Impact of carbohydrate timing on glucose metabolism and substrate oxidation following high-intensity evening aerobic exercise in athletes: a randomized controlled study

ABSTRACT

Objective

The study aimed to investigate the impact of nutrient timing in relation to evening exercise. Specifically, it examined the effects of pre- or post-exercise carbohydrate (CHO) ingestion on glucose metabolism, glucose regulation, and overall substrate oxidation in well-trained athletes during and after physical exercise (PE), spanning the nocturnal period and the subsequent morning.

Methods

Ten male endurance cyclists participated in the study. The initial assessments included body composition measurements and an incremental cycle test to determine maximal oxygen uptake ($\dot{\mathrm{V}}$O2 max) and maximum power output (Wmax). Following this, participants underwent a control (rest previous day) oral glucose tolerance test (OGTT) and a familiarization exercise trial that had two objectives: (1) to establish the appropriate amount of CHO to use in the pre- or post-exercise drink during the experimental trials, and (2) to familiarize participants with the equipment and study protocol. In the three days prior to both the control and experimental trials, participants followed a standardized, individualized diet designed to meet their energy needs. During the experimental trials, participants completed two separate evening exercise sessions (50 min@70%Wmax +  ~24 min time-trial (TT)) with either pre- or post-exercise CHO ingestion (253 ± 52 g), matching the CHO oxidized during exercise. The CHO drink and a volume-matched placebo (PLA) drink (containing no energy) were randomly assigned to be consumed two hours before and directly after the experimental exercise sessions. Post-exercise nocturnal interstitial glucose levels (24:00–06:00) were continuously monitored, and a 120-min OGTT was conducted the following morning to assess substrate oxidation rates and glucose control.

Results

Pre-exercise CHO intake significantly lowered capillary glucose levels during steady-state exercise (mean difference 0.41 ± 0.27 mmol/L, p = 0.001) without affecting perceived exertion and TT-performance. No difference was observed in nocturnal glucose regulation (00:00–06:00) regardless of whether CHO was consumed before or after exercise. Post-exercise CHO ingestion reduced glucose tolerance during the OGTT compared to the iso-caloric pre-exercise CHO intake (mean difference 0.76 ± 0.21 mmol/L, p = 0.017). However, a post-exercise CHO intake improved respiratory exchange ratio/metabolic flexibility (MetF) significantly. Enhanced MetF during the first OGTT hour after post-exercise CHO ingestion resulted in 70% and 91% higher CHO oxidation compared to pre-exercise CHO and control, respectively (p ≤ 0.029). Average 120-min OGTT fat oxidation rates were higher with both pre- and post-exercise CHO ingestion compared to control (p ≤ 0.008), with no difference between pre- and post-exercise CHO intake.

Conclusion

Morning glucose tolerance was markedly reduced in healthy athletes when CHO was ingested after evening exercise. However, the observed improvements in MetF during the OGTT compared to placebo post-exercise suggest a potential for enhanced athletic performance in subsequent exercise sessions. This opens exciting possibilities for future research to explore whether enhanced MetF induced by CHO-timing can translate to improved athletic performance, offering new avenues for optimizing training and performance.

1. Introduction

Physical exercise (PE) significantly influences insulin sensitivity (IS), glucose control and overall metabolic health [1]. Prior studies have demonstrated enhancement of glucose uptake in skeletal muscle during prolonged moderate-intensity exercise [2] and high-intensity interval training (HIIT) [3]. This improvement in IS have been shown to persist for several days post-exercise [4,5]. While PE effectively enhances IS and glucose regulation, the timing of nutrient intake, before and after exercise, plays a crucial role in modulating its effects[6,7].

