Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jul 5;604(15):6391–6404. doi: 10.1113/JP291000

Fasting‐induced glucose intolerance is not impacted by the addition of daily exercise: A randomized crossover study

Tori Bouck 1, Alexis Marcotte‐Chénard 1,2,3, Darren Barg 1, Charlie Kanuka 1, Roderick E Sandilands 1, Spencer Ursel 1, Garett Jackson 1, Arwyn MacIntosh 1, Sean McIntyre 1, Sriyashika Nekkanti 1, Kimberly Day 1, Hashim Islam 1,2, Jonathan P Little 1,2,✉
PMCID: PMC13423017  PMID: 42402811

Abstract

Abstract

Short‐term fasting (48–72 h) impairs insulin sensitivity and glucose tolerance, but it is unclear if aerobic exercise – which is known to have insulin‐sensitizing effects – can attenuate fasting‐induced decrements in glucose homeostasis. To determine whether the addition of exercise to a short‐term fast impacts glycaemic responses to refeeding, we implemented a randomized crossover design where 16 healthy adults (8 males and 8 females, 24±4 years) fasted for 48 h with and without the addition of daily cycling (50‐min moderate‐intensity continuous training + 5×1‐min high‐intensity interval training) performed at baseline and 24 h into the fast. Glycaemic responses to a mixed meal tolerance test (MMTT) were assessed via capillary measurements and continuous glucose monitoring (CGM) before and after the fast, as well as 24 h after refeeding. Capillary glucose and CGM incremental area under the curve (iAUC) during the 2‐h mixed meal tolerance test increased at 48 h fasted relative to baseline (P<0.001), which was unaffected by the addition of exercise (condition × time interaction, P>0.364). The 2‐h CGM iAUC reduced 24 h after refeeding compared to 48 h fasted (with or without exercise, P<0.001) but remained elevated relative to baseline (P = 0.01). Exploratory analyses revealed that 2‐h capillary (P = 0.020) and CGM (P<0.001) glucose iAUC and plasma glucose (P<0.001) were elevated to a greater extent after refeeding in females compared to males, and females exhibited greater impairments in CGM metrics in the 24 h after refeeding (P<0.01). Our results suggest that combined moderate‐intensity continuous training and high‐intensity interval training exercise does not attenuate fasting‐induced perturbations in glucose homeostasis and that glucose intolerance persists 24 h after refeeding.

graphic file with name TJP-604-6391-g003.jpg

Key points

  • Short‐term fasting reduces insulin sensitivity and glucose tolerance, which manifests as exaggerated glycaemic responses to refeeding.

  • Performing exercise during a fast could mitigate fasting‐induced glucose intolerance, due to the well‐established insulin‐sensitizing effects of acute exercise.

  • Here we report that a 48‐h fast performed with or without the addition of daily aerobic exercise similarly impaired glucose tolerance upon refeeding and resulted in comparable decrements in continuous glucose monitoring–derived metrics of glycaemic regulation over the 24 h after refeeding.

  • Exercise appears ineffective in reversing fasting‐induced glucose intolerance, which could reflect the mechanistic distinction between fasting‐induced and pathological forms of insulin resistance.

Keywords: CGM, continuous glucose monitor, fast, high‐intensity interval training (HIIT), insulin resistance

Abstract figure legend In a randomized crossover design 16 healthy young adults performed a 48‐h fast with or without the addition of daily cycling (50‐min moderate‐intensity continuous training (MICT) + 5×1‐min high‐intensity interval training (HIIT)) at baseline and 24 h into the fast. Glycaemic responses to a mixed meal tolerance test (MMTT) were assessed before and after the fast. As expected fasting elicited a significant reduction in glucose tolerance, which was unaffected by the addition of daily exercise. Exploratory sex difference analyses suggested glucose tolerance was impaired to a greater extent after fasting in females compared to males.

graphic file with name TJP-604-6391-g003.jpg

Introduction

Fasting has gained increasing attention among the scientific community and the general public, with preliminary evidence from preclinical models and clinical research suggesting the therapeutic potential of various fasting regimens for the prevention and management of cardiometabolic, neurodegenerative, inflammatory and oncological chronic diseases. One type of fasting regimen that is increasing in popularity is short‐term fasting, which is defined as a period of 2–3 days without caloric intake (Koppold et al., 2024). Much of the potential therapeutic benefits of short‐term fasting have been attributed to the ‘metabolic switch’ that occurs in the absence of exogenous nutrient availability. Specifically short‐term fasting is characterized by a significant increase in adipose tissue lipolysis (Anton et al., 2018), which subsequently induces a shift in whole‐body metabolism towards fatty acid and ketone body oxidation, with an associated decrease in carbohydrate utilization (Andriessen et al., 2023; Kolb et al., 2021). It has been suggested that this shift in substrate utilization may facilitate favourable changes in various cellular processes, which could have therapeutic implications for the prevention and management of chronic diseases (Kolb et al., 2021).

One aspect of fasting physiology that has received limited attention is the refeeding period that inevitably follows a short‐term fast. Interestingly it appears that the metabolic adaptations to prioritize lipids that characterize short‐term fasting are maintained upon refeeding, as indicated by significant reductions in insulin‐stimulated glucose disposal (Castillo et al., 1991; Hoeks et al., 2010; Mansell & Macdonald, 1990; Soeters et al., 2008; van der Crabben et al., 2008; Webber et al., 1994; Bak et al., 2018) and glucose tolerance (Frank et al., 2013; Green et al., 2011; Johnson et al., 2006; Jørgensen et al., 2021) after short‐term fasting. This phenomenon is broadly viewed as an adaptive response to spare glucose for the brain during periods of starvation, which is likely secondary to the significant (∼2‐ to 5‐fold) increase in circulating free fatty acids (FFAs) in response to 2–3 days of fasting (Frank et al., 2013; Hoeks et al., 2010; Jensen et al., 2001). Elevated FFA can impair intramuscular glucose metabolism through direct competition between substrates (Hue & Taegtmeyer, 2009; Randle et al., 1964; Sugden et al., 2001) and/or lipid‐mediated inhibition of insulin signalling (Soeters et al., 2008; Vendelbo et al., 2012; Yu et al., 2002).

Contrary to fasting acute exercise is generally considered to produce pronounced insulin‐sensitizing and glucose‐lowering effects (Richter & Hargreaves, 2013). In the initial postexercise period increases in skeletal muscle glucose uptake occur independently of the classic insulin signalling pathway and are largely attributable to contraction‐mediated upregulation of GLUT4 trafficking to the cell membrane (Richter & Hargreaves, 2013). In the subsequent hours to day(s) after acute exercise, the contraction‐mediated increase in skeletal muscle glucose uptake is replaced by an enhanced sensitivity to insulin (Mikines et al., 1988) likely attributable to the effects of exercise on downstream targets of insulin signalling (Kjøbsted et al., 2019). However whether performing exercise during a short‐term fast can offset fasting‐mediated impairments in insulin sensitivity and glucose tolerance remains unclear. Because elevated lipid availability is a likely contributor to fasting‐mediated decrements in insulin sensitivity, the modulating effects of acute exercise on intramuscular lipid metabolism may be protective against lipid‐mediated insulin resistance in this context (Newsom et al., 2013; Schenk & Horowitz, 2007). Additionally the persistent reduction in skeletal muscle glycogen content with combined fasting and exercise may serve as a potent stimulus for glucose disposal upon refeeding, as glycogen content and/or resynthesis are key mediators of exercise‐associated improvements in insulin action (Jensen et al., 2011). It is therefore speculated that performing daily aerobic exercise during a short‐term fast could mitigate fasting‐induced perturbations in glucose homeostasis.

