Eating Carbohydrates Before and During Maximal Sprint Cycling Does Not Improve Performance

Fifteen physically active males (n = 13) and females (n = 2) (mean±SD: age: 22 ± 2 yrs; height: 176 ± 11 cm; mass: 76.8 ± 11.3 kg, BMI 24.7 ± 2.4 kg/m²) volunteered to participate in the study, and were recruited from the local university campus using posters, e-mails and word of mouth. To be eligible for the study, individuals must have been aged between 18–39 years old, and had to participate in moderate intensity physical activity at least 2–3 times per week. All participants took part in recreational athletic endeavours, e.g., university or amateur club level sports. Self-reported activity levels of participants were 2–3 training sessions of moderate-high intensity activity lasting ~90 min each per week, in addition to 1–2 competitive matches per week. This varied slightly depending on each individual participant’s sport and point in their relative competitive season. Exclusion criteria included not having any musculoskeletal or neurological disorders, free from injury and not currently supplementing with any ergogenic aids either 3 months prior to or during the study. Following a pre-screening physical activity questionnaire to ensure eligibility, participants were provided with information sheet, outlining the full experimental procedure and risks involved. All participants gave their written informed consent to participate. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ulster University School of Sport Ethics Committee (MSc321 2018–2019).

The study utilised a double-blind, placebo-controlled, randomised, cross-over design, with two experimental conditions: carbohydrate supplementation and placebo. Participants each visited the laboratory on two occasions to perform a bout of SIE (4 × 30 s maximal sprints). All participants self-reported as being familiar with SIE through routine training in their sports. Ten minutes prior to and during the bout of SIE, participants consumed either a carbohydrate solution or a placebo (described below). The order of conditions was randomised with ~10 days washout between the two trials. All participants were instructed to refrain from exercise 48 h prior to their lab visit and to abstain from alcohol and caffeine for 24 h before testing. All participants verbally confirmed on the morning of the trial that they avoided exercise for 48 h and consumption of alcohol and caffeine for 24 h. Participants were asked to maintain an identical nutritional intake 24 h before each visit to the lab, with participants’ last food intake (breakfast) ~3 h prior to exercise. Participants confirmed that they had maintained nutritional intakes in the 24-h period prior to exercise between trials at their second laboratory visit. The timing of each trial was standardised within and between participants, and the lab conditions were similar (temperature between 19–21 °C) during each visit. A member of the research team prepared each of the solutions and provided them to the participants, reaffirming the ingestion instructions. Neither the researcher recording/analysing performance and physiological variables, nor the participant, knew the contents of the solution. The blinding continued to the stage of statistical analysis, where the trial conditions were revealed.

Participants reported to the laboratory where physical characteristics were assessed. Height and body mass were measured using a stadiometer with integrated scales (SECA, Birmingham, UK). Body mass was recorded on each of the two laboratory visits. Participants then performed a 5 min warm-up on a cycle ergometer (Wattbike Pro, Wattbike, Nottingham, UK) at a self-selected intensity, perceived to be 10 on the Borg scale (i.e., ‘light/fairly light’). Pre-exercise blood lactate and heart rate were taken at rest ~3–5 min, following completion of the warm-up and approximately 1 min prior to SIE. Participants then performed four bouts of SIE, consisting of repeated maximal effort for 30 s at a standardised air resistance of level 6, followed by 3.5 min of passive recovery following each interval. During recovery, participants were allowed to disembark the ergometer, but remain stationary beside it, with no walking etc. permitted. Participants ingested a carbohydrate solution consisting of 8% carbohydrate (100% Maltodextrin, MyProtein, THG, Manchester, UK) and a single lemon-flavoured sweetener tablet per litre water (Splenda, Heartland Food Products Group, PA, USA), or a placebo consisting of water and a single lemon sweetener tablet per litre. The volume of solution ingested was 5 mls/kg/body mass. Exactly 50% of the solution was consumed approximately ten minutes prior to the first interval, with the remaining solution to be ingested during the rest periods of the first and second interval. To reduce the possibility of participants identifying the beverage they were consuming, both condition solutions were lemon-flavoured, colourless and presented in opaque bottles. An exit question was posed upon completion of the study to ascertain whether participants detected any differences in solutions between trials. All participants reported being unable to identify any differences between the experimental solutions. Performance and physiological measures were recorded during each sprint interval. Verbal encouragement was provided during each experimental trial, to encourage the participants to maintain maximal effort on the cycle ergometer. Each session lasted approximately 30 min in total.