For competitive and recreational athletes, nutrient timing involves strategically adjusting nutrient intake around exercise sessions to enhance performance, support recovery, and promote adaptation.⁷ Pre-exercise carbohydrate (CHO) timing aims to ensure sufficient liver and muscle glycogen levels for the forthcoming physical activity. A post-exercise meal aims to transition the exercise-induced catabolic state into an anabolic state [8]. A recent meta-analysis [9] draws attention to replenishing endogenous substrate stores (e.g. muscle- and liver glycogen) following exercise for subsequent recovery and optimal physical performance. Furthermore, the depletion of liver glycogen is associated with an increase in the risk of hypoglycemia [10,11], coupled with the subsequent increase in cortisol secretion, which may induce short-term insulin resistance [12] (IR) and skeletal muscle proteolysis [13–17], and emphasizes the significance of post-exercise CHO availability. Attention to CHO intake timing may be crucial when limited recovery time exists between exercise bouts or when disrupted meal intake occurs between training sessions due to overnight fasting [8].

In contrast to studies demonstrating the benefits of rapid nutritional recovery, several studies show that immediate post-exercise energy/CHO intake could attenuate the exercise-induced improvements in IS and glucose control 12-15 h after exercise compared to retaining the PE-induced energy/CHO deficit [18–21]. The underpinning mechanisms for impaired glucose control the day after a single bout of exercise is not yet fully understood [22]. These studies did not individualize the total daily energy and CHO intake levels, so their results should be interpreted cautiously [23].

The present study aims to evaluate how pre- or post-evening exercise CHO ingestion influences glucose metabolism and substrate oxidation (fat/CHO) during exercise and after exercise in athletes during the nocturnal period and the morning after PE.

Although each participant followed an identical, individualized diet (carbohydrates, protein, and fat) that maintained energy balance during all trials, we hypothesized that glucose tolerance and metabolic flexibility (MetF) would be reduced if CHO were consumed after exercise during an oral glucose tolerance test (OGTT) the following morning, compared to an iso-caloric CHO intake before the evening exercise.

2. Material and methods

Ten healthy, well-trained male endurance cyclists/triathletes, aged 19–45 years, were recruited via local contacts from the Gothenburg region, Sweden. The study cohort characteristics are shown in Table 1. Eligible participants were adults aged 18 to 50 years with an endurance fitness level ≥50 mL O₂/kg/min who provided signed consent. The study was approved by the Swedish ethical review authority (Dnr: 2022 -03,056-01) and was registered at ClinicalTrials.gov (NCT06400836). All procedures were carried out in accordance with the ethical standards of the Helsinki Declaration of 2013.

Table 1.

Characteristics of study population (n = 10).

	mean±SD
Age (years)	37.2 ± 6.3
Height (cm)	179.1 ± 5.8
Weight (kg)	77.1 ± 4.3
BMI (kg/m2)	24.0 ± 1.1
Lean mass (kg)	60.2 ± 4.3
Fat mass (%)	18.5 ± 4.5
Wmax (W) at $\dot{\mathrm{V}}$O2max	357 ± 46.6
70% Wmax (W)	250 ± 32.6
$\dot{\mathrm{V}}$O2max (mL/kg/min)	62.0 ± 6.5

Abbreviations: BMI, Body Mass Index.

2.1. Experimental design

The study was conducted using a double-blind, randomized, placebo-controlled crossover design.

The primary outcome measure was to investigate glucose tolerance assessed during a 75 g glucose, 120-min OGTT. The secondary objective was to evaluate substrate oxidation (CHO and fat) during the OGTT. The study included seven visits to an exercise laboratory. Visit 1 involved body composition assessment using dual-energy X-ray absorptiometry (DXA) and determination of maximal oxygen uptake ($\dot{\mathrm{V}}$O₂max) and maximum power output (Wmax) through an incremental cycle test. A detailed description of body composition estimation and exercise capacity tests is available in Supplementary Table S1. Visit 2 featured a control OGTT and a familiarization exercise trial. On Visit 3, three days prior to the experimental exercise trial, subcutaneous insertion of a glucose sensor for continuous glucose monitoring (CGM) took place. The experimental exercise trials occurred on Visits 4 and 6, with OGTTs on Visits 5 and 7 (Table S1). The dietary intake for each participant was calculated to provide the appropriate amount of energy, carbohydrates, protein, and fat per kg/body weight. The energy expenditure estimations during the trial are presented in Table S2, while the data on energy and macronutrient intake are available in Table S3. Three days preceding both control and experimental trials, participants received pre-portioned food that was distributed to them, ensuring that all consumed food was provided by the study. They were also instructed on the specific times to consume their meals. Current compliance with meal timing was checked using CGM and analysis of glucose data. Participants were allowed to drink coffee and tea the day before the familiarization- and testing sessions. Water intake was managed by the participants themselves. Participants were not permitted to take any medications or dietary supplements. For the study design (schematic), see Figure 1.