In the fasting and refeeding literature, the persistence of fasting‐induced perturbations in glucose tolerance remains largely unexplored, as a few studies have examined glucose homeostasis beyond the initial reintroduction of feeding/glucose. Because the mechanistic basis of fasting‐induced glucose intolerance is likely multifactorial, it is difficult to predict how quickly this effect is reversed upon refeeding. However preliminary data suggest that metabolic perturbations may persist in the hours and days after the initial refeed (Grey et al., 1989; Norton et al., 2007; Solianik et al., 2023; Scharf et al., 2022). Continuous assessment of glycaemia after a short‐term fast is therefore warranted to explore the time course of the reversal of fasting‐induced impairments in glucose homeostasis.

Therefore the primary aim of this study was to determine whether daily exercise can mitigate short‐term fasting‐induced reductions in glucose tolerance. The secondary aim was to explore the persistence of perturbations in glucose homeostasis after short‐term fasting and whether this is influenced by the addition of exercise during the fast. Furthering our understanding of the effects of short‐term fasting with and without the addition of daily exercise on the magnitude and persistence of decrements in glucose homeostasis upon reintroduction of food intake is critical in interrogating the broader implications of short‐term fasting as a therapeutic strategy.

Methods

Participants

Sixteen participants were recruited based on the following inclusion criteria: (1) between 19 and 35 years of age and (2) engaging in ≥150 min of moderate‐to‐vigorous physical activity (MVPA) per week. Exclusion criteria were as follows: (1) history of cardiometabolic diseases or inflammatory diseases (e.g. chronic obstructive pulmonary disease, rheumatoid arthritis, cardiovascular disease); (2) currently following a ketogenic diet, low‐calorie diet, periodic fasting regimen, or consuming ketogenic supplements (e.g. exogenous ketone drinks); (3) currently smoking cigarettes or vaping; (4) physical limitation that would impair the ability of the participant to perform exercise; (5) body mass index (BMI) >30 kg/m2; (6) diagnosed with cancer in the past 5 years; (7) regularly taking anti‐inflammatory drugs (e.g. ibuprofen, aspirin, naproxen); and (8) being unable to read or communicate in English. Physical activity readiness and self‐reported weekly minutes of MVPA were assessed using the Canadian Society for Exercise Physiology (CSEP) Get Active Questionnaire (GAQ). All data collection was performed at the University of British Columbia (Okanagan) in Kelowna, British Columbia, Canada. Verbal and written consent was obtained from all participants prior to data collection, and the study was approved by UBC's clinical research ethics boaard (H24‐01732) and was registered on ClinicalTrials.gov (NCT06737224). The study conformed with ethical standards set by the Declaration of Helsinki.

Experimental design

As outlined in Fig. 1, a randomized crossover design was implemented, with participants performing a 48‐h fast on two separate occasions separated by a washout period of 5 days between fasts. This allowed each experimental condition to begin on the same day of the week. Randomization was performed using the Sealed Envelope randomization tool with variable block sizes (2×4). In one condition participants performed an exercise protocol on a stationary bike (60 min total) at baseline (2 h postprandial) and after 24 h of fasting (FAST+EX), whereas in the alternate condition, participants abstained from exercise throughout the duration of the fast (FAST). In each condition participants arrived at the laboratory after an overnight fast and underwent a mixed meal tolerance test (MMTT; Ensure Plus Calories: 350 kcal, 50 g of carbohydrates, 11 g of fat, 13 g of protein), with capillary blood glucose and β‐hydroxybutyrate (BHB) measurements (Abbot Freestyle Precision Neo) taken at 0, 30, 60, 90 and 120 min. This MMTT allowed for a baseline assessment of glucose tolerance in the overnight fasted state, in addition to serving as a standardized meal to begin the 48‐h fast. Two hours after the MMTT a venous blood sample was taken to obtain standardized postprandial baseline measures for plasma glucose, insulin and FFAs. This baseline venous blood sample taken 2 h after the standardized drink (Ensure Plus Calories) marked the beginning of the 48‐h fast. In the FAST+EX condition participants performed the first exercise session after this baseline venous blood sample. The exercise session consisted of 50 min on a cycle ergometer (Lode Excalibur Sport, version 2.0, Groningen, the Netherlands) at a moderate intensity, targeting 60% of age‐predicted heart rate reserve (HRR). Participants then performed 5 × 1‐min intervals targeting 90% of HRR, with 1 min of passive recovery between intervals. Heart rate was continuously measured using the Polar H10 monitor (Polar, Kempele, Finland), and workload was adjusted throughout the exercise session to maintain the target heart rate. Capillary glucose and BHB were measured 24 h after the baseline blood draw, and in the FAST+EX condition, participants completed the same exercise session as the previous day. Forty‐eight hours after the baseline blood draw, participants returned to the laboratory and broke their fast, with the same MMTT performed at the beginning of the fast to assess glucose tolerance after fasting. Venous blood samples were taken before and 2 h after the MMTT. Participants were instructed to abstain from exercise for the remainder of that day, as well as to record their dietary intake and replicate this to the best of their abilities between conditions. However dietary intake was not explicitly controlled or monitored between conditions. Participants were also given an Ensure Plus Calories to consume in an overnight fasted state ∼24 h after breaking their fast as a final MMTT to assess glucose tolerance in free‐living conditions. Glycaemic responses to this MMTT were recorded via a continuous glucose monitor (CGM). Participants were asked not to exercise or consume any other food for 2 h after consuming the beverage.

Figure 1. Experimental design.

Figure 1

Open in a new tab

Capillary glucose and BHB

Capillary glucose and BHB were measured at 0, 30, 60, 90 and 120 min of the MMTT performed at the beginning and end of each fast. Capillary glucose measurements were used to calculate the 2‐h glucose incremental area under the curve (iAUC) during each MMTT using the trapezoidal method (Allison et al., 1995).

