An investigative study into the influence of a commercially available carbohydrate-protein-electrolyte beverage on short term repeated exercise performance

Participants

Seventeen healthy males volunteered for participation in the study. Participants were required to be 18-40 years of age, and to be recreationally active. For study inclusion, participants were required to: satisfactorily complete a health screen questionnaire; not be taking any other supplementation; not have dysglycemia or known diabetic conditions; and have a maximal oxygen uptake between 40-59 ml·kg^-1∙min^-1. Following a study briefing, all participants provided written, informed consent for inclusion. Ethical approval for the study was provided by the University of Hertfordshire Life Sciences Ethics Committee.

Preliminary testing

All testing was undertaken in the Human Physiology Laboratory, Division of Sport, Health and Exercise, University of Hertfordshire. Upon entry to the laboratory, nude body mass (Seca 780, Hamburg, Germany) and height were recorded. All participants then performed a maximal oxygen uptake (VO_2max) test on a Computrainer cycle-ergometer (RaceMate Inc., Seattle, US) to assess against inclusion criteria. After a 5 minute warm-up at a standardised 100 W workload, a continuous ramp protocol (starting at 150 W) was employed with workload increasing at a rate of 15 min^-1.

Expired air was sampled throughout all tests with an online gas analyser (Metalyser 3B, Cortex Biophysik, Leipzig, Germany) to assess VO_2max and other respiratory variables. Heart rate (HR) was measured by means of a telemetric system (Polar Electro Oy, Kempele, Finland). Ratings of perceived exertion (RPE) were collected at 1 minute intervals, using the Borg 6-20 subjective exertion scale [12]. The test concluded when participants reached volitional exhaustion or were unable to maintain the required power output.

VO_2max was defined when a minimum of two of the following criteria were attained: 1) an increase of no more than 2 ml·kg^-1·min^-1 in oxygen consumption with additional workload, 2) attainment of maximal predicted heart rate (± 10 beats.min^-1) and 3) a respiratory exchange ratio (RER) of > 1.05. Maximal power (W_max) was calculated by adding the fraction of time spent in the final non-completed workload, multiplied by the 15 W increment, to the final completed workload.

Experimental trials

Experimental trials were conducted under laboratory and temperature controlled conditions (temperature: 20.2 ± 0.5°C; barometric pressure - range: 904-1015 mBar; and relative humidity -range: 24-47%), with no statistically significant differences demonstrated between trials (P > 0.05) for any of the environmental variables.

A randomised, double-blind, placebo controlled design was employed, with participants being required to attend the laboratory at the same time of day over two trials (separated by one week). Participants were requested to arrive at the laboratory having overnight fasted (12 hours) and having refrained from strenuous activity for the previous 72 hours. Additionally, individual food diaries for the 72 hours prior to each trial were provided by all subjects to assess for dietary compliance. On arrival to the laboratory, participants were required to complete a subjective muscle soreness questionnaire for the knee extensors and hamstring areas, as well as a daily analysis of life demands for athletes questionnaire (DALDA [13]).

Each trial consisted of two exercise bouts separated by a two hour recovery period. For each exercise bout, participants were required to complete a 45 minute submaximal exercise period (ST), followed immediately by a 45 minute time trial performance test (PT). A standardised warm up of 5 minutes at 100 W on the same Computrainer cycle-ergometer used in pre-testing conditions was employed for all participants prior to each exercise bout. At the end of the warm up period, participants were provided with an opaque drinks bottle containing 500 ml of either the test drink (40 g of a combined dextrose, maltodextrin and hydrolysed whey protein formula (VIPER^®ACTIVE, Maxinutrition Ltd.) delivering an 8% concentrated solution) or a taste/appearance matched citrus fruit concentrate placebo. A fixed volume of 100 ml was consumed by the participants at 0, 10, 20, 30 and 40 minutes of the submaximal exercise period. The test beverage per 100 g comprised: 7.1 g of protein; 88.4 g of total carbohydrate (of which 50.6 g glucose); 0.4 g of total fat; 0.53 g of sodium; 0.03 g of magnesium; 0.17 g of potassium and 0.14 g of calcium, and delivered 386 kcal. Conversely the placebo beverage per 100 g comprised: 0.6 g of total carbohydrate; 0.2 g of protein; trace amounts of total fat and sodium, and delivered only 8 kcal.