Figure 1.

Schematic study design.

Day -3: Application of intermittent scanned continuous glucose monitoring (isCGM) sensors. Days -2 to 1: Standardized diet, with a 60-min light bicycle exercise on day -2. Day 1: At 17:00, randomized ingestion of carbohydrate (CHO) or placebo drink. At 19:30, exercise at 70% Wmax and time trial performance. Post-exercise, participants switched to the opposite drink (CHO or placebo). Day 2: At 07:15, measurement of substrate oxidation followed by a 120-min oral glucose tolerance test (OGTT) after an overnight fast. These procedures were repeated the following week with reversed drink order (CHO or placebo).

2.2. Familiarization exercise trial

The familiarization trial served two purposes: Determining CHO amount for pre- or post-drink and familiarizing participants with the equipment and procedures. It involved a 50-minute submaximal workload and an individual time trial (TT) with continuous oxygen uptake measurement, where participants aimed to complete a workload (kJ) based on their Wmax, as quickly as possible, equivalent to 25 min of cycling at 80% $\dot{\mathrm{V}}$O₂max, using a formula inspired by Jeukendrup et al. [24]

Target work (kJ) = W@80% of $\dot{\mathrm{V}}$O₂max × 1.5.

2.3. Experimental exercise trials and pre/post exercise drink content

In this double-blind cross-over study, a random sequence generator provided the order of CHO/placebo drinks, where each individual received both drinks, but in random order.

Two hours before the exercise trial, which started at 19:30 h, the participants consumed either a CHO drink or a placebo drink (no energy). Immediately after completing the exercise trial, the participants switched to the opposite condition. The CHO drink was a commercial soda (Freeway Orange, Lidl AB, Germany) containing 11.2% CHO. To match CHO intake with the CHO oxidized during exercise, the soda was enriched with a nonsweet CHO powder (112 ± 23 g maltodextrin; Body Science, MM sport, Gothenburg, Sweden). Thus, the final maltodextrin-supplemented soda contained up to 20% of CHO with a mean CHO content of 253 ± 51 g. The volume-, flavor- and color-matched placebo drink (Premier Orange, Willys, Sweden) contained no energy/CHO.

The trial started with a 5-minute warm-up at 40% of Wmax (~50% $\dot{\mathrm{V}}$O₂max), followed by 50 minutes of cycling at 70% of Wmax (~72% of $\dot{\mathrm{V}}$O₂ max). Capillary glucose levels (Biosen C-line, EKF diagnostics GmbH, Magdeburg, Germany), HR and rate of perceived exertion (RPE; Borg category scale 6–20) were measured 5, 20, 35 and 50 minutes after the start of the exercise bout. Respiratory gases were collected during 5 min periods at every 15 min interval and mean values during the final 60 s at each measurement point were used to assess substrate utilization via RER. Rates of total CHO and fat oxidation were calculated from $\dot{\mathrm{V}}$O₂ and $\dot{\mathrm{V}}$CO₂ (L/min) using stoichiometric equations [25,26] (Table S4) with the assumption that protein oxidation during exercise was negligible.

Following the 50-min submaximal exercise, participants had a 1-minute rest period followed by a 5-minute cool-down at 40% of Wmax, culminating in the TT as described for the familiarization trial. Three min post-TT, a capillary blood sample was collected.

