A Five-Week Carb Cycling Diet Does Not Improve High-Intensity Exercise Performance or Fuel Use Compared to a High-Carb Diet in Trained Cyclists

A nutritional strategy commonly used in endurance sports is to increase the consumption of carbohydrates [1], as low muscle glycogen availability can contribute to decreased exercise capacity [2]. Previous research has also associated high carbohydrate intake for athletes’ diets with improving specific parameters related to recovery, muscle damage, and performance [2,3,4]. In contrast to traditional guidelines [5], recent evidence suggests that the periodization of carbohydrates may benefit athletes due to the potential muscle cellular adaptations [6]. This strategy requires limiting exogenous carbohydrate intake before and/or during specific training sessions by a variety of techniques such as “sleep-low” (i.e., glycogen-depleting session of training is followed by overnight carbohydrate (CHO) restriction and a moderate-intensity session the subsequent morning), “fasted exercise” (i.e., performing endurance exercise in a fasted state, after an overnight fast and without CHO intake during session), or “twice-a-day training” (i.e., two sessions on the same day; the second is commenced with reduced muscle glycogen) [7].

Some studies found improvements in performance after the application of these nutritional protocols [8,9]. On the contrary, other studies do not report superior effects on performance [10,11,12,13,14,15,16]. The question arises whether a nutritional strategy in which carbohydrate intake is manipulated based on the goal, duration, and intensity of the training session, known as “fuel for the work required” [17], can be beneficial for endurance performance and whether it may produce an effect on substrate utilization in well-trained athletes during a 5-week period. Studies in which performance was measured following different carbohydrate (CHO) feeding strategies have usually tested recreational or trained athletes over 1–4 weeks of treatment [11,12,15]. Still, the study of prolonged carbohydrate periodization (5 weeks) in well-trained athletes (~70 VO_2peak) is yet to be explored.

Similarly, most studies that applied CHO periodization performed a submaximal test followed by a maximal test to assess the participants’ performance [8,9,10,11,12,14,15]. The present study showed novelty in performing measurements at maximal lactate steady-state (MLSS) intensity. It is suggested to be the gold standard for assessing the lactate threshold, which is directly associated with physical capacity and performance, and accurately investigating the consequences of diet at the same physiological milestone [18,19].

Regarding substrate utilization and body composition, one of the most interesting and discussed outcomes when applied to CHO-restricted diets is increasing fat as fuel utilization to enhance endurance sports performance [11]. However, significant fat oxidation capacity improvements have been shown to be insufficient to observe improved performance in elite athletes compared with a high-CHO condition [13]. Further, low-carbohydrate diets are often used as weight-loss strategies by exercising individuals and athletes [8]. Several researchers reported a decrease in fat-free mass after low-carbohydrate diets for extended periods (>six months) [20]. However, when both interventions were performed over a short-term period (e.g., three weeks) with analogous daily calorie intake, markers of body composition were altered to a similar extent [21].

Therefore, this study aimed to investigate whether periodizing the carbohydrate intake during a 5-week period has any benefit compared to a traditional high-carbohydrate diet regarding performance, substrates used during exercise, and body composition in well-trained cyclists.

The main findings of the present study were as follows: (1) Both groups improved in similar pre–post-MLSS tests expressed in watts and watts/kg. However, the periodized carbohydrate diet did not lead to a greater improvement pre–post-intervention -time-to-exhaustion tests when compared to a traditional high-carbohydrate diet; (2) there were no significant differences in CHO and LIP oxidation and RER during the time-to-exhaustion test at MLSS intensity pre–post-intervention; (3) there was a significant improvement in muscle mass or decrease in fat percentage for both conditions; (4) no significant differences were found between groups during the performance tests in any of the parameters studied (final lactate, mean heart rate, mean cadence, and before and after time-to-exhaustion test and body weight) before or after the intervention.

Based on the current literature, most studies have investigated the relevance of low CHO periodization on well-trained endurance athletes (i.e., cyclists, triathletes, and race walkers with a VO_2max of ~60 mL/kg/min) for three weeks following different CHO periodizing strategies including “fasted training”, “twice-a-day”, and “sleep-low”. They did not show performance enhancements regarding endurance performance in comparison to high CHO availability after performing a time trial endurance test [8,10,12,16,40]. Even though two studies found beneficial effects of CHO periodization on endurance performance, applying a sleep-low approach during one [9] and three weeks [8], a recent review and meta-analysis, which bring together most of the studies published to date investigating between one and four weeks of periodizing CHO availability in well-trained endurance athletes (≥55–60 VO_2max), revealed that the collective evidence does not support a beneficial or superior approach for enhancing endurance performance compared to a high-CHO diet [41]. The present study extends the period of nutritional diet intervention, applying 5 weeks of intervention because the intervention period (i.e., one to four weeks) is part of the limitations in existing research; more is needed to draw definitive conclusions [9,11]. Another novelty of the present investigation was in performing time-to-exhaustion exercise at steady-state intensity until fatigue to assess the time completed. Detection of MLSS intensity is crucial since a substantial portion of aerobic training in athletes is carried out at MLSS intensities, specifically in well-trained cyclists [19,42]. In the present investigation, both groups improved in the MLSS test pre–post-intervention regardless of the nutritional intervention of the protocol. Furthermore, we did not find any improvement in total time-to-exhaustion exercise in either group (Figure 2a). These results suggest that the MLSS improvement observed in the present study may be due to the training, not the dietary intervention. As described in the study by Mendes et al. (2013), training per se increases the MLSS but does not improve the time to fatigue at MLSS in athletes [43]. This finding suggests that in well-trained athletes, periodized CHO intake does not translate into superior sustained adaptations for enhancing endurance performance compared to a high-CHO diet.