The primary outcomes of interest in this study were peak and mean power output during each 30 s sprint. Power output was measured (100 Hz) continuously during each sprint (Wattbike Pro, Wattbike, Nottingham, UK), saved, and subsequently downloaded for analysis using the WattBike Expert Software. Relative peak power output (RPPO) and relative mean power output (RMPO) were calculated by dividing the absolute power outputs by body mass.

Physiological responses, including oxygen uptake, respiratory exchange ratio (RER), blood lactate concentrations, heart rate, and ratings of perceived exertion (RPE) were secondary outcomes of this study.

To measure oxygen uptake () and RER, participants were fitted with a silicone face mask connected to a metabolic cart, with an online gas analysis system to enable breath by breath analysis of expired air (Cosmed, Quark CPET, Rome, Italy). The metabolic cart was calibrated prior to each participant trial. Breath by breath measurements ofand RER were recorded from 60 s before the first 30 s sprint and terminated upon cessation of the final 30 s sprint. Only the data from each of the 30 s exercise intervals were analysed. V ˙ O 2 V ˙ O 2

To obtain blood lactate, an alcohol wipe (Alcotip swab, Universal Hospital Supplies, London, UK) was used to cleanse the participant’s finger, and a sterile lancet (Accu Chek, Roche, Mannheim, Germany) was used to pierce the skin of the finger to produce blood. A Lactate Pro 2 lactate strip (Arkray, KDK Corp., Shiga, Japan) collected the blood, which was then inserted into a Lactate Pro 2^TM lactate analyser (Arkray, KDK Corp., Shiga, Japan), to determine the lactate content of the participant’s blood. Lactate measures were taken during the first 30 s, following the completion of the interval.

Heart rate was recorded using a Polar H7 heart rate monitor (Polar, Kempele, Finland) and a Polar RS400 watch, and as taken immediately, following the cessation of the interval. RPE (rate of perceived exertion) was provided by each participant via the Borg scale, and was taken immediately following cessation of the interval.

All data were analysed using SPSS software version 25 (IBM, Armonk, NY, USA). A Shapiro–Wilk test was used to check each variable for any deviation from a normal distribution. Data were revealed to be normally distributed and parametric. A two-way (condition × time) repeated measures ANOVA, with sex (male/female) included as a covariate, was applied to determine the effect of CHO supplementation compared to placebo on performance and physiological outcomes during SIE. If appropriate, post-hoc comparisons were analysed via paired sample t-tests with a Bonferroni correction. In order to assess potential order effects on performance variables, a two-way (condition × time) repeated measures ANOVA was applied, with trial order as the between factor [27]. Statistical significance was set a priori at p < 0.05. The effect size for the main effects and interactions was estimated by calculating partial eta squared values (η_p²), with 0.01, 0.06 and 0.14 used to denote small, moderate and large effects sizes, respectively. All data are reported as mean ± SD.

For RPPO, during the sprints, there was a main effect of time (p < 0.0001, η_p² = 0.85), but no effect of condition (p = 0.79, η_p² = 0.03), and no condition × time interaction (p = 0.84, η_p² = 0.35). RPPO progressively decreased with each repeated sprint, and this effect did not appear to be impacted by CHO supplementation (Figure 1). Similarly, in terms of RMPO, there was also a main effect of time (p < 0.0001, η_p² = 0.83), but no effect of condition (p = 0.79, η_p² = 0.03) or condition × time interaction (p= 0.77, η_p² = 0.41). There were marked reductions in RMPO with each interval, however, this effect was not altered by CHO supplementation (Figure 2).