2.4. OGTT and substrate oxidation measurements

The participants arrived at the lab in the morning after approximately a 9-hour fast, and their hydration status was initially examined using a refractometer. In accordance with the American College of Sports Medicine’s (ACSM) hydration testing guidelines [27], each participant was instructed to provide a mid-flow urine sample from their first void in the morning upon waking (i.e. the morning of the OGTT). The Urine Specific Gravity (USG), reflecting hydration status, was assessed by refractometry (Atago PAL-10S, Tokyo, Japan), with euhydration defined as USG < 1.025. Substrate oxidation was measured at 07:15 in a seated position.

Following this, study participants consumed a 300 mL solution containing 75 g glucose (dextrose monohydrate) (Topstar 75 lemon, Esteriplas Ltd, Portugal) for a 120-min OGTT. Capillary blood samples were collected at baseline and every 15 minutes during the OGTT. Gas exchange was measured for five-minute periods every 15 minutes, resulting in a total of 8 measurements. The stoichiometric equations used during rest (OGTT) were: [25]

Carbohydrate utilization at rest (g/min) = (4.585 • $\dot{\mathrm{V}}$CO₂) ˗ (3.226 • $\dot{\mathrm{V}}$O₂)

Fat utilization at rest and during exercise (g/min) = (1.695 • $\dot{\mathrm{V}}$O₂) ˗ (1.701 • $\dot{\mathrm{V}}$CO₂)

Mean values from the final 60 seconds of each measurement point were used to assess substrate utilization (CHO and fat) via RER. MetF was determined as the change in RER from time point 0 to 60 minutes after glucose ingestion (δRER = RER 60 min – RER 0 min). Fat oxidation rates during the OGTT were calculated as previously described, and CHO oxidation was determined using a stoichiometric equation (Table S5).

2.5. Continuous glucose monitoring

An intermittently scanned continuous glucose monitoring (isCGM) system was utilized to monitor nocturnal interstitial glucose levels (FreeStyle Libre, Abbott, Berkshire, UK). On day −3 (Figure 1), a factory-calibrated [28] glucose sensor was subcutaneously inserted into the back of the right upper arm. The interstitial glucose data were subsequently uploaded to the LibreView software (Abbott, Berkshire, UK) for analysis.

2.6. Sample size estimation

Sample size estimation was performed using G*Power software (v3.1.9.7) based on postprandial glycemic responses (1.0 ± 0.2 mmol/L) after exercise with and without energy deficit replenishment (350 kcal) [20]. With 90% power and α = 5%, it was determined that n = 3 participants would be sufficient to detect differences in the 0–120 min mean capillary glucose AUC following intense 75-minute cycling exercise. However, since the estimation focused on energy deficit replenishment rather than the present study’s objective of carbohydrate (CHO) timing, a larger sample size (n = 10) was deemed necessary. Consequently, we included 10 participants, a sample size consistent with previously published research that investigated how macronutrient intake following exercise influences next-day insulin sensitivity (n = 7–9) [29–31].

2.7. Statistical analysis

All data were checked for normality using a Shapiro – Wilk test. Two-way repeated measures ANOVA (time × treatment) with Bonferroni post-hoc adjustment was utilized to identify the location of significant differences (p < 0.05) when the analysis of variance yielded a significant F-ratio in time-dependent variables (glucose, CHO and fat oxidation, HR, RER, rating of perceived exertion (RPE), and lactate levels, respectably). Non-time-dependent variables were analyzed with two-tailed paired sample t-tests. Nocturnal glucose AUC and AUC during the OGTT were calculated using the trapezoidal rule in Excel [32]. Group mean differences were assessed through one-way repeated measures ANOVA, with Bonferroni correction for multiple comparisons when needed. Missing isCGM glucose values were imputed using the mean value for that specific time point across all participants. Results are expressed as mean ± standard deviation. All statistical analyses (except AUC) were performed with SPSS for Windows version 28 (Armonk, NY, USA), and figures were generated using GraphPad Prism, Version 9 (San Diego, CA, USA).