Continuous glucose monitoring

The day prior to the first experimental visit, a continuous glucose monitor (CGM) (Freestyle Libre 2) was placed on the participant's non‐dominant upper arm, which allowed for continuous measurement of interstitial glucose levels in 15‐min increments. Raw CGM files were extracted from LibreView software and cleaned by visual inspection. The 2‐h glucose iAUC beginning at the time the drink was consumed during each MMTT performed (beginning of fast, end of fast and 24‐h post fast) was calculated using the trapezoidal method (Allison et al., 1995). Viable data were then organized by date and time corresponding to each period of interest (i.e. during the three 24‐h periods after the reintroduction of normal eating; the first of which the diet was controlled between conditions, whereas the latter two 24‐h periods were in free‐living conditions). Timepoints with less than 70% viable data were excluded from the analysis. CGM metrics of interest, including mean glucose values, time in tight range (i.e. 3.9–7.8 mmol/L) and glycaemic variability (standard deviation (SD) and coefficient of variation (CV) of mean glucose), were calculated using the Diametrics platform (University of Exeter).

Blood sample collection and analysis

Venous blood samples were taken at baseline (2 h after first MMTT) and 2 h after the MMTT performed at 48 h to break the fast. To assess the metabolic responses to refeeding, baseline samples were compared to samples obtained 2 h after the MMTT at 48 h, allowing for a direct comparison of metabolic outcomes at 2 h postprandial in the overnight fasted versus 48‐h fasted state. Three millilitres of blood was centrifuged (20 min, 2000 g at 4°C), and plasma was carefully removed and frozen at –80°C for batch analysis. Plasma was analysed for glucose (Infinity Glucose Hexokinase, Thermo Fisher Scientific Inc., Middletown, VA, United States) according to the manufacturer's instructions. Plasma insulin was measured using an enzyme‐linked immunosorbent assay (ELISA) (Insulin ELISA Kit, Crystal Chem, Elk Grove Village, IL, United States). Plasma FFA were quantified using enzymatic assay (Wako NEFA‐HR(2) Assay, FujiFilm, Lexington, MA, United States).

Statistical analyses

Data from all participants were included in an intention‐to‐treat analysis. Normality was assessed through visual inspection of Q–Q plots and Shapiro–Wilk tests. Outcome variables that deviated significantly from a normal distribution were log transformed where appropriate. Data were assessed for influential data points using the interquartile range with a multiplier of 2.2 (Hoaglin & Iglewicz, 1987). Sensitivity analyses conducted with statistical outliers removed did not change the interpretation of results. Additionally all data points fell within the range of physiological plausibility and were therefore included in the final analysis. Data were analysed using linear mixed effects models, with the interaction between condition (FAST vs. FAST+EX) and timepoint as fixed effects; sex, order of conditions (i.e. FAST‐FAST+EX vs. FAST+EX‐FAST) and period (i.e. first vs. second condition performed) as covariates; and participant as a random intercept. CGM metrics in the 24 h after the fast (i.e. when diet was controlled between conditions) were analysed via a linear mixed effects model with condition as a fixed effect; sex, order of conditions and period as covariates; and participant as a random intercept. Capillary BHB and glucose responses to the MMTT at baseline and 48 h fasted were analysed via linear mixed effects model, with the interaction between condition (FAST vs. FAST+EX), visit (baseline vs. 48 h fasted) and timepoint (baseline, 30, 60 and 120 min post‐MMTT) as fixed effects; sex, order of conditions and period as covariates; and participant as a random intercept. Significant main effects of time and interactions were followed up with pre‐planned pairwise comparisons within conditions, with Tukey's correction applied post hoc. Outcome variables that exhibited a significant effect of sex in the model were followed up with exploratory analysis (Condition × Time × Sex) to determine where these sex‐based differences might lie. Data are presented as model‐derived estimated marginal means with 95% confidence intervals (CIs), and/or effect estimates for pairwise comparisons of interest with 95% CI. Data were analysed in R (version 4.5.1) using lme4, lmerTest and emmeans packages (Bates et al., 2015; Kuznetsova et al., 2017). Baseline characteristics are presented as means ± SD. Significance was set at P < 0.05.

Results

Participant characteristics

Participant characteristics are presented in Table 1. All participants completed the study with 100% adherence to the experimental protocol, with the exception of one person who experienced adverse side effects (nausea and vomiting) during the first fasting condition.

Table 1.

Participant characteristics (n = 16: 8 females and 8 males)

All Male Female
Age (years) 24.1 (3.9) 25.5 (3.9) 22.8 (3.7)
Body mass (kg) 71.5 (11.8) 76.1 (9.1) 66.8 (12.9)
Body mass index (kg/m2) 23.6 (2.4) 24.3 (1.8) 22.9 (2.7)
Self‐reported MVPA (min/week) 321 (178) 320 (196) 322 (174)
Open in a new tab

Data are presented as mean (SD).

Abbreviation: MVPA, moderate‐to‐vigorous physical activity.

Metabolic responses to fasting and refeeding

Metabolic responses to the MMTT performed at the beginning of the fast and 48 and 24 h after the reintroduction of normal eating are shown in Fig. 2. Significant main effects of time were observed for 2‐h capillary glucose and CGM iAUC, plasma FFA, plasma glucose, insulin and the ratio of plasma glucose to insulin, as well as capillary glucose and BHB levels over the course of the MMTT (P < 0.001). The 2‐h iAUC, as measured via CGM, was elevated during the MMTT performed at 48 h relative to baseline in both conditions (P < 0.001), and was significantly lower 24 h after the reintroduction of normal eating compared to the 48‐h timepoint (P < 0.001). However the 2‐h CGM iAUC in response to the MMTT 24 h after breaking the fast remained elevated relative to baseline (P = 0.014). The 2‐h capillary glucose iAUC was also significantly higher after the fast (P < 0.001); 2‐h postprandial plasma glucose and insulin were both elevated after the 48‐h fast relative to baseline (both P < 0.001), whereas the plasma glucose to insulin ratio was significantly lower (suggesting reduced insulin sensitivity, P < 0.001). A significant Condition × Time interaction was observed for plasma FFA (P = 0.044), with the FAST + EX condition exhibiting higher 2‐h postprandial plasma FFA at 48 h (P < 0.01). Plasma FFA at 2 h postprandial appeared to be marginally elevated at 48 h relative to baseline in both conditions, although this difference reached statistical significance only in the FAST + EX condition (FAST: P = 0.087, FAST+EX: P < 0.001).

Figure 2. Changes in the 2‐h CGM iAUC (incremental area under the curve) (A), 2‐h capillary glucose iAUC (B), plasma FFA at 2 h postprandial (C), plasma glucose at 2 h postprandial (D), plasma insulin at 2 h postprandial (E) and the plasma glucose to insulin ratio at 2 h postprandial (F), in response to an MMTT (mixed meal tolerance test) performed at baseline and after the 48‐h fast during each experimental condition. *Condition × Time interaction, P < 0.05; †main effect of time, P < 0.05; abaseline (BSL) versus 48 h, P < 0.05; dFAST versus FAST + EX at 48 h, P < 0.05. Data are disaggregated by sex, with open circles (°) indicating female participants and filled circles (•) indicating male participants. Data are presented as model‐derived estimated marginal means with 95% confidence intervals (CIs). n = 16.