In the current study, both peak (PPO) and mean power outputs (MPO) systematically declined with each interval, relative to the previous interval, which is in agreement with observations from previous SIE studies [18,20,24,25,28]. Bogdanis et al. [25] demonstrated that following performance of a second consecutive 30 s sprint with either 1.5, 3 or 6 min of passive recovery, neither of these recovery periods were sufficient to maintain PPO or MPO compared to the first sprint. Based off this evidence, it is not entirely surprising that following the current protocol of 4 × 30 s sprints with 3.5 mins of passive recovery was insufficient to mitigate performance declines. However, CHO ingestion also did not appear to have any further effect on protecting against exercise-induced reductions in performance, compared to placebo. In order for exogenous CHO to mediate its effect as an ergogenic aid, it needs to be consumed, absorbed, delivered, taken up and preferentially oxidised over other substrates available by the skeletal muscle for use in ATP turnover. Hulston et al. [29] showed that, with a moderate (6%) CHO solution, plasma glucose and exogenous glucose oxidation were increased by 15–30 min at the onset of exercise. In the current study, 50% of an 8% solution was ingested 10 min prior to the first interval, with the remaining solution consumed before the third interval commenced. A total of 16.5 min elapsed from the beginning of sprint one until the completion of sprint four. Therefore, within the total 26.5 min timeframe, CHO may have been available for oxidation, although without mechanistic measures, this cannot be confirmed. It is also worth noting that exercise at very high intensities are a limiting factor on gastric emptying rates following glucose ingestion [30]. Therefore, it is also unclear how much of the CHO ingested would be available to the muscle for oxidation during SIE in the current study. Previous studies investigating exogenous CHO ingestion on SIE using either a 7% CHO solution consumed 30 min pre SIE [18] or 10% CHO solution consumed 5 min pre- and before each sprint [20], reported beneficial effects on performance compared to placebo or CHO mouth rinse respectively (see the next section for a further discussion). Therefore, it is entirely plausible that, if CHO ingestion mediated a performance effect in these studies, our ingestion schedule would also have provided at least the opportunity for performance enhancement. Although with recent evidence pointing to high ingestion rates of 120 g/h of CHO improving exercise performance [22,23], each of the SIE-based protocols may have benefited from a greater CHO intake. It has been shown that the phosphocreatine (PCr) and glycolytic systems are primarily responsible for the metabolism of substrates to meet the maximal demand for ATP during the first bout of maximal sprint exercise [24,25,31]. As sprint exercise progresses, muscle pH is rapidly decreased, halting glycolysis through inhibition of key glycolytic enzymes, such as phosphofructokinase (PFK), resulting in an increased demand for ATP turnover via aerobic oxidation [24,25], with increased pyruvate dehydrogenase activation and oxygen uptake, reducing lactate accumulation and increasing pyruvate oxidation [32]. Bogdanis et al. [24,25] demonstrated the mobilisation of large quantities of glycogen, with increased intramuscular glucose levels increasing substantially following initial maximal sprints. It may in fact be that preferential oxidation of existing substrates within the muscle provide sufficient energy, to fuel not only the initial bout, but also the subsequent bouts. In this case, subsequent re-oxidation of lactate and pyruvate following increased oxygen uptake during the recovery phase, is enough to fuel ATP turnover, and that the decrement in sprint performance may be more aligned to PCr recovery kinetics [24,25,28]. As such, provision of exogenous CHO in the current study, assuming delivery and availability to the muscle for oxidation, may not have been used to increase ATP turnover, and thus improve cycling performance compared to placebo.

At first glance, the results of the current investigation appear to contrast with previous studies, which have concluded that CHO supplementation improves SIE performance [18,19,20]. However, there are aspects of each of these three studies which can be critiqued. For example, Lee et al. [19] found that ingestion of an 0.8 g/kg⁻¹ solution had a significant performance effect on peak and mean power output, and total work versus placebo, in young, female athletes. However, on further examination, this performance effect was only evident during one of the sets (set 6), out of the total of 10 sets completed, with negligible effects observed throughout the rest of the bouts. Another study by Pomportes et al. [18] also reported a beneficial impact of acute CHO ingestion (7% CHO solution) on peak power output and perceived tiredness (a surrogate index of fatigue). However, it is worth noting that these conclusions were based on inferences made using magnitude based inferences (MBI), with ‘most likely positive’ and ‘likely positive’ effects of CHO supplementation reported for peak power and tiredness, respectively. The MBI approach has recently received substantial scrutiny for failing to adequately control type 1 error rates (i.e., false positives), particularly for effects characterised as ‘likely’ [33]. Furthermore, the study of Pomportes et al. [18] also reported (alongside MBI) no significant effect on either variable when assessed using a repeated measures ANOVA. Finally, Krings et al. [20] also investigated the influence of acute CHO ingestion (300 mL of a 10% CHO solution) on repeated maximal sprint interval performance. The results of this study reported significantly greater mean power output and total work, and significantly lower fatigue index, in the CHO ingestion condition, compared with a CHO mouth rinse condition. Critically, however, there were no differences in performance variables, compared with either an ingested or mouth rinse placebo condition, thus, inferring that other factors aside from the ingested CHO may have been associated with the performance differences between the conditions. Therefore, on the basis of these critiques, caution should be applied when interpreting such data in the context of CHO ingestion and perceived ergogenic effects on repeated sprint performance.