3. Results

3.1. Blood metabolites, substrate oxidation, heart rate and RPE during exercise

Pre-exercise glucose levels significantly differed between CHO Pre-EX (6.21 ± 1.10 mmol/L) and CHO Post-EX (4.47 ± 0.25 mmol/L) (p < 0.001). All participants in the CHO Pre-EX trial experienced hypoglycemia at some point during the submaximal steady-state exercise (hypoglycemia was defined as a glucose value <3.9 mmol/L [33]). The average glucose value during 5–50 minutes of submaximal exercise was 3.60 ± 0.37 mmol/L for CHO Pre-EX compared to 4.00 ± 0.38 mmol/L for CHO Post-EX (p = 0.001). Mean glucose concentrations were significantly lower in CHO Pre-EX compared to CHO Post-EX (mean difference 0.41 ± 0.27 mmol/L, 95% CI = 0.21 to 0.60, p = 0.001) (Figure 2).

Figure 2.

Plasma glucose concentration during submaximal exercise at 70% wmax. Panel (a) illustrate absolute values, while panel (b) present the mean value over the 50-minute period. Significant differences between groups are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as mean±SD.

Figure S1 displays mean RER values and substrate oxidation rates (CHO/fat). Exercise onset caused a gradual, significant decrease in RER across trials (main effect of time: p < 0.001). RER remained consistently higher in CHO Pre-Ex vs. CHO Post-EX (mean difference 0.04, 95% CI = 0.03 to 0.05, interaction effect p < 0.001), leading to significantly lower FAT oxidation (p < 0.001) and significantly higher CHO oxidation (p < 0.001) with pre-exercise CHO intake compared to pre-exercise PLA intake (CHO Post-EX).

At exercise onset, CHO Pre-EX showed higher HR compared to CHO Post-EX, mean difference 6 ± 5 BPM/min (p = 0.004) (Figure S3). This significant HR difference persisted throughout the 50-min submaximal exercise (p = 0.032), with no significant RPE difference between trials (p = 0.338).

CHO Pre-EX exhibited 20% higher absolute CHO oxidation than CHO Post-EX during 50 min at 70% Wmax +TT exercise (276 ± 34 g and 230 ± 34 g, respectively; p = 0.022). Conversely, fat oxidation was significantly lower (−63.0%) in CHO Pre-EX compared to CHO Post-EX (mean difference 17.5 ± 5.8 g; 95% CI = 13.0 to 21.9 g; p < 0.001). However, these substrate utilization differences did not affect TT performance (p = 0.189), with average completion times of 1477 ± 100 s and 1442 ± 78 s for CHO Pre-EX and CHO Post-EX, respectively.

3.2. Nocturnal glucose regulation

Glucose AUC significantly increased in CHO Post-EX during 21:00–24:00 compared to control (p < 0.001) and CHO Pre-EX (p = 0.007) (Figure 3a,b). However, no significant differences were observed in AUC across the three trials preformed between 24:00–03:00 and 03:00–06:00.

Figure 3.

Nocturnal glucose control after high intensity evening exercise. Panel (A) illustrate absolute glucose values during the period 21:00–06:00. Panel (B) shows the glucose area under the curve (AUC) for 3 h periods throughout the night. Significant difference (p < 0.05) between groups in panel (A) are indicated as; control and CHO pre-EX (a), control and CHO post-EX (b), CHO-Pre-EX and CHO post-EX (c) respectively. Significant differences between groups in panel (B) are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as mean±SD.

3.3. Glucose control and substrate oxidation during OGTT

Pre-OGTT glucose levels were similar across control, CHO Pre-EX, and CHO Post-EX (4.7 ± 0.2, 4.6 ± 0.2, and 4.6 ± 0.4 mmol/L respectively (p > 0.05)), as shown in Figure 4a. Glucose concentrations were consistently higher in CHO Post-EX between 60 and 90 min compared to CHO Pre-EX (p < 0.05). Moreover, mean glucose concentrations during OGTT were 13.1% and 9.5% higher when CHO were consumed post-exercise compared to both CHO Pre-EX (p = 0.017) and control (p = 0.049), with no significant difference observed between control and CHO Pre-EX (p = 0.875).

