Training programs for competitive soccer players include activities of varying intensities, which have been shown to deplete muscle glycogen stores [25, 26]. In addition, plyometric exercises such as vertical jumping, which are a common component of soccer training, have been associated with increased muscle soreness, elevated blood CK levels and impaired performance in subsequent exercise . Thus, the utilization of post-exercise nutrition interventions that influence these variables could potentially affect recovery in soccer players. The purpose of this investigation was to assess the efficacy of CM as a post-exercise recovery beverage in soccer players, compared to a carbohydrate-only beverage. The recovery drinks were matched in total caloric content (504 kcal/serving), and both beverages contained carbohydrate in amounts that approached (CM: 1.1 g/kg) or exceeded (~CHO: 1.5 g/kg) levels associated with optimal post-exercise glycogen repletion [34, 35]. Although few studies have investigated the specific effects of CM on post-exercise recovery, our findings can also be compared with studies investigating CHO+Pro recovery beverages, which contain carbohydrate and protein in similar proportions to CM.
Overall, the isocaloric CM and CHO supplements provided similar effects on markers of post-exercise recovery over the four-day period of ITD. No significant treatment*time interactions were observed for muscle soreness, ratings of energy/fatigue and muscle function (MVC). Similarly, there were no treatment effects on serum Mb. However, serum CK levels were significantly lower following four days of ITD with CM supplementation versus CHO supplementation. Numerous studies of CHO+Pro beverages have reported attenuated post-exercise plasma/serum CK levels after heavy endurance or resistance exercise [4, 5, 7–10], though this finding has not be observed in all studies [11, 12]. The reduced CK levels observed in this investigation is also consistent with Cade et al.  and Luden et al. , who reported lower plasma CK levels with CHO+Pro ingestion over the course of multiple days of training in free-living swimmers and runners, respectively. Our findings similarly suggest that CM may attenuate blood CK levels in athletes performing heavy soccer training.
Plasma/serum CK is often used as a broad indicator of muscle damage. However, CK levels can be poorly correlated with direct measures of muscle damage or muscle function [36, 37]. Thus, the practical significance of modestly lower serum CK levels (~115 U/L) with CM is not clear. This is particularly notable here because other measures of recovery, such as muscle soreness and Mb were not significantly different between treatments. Findings from other studies have reported mixed findings with respect to the influence of CHO+Pro on these variables. Some have reported attenuated muscle soreness ratings or Mb levels following heavy endurance [6–8, 10, 11] or resistance exercise [4, 38], while others have reported no differences between treatments .
Though it cannot be concluded that recovery was different between treatments based on the CK data alone, other information from this study could suggest a potential tendency towards augmented recovery with CM. For example, increases in MVC over the four days of ITD were slightly greater with CM ingestion (53 ± 75 N) than with CHO (16 ± 93 N). This observation is consistent with findings from Valentine et al. , who reported that CHO+Pro ingestion improved muscle function versus CHO and placebo beverages following heavy endurance exercise. The difference in MVC levels between treatments in the present study was not statistically significant (p = 0.295), but may warrant investigation in future studies in light of the relatively small effect of our ITD protocol on symptoms of overreaching, as discussed below.
From a functional perspective, the most important measure of 'recovery' for athletes is performance in subsequent exercise. Some recent investigations have reported that CHO+Pro co-ingestion during/following heavy endurance exercise may improve subsequent exercise performance versus CHO [9, 14–18]. However, a similar number of studies have reported no differences in subsequent performance between CHO and CHO+Pro recovery beverages [6–8, 11, 19–21]. Subsequent exercise performance was not assessed in the present study, as it was not possible to perform repeated sport-specific exercise testing within each training period without interfering significantly with the prescribed training programs from the coaching staff. However, sport-specific exercise tests (T-drill, vertical jump) were conducted within the ITD periods, and compared between treatments. Performance test results were virtually identical between treatment periods, suggesting that post-exercise CM consumption did not have a preferential effect on short-duration, high-intensity whole-body exercise performance versus CHO. Our findings suggest that isocaloric CHO and CM beverages provide similar effects on whole body exercise recovery during short periods of heavy soccer training.