The present study found no differences in substrate utilization and metabolism during the MLSS test (Table 4). Most of the existing studies applied a submaximal test at a certain percentage of the PPO followed by a test until exhaustion [8,9,11,12,14,15], which makes it hard to assess whether the results obtained in the measurements of the use of energy substrates during the submaximal tests were carried out at the same physiological milestone measured with lactate [44]. Therefore, in the present study, we tried to fill this gap by measuring the use of energy substrates for a specific physiological milestone during the TTE test (MLSS intensity, verified with the final lactate of the TTE). Previous researchers obtained similar results to the present study without observing changes in substrate utilization [13,14,16]. In all of them, it was possible to observe no change in the oxidation of substrates after the previous ingestion of a meal rich in CHO (usually carried out before races) before and after main performance tests. For instance, Riis et al. (2019) described that after CHO intake, there is hyperinsulinemia which attenuates the use of fat as fuel in “low” groups [15]. In contrast, other authors reported that carbohydrate and lipid oxidation changes, seeing increased oxidation of lipids at submaximal intensity in groups following a low-carbohydrate, high-fat diet [10,11,12]. Furthermore, studies such as that of Impey et al. (2016) [45] found greater fat oxidation during exercise after carrying out the “sleep-low” strategy. The factors mentioned above, such as previous intake (low in carbohydrates), could condition the oxidation of test substrates, unlike this other study [46]. Lastly, a low-carbohydrate, high-fat diet results in a decrease in the oxidation of CHO during steady-state exercise [11]. This suggests that training with low CHO may be counterproductive for athletes who compete in high-intensity events where CHO oxidation plays a significant role in performance [47]. Finally, despite the different studies that have tried to show that low-carb diets have their place with the premise of increasing fat oxidation during exercise [45] and saving glycogen as a consequence, most of the articles do not endorse it, above all, when a previous intake of CHO (real conditions) is carried out before exercise [48]. Therefore, periodized CHO intake does not show superior metabolism adaptations compared to a high-CHO diet and may also impair training sessions and following training improvements due to lower glucose oxidation and carbohydrate utilization, which may affect the duration and quality of training [2,49,50,51,52].

Notably, we found decreased fat and increased muscle mass in both groups (Table 5). Marquet et al. (2016) [8], matching the caloric intake between groups, obtained significant body fat losses only in the sleep-low (SL) group, which, unlike the current study, was attributed to greater fat oxidation during exercise. In the present study, the general improvement in the anthropometric parameters was as expected since, during the five weeks of intervention, there was a caloric reduction in both groups because the caloric intake was carefully controlled. In contrast to the Marquet study [8], the BM decreased (by ~1 kg) only in the SL group, which was mainly attributed to a 1.1% decrease in fat mass. We found significant weight loss in both groups, which was one of the possible explanations for the performance improvement observed in the SL group of that study. This could suggest that, given the difficulty in controlling the diet, the two groups did not consume the same number of calories because one of the main factors when it comes to improving body composition is the caloric intake in the diet [51].

Lastly, this study faced some limitations. When conducting the research with well-trained participants, obtaining muscle biopsies was denied due to the invasive technique. Therefore, the potential underlying physiological mechanisms were not investigated. Another limiting factor was that the participants had to follow a controlled diet (weighing all foods) and a structured training plan for five weeks, which reduced and limited the number of participants. Furthermore, a periodized CHO diet is not a “natural” nutrition approach to training in competitive athletes due to their ad libitum nutritional characteristics as well as the requirements of both training and competition. Further cellular and molecular studies are warranted to prove any advantages of a periodized CHO intake over a normal or higher CHO diet in well-trained athletes.

The results of the present study show that periodization of CHO vs. a high-CHO diet during five weeks of supervised exercise training in well-trained athletes does not influence MLSS and does not change substrate oxidation (CHO and LIP) during a time-to-exhaustion test at MLSS intensity. Similarly, it can be concluded that both diets effectively improve anthropometric parameters and exercise performance (watts in MLSS) if caloric intake and training are controlled. Further studies are needed to identify the specific cellular responses to different nutritional interventions and the timing of such interventions deployed to athletes and populations with chronic diseases.

We gratefully acknowledge and thank the individuals who volunteered to participate in this study. We also thank Carlos Jimenez for his assistance with data collection and material preparation.

All authors G.P.-B., J.D.-L., D.J.B., J.F.-S., J.A.-V., I.S.-M., and J.S.-C. contributed to the study’s conception and design G.P.-B. and J.S.-C. performed material preparation, data collection, and analysis. G.P.-B. wrote the first draft of the manuscript, and all authors commented on previous versions. All authors have read and agreed to the published version of the manuscript.

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of University of the Basque Country UPV/EHU (Spain) (CEISH 113/2019, 23 May 2019) for studies involving humans.

Informed consent was obtained from all subjects involved in the study. Written informed consent was obtained from the patients to publish this paper.

Data are contained within the article.

The authors declare no conflicts of interest.

This research received no external funding.