Figure 4.

Postprandial glucose response during a 120-min oral glucose tolerance test (OGTT) expressed as absolute concentration (a) and as a mean glucose value during the 120-min period (b). Oxidation rates of carbohydrates and fat during the OGTT are illustrated in (c) and (d) respectively. Significant differences between groups in panel (b), (c) and (d) are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as mean±SD.

RER, CHO, and fat oxidation rates (g/min) during OGTT are shown in Supplementary Figure S4. In the fasted state, RER was significantly lower (p = 0.008) when CHO was consumed post-exercise the evening before (0.77 ± 0.01), compared to control (RER = 0.82 ± 0.10), with no difference between CHO Post-EX and CHO Pre-EX (p = 0.214). Following glucose ingestion, RER gradually increased (main effect of time p < 0.001), but no significant time × trial interaction effect was observed (p = 0.250). However, the mean 120-min RER was significantly higher in control compared to both CHO Pre-EX (p = 0.027) and CHO Post-EX (p = 0.015).

Mean CHO oxidation during the OGTT didn’t differ significantly among the three trials. Time (p < 0.001) and treatment (p = 0.034) had significant effects, but no significant interaction effect (p = 0.308). Mean 120-min CHO oxidation rates (Figure 4c) for control (0.23 ± 0.06 g/min), CHO Pre-EX (0.20 ± 0.06 g/min), and CHO Post-EX (0.19 ± 0.07 g/min) were not significantly different.

Fasted state fat oxidation rate pre-OGTT was 31% higher in CHO Post-EX compared to control (p = 0.003), but not significantly different compared to CHO Pre-EX (p = 0.250). Average 120-min OGTT fat oxidation rates were 37.7% higher in CHO Post-EX group and 29.5% higher in CHO Pre-EX group (both p = 0.008), compared to control group, but no significant difference between CHO Pre-EX and CHO Post-EX was observed (p = 0.749) (Figure 4d). Two-way ANOVA revealed significant main effects for time (p < 0.001) and treatment (p < 0.001), but no significant interaction effect was observed between time and treatment (p = 0.193).

MetF (δRER) after glucose ingestion during 0-60 min was 80% higher in CHO Post-EX (0.11 ± 0.28) compared to CHO Pre-EX (0.06 ± 0.31) (p = 0.001), with no significant difference compared to control (0.06 ± 0.46) (Figure S4). This increase in MetF during the first hour of OGTT is reflected by a 70% and 91% higher increase in glucose oxidation for CHO Post-EX compared to CHO Pre-EX (p = 0.010) and control (p = 0.029) (Figure 5). Additionally, there was an 89% increase in the suppression of lipid oxidation for CHO Post-EX compared to CHO Pre-Ex (p = 0.006), but no difference between control and CHO Pre-EX (p = 0.098).

Figure 5.

Metabolic flexibility illustrated as the changes in substrate oxidation, specifically the increase in carbohydrate oxidation and the suppression of fat oxidation following the ingestion of a 75 g glucose load during an oral glucose tolerance test (OGTT), measured from time point 0 to 60 minutes after glucose ingestion. Panel a presents the increase in carbohydrate oxidation, while panel B depicts the suppression in fat oxidation. Significant differences between groups are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001. Data are presented as mean ± SD.

4. Discussion

This study has demonstrated that well-trained athletes who meet their daily energy and CHO requirements on a day with evening exercise experience reduced glucose tolerance during an OGTT the following morning when CHO is replaced immediately after exercise compared to a placebo post-exercise. However, CHO replacement after evening exercise resulted in a significantly improved MetF the following morning during the OGTT, as evidenced by a significant increase in CHO oxidation and suppression of fat oxidation during the first hour of the OGTT, compared to both CHO Pre-EX and the control (no exercise).