Figure 2

Open in a new tab

BSL, baseline; CGM, continuous glucose monitor; FFA, free fatty acid; refeed, 2 h postprandial at 48 h; 48 h, MMTT performed at 48 h fasted; 24 h post, MMTT at 24 h after reintroduction of normal eating.

Capillary glucose and BHB levels during the MMTT (0, 30, 60, 90 and 120 min) at baseline and 48 h are shown in Fig. 3. A near‐significant Condition × Timepoint × Visit interaction was observed for capillary glucose responses to the MMTT (P = 0.064). A significant Condition × Visit interaction was observed for BHB (P = 0.001), with significantly higher BHB levels observed in FAST+EX versus FAST prior to the MMTT at 48 h in FAST+EX versus FAST (P < 0.01).

Figure 3. Capillary glucose (A) and BHB (β‐hydroxybutyrate) (B) at 0, 30, 60 and 120 min of aMMTT (mixed meal tolerance test) performed at baseline and after the 48‐h fast in each experimental condition. Data are disaggregated by sex, with open circles (°) indicating female participants and filled circles (•) indicating male participants. Data are presented as model‐derived estimated marginal means with 95% confidence intervals (CI). n = 16.

Figure 3

Open in a new tab

Baseline, MMTT performed in overnight fasted state to begin the fast; 48 h; MMTT performed at 48 h fasted.

Exploratory sex differences

A main effect of sex was observed for 2‐h CGM iAUC, with exploratory analyses (Time × Condition × Sex) suggesting this difference occurred only at 48 h (P < 0.001). Although the effect of sex in the model did not reach statistical significance for other outcomes (all P ≥ 0.1), exploratory Time × Condition × Sex analyses suggested plasma glucose (P < 0.001), insulin (P = 0.050) and the 2‐h glucose iAUC (P = 0.020) after the MMTT at 48 h were significantly higher in females compared to males. Females also appeared to have slightly elevated plasma FFA compared to males after refeeding (Time × Sex interaction, P = 0.016), although post hoc analyses did not reveal a statistically significant difference between sexes (P = 0.138).

Glucose homeostasis in the 24 h after the fast

Between‐condition differences in CGM‐derived metrics of glucose control in the 24‐h period after the reintroduction of normal eating are shown in Fig. 4. Thirteen of the 16 participants had greater than 70% viable CGM data in 24 h after at least one of the fasts and were therefore included in the final analysis. There was no significant effect of condition on mean glucose, glycaemic variability (SD and CV of mean glucose) or time in tight range (3.5–7.8 mmol/L) (all P > 0.1). A significant effect of sex was observed for SD, CV and time in tight range (all P < 0.01), with females exhibiting significantly higher SD, CV, and lower time in tight range relative to males. A CGM curve (averaged from all participants) over the course of the 24 hours following the reintroduction of nomal eating is shown in Fig. 5.

Figure 4. CGM metrics measured across 24 hours of refeeding.

Figure 4

Open in a new tab

Average glucose (A), time in tight range (3.9–7.8 mmol/L) (B), coefficient of variation of mean glucose (C) and standard deviation of mean glucose (D) in the 24 h after reintroduction of normal eating in each experimental condition. Data are disaggregated by sex, with open circles (°) indicating female participants and filled circles (•) indicating male participants. Data are presented as model‐derived estimated marginal means with 95% confidence intervals (CI). n = 13.

Figure 5. CGM curve across 24 hours of refeeding.

Figure 5

Open in a new tab

CGM curve (averaged from all participants) with 95% confidence intervals (CIs) in the 24 h after the reintroduction of normal eating after a 48‐h fast in each experimental condition.

Glucose homeostasis in the 72 h after the fast

Estimated marginal means for CGM‐derived metrics of glycaemic control in the 0–24 h (day 1), 24–48 h (day 2) and 48–72 h (day 3) after the reintroduction of normal eating are presented in Table 2. Within‐condition effect estimates and P‐values are presented in Table 3. Significant main effects of time were observed for mean glucose, SD and CV of mean glucose and time in tight range (3.5–7.8 mmol/L, P ≤ 0.001). In both FAST and FAST+EX, mean glucose, SD and CV were significantly elevated on day 1 relative to days 2 and 3, whereas time in tight range was significantly lower on day 1. There were no significant differences between days 2 and 3 for any CGM metric. No significant Condition × Time interactions were observed for any CGM metric. A significant effect of sex was observed for SD, CV and time in tight range. Exploratory analyses (Time × Condition × Sex) suggested that these sex‐based differences were driven by significantly higher SD and CV and lower time in tight range in females on day 1 (all P < 0.001), whereas there were no significant differences between sexes on days 2 and 3 (P > 0.3).

Table 2.

CGM metrics across 72 hours of refeeding

FAST FAST+EX P‐value
Day 1 Day 2 Day 3 Day 1 Day 2 Day 3 Condition x Time Time Condition
Mean glucose (mmol/L) 6.8 6.0 5.8 6.4 5.9 5.9 0.230 <0.001 0.332
(6.5, 7.1) (5.7, 6.2) (5.4, 6.1) (6.1, 6.7) (5.6, 6.3) (5.5, 6.2)
SD (mmol/L) 1.3 0.8 0.7 1.1 0.7 0.7 0.315 <0.001 0.354
(1.2, 1.5) (0.7, 0.9) (0.6, 0.8) (1.0, 1.3) (0.6, 0.9) (0.6, 0.8)
CV (%) 20.3 13.8 11.9 18.7 13.0 12.6 0.563 <0.001 0.547
(18.1, 22.4) (11.6, 16) (9.5, 14.2) (16.5, 20.9) (10.6, 15.3) (10.2, 15.0)
Percentage time in tight range (3.9–7.8 mmol/L) 77.1 95.7 97.5 82.1 96.9 96.1 0.490 <0.001 0.462
(72.0, 82.1) (90.5, 100) (92.0, 100) (76.9, 87.3) (91.5, 100) (90.4, 101.8)
Open in a new tab

Day 1, 0–24 h post breaking fast; day 2, 24–48 h post breaking fast; day 3, 48–72 h post breaking fast; CV, coefficient of variation of mean glucose; SD, standard deviation of mean glucose. Data by group presented as estimated marginal means (95% CI).

Table 3.