Submaximal exercise (ST1) comprised 45 minutes cycling at a workload equivalent to 60% VO_2max. During the ST period, capilliarised fingertip blood sampling (100 μl) was undertaken at 10 minute intervals for the assessment of blood lactate and glucose (Biosen C, EKF Diagnostics, Barleben, Germany). Respiratory measurements were ascertained at 10 minute intervals during ST to confirm intensity adherence utilising expired air analysis. RPE and HR measurements were collected at 5 minute intervals. Mean power output (W), speed (km.hr^-1) and distance covered (km) were also assessed during ST.

On completion of the ST protocol, participants immediately undertook a 45 minute maximal time trial performance test (PT1). Only time elapsed was visible to the participants. No test drinks were administered during PT1. Mean power output (W), speed (km.hr^-1), distance covered (km), RPE and HR were assessed at 10 minute intervals during PT1.

At the end of the first 90 minute exercise period, participants undertook a 2 hour superivsed recovery period. During this period participants were provided with 500 ml of the test drink at 0 and 60 minutes into recovery. In addition, all participants received a standard protein meal bar (Promax™ Meal Bar, Maxinutrition Ltd.) at 60 minutes into recovery. This was to avoid any unnecessary risks of severe hypoglycaemia occurring during the placebo trial. The standard protein bar comprised 206 kcal, containing 21.6 g of protein, 17.0 g of carbohydrate (of which 9.5 g sugars), 5.7 g of total fat, and 0.05 g of sodium.

At the end of the recovery period, all participants underwent a second exercise period, comprising the same protocol for both submaximal (ST2) and time trial performance (PT2) previously described. Participants returned to the laboratory one week later to complete the same experimental procedure on the alternate drink. On completion of each trial, participants were provided with three muscle soreness/DALDA questionnaires for completion on waking on days 1, 2, and 3.

Calculations and statistical analyses

Statistical analyses were performed using SPSS Statistics for Windows version 17 (SPSS, Chicago, USA). A two-way analysis of variance (ANOVA) for repeated measures was used to assess interactions between trial (ST or PT), condition (beverage used) and where applicable, time, for all variables. Where F ratios were found to be significant a Bonferroni post hoc test was applied. An alpha level of 0.05 was employed for assessment of statistical significance. All data are reported as means ± SE.

Submaximal exercise trials (ST)

Distance, speed and power output

Data for distance covered (km) and average speed output (km.hr^-1) are represented in Table 2. There was a significant interaction effect for total distance covered during submaximal exercise (F = 8.054; P = 0.013). Whereas total distance covered with CPE was not different between trials; there was a significant reduction in mean distance covered with PL (20.18 ± 0.28 km in ST1 v 18.34 ± 0.36 km in ST2; P = 0.0001). This represented a 9.12% decrease in submaximal performance with PL. In addition, reduced distance covered in ST2 for the PL condition was specifically noted in the last 15 minutes of the trial (P = 0.0001). Accordingly, there was a similar interaction effect for average speed output during submaximal exercise between trials and conditions (F = 8.724; P = 0.010).

