MPS and muscle protein breakdown (MPB) which contribute to these [65]. Two studies that reported a greater increase in muscle strength reported a larger reduction in fat-mass with protein intakes ≥2.2 g·kg −1 ·day −1 compared to a placebo [51,55]. This suggests that individuals undergoing arduous concurrent training may benefit from protein intakes higher than the current recommendation of 1.8-2.2 g·kg −1 ·day −1 [72]. Mechanistically, it is speculative as to how an elevated protein intake promoted a greater loss in fat-mass, but previous work suggests that the greater thermic effect of protein may play a key role [73]. However, despite similar daily energy intakes between groups in both studies, neither included a measure of energy expenditure, and therefore, it is unclear if participants were in energy balance. Consuming a protein intake >2.2 g·kg −1 ·day −1 and possibly higher than 3.0 g·kg −1 ·day −1 while restricting energy intake has been suggested to maximise the loss of fat-mass and promote the maintenance of FFM [72]. It is unclear if the greater reduction of fat-mass promoted greater improvements in muscle strength performance in studies included in this review [51,55]. More work is needed to better determine the impact of protein intakes higher than the current recommendations (1.7-2.2 g·kg −1 ·day −1 ) on body composition adaptations, and how this may influence exercise performance in individuals undergoing arduous concurrent training. The studies identified in this review suggest that protein supplementation may be an effective strategy at augmenting muscle strength and body composition adaptations in healthy adults undertaking concurrent training. It is likely that this effect is facilitated by maximising the MPS post-exercise and attenuating the potential interference effect of endurance training on muscle strength and FFM adaptations. However, more work which includes measures of nitrogen balance or protein turnover are needed to confirm this. Furthermore, future work should also consider factors such as the timing of protein intake around exercise, energy intake/expenditure, and the duration of the training intervention.

Aerobic Adaptations
Protein requirements are elevated in endurance athletes to 1.6-1.8 g·kg −1 ·day −1 [74] but may be higher (1.7-2.2 g·kg −1 ·day −1 ) during periods of intense and/or high volume training [72]. Protein feeding has been shown to facilitate recovery and performance adaptations to endurance training [17,64]. Nonetheless, the effects of protein supplementation on aerobic performance adaptations during an arduous concurrent training programme are unknown. Similar improvements in run time performance was observed in military recruits with total daily protein intake of 2.8 ± 0.5 and 1.6 ± 0.4 g·kg −1 ·day −1 in the protein and placebo conditions, respectively [51]. As such, the dietary protein requirements to facilitate endurance-based adaptations were likely met in both groups, therefore, between groups differences were not observed. This suggests that to promote endurance-based performance adaptations, additional protein intake is not warranted when total habitual intake is ≥1.6 g·kg −1 ·day −1 . Walker et al. [58] also observed no between group differences in run time performance in military recruits supplemented with whey protein or CHO for eight-weeks. However, the total daily protein intakes were not reported, and it is unknown if protein requirements were met. In contrast, one study found leucine supplementation (45 mg·kg −1 ·day −1 ) for six-weeks improved exercise time to exhaustion in canoeists [53]. The reported total daily protein intake was 0.85 ± 0.06 g·kg −1 ·day −1 and 0.85 ± 0.05 g·kg −1 ·day −1 in the protein and placebo groups, therefore, participants likely benefited from the elevated leucine intake given that this amount of protein per day is much lower than the recommended amount [72,74]. Leucine is the key amino acid which stimulates MPS through the mTOR pathway [39,75], but given that the other essential amino acids are required to support this process [36,39], it is unclear how leucine supplementation improved exercise time to exhaustion. The results of the meta-analysis suggest that there is limited evidence to support the use of protein supplementation for aerobic endurance adaptations in response to concurrent training compared to placebo. However, based on the limited number of studies identified, more work is needed to confirm this.

Muscle Recovery Adaptations
Protein supplementation has been shown to improve muscle function recovery following resistance training [19] and other modes of exercise including cycling, running, eccentric exercise and resistance training [62], yet no review has evaluated the effects following arduous concurrent exercise training specifically. Protein consumption in close proximity to exercise prevents a decrease in myogenin messenger RNA expression, which can accelerate the remodelling and recovery of skeletal muscle [76]. Specifically, leucine may be a key component at initiating this process post-exercise through the activation of mTOR and MPS [39,75]. One study found post-exercise protein ingestion reduced muscle soreness in U.S. Marines following a loaded march [61]. However, Flakoll et al. [61] failed to provide the total daily protein intake, which therefore, limits our understanding of the impact of protein supplementation specifically on recovery adaptations [77]. The remaining studies all failed to find an effect of protein supplementation on muscle recovery in the days following arduous concurrent exercise [52,59,60]. Specifically, no impact was observed on markers of muscle damage [59], soreness [52] or function [52,60]. Jimenez-Flores et al. [59] observed no differences in markers of muscle inflammation or damage between protein and placebo conditions. However, some of the markers which were chosen may be questionable. For example, cortisol is a stress hormone [78,79] which can indicate changes in whole-body catabolism [79]. C-reactive protein is a marker of whole-body inflammation and is not necessarily specific to skeletal muscle [78]. The data also suggest a large inter-participant variability, which is a known limitation of such markers, particularly creatine kinase [79].
Eccentric exercise initiates a chain of events which leads to myofibrillar damage, degradation of structural proteins and membrane damage, thus inhibiting muscle function especially if individuals are unaccustomed to the exercise bout [80]. The participants in the study by Eddens et al. [52] completed a bout of concurrent endurance and eccentric exercise. The participants consumed a similar total daily protein intake, which corresponds to current recommendations [64]. As such, it is possible that the additional protein consumed post-exercise in the experimental group did not accelerate muscle recovery, due to protein requirements already being met by the participants. It was acknowledged by Eddens et al. [52] that the decrement in muscle function over the 24 h post-exercise was~15%, which is lower than that observed with other eccentric exercise protocols, with decrements of between 10-65% reported elsewhere [81]. Therefore, it cannot be excluded that the muscle damaging protocol may not have been arduous enough, which might explain the lack of statistical difference between conditions [52]. Blacker et al. [60] found no statistically significant difference between protein and placebo conditions on acute muscle function recovery following arduous concurrent exercise. However, both supplement conditions accelerated recovery of muscle function compared to the control condition. Similarly, participants consumed a standardised total daily amount of protein (0.9 ± 0.3 g·kg −1 ·day −1 ) across conditions. Although the amount of protein is lower than the current general recommendations (1.2-2.0 g·kg −1 ·day −1 ) [64], no effect of protein supplementation post-exercise was observed by Blacker et al. [60] when compared to an isocaloric placebo [60]. Jimenez-Flores et al. [59] found no impact of protein supplementation compared to an isocaloric placebo on markers of muscle damage following arduous concurrent exercise, however, high-variability between study participants was observed and likely influenced the ability to detect statistical differences [59]. However, unlike Blacker et al. [60], there was no control group, therefore, it is unknown if the additional energy intake accelerated muscle recovery [59]. The limited number of studies and the differences between methodologies and outcome measures make it difficult to determine whether protein supplementation does improve muscle recovery, thus warranting further research. Additionally, due to the lack of women included in studies to date, future research is needed in women to examine the impact of protein on muscle recovery, particularly given the known difference in the rate of muscle function recovery post-exercise between men and women [82].