Effect estimates for CGM metrics in each 24‐hour period in the 72 hours following refeeding

FAST
Effect estimate (day 1–day 2) P‐value Effect estimate (day 1– day 3) P‐value Effect estimate (day 2– day 3) P‐value
Mean glucose (mmol/L) 0.8 (0.4, 1.2) <0.001 1.0 (0.6, 1.5) <0.001 0.2 (−0.2, 0.7) 0.523
SD (mmol/L) 0.5 (0.3, 0.7) <0.001 0.7 (0.4, 0.9) <0.001 0.2 (0.0, 0.4) 0.147
CV (%) 6.5 (2.9, 10.1) <0.001 8.4 (4.7, 12.1) <0.001 1.9 (−1.9, 5.7) 0.444
Percentage time in tight range (3.9–7.8 mmol/L) −18.6(–27.3, −9.9) <0.001 −20.4 (−29.3, −11.5) <0.001 −1.8 (−10.9, 7.24) 0.880
FAST+EX
Mean glucose (mmol/L) 0.5 (0.0, 0.9) 0.0346 0.6 (0.1, 1.0) 0.0117 0.1 (−0.4, 0.6) 0.880
SD (mmol/L) 0.4 (0.2, 0.7) <0.001 0.5 (0.2, 0.7) <0.001 0.0 (−0.2, 0.3) 0.881
CV (%) 5.7 (2.0, 9.5) <0.001 6.1 (2.2, 9.9) <0.01 0.3 (−3.6, 4.3) 0.976
Percentage time in tight range (3.9–7.8 mmol/L) −14.9(−23.9, −5.8) <0.001 −14.0 (−23.3, −4.8) <0.001 0.1 (−8.6, 10.2) 0.976
Open in a new tab

Day 1, 0–24 h post breaking fast; day 2, 24–48 h post breaking fast; day 3, 48–72 h post breaking fast; CV, coefficient of variation of mean glucose; SD, standard deviation of mean glucose. Within‐group effect presented as effect estimates (95% CI).

Data were analysed via linear mixed effects models with condition (FAST vs. FAST+EX), time (day 1 vs. day 2 vs. day 3) and the interaction between group and timepoint as fixed effects; sex, order of conditions (i.e. FAST‐FAST+EX vs. FAST+EX‐FAST) and period (i.e. first vs. second condition performed) as covariates; and participant as a random intercept. n = 13.

Data were analysed via linear mixed effects models with condition (FAST vs. FAST+EX), time (day 1 vs. day 2 vs. day 3) and the interaction between group and timepoint as fixed effects; sex, order of conditions (i.e. FAST‐FAST+EX vs. FAST+EX‐FAST) and period (i.e. first vs. second condition performed) as covariates; and participant as a random intercept. n = 13.

Discussion

The present study aimed to determine the effects of a 48‐h fast alone or combined with daily exercise on the magnitude and persistence of perturbations in glucose homeostasis upon refeeding. As expected fasting induced a significant reduction in glucose tolerance, which was still partially evident 24 h after the reintroduction of normal eating. Contrary to our hypothesis the observed glucose intolerance produced by fasting was not effectively mitigated by the addition of daily exercise.

As anticipated the 48‐h fast produced a significant degree of glucose intolerance, evident by an ∼3‐ to 4‐fold increase in both the CGM and capillary glucose iAUC response to the MMTT performed at 48 h fasted relative to the baseline MMTT performed in the overnight fasted state. Interestingly we found that performing exercise during the fast did not appear to influence glucose tolerance, as values were largely similar between conditions. Capillary glucose appeared to be slightly lower at the 2‐h timepoint during the MMTT performed at 48 h fasted in FAST+EX relative to FAST, with a medium effect size (d = 0.5), although the interaction effect did not reach statistical significance (P = 0.064). This finding could point towards a slight protective effect of exercise against fasting‐induced perturbations in glucose tolerance (as can be seen by the shape of the glucose curve in Fig. 3). However as plasma glucose at 2 h after the MMTT was not different between conditions, and no other metrics of glucose tolerance or glycaemic regulation appeared to be affected by the addition of exercise, it seems unlikely that this finding is indicative of any meaningful improvements in glycaemic regulation with combined fasting and exercise versus fasting alone. Although this result is somewhat surprising due to the well‐established glucose‐lowering and insulin‐sensitizing effects of acute exercise (Richter, E.A. & Hargreaves; Kjøbsted et al., 2019; Mikines et al., 1988; Schenk & Horowitz, 2007; Newsom et al., 2013), it could be hypothesized that the additional metabolic stress of exercise could have an additive effect on some of the fasting‐associated metabolic changes that contribute to glucose intolerance. Specifically it is possible elevations in FFA flux (Kersten, 2023) and/or skeletal muscle PDK4 transcription (Kolnes et al., 2025) that mediate fasting‐induced reductions in glucose oxidation, as well as changes in glucoregulatory hormones (Kersten, 2023), could be augmented by the addition of exercise during fasting (Lundsgaard et al., 2020; Athanasiou et al., 2023; Magkos et al., 2009; Pilegaard & Neufer, 2004) and therefore counteract any insulin‐sensitizing effects of exercise. Whereas we did not observe significant between‐condition differences in plasma FFA concentration during the fast, plasma FFA levels after refeeding were higher in the FAST+EX conditions, indicating a potential persistent elevation in lipid flux attributable to the prior exercise session (Magkos et al., 2009). Our results are in line with the observations of Green et al. (2011), who found that performing two bouts of exercise during a short‐term fast (80‐min treadmill running at 50% VO2max) failed to rescue insulin sensitivity and glucose tolerance as assessed by an intravenous glucose tolerance test (IVGTT). Although we anticipated the higher relative intensity of the moderate‐intensity continuous training (MICT) protocol implemented, as well as the addition to the high‐intensity interval training (HIIT) at the end of the exercise bout (∼60% VO2max), would elicit a more robust insulin‐sensitizing response due to greater depletion of muscle glycogen stores, it is possible that this stimulus for glucose disposal was not sufficient to compensate for the exercise‐mediated alterations in lipid flux that may act to augment fasting‐induced insulin resistance. Our results contradict those of Frank et al. (2013), who did observe that a bout of cycling exercise (5×10 min at 50%–70% VO2max with 2–4 min of rest) elicited a degree of protection against fasting‐induced glucose intolerance and insulin resistance. However in this study the exercise bout was performed only 3 h prior to the measurement of insulin sensitivity, and therefore contraction‐mediated stimulation of glucose uptake may have contributed to the observed insulin‐sensitizing effects of exercise.

Interestingly we observed potential sex‐based differences in the magnitude of fasting‐induced glucose intolerance, with females generally demonstrating greater perturbations in glucose homeostasis compared to males, despite having similar glucose tolerance at baseline. This may reflect sexual dimorphism in lipid metabolism during fasting, as an increased lipolytic rate, higher plasma FFA concentrations (Browning et al., 2012; Lundsgaard & Kiens, 2014; Soeters et al., 2007) and increased lipid deposition in skeletal muscle have been observed after short‐term fasting in females compared to males (Browning et al., 2012). Although we did not observe robust sex‐based differences in plasma FFA levels, this does not discount that lipid flux may be significantly different between sexes, particularly because the capacity for intramuscular fatty acid uptake and oxidation is greater in females (Lundsgaard & Kiens, 2014; Tarnopolsky, 2008). Previous studies that have compared metabolic responses to refeeding between males and females have not observed sex‐based differences in glycaemia; however higher insulin responses have been reported in females compared to males (Browning et al., 2012; Solianik et al., 2023), suggesting insulin sensitivity may be compromised to a greater extent in females. Although not explicitly designed or powered for sex differences, our results suggest additional research is certainly warranted to further characterize sex‐based differences in fasting‐induced glucose intolerance and insulin resistance, and elucidate the potential underlying mechanisms that contribute to these divergent metabolic responses between sexes.