	PL		CPE
	ST 1	ST 2	ST 1	ST 2
Total Distance Covered (km)	20.18 ± 0.28	18.34 ± 0.36*	20.03 ± 0.32	19.17 ± 0.44
Distance in last 15 mins(km)	6.65 ± 0.11	5.84 ± 0.16*	6.67 ± 0.12	6.32 ± 0.18
Average Speed (km.hr-1)	26.89 ± 0.39	24.67 ± 0.46*	26.54 ± 0.36	25.70 ± 0.56
Average Speed - last 15 mins (km.hr-1)	27.05 ± 0.39	24.75 ± 0.49*	26.72 ± 0.43	25.64 ± 0.58

Values are presented as mean ± SE; n = 16; PL, Placebo; CPE, carbohydrate-protein-electrolyte; ST1, submaximal exercise trial 1, ST2, submaximal exercise trial 2. * denotes significant difference (P = 0.0001) between trials within condition only.

Data for average power output are shown in Figures 1 and 2. During submaximal exercise, there was a significant interaction effect for average power output (F = 7.637; P = 0.015). Over the full 45 minute trial, power output significantly decreased by 10.9% from 128.89 ± 3.61 W in ST1 to 114.82 ± 4.04 W in ST2 (P = 0.002) for PL only. A similar pattern was observed for the last 15 minutes of the exercise trial, with average power output being significantly lower in ST2 (112.38 ± 4.22 W) compared to ST1 (128.38 ± 3.85 W) for PL only (P = 0.0001). No significant differences were found for the CPE beverage between trials.

Wholeblood data

Data for blood glucose are represented in Figure 3. No significant differences were found between trials or conditions for resting values (P = 0.327). There was, however, a significant interaction effect over both time and condition (F = 3.654; P = 0.01). Mean blood glucose was significantly greater over the first exercise bout with CPE compared to PL (5.06 ± 0.13 mmol.L^-1 and 4.53 ± 0.08 mmol.L^-1 respectively; P = 0.002).

During recovery between exercise bouts, there was a significant interaction effect (P < 0.001), with mean blood glucose being significantly lower with PL both at 30 minutes (6.30 ± 0.30 mmol.L^-1 for CPE and 3.87 ± 0.12 mmol.L^-1 for PL, P < 0.01) and 60 minutes (5.47 ± 0.27 mmol.L^-1 for CPE and 3.82 ± 0.12 mmol.L^-1 for PL, P < 0.01). Mean blood glucose in ST2 was maintained with CPE compared to ST1; and was significantly higher than with PL during ST2 (4.77 ± 0.08 mmol.L¹ for CPE compared with 4.18 ± 0.06 mmol.L^-1 for PL, P < 0.001).

Data for blood lactate are represented in Figure 4. Whilst there were no significant differences for resting lactate between conditions, blood lactate was elevated at the beginning of the second exercise bout with CPE compared to the first bout only (1.74 ± 0.21 mmol.L^-1 compared to 1.04 ± 0.12 mmol.L^-1, P = 0.04). Mean data demonstrated a significant decrease in blood lactate between exercise bouts for CPE (2.47 ± 0.20 mmol.L^-1 compared to 1.78 ± 0.18 mmol.L^-1, P = 0.005) and for PL (2.75 ± 0.26 mmol.L^-1 compared to 1.67 ± 0.17 mmol.L^-1, P = 0.009). There were no other significant differences reported between conditions.

Time trial performance data

Data for overall distance covered during the time trial performance tests (PT) are shown in Figure 5. A significant interaction effect was found for total distance covered (F = 12.231; P = 0.004). No differences were reported between conditions for PT1. However, with PL, average distance covered fell from 21.64 ± 0.58 km in PT1 to 17.27 ± 0.62 km in PT2 (P = 0.0001), representing a 20.2% reduction in performance. Total distance covered was also lower in PT2 compared to PT1 with CPE (20.23 ± 0.65 km v 22.55 ± 0.34 km respectively; P = 0.02), representing a 10.3% reduction in performance. However, there was a significant difference between conditions following PT2, with the CPE group cycling on average 2.96 km further than the PL group (P = 0.003) representing a 17.1% difference between conditions.