Few studies have examined the specific effects of CM on recovery from heavy endurance-based exercise. Karp et al.  compared three recovery beverages consumed following a glycogen-depleting session of cycling intervals. In a time-to-exhaustion test performed four hours later, cyclists rode significantly longer with CM compared to a commercial CHO+Pro beverage, but had similar performances as compared to a commercial CHO beverage. These findings are difficult to interpret, as CM contained similar carbohydrate and protein amounts than the CHO+Pro beverage, but the CHO beverage contained no protein and less than half the carbohydrate and caloric content than CM. More recently, Thomas et al.  conducted a similar study, comparing isocaloric CM and CHO+Pro beverages and a CHO beverage comparable to that used by Karp et al. . Time to exhaustion in the subsequent exercise bout was significantly longer with CM than either comparison beverage. Although the potential mechanisms for these findings are not clear, these studies support the potential efficacy of CM as a post-exercise recovery beverage following heavy endurance exercise.
The present study was designed to compare recovery beverages in free-living athletes within a collegiate team setting. Although this maximizes the generalizability of our findings for athletes, there were some relevant limitations to this design. Firstly, the free-living environment may have increased measurement error over the course of the study. Great care was taken throughout the study to insure that training/nutritional conditions were virtually identical between the two treatment periods. However, it is possible that activities outside the experimental protocols may have influenced the outcomes of the study. For example, four of the seventeen participants who completed the study were removed from statistical analyses (as described in Methods) due to large variations in baseline measurements (i.e. prior to ITD and beverage treatments), possibly due to activities outside of the study parameters. Six subjects failed to return completed dietary recall questionnaires, and thus we cannot be certain that nutrient intake did not vary between treatment periods for the entire sample. In addition, subjects were instructed to replicate the same dietary habits between treatment periods, but were not required to arrive at the laboratory in a fasted state. Thus, differences in nutrient timing between treatment periods could also have influenced some of the study outcomes.
Another limitation was the NCAA regulation limiting out-of-season practice time to a maximum of 8 hrs per week of 'athletically related activities' (NCAA Playing and Practice Limitations, Bylaw 18.104.22.168). As a result, it was not possible to implement an ITD period greater than 4 days in the present study. The prescribed training program was designed to increase daily training time by >25% per day between baseline and ITD periods (Table 1). However, due to adjustments in training plans to accommodate for inclement weather on two days (and maintain consistency between treatment periods), the ITD period increased daily training times by only 12% (Table 3). This training stimulus produced significant increases in muscle soreness ratings, and serum CK levels over the four-day period. However, MPSTEFS ratings and serum Mb were not significantly altered over time, and MVC actually improved over the four days of ITD. These findings suggest that the relatively modest increase in training volumes (combined with the daily consumption of post-exercise recovery beverages) may have been an inadequate stimulus to substantially impair muscle recovery. Without a relatively robust effect on these markers following exercise, it may be difficult to assess differences in recovery between treatments, especially with a relatively small sample of subjects, as described by Luden et al. . This issue is particularly relevant with regards to our measurements of vertical jump performance. Byrne and Eston  reported that vertical jump performance declined to 90% of initial levels one day following muscle damaging exercise. However, their exercise protocol produced elevations in CK that were approximately 3-4 times greater than the present study. Because our vertical jump device assessed only 0.5 inch increments, our instrument potentially lacked the sensitivity to detect realistic changes in vertical jump height. Other investigators have reported significant decrements in physical performance, fatigue and/or muscle soreness following periods of ITD [3, 39]. However, these studies provided 8-11 days of ITD (and relatively low post-exercise carbohydrate intake), which represented a much greater alteration in training stimulus than the present study. Thus, it may be worthwhile for future researchers to investigate the efficacy of CM during longer, more demanding periods of ITD.
Due to the practical restrictions of studying collegiate athletes, it was also not possible to add a placebo trial to the present study design. This prevented us from establishing the direct effects of the ITD period, independent of supplementation. Recovery beverages were provided immediately post-exercise, and both contained high doses of carbohydrate (>1.1 g/kg). As a result, both beverages probably produced high rates of post-exercise glycogen resynthesis , and potentially sustained muscle recovery and performance levels to a greater degree than if inadequate carbohydrate were provided [3, 39]. However, the relative efficacy of the 'control' beverage in this study (CHO) cannot be quantified without a placebo trial for comparison.