We also aimed to explore the reversal of fasting‐induced glucose intolerance by assessing glycaemic responses to an MMTT in free‐living conditions ∼24 h after the reintroduction of normal eating via CGM, which revealed fasting‐associated glucose intolerance was largely but not completely reversed after 24 h of refeeding. We also explored changes in CGM‐derived metrics of glycaemic regulation in three separate 24‐h periods after the fast (i.e. from 0 to 24, 24 to 48 and 48 to 72 h after the fast) to further explore the reversal/recovery of fasting‐induced impairments in glucose homeostasis under free‐living conditions. We found that all CGM metrics were significantly impaired in the first 24 h of refeeding relative to the subsequent two 24‐h periods. However there were no differences in any CGM metrics between day 2 (24–48 h post fast) and day 3 (48–72 h post fast). Although the study design did not permit a baseline measure of CGM glycaemic regulation for these exploratory outcomes, the lack of difference in CGM metrics after the initial 24 h of refeeding could suggest glucose homeostasis was largely restored on days 2 and 3 post fasting.

Our results complement prior evidence suggesting fasting‐induced metabolic perturbations are incompletely reversed 24 h after a short‐term fast, as Norton et al. observed that short‐term fasting‐induced reductions in insulin sensitivity and GLUT4 content remained marginally impaired relative to baseline after 24 h of high carbohydrate refeeding (Norton et al., 2007). The mechanistic basis for persistent impairments in glucose homeostasis after a short‐term fast is unclear, because plasma FFA levels return to baseline within a few hours of refeeding (Grey et al., 1989). However sensitivity to lipolytic stimuli (i.e. hypoinsulinaemia, epinephrine) is elevated after fasting (Jensen et al., 1987; Wolfe et al., 1987), which could contribute to residual elevations in FFA flux that may interfere with intracellular glucose metabolism. Additionally although the contribution of lipid accumulation in skeletal muscle and subsequent inhibition of insulin signalling to fasting‐mediated insulin resistance is not clear, it could be speculated that some degree of skeletal muscle lipotoxicity due to elevated lipid load during short‐term fasting might lead to more persistent perturbations in skeletal muscle glucose disposal.

A few limitations of the present study should be noted. For one we were only able to determine changes in glucose tolerance in response to short‐term fasting and therefore can only speculate on the mechanisms involving substrate flux, insulin sensitivity and metabolic control. Additionally while we asked participants to replicate their diet between conditions in the initial 24 h after fasting, we did not explicitly control or monitor dietary intake during the refeeding period. It is therefore difficult to make any definitive conclusions with regard to the reversal of fasting‐induced perturbations in CGM glucose homeostasis, as well as the effect of exercise performed during the fast on 24‐h glucose regulation. It is also unclear whether the washout period of 5 days between fasts was sufficient for metabolic markers to return to baseline, although we expect the randomized order of conditions would account for this. We also controlled for a period effect in the statistical model (i.e. first vs. second condition performed), which did not reveal any systematic differences in the main outcomes based on the period in which each condition was performed, suggesting that the 5‐day washout period was likely sufficient. Finally although we reported potential sex differences due to the consistency in responses observed across glycaemic outcomes, the study was not explicitly designed or powered for these analyses.

In conclusion our results suggest that performing daily exercise during a short‐term fast does not attenuate fasting‐associated decrements in glucose homeostasis upon refeeding. We also found that glucose tolerance was largely, but not completely, restored 24 h after the reintroduction of normal eating. The significant perturbations in glucose homeostasis in response to refeeding after a short‐term fast may warrant consideration in interrogating the broad health implications of certain fasting regimens, as even short‐term exposure to hyperglycaemic conditions has been linked to oxidative stress and associated impairments in vascular function (González et al., 2023; Kawano et al., 1999). Ultimately future research on short‐term fasting should aim to characterize the effects of refeeding on various cellular processes to properly inform the therapeutic applications of fasting‐based strategies, and determine how fasting regimens can be optimized to mitigate some of the potentially adverse metabolic outcomes associated with refeeding.

Additional information

Competing interests

The authors declare no conflicts of interest.

Author contributions

A.M.‐C., H.I. and J.P.L. conceived and designed the study. T.B., A.M.‐C., D.B., C.K., R.E.S., G.J., S.U., S.M., A.M., S.N., K.D. and H.I. contributed to data acquisition. T.B. performed statistical analyses and drafted the manuscript, with assistance and oversight from H.I. and J.P.L. J.P.L. secured funding for the trial. All authors contributed to revising the manuscript and approved the final version of the manuscript.

Funding

This research was funded by an NSERC Discovery Grant (RGPIN‐2019‐05204) awarded to J.P.L. and institutional start‐up funds awarded to H.I.

Generative AI statement

The authors declare that no generative artificial intelligence (AI) was used in the preperation of this manuscript.

Supporting information

Peer Review History

TJP-604-6391-s001.pdf (545KB, pdf)

Acknowledgements

The authors express their gratitude to the study participants for volunteering their time and effort, as this research would not have been possible without them.

Biography

Tori Bouck completed a master's degree at the University of British Columbia Okanagan. Her work focuses on metabolic and immunological responses to nutrition and exercise interventions. She is particularly interested in exploring sex‐based differences in skeletal muscle metabolism in the context of fasting and exercise.

graphic file with name TJP-604-6391-g005.gif

Handling Editors: Bettina Mittendorfer & Zachary Schlader

The peer review history is available in the Supporting Information section of this article (https://doi.org/10.1113/JP291000#support‐information‐section).

Data availability statement

All data are available upon reasonable request made to the corresponding author.