Additionally, assessment of distance covered in the last 15 minutes of the PT revealed a significant interaction effect (F = 6.288; P = 0.024), with mean distance reducing from 7.29 ± 0.21 km to 5.81 ± 0.24 km with PL across trials (P = 0.0001), and from 7.76 ± 0.15 km to 6.74 ± 0.25 km with CPE across trials (P = 0.007). The reduction observed with PL was 16.0% greater than with CPE for PT2 (P = 0.01).

A significant interaction effect was found for heart rate data across trials (F = 17.220; P = 0.001), with average heart rate shown to be significantly lower in PT2 compared to PT1 for PL only (P = 0.005). No other differences were observed for heart rate between trials or conditions. Data for RPE demonstrated consistent hard to very hard exertion across PT tests, and was significantly higher in PT2 compared to PT1 only (F = 4.752; P = 0.047). No other differences were observed for RPE within or between conditions.

Post trial questionnaire and subjective muscle soreness assessment

No significant differences were found for any of the pre trial muscle soreness assessments (P > 0.05). Post trial muscle soreness assessment data are represented in Table 5. Mean quadriceps soreness was significantly different post trial (F = 7.824; P = 0.013), with soreness ratios only different between days 1 and 3 (P = 0.05). Data was not different between conditions (P > 0.05). Likewise, mean vastus lateralis (VL), and mean distal VL soreness assessment was significantly different between days 1 and 2, and 1 and 3 post trial only (P < 0.05). No other differences were observed for soreness data.

Submaximal exercise

One of the key findings from this study was that the ingestion of a CPE beverage maintained total distance, average speed and power output in ST2 when compared to PL. At a prescribed exercise intensity, total distance covered significantly decreased by 9.12% from 20.18 ± 0.28 km in ST1 to 18.34 ± 0.36 km in ST2 when participants consumed a fruit concentrate PL. In contrast, there was no significant difference between ST1 and ST2 for total distance covered when participants consumed a CPE beverage. Whilst there were no differences found between conditions for ST1 or ST2, the significant reduction in work output for the PL group does support previous research indicating that CHO ingestion is likely to be more beneficial for longer duration [15, 16] or subsequent high intensity exercise bouts [17].

Additionally, as there has been previous interest in the effect of CHO beverages in the latter stages of exercise, the decrement in exercise maintenance with the ingestion of PL was also observed in the last 15 minutes of ST2. When participants undertook ST2 during the PL condition, average speed significantly reduced from 27.05 ± 0.39 km.hr^-1 in ST1 to 24.75 ± 0.49 km.hr^-1 in ST2. This was replicated with a significant reduction in average power output in the final 15 minutes of ST2 of 16.0 W in the PL condition. As the degree of statistical significance was greater at 45 minutes compared with 30 minutes, it can be inferred that the level of fatigue was exacerbated in the last 15 minutes without ingestion of CPE.

The maintenance of submaximal work output observed with CPE indicates the beneficial effects of such beverages on single day repeated training sessions. It is probable that such replication of work output is explained by the maintenance of plasma glucose, especially in ST2. Interestingly, the ingestion of CPE resulted in a greater mean blood glucose in the first exercise bout compared with PL (5.06 ± 0.13 mmol.L^-1 and 4.53 ± 0.08 mmol.L^-1 respectively), but this did not impact on short term work output in ST1. The maintenance of a higher mean blood glucose was further apparent with CPE in ST2 (4.77 ± 0.08 mmol.L^-1 compared with 4.18 ± 0.06 mmol.L^-1 for PL), which potentially contributed to overall and end stage work output.