The current study shares some of the study design features of the three aforementioned studies that reported ergogenic effects for CHO supplementation in SIE models [18,19,20]. For example, the current study, and that of Lee et al. [19] and Krings et al. [20], were each randomised, double-blind, placebo-controlled studies, with Pomportes et al. [18] not reporting the blinding of participants to conditions. A strength of the current study’s design compared to the earlier studies was that the placebo was identical in taste, appearance, presented in opaque bottles, with all participants reporting they were unable to detect a difference between trial solutions. The studies of Pomportes et al. [18], Lee et al. [19] and Krings et al. [20] provided similar, but not identical placebo solutions to the CHO solution. Neither of these studies specifically report whether study participants were able to detect differences between solutions per condition. The description of non-identical solutions use raises the possibility that participants may have been able to identify the solution, and thus potentially confound results. The studies by Lee et al. [19], Pomportes et al. [18], and the current study reported that participants ingested breakfast meals several hours (~3 h) before the exercise protocols. However, the participants in the study of Krings et al. [20] were fasted, but no details are provided of the fasting duration. Participation in such activities in the fast stated is not reflective of ‘real-world’ practices [34], unless specifically attempting to modulate adaptations through AMPK-regulated influences on aerobic adaptations [35], although the value of SIE in these condition is questionable, in our view. Therefore, the approach used in the current study was adopted to maintain both scientific robustness [36] and real world application [34].

The novel aspect of the current study was the extended interval duration with respect to the current literature on SIE and CHO ingestion. The previous three studies employed sprint interval durations of 4, 5 and 15 s, with 20 s (passive), 25 s (passive) and 4 min (active) recovery, respectively, between sprints. The total number of sprints also ranged from 50, to 6 and 5 respectively. The current study employed 4 maximal sprints, 30 s in duration, with 3.5 min of passive recovery. The interaction of sprint interval duration with mode and duration of recovery period will result in very different energetic demands. It is likely that shorter intervals (≤15 s) derive much greater ATP turnover via PCr hydrolysis, with some contribution from glycolysis, to fuel maximal work during sprint intervals [28], whereas sprints lasting up to 30 s would derive much more ATP turnover via anaerobic glycolysis and glycogenolysis [24]. Therefore, the reliance on CHO as a fuel may have been more pertinent in the current study compared to the previous studies. Nevertheless, the volume of work completed in the study of Lee et al. [19], despite a short sprint interval duration, may also have required significant CHO contribution. Concomitantly, it is also important to consider that the total exercise durations were markedly different between the current study and previous SIE studies. In the current study, the total exercise duration was 120 s, with previous studies employing as little as 30 s [18], 75 s [20], and up to 200 s [19]. Therefore, in addition to interval length, one must also consider total volume of exercise duration when appraising the potential impact of CHO ingestion. The exogenous CHO ingested in the current study provided no additional performance benefit, and could not postpone the decline in performance during the SIE, despite other studies of shorter interval length and total exercise duration being able to report ergogenic benefits.

The total amount of work completed (volume) or higher rate of work done (intensity), evident through the production of higher mean power outputs, may be important for adaptations associated with high-intensity interval exercise, as each of these factors have been independently shown to be associated with modulation of various adaptive signalling pathways [7,12]. Conversely, CHO ingestion may potentially attenuate cell signalling responses associated with aerobic adaptations to SIE. Guerra et al. [37] investigated the effects of either 75 g glucose ingestion (n = 8) versus placebo (n = 7), 60 min before a single maximal sprint. Interestingly, there were also no differences in relative peak or mean power between groups during the sprint, although during the recovery period, AMPKα phosphorylation was modulated in the glucose group. Despite this attenuation of AMPKα phosphorylation, there was no effect between groups on PGC-1α protein content. Therefore, the effects of glucose ingestion on such factors remain unclear. Acute ingestion of CHO did not improve mean power output in any of the intervals in the current study, nor mitigate the exercise-induced reduction in power output following the completion of further successive intervals, compared to placebo. In terms of sporting performance, many cycling events are determined by sprint performance. In the context of training and/or performance, this type of CHO ingestion does not appear to provide any ergogenic benefits.

There are limitations present in the current study that warrant attention. While the participants in the current study reported being familiar with SIE through their own training and competition in sports, we did not perform a specific familiarisation with this cohort. Therefore, we cannot discard completely that there may have been some learning effects that could have added variability to the current data set. Additionally, we did not quantify/analyse/prescribe nutritional intake in the 24-h period prior to each trial, despite instructions to the participants to consume the same intake. This could have meant that glycogen stores and blood glucose levels may have been different upon commencing exercise between trials. However, as discussed already, glycogenolysis is attenuated relatively early during SIE, and performing SIE, even in a low glycogen store status, may not be a critical source of performance variability. Objective measures of participant training/fitness status were not conducted, which could affect responses and recovery to SIE, and also add variability to the data. Lastly, we did not perform any mechanistic/biochemical measurements. This meant that we were unable to detect the fate of exogenously consumed CHO and its ability to potentially alter ATP turnover in this condition. Further studies may wish to address these limitations in study design.