References

  1. Allison, D. B. , Paultre, F. , Maggio, C. , Mezzitis, N. , & Pi‐Sunyer, F. X. (1995). The use of areas under curves in diabetes research. Diabetes Care, 18(2), 245–250. [DOI] [PubMed] [Google Scholar]
  2. Andriessen, C. , Doligkeit, D. , Moonen‐Kornips, E. , Mensink, M. , Hesselink, M. K. C. , Hoeks, J. , & Schrauwen, P. (2023). The impact of prolonged fasting on 24h energy metabolism and its rhythmicity in healthy, lean males. Clinical Nutrition, 42(12), 2353–2362. [DOI] [PubMed] [Google Scholar]
  3. Anton, S. D. , Moehl, K. , Donahoo, W. T. , Marosi, K. , Lee, S. A. , Mainous, A. G. , Leeuwenburgh, C. , & Mattson, M. P. (2018). Flipping the metabolic switch: Understanding and applying the health benefits of fasting. Obesity (Silver Spring), 26(2), 254–268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Athanasiou, N. , Bogdanis, G. C. , & Mastorakos, G. (2023). Endocrine responses to different types of exercise. Reviews in Endocrine & Metabolic Disorders, 24(2), 251–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bak, A. M. , Vendelbo, M. H. , Christensen, B. , Viggers, R. , Bibby, B.o M. , Rungby, J. , Jørgensen, J. O. L. , Møller, N. , & Jessen, N. (2018). Prolonged fasting‐induced metabolic signatures in human skeletal muscle of lean and obese men. PLoS ONE, 13(9), e0200817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bates, D. , Mächler, M. , Bolker, B. , & Walker, S. (2015). Fitting linear mixed‐effects models using lme4. Journal of Statistical Software, 67(1), 1–48. [Google Scholar]
  7. Browning, J. D. , Baxter, J. , Satapati, S. , & Burgess, S. C. (2012). Effects of short‐term fasting on liver and skeletal muscle metabolism. Journal of Lipid Research, 53(3), 577–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Castillo, C. E. , Katz, A. , Spencer, M. K. , Yan, Z. , & Nyomba, B. L. (1991). Fasting inhibits insulin‐mediated glycolysis and anaplerosis in human skeletal muscle. American Journal of Physiology, 261(5 Pt 1), E598–E605. [DOI] [PubMed] [Google Scholar]
  9. Frank, P. , Katz, A. , Andersson, E. , & Sahlin, K. (2013). Acute exercise reverses starvation‐mediated insulin resistance in humans. American Journal of Physiology‐ Endocrinology and Metabolism, 304(4), E436–E443. [DOI] [PubMed] [Google Scholar]
  10. González, P. , Lozano, P. , Ros, G. , & Solano, F. (2023). Hyperglycemia and oxidative stress. International Journal of Molecular Sciences, 24(11), 9352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gray, D. S. , Takahashi, M. , Fisler, J. S. , LoPresti, J. , Nicoloff, J. T. , & Bray, G. A. (1989). Effect of carbohydrate refeeding after fasting. Metabolism, 38(3), 208–214. [DOI] [PubMed] [Google Scholar]
  12. Green, J. G. , Johnson, N. A. , Sachinwalla, T. , Cunningham, C. W. , Thompson, M. W. , & Stannard, S. R. (2011). Exercise prevents starvation‐induced intramyocellular lipid accumulation but not insulin resistance. Metabolism, 60(8), 1051–1057. [DOI] [PubMed] [Google Scholar]
  13. Hoaglin, D. C. , & Iglewicz, B. (1987). Fine‐tuning some resistant rules for outlier labeling. Journal of the American Statistical Association, 82(400), 1147–1149. [Google Scholar]
  14. Hoeks, J. , van Herpen, N. A. , Mensink, M. , Moonen‐Kornips, E. , van Beurden, D. , Hesselink, M. K. C. , & Schrauwen, P. (2010). Prolonged fasting identifies skeletal muscle mitochondrial dysfunction as a consequence rather than cause of insulin resistance. Diabetes, 59(9), 2117–2125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hue, L. , & Taegtmeyer, H. (2009). The Randle cycle revisited. American Journal of Physiology‐ Endocrinology and Metabolism, 297(3), E578–E591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jensen, J. , Rustad, P. I. , Kolnes, A. J. , & Lai, Y. C. (2011). Role of glycogen breakdown in regulating insulin sensitivity. Frontiers in Physiology, 2, 112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jensen, M. D. , Ekberg, K. , & Landau, B. R. (2001). Lipid metabolism during fasting. American Journal of Physiology‐ Endocrinology and Metabolism, 281(4), E789–E793. [DOI] [PubMed] [Google Scholar]
  18. Jensen, M. D. , Haymond, M. W. , Gerich, J. E. , Cryer, P. E. , & Miles, J. M. (1987). Lipolysis during fasting. Journal of Clinical Investigation, 79(1), 207–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Johnson, N. A. , Stannard, S. R. , Rowlands, D. S. , Chapman, P. G. , Thompson, C. H. , O'Connor, H. , Sachinwalla, T. , & Thompson, M. W. (2006). Effect of short‐term starvation vs high‐fat diet on intramyocellular triglycerides and insulin resistance in fit men. Experimental Physiology, 91(4), 693–703. [DOI] [PubMed] [Google Scholar]
  20. Jørgensen, S. W. , Hjort, L. , Gillberg, L. , Justesen, L. , Madsbad, S. , Brøns, C. , & Vaag, A. A. (2021). Impact of prolonged fasting on insulin secretion and insulin action in healthy young males. American Journal of Physiology‐ Endocrinology and Metabolism, 320(2), E281–E290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kawano, H. , Motoyama, T. , Hirashima, O. , Hirai, N. , Miyao, Y. , Sakamoto, T. , Kugiyama, K. , Ogawa, H. , & Yasue, H. (1999). Hyperglycemia suppresses flow‐mediated dilation. Journal of the American College of Cardiology, 34(1), 146–154. [DOI] [PubMed] [Google Scholar]
  22. Kersten, S. (2023). The impact of fasting on adipose tissue metabolism. Biochimica et Biophysica Acta Molecular and Cell Biology of Lipids, 1868(3), 159262. [DOI] [PubMed] [Google Scholar]
  23. Kjøbsted, R. , Chadt, A. , Jørgensen, N. O. , Kido, K. , Larsen, J. K. , de Wendt, C. , Al‐Hasani, H. , & Wojtaszewski, J. F. P. (2019). TBC1D4 is necessary for enhanced insulin sensitivity from contraction. Diabetes, 68(9), 1756–1766. [DOI] [PubMed] [Google Scholar]
  24. Kolnes, K. J. , Nilsen, E. T. F. , Brufladt, S. , Meadows, A. M. , Jeppesen, P. B. , Skattebo, Ø. , Johansen, E. I. , Birk, J. B. , Højlund, K. , Hingst, J. , Skålhegg, B. S. , Kjøbsted, R. , Griffin, J. L. , Kolnes, A. J. , O'Rahilly, S. , Wojtaszewski, J. F. P. , & Jensen, J. (2025). Seven days’ fasting and exercise performance. Nature Communications, 16(1), 122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Koppold, D. A. , Breinlinger, C. , Hanslian, E. , Kessler, C. , Cramer, H. , Khokhar, A. R. , Peterson, C. M. , Tinsley, G. , Vernieri, C. , Bloomer, R. J. , Boschmann, M. , Bragazzi, N. L. , Brandhorst, S. , Gabel, K. , Goldhamer, A. C. , Grajower, M. M. , Harvie, M. , Heilbronn, L. , Horne, B. D. , … Michalsen, A. (2024). International consensus on fasting terminology. Cell Metabolism, 36(8), 1779–1794.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kolb, H. , Kempf, K. , Röhling, M. , Lenzen‐Schulte, M. , Schloot, N. C. , & Martin, S. (2021). Ketone bodies: From enemy to friend and guardian angel. BioMed Central Medicine [Electronic Resource], 19(1), 313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kuznetsova, A. , Brockhoff, P. B. , & Christensen, R. H. B. (2017). lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software, 82(13), 1–26. [Google Scholar]