4.1. Oral glucose tolerance test

The present study observed a significant deterioration in glucose tolerance during the OGTT the morning after CHO ingestion post-exercise. This finding is in agreement with previous research showing that post-exercise CHO ingestion, but not post-exercise fat intake [31], leads to attenuated IS and poorer glycemic control the following morning [18,21,29]. A mechanistic explanation of why Post-EX CHO ingestion negatively affected next-morning glucose tolerance is not clear. Flockhart et al. [34] recently demonstrated that endurance-trained athletes, unlike untrained controls, experienced reduced glucose tolerance and increased IR during an OGTT the day after prolonged continuous exercise. This finding was linked to the athletes’ enhanced fat utilization during and after exercise. They found the factors closely associated with the marked glucose tolerance difference included RER, fasting serum FFA, and serum β-hydroxybutyrate values. The increased fat oxidation during the OGTT is in line with what was observed in the present study following the CHO-Pre and CHO-Post trials, but not during control.

Interestingly, one previous study [35], with similar design as the present study, wherein the exercise-induced CHO/energy deficit was matched with either a pre- or post-exercise nutritional intake, found no significant increase in fat oxidation before and during glucose infusion when CHO was consumed pre-exercise. However, contrary to our findings, Stephan et al. observed greater improvements in next-morning IS following post-exercise energy replacement compared to ingesting an iso-caloric meal before exercise. The divergent results in next-day glucose control may be attributed to differences in fat oxidation. The exercise-induced energy expenditure was significantly higher in the present research and Flockhart et al.‘s study compared to the results reported by Stephan et al.

In the present study, post-exercise CHO intake worsened OGTT glucose tolerance compared to pre-exercise CHO intake, despite similar fat oxidation rates during the OGTT. Some studies suggest minor changes in glycogen content can improve IS and glycemic control post-exercise [36]. Estafanos et al. [18] found improved post-meal glucose levels in untrained women after a low-volume HIIT session expending merely 173 kcal and 43 g CHO, without CHO replacement afterward. In the present study, all participants consumed an identical individualized diet before and on the day of the trials, so it can be assumed that they had similar glycogen content before receiving the pre-exercise drink. When CHO were consumed before evening exercise, the participants most likely had higher glycogen content during exercise compared to when they received a placebo. This may explain the higher CHO oxidation (+46 g) observed compared to consuming a placebo before exercise. Higher CHO oxidation during exercise with pre-exercise CHO consumption suggests lower muscle glycogen content during the OGTT. Reduced glycogen content may explain why CHO Pre-EX didn’t show reduced glucose tolerance, possibly due to increased non-oxidative disposal (glucose storage). Prior research has shown reduced glycogen content can enhance glycogen synthase activity and accelerate muscle glycogen resynthesis[37].

An intriguing finding in current study is that post-exercise CHO intake enhanced MetF during OGTT compared to pre-exercise CHO intake and control. The partially replenished glycogen content as in the CHO Post-Ex condition reduces glucose-to-glycogen conversion [38], promoting early OGTT glucose oxidation. Elevated CHO oxidation decreases fat oxidation rates due to diminished carnitine and carnitine palmitoyl transferase 1 (CPT1) levels[39].

The observation that endurance athletes with high metabolic fitness exhibit a decrease in MetF when consuming CHO before exercise may seem unexpected. However, this shift in metabolism could be interpreted as an adaptive response to the metabolic conditions and demands that arise post-exercise in well-trained endurance athletes, particularly favoring lipid metabolism and sparing glucose. Decreased muscle glycogen in the CHO Pre-Ex trial enhances GS activation [37], favoring glycogen storage over glucose oxidation, thereby altering RER. In our study, fat oxidation was significantly increased during OGTT in both CHO Pre-Ex and CHO post-Ex trials compared to control (rest). In a study by Phielix et al. [40] lipid infusion reduced both oxidative and non-oxidative glucose disposal in untrained, whereas athletes with high mitochondrial oxidative capacity increased glycogen synthesis and reduced glucose oxidation in favor of lipid oxidation, supported by dephosphorylation of glycogen synthase.