The ingestion of a PL beverage clearly resulted in increased levels of fatigue, demonstrated by significant reductions in power output and total distance covered during ST2 relative to ST1. Concomitant reductions in VCO₂, RER and CHO_TOT suggest that depletion of endogenous energy stores may be the major mechanism contributing to short term fatigue, particularly in a glycogen-fasted state. With increased utilisation of endogenous carbohydrate, there will be a decreased reliance on glycolytic flux and hence reduced lactic acid production, as demonstrated in the PL condition. With a reduced demand to buffer hydrogen ion production, this likely explains the significantly lowered VCO₂ levels observed in ST2 for PL. Whilst mean CHO_TOT was observed to decrease in ST2 with CPE (from 2.615 ± 0.216 g.min^-1 in ST1 to 2.159 ± 0.132 g.min^-1 in ST1), the reduction was not significant, and indicates a relative maintenance of CHO_TOT throughout the repeated submaximal exercise. The absolute reduction between submaximal bouts for CHO_TOT in the CPE trial could be explained by low carbohydrate ingestion rates used in the study.

Whilst CHO_TOT was not assessed during the recovery period, the inclusion of a double bolus of the test beverage at 0 and 60 minutes of recovery resulted in significant differences in mean blood glucose between conditions at 30 minutes (6.30 ± 0.30 mmol.L^-1 for CPE and 3.87 ± 0.12 mmol.L^-1 for PL) and 60 minutes (5.47 ± 0.27 mmol.L^-1 for CPE and 3.82 ± 0.12 mmol.L^-1 for PL) of the recovery period. The inclusion of a standard protein bar at 60 minutes into the recovery period was employed to minimise any absolute risk of hypoglycaemia in the PL condition, and to replicate strategies often employed by athletes in daily practice. In the CPE condition a total of 123.1 g of CHO was therefore ingested prior to the start of ST2 in comparison to 17.7 g ingested with the PL condition. Prior to the start of ST2, this would have equated to a total CHO ingestion rate of 0.59 g.min^-1 for the CPE condition. This is considerably below the 1.0-1.2 g.min^-1 suggested saturation range of intestinal glucose transporters [16, 18], yet still infers an ergogenic benefit.

Performance exercise

There has been much, and often controversial interest, in the potential performance ergogenic effects of CHO beverages both for shorter duration exercise sessions, as well as repeated bouts. It is widely known that in the absence of sufficient CHO, absolute work output will gradually decline with exercise duration and intensity, based on both liver and muscle glycogen depletion rates, and associated mechanisms of intracellular fatigue. In this study, the use of a CPE beverage did not confer performance advantages in PT1 compared to PL, with average power outputs being comparable (134.21 ± 4.79 W for PL and 136.82 ± 3.80 W for CPE). Interestingly, in PT1, mean distance when consuming CPE was 0.91 km greater than PL, which comprised a 4.2% overall improvement comparable to other studies [19].

The lack of statistical significance between conditions for PT1 however do conflict with other studies both for cycling [20] and running tests [21]. In the latter study, the ingestion of a 6.4% CHO-E solution 30 minutes before and at 15 minute intervals during a 1-hr treadmill run, significantly improved performance by 2.7%. Both studies proposed that the inclusion of carbohydrate prior to exercise resulted in higher CHO_TOT which conveyed the performance increments in the latter stages of exercise. In the current study, carbohydrate ingestion preceded PT1, but not under resting conditions. The lack of difference in CHO_TOT between conditions for ST1 suggests that ingestion rates were not of sufficient magnitude to elicit short term performance gains. In the previous study [21], participants ingested a total of 67.1 g of CHO prior to completion of a time trial (effectively an ingestion rate of 0.75 g.min^-1). In the current study, participants ingested a total of 35.4 g CHO prior to completion of PT1 (an effective ingestion rate of 0.39 g.min^-1). It is therefore possible that higher ingestion rates either pre exercise and/or during PT1 may have resulted in significant short term gains.

However, when repeated bouts of exercise are undertaken, the beneficial effects of CPE ingestion appear to be more pronounced. Total distance covered in PT2 was 17.1% greater with the ingestion of CPE compared to PL. The demanding nature of the trials was observed, with a significant 10.3% reduction in total distance covered between trials for the CPE condition (22.55 ± 0.34 km for PT1 compared to 20.23 ± 0.65 km in PT2), potentially explained by the relatively low total CHO ingestion rates employed.