  28. Lundsgaard, A. M. , Fritzen, A. M. , & Kiens, B. (2020). Importance of fatty acids in post‐exercise recovery. Nutrients, 12(2), 280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lundsgaard, A. M. , & Kiens, B. (2014). Gender differences in muscle substrate metabolism. Frontiers in Endocrinol (Lausanne), 5, 195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Magkos, F. , Mohammed, B. S. , Patterson, B. W. , & Mittendorfer, B. (2009). FFA kinetics late in post‐exercise recovery. Metabolism, 58(9), 1248–1255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mansell, P. I. , & Macdonald, I. A. (1990). The effect of starvation on insulin‐induced glucose disposal and thermogenesis in humans. Metabolism, 39(5), 502–510. [DOI] [PubMed] [Google Scholar]
  32. Mikines, K. J. , Sonne, B. , Farrell, P. A. , Tronier, B. , & Galbo, H. (1988). Effect of exercise on insulin sensitivity. American Journal of Physiology, 254(3), E248–E259. [DOI] [PubMed] [Google Scholar]
  33. Newsom, S. A. , Everett, A. C. , Hinko, A. , & Horowitz, J. F. (2013). Low‐intensity exercise enhances next‐day insulin sensitivity in obese adults. Diabetes Care, 36(9), 2516–2522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Norton, L. , Parr, T. , Bardsley, R. G. , Ye, H. , & Tsintzas, K. (2007). GLUT4 and calpain expression during fasting and refeeding. Acta Physiologica (Oxford), 189(3), 233–240. [DOI] [PubMed] [Google Scholar]
  35. Pilegaard, H. , & Darrell Neufer, P. (2004). Regulation of PDK4 during and after exercise. Proceedings of the Nutrition Society, 63(2), 221–226. [DOI] [PubMed] [Google Scholar]
  36. Randle, P. J. , Newsholme, E. A. , & Garland, P. B. (1964). Regulation of glucose uptake by muscle: Effects of fatty acids, ketones, and starvation. Biochemical Journal, 93(3), 652–665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Richter, E. A. , & Hargreaves, M. (2013). Exercise, GLUT4, and muscle glucose uptake. Physiological Reviews, 93(3), 993–1017. [DOI] [PubMed] [Google Scholar]
  38. Scharf, E. , Zeiler, E. , Ncube, M. , Kolbe, P. , Hwang, S.‐Y. , Goldhamer, A. , & Myers, T. R. (2022). The effects of prolonged water‐only fasting and refeeding on markers of cardiometabolic risk. Nutrients, 14(6), 1183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Schenk, S. , & Horowitz, J. F. (2007). Acute exercise increases triglyceride synthesis and prevents lipid‐induced insulin resistance in muscle. Journal of Clinical Investigation, 117(6), 1690–1698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Soeters, M. R. , Sauerwein, H. P. , Dubbelhuis, P. F. , Groener, J. E. , Ackermans, M.¨ T. , Fliers, E. , Aerts, J. M. , & Serlie, M. J. (2008). Muscle adaptation to short‐term fasting in healthy lean humans. Journal of Clinical Endocrinology and Metabolism, 93(7), 2900–2903. [DOI] [PubMed] [Google Scholar]
  41. Soeters, M. R. , Sauerwein, H. P. , Groener, J. E. , Aerts, J. M. , Ackermans, M.¨ T. , Glatz, J. F. C. , Fliers, E. , & Serlie, M. J. (2007). Gender‐related differences in the metabolic response to fasting. Journal of Clinical Endocrinology and Metabolism, 92(9), 3646–3652. [DOI] [PubMed] [Google Scholar]
  42. Solianik, R. , Židonienė, K. , Baranauskienė, N. , & Brazaitis, M. Fasting for 48 h induced similar glucose intolerance in both sexes despite greater perceived stress and decreased estradiol levels in females. European Journal of Applied Physiology (2024). 124(5), 1449–1459. [DOI] [PubMed] [Google Scholar]
  43. Solianik, R. , Židonienė, K. , Eimantas, N. , & Brazaitis, M. (2023). Prolonged fasting improves glucose tolerance and insulin release. British Journal of Nutrition, 130(9), 1500–1509. [DOI] [PubMed] [Google Scholar]
  44. Sugden, M. C. , Bulmer, K. , & Holness, M. J. (2001). Fuel‐sensing mechanisms integrating lipid and carbohydrate use. Biochemical Society Transactions, 29(Pt 2), 272–278. [DOI] [PubMed] [Google Scholar]
  45. Tarnopolsky, M. A. (2008). Sex differences in exercise metabolism. Medicine and Science in Sports and Exercise, 40(4), 648–654. [DOI] [PubMed] [Google Scholar]
  46. van der Crabben, S. N. , Allick, G. , Ackermans, M. T. , Endert, E. , Romijn, J. A. , & Sauerwein, H. P. (2008). Prolonged fasting induces peripheral insulin resistance not ameliorated by salicylate. Journal of Clinical Endocrinology and Metabolism, 93(2), 638–641. [DOI] [PubMed] [Google Scholar]
  47. Vendelbo, M. H. , Clasen, B. F. F. , Treebak, J. T. , Møller, L. , Krusenstjerna‐Hafstrøm, T. , Madsen, M. , Nielsen, T. S. , Stødkilde‐Jørgensen, H. , Pedersen, S. B. , Jørgensen, J. O. L. , Goodyear, L. J. , Wojtaszewski, J. F. P. , Møller, N. , & Jessen, N. (2012). Insulin resistance after a 72‐h fast is associated with impaired AS160 phosphorylation. American Journal of Physiology‐ Endocrinology and Metabolism, 302(2), E190–E200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Webber, J. , Taylor, J. , Greathead, H. , Dawson, J. , Buttery, P. J. , & Macdonald, I. A. (1994). Effects of fasting on fatty acid kinetics and cardiovascular, thermogenic and metabolic responses to the glucose clamp. Clinical Science (London), 87(6), 697–706. [DOI] [PubMed] [Google Scholar]
  49. Wolfe, R. R. , Peters, E. J. , Klein, S. , Holland, O. B. , Rosenblatt, J. , & Gary, H., Jr. (1987). Effect of short‐term fasting on lipolytic responsiveness in lean and obese human subjects. American Journal of Physiology, 252(2), E189–E196. [DOI] [PubMed] [Google Scholar]
  50. Yu, C. , Chen, Y. , Cline, G. W. , Zhang, D. , Zong, H. , Wang, Y. , Bergeron, R. , Kim, J. K. , Cushman, S. W. , Cooney, G. J. , Atcheson, B. , White, M. F. , Kraegen, E. W. , & Shulman, G. I. (2002). Fatty acid inhibition of IRS‐1–associated PI3K activity. Journal of Biological Chemistry, 277(52), 50230–50236. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Peer Review History

TJP-604-6391-s001.pdf (545KB, pdf)

Data Availability Statement

All data are available upon reasonable request made to the corresponding author.

Articles from The Journal of Physiology are provided here courtesy of Wiley

RESOURCES