Limitations
Only one study identified in this review included a measure of protein balance or turnover [57]. Therefore, it is unknown if participants were in a positive protein balance before, during or after the interventions in the remaining studies. Including a measure of protein requirements in future studies can allow for a better understanding of how much additional protein is potentially needed during arduous concurrent training by estimating changes in MPS and whole-body protein balance. The majority of studies failed to include markers of skeletal muscle damage or inflammation when focusing on muscle recovery and therefore, the mechanisms of the effects observed are speculative. The control of dietary intake is critical for comparison between studies involving nutritional interventions. Four studies failed to report the total daily protein intake during the intervention [56,58,59,61], making the comparisons between studies even more challenging, given that this is considered more important than the timing of protein intake [71,83]. It is recommended that these methodological considerations be factored into future studies aimed at investigating the influence of protein supplementation on arduous concurrent training adaptations. The heterogeneity of the meta-analysis results should be acknowledged as this may make it difficult to apply these findings to a specific population. Nonetheless, the findings of this review infer that protein supplementation can support muscle strength, aerobic endurance, and body composition adaptations during concurrent training. However, more population specific randomised controlled trials (RCTs) are needed to build upon these findings.

Military Research Applications
Five of the eleven studies included in this review quantified the effects of protein supplementation in those during military training, or in response to a military training-based activity [51,[58][59][60][61]. Given the potential for concurrent endurance training to inhibit muscle strength adaptations [27,28], military recruits may be one population who can benefit from strategies which aim to promote gains in FFM and muscle strength during arduous concurrent military training. The findings of this systematic review suggest that protein supplementation may be an effective strategy to support body composition and muscle strength development. However, to better understand the effects protein supplementation has on adaptation and performance outcomes of military recruits, additional population specific RCTs are needed. Future RCTs should consider investigating the effects of elevated protein intakes on training adaptations during arduous military training, including muscle strength, body composition and muscle recovery. Additionally, future work may want to consider adaptations not included in this review, such as bone adaptations, given that bone health and stress fracture incidence are important areas of military research [84][85][86] and which protein supplementation may be able to support [70,87]. The lack of women studied to date in this area also highlights a gap in the current literature. Therefore, future work should aim to include data in women since they now take-up more arduous (ground close combat) roles in the military [88]. Furthermore, including measures of protein metabolism, such as nitrogen balance and protein turnover, in future work should be considered as a means of better understanding the effects of protein supplementation during military recruit training.

Conclusions
This is the first systematic review and meta-analysis to investigate the effects of protein supplementation on arduous concurrent training adaptations. Based on the limited number of studies identified, more work in this area is clearly warranted, particularly given the importance of developing aerobic fitness and muscle strength concurrently in many exercising populations. The findings of this review suggest that protein supplementation may be an effective strategy at supporting lean-mass accretion and muscle strength adaptations in healthy adults, whilst considering the impact that training programme duration, total energy and protein intake of participants has on outcome measures. From the existing literature, it is reasonable to recommend that individuals aim for total daily protein intakes between 1.7 and 2.2 g·kg −1 ·day −1 whilst ingesting 20-40 g of protein immediately post-exercise to maximise MPS and support muscle strength and FFM adaptations. However, the disassociation between MPS and chronic physiological adaptations should be acknowledged. Based on the novel data included in this review, subsequent research may consider investigating the potential benefits of higher total daily protein intakes during arduous concurrent training on adaptation and performance outcomes.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/ 10.3390/nu13051416/s1, Figure S1: Funnel plot of the comparison of the effect of protein supplementation vs. placebo on muscle strength, aerobic endurance, FFM and fat-mass adaptations. All studies are included. Figure S2: Forest plot showing the effect sizes of the studies which assessed the effects of protein supplementation on muscle strength adaptations. Figure S3: Forest plot showing the effect sizes of the studies which assessed the effects of protein supplementation on aerobic endurance adaptations. Figure S4: Forest plot showing the effect sizes of the studies which assessed the effects of protein supplementation on fat-free mass (FFM) changes in response to concurrent training. Figure