Average power output during (and in the final 15 minutes) of PT2 were significantly reduced in PL, demonstrating the contrasting benefits of CPE. Whilst the type and quantity of CHO has been shown to enhance exogenous CHO oxidation rates [3, 7, 18], late stage performance enhancement may still occur with more conservative ingestion rates. By the start of PT2, during the CPE trial, participants had consumed a total of 158.5 g CHO or 37.3 g.hr^-1. Comparable ingestion rates have been shown to enhance late stage exercise performance elsewhere [22] despite being below known optimal delivery rates of 1-1.2 g.min^-1 or 60-70 g.hr^-1[16].

It is most likely that any ergogenic or recovery effects from the CPE beverage are explained by the combination of the maltodextrin and dextrose formulation. It has been demonstrated that the inclusion of multiple carbohydrates will result in higher exogenous carbohydrate oxidation (CHO_EXO) rates [23]. The combined uptake of total sugars from the sodium dependent glucose transporter (SGLT1) and GLUT5 intestinal transport mechanisms provides potential for maximal exogenous oxidation rates [3]. Whilst the oxidation rates of both dextrose and maltodextrin are similar, the inclusion of maltodextrin reduces beverage osmolarity, hence increasing the potential for carbohydrate delivery to the intestinal lumen, as well as fluid uptake.

Furthermore, the inclusion of sodium to the test beverage is known to enhance carbohydrate bioavailability [24]. Despite relatively low CHO ingestion rates employed in the current study, an enhancement in both CHO delivery and CHO_EXO would still have a resultant sparing or even suppressing effect on endogenous CHO utilisation [25], as well as maintaining the CHO_TOT observed between performance bouts. As CHO_EXO rates have typically been shown to plateau after 90 minutes of steady state exercise, this in part explains the ergogenic potential observed in PT2 with CPE.

Alternatively, as CHO ingestion rates were below optimal delivery levels, it is possible that the co-ingestion of protein may have provided additional ergogenic value through increased caloric content. Whilst it has been suggested the addition of approximately 2% protein to a CHO beverage has minimal effect on subsequent performance, or glycogen resynthesis [26, 27], other studies have demonstrated a positive effect of co-ingestion of protein on endurance performance [8, 9, 28, 29] and short term recovery [30]. When carbohydrate-protein beverages have been administered during acute recovery (in comparison to an iso-energetic carbohydrate beverage), there is supporting evidence that the addition of protein positively enhances repeated same day time to exhaustion trials [31, 32]. The most likely explanation for this is the higher caloric content of the beverages employed, in comparison to lower dose carbohydrate only beverages [32]. In the current study, as the protein intake for CPE was only 0.6% or 2.84 g per 40 g serve, any enhancement of acute recovery through insulin-mediated pathways would most likely be explained via the inclusion of a standard protein bar between exercise trials.

In terms of short term recovery post trials, the only significant observations from this study were reductions in mean quadriceps soreness, mean vastus lateralis soreness and mean distal vastus lateralis soreness by day 3. This was expected considering subjects had a 7 day rest period between trials, hence explaining the gradual reduction in perceived soreness for both conditions. As no differences were found between conditions for post exercise muscle soreness or DALDA responses, the inclusion of early protein feeding (mainly in the form of a protein meal bar) may have assisted recovery in both conditions, as demonstrated elsewhere [33]. It has been suggested that the inclusion of protein to a carbohydrate beverage during early recovery, particularly in higher dosages than the present study, may facilitate intracellular rps6 and mTor signalling pathways leading to enhanced protein resynthesis and hence recovery [34–36]. However, beneficial effects of such beverages on acute glycogen resynthesis is most likely accounted for by underlying carbohydrate dosage and content [37].