To optimize physical performance, endurance athletes depend on adequate glycogen stores and a high MetF, as it enables them to adapt to varying energy demands during training and competition [41]. The present study has shown that a single acute post-exercise CHO replacement meal after evening exercise promotes next-day MetF during rest in well-trained endurance athletes.

From a health perspective, MetF has been linked to insulin resistance [42]. Similar to research conducted on patients or in at-risk subjects, MetF in the present study was measured at rest. However, more subtle differences in fuel selection might be revealed when the oxidation rates are high during training or competition, and it is unknown whether an increased MetF would affect physical performance. Further research is needed to explore the implications of this finding on substrate utilization during subsequent exercise sessions.

4.2. Nocturnal glucose regulation

This study indicates that to achieve stable nocturnal glucose levels, nutrient timing, specifically CHO replacement before or after evening exercise, is of secondary importance when daily energy and CHO needs are met. There was no difference in nocturnal glucose regulation between the hours of 00:00–06:00, regardless of whether CHO replacement (253 ± 52 g of CHO) was consumed before or after evening exercise. Jeukendrup et al. [43] have shown that increased CHO availability during exercise reduces or completely suppresses endogenous glucose production by the liver, similar to the CHO Pre-EX condition in the current study, leading to higher liver glycogen content at the end of exercise, a factor that promotes nocturnal glucose regulation after exercise.

4.3. Strengths and limitations

The strength of the present study lies in its standardized exercise and diet interventions. The participants were given a diet to match their total daily energy and macronutrient requirements. The majority of previous studies investigating energy and CHO replacement in conjunction with exercise and the subsequent effect on IS have not controlled the diet to ensure that energy and macronutrient balance (CHO, fat, protein) was maintained, often resulting in an energy or CHO deficit following exercise. Limitations of our study are the small sample size and that the study was performed only in young to middle-aged male athletes. The absence of biopsies to determine muscle glycogen levels was a limitation, as glycogen levels affect IS. Given that the proximity of meals to bedtime has been shown to negatively affect sleep quality [44] and since sleep quality can affect insulin sensitivity [45] future research should examine the influence of sleep quality on next-day glycemic control. Additional measurements such as: FFA, ketones, and insulin, could have been made. Considering well-trained individuals’ enhanced ability to utilize lipids both during and after exercise, and its impact on post-exercise glucose homeostasis, our findings may have limited generalizability both to sedentary individuals and clinically to those with metabolic inflexibility linked to obesity and type 2 diabetes [34].

5. Conclusion

The present study demonstrated that in well-trained healthy athletes, CHO ingestion after intense prolonged (~1.5 h) evening exercise hampered glucose tolerance at rest the following morning. Furthermore, post-exercise CHO replacement meals not only replenish glycogen content but may also positively enhance performance in subsequent exercise sessions through an enhanced MetF. Future research should explore whether changes in MetF, induced by nutrient timing, have implications on athletic performance during subsequent exercise sessions.

Funding Statement

This research received no specific grant from any funding agency in the public, commercial or non-profit sectors. Open access funding provided by Örebro University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Abbreviations

BM

Body Mass

CPT1

Carnitine Palmitoyl Transferase 1

CHO

Carbohydrates

CGM

Continuous Glucose Monitoring

HR

Heart Rate

IS

Insulin Sensitivity

IR

Insulin Resistance

isCGM

Intermittently Scanned Continuous Glucose Monitoring

MefF

Metabolic Flexibility

OGTT

Oral Glucose Tolerance Test

PAL

Physical Activity Level

PE

Physical Exercise

RPE

Rate of Perceived Exertion

RER

Resting Energy Requirements

TT

Time Trial

T1D

Type 1 Diabetes

Contributors

The study was designed by SP, FE and SM. Experimental data were collected by FE, JT and SP. Statistical analyses and data interpretation were performed by SM and the manuscript was prepared by SM, JJ, PA and SP. All authors (SM, FE, JT, PA, JJ and SP) read and approved the final version of the manuscript.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, reporting, or dissemination plans of this research.

Provenance and peer review

Not commissioned, externally peer reviewed.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15502783.2025.2494839