Impact of Dietary Carbohydrate Restriction versus Energy Restriction on Exogenous Carbohydrate Oxidation during Aerobic Exercise

ABSTRACT

Individuals with high physical activity levels, such as athletes and military personnel, are likely to experience periods of low muscle glycogen content. Reductions in glycogen stores are associated with impaired physical performance. Lower glycogen stores in these populations are likely due to sustained aerobic exercise coupled with suboptimal carbohydrate or energy intake. Consuming exogenous carbohydrate during aerobic exercise may be an effective intervention to sustain physical performance during periods of low glycogen. However, research is limited in the area of carbohydrate recommendations to fuel performance during periods of suboptimal carbohydrate and energy intake. Additionally, the studies that have investigated the effects of low glycogen stores on exogenous carbohydrate oxidation have yielded conflicting results. Discrepancies between studies may be the result of glycogen stores being lowered by restricting carbohydrate or restricting energy intake. This narrative review discusses the influence of low glycogen status, resulting from carbohydrate restriction versus energy restriction, on exogenous carbohydrate oxidation and examines the potential mechanism resulting in divergent responses in exogenous carbohydrate oxidation. Results from this review indicate that rates of exogenous carbohydrate oxidation can be maintained when glycogen content is lower following carbohydrate restrictions but may be reduced following energy restriction. Reductions in exogenous carbohydrate oxidation following energy restriction appear to result from lower insulin sensitivity and glucose uptake. Exogenous carbohydrate may thus be an effective intervention to sustain performance following short-term energy-adequate carbohydrate restriction but may not be an effective ergogenic aid when glycogen stores are low due to energy restriction.

Statement of Significance: Exogenous carbohydrate may be an effective intervention to sustain physical performance following short-term energy-adequate carbohydrate-restriction diets. Conversely, reductions in exogenous carbohydrate oxidation due to reduced glucose uptake indicate carbohydrate supplements may not be an effective ergogenic aid when glycogen stores are low due to energy restriction.

Introduction

Endogenous carbohydrate stores (e.g., liver and muscle glycogen) are an important, readily available energy source used to fuel physical performance (1). Reductions in glycogen content through prolonged aerobic exercise and suboptimal daily carbohydrate or energy intake are associated with declines in physical performance (2–4). A recent meta-analysis of 181 studies assessing muscle glycogen content reported that normal/adequate resting glycogen content was ∼450 mmol/kg dry muscle wt, high glycogen content was ∼550 mmol/kg dry muscle wt, and low glycogen content was ∼200 mmol/kg dry muscle wt (5). Individuals with high physical demands, such as athletes and military personnel, are likely to experience intentional or unintentional periods of low glycogen content (6–9). Intentional low glycogen content may be due to practices of periodized or chronic dietary carbohydrate restriction with the goal of increasing fat oxidation and decreasing endogenous carbohydrate oxidation during aerobic exercise (10–12). Unintentional low glycogen content may result from insufficient energy intake to match high daily energy expenditures, resulting in energy deficit and low energy availability (13, 14). Regardless of the reason, without intervention, low glycogen content may lead to compromised physical performance during prolonged aerobic exercise.

The current nutritional recommendation is to consume carbohydrate during prolonged exercise, particularly when glycogen stores are low, to sustain submaximal aerobic performance (1). Consuming carbohydrate before and during aerobic exercise allows for oxidation of exogenous carbohydrate, while reducing reliance on endogenous carbohydrate for fuel to spare glycogen stores. However, as noted in the current joint position stand on “Nutrition and Athletic Performance” by the American College of Sports Medicine and Academy of Nutrition and Dietetics (1), there has been limited research conducted in the area of dietary carbohydrate requirements during periods of suboptimal energy and carbohydrate intake (15–19). Furthermore, the few studies that have assessed the influence of low glycogen stores on exogenous carbohydrate oxidation during exercise have yielded conflicting results, with some reporting an increase (15), decrease (16, 17), or no change (18, 19) in exogenous carbohydrate oxidation (Table 1). A primary factor resulting in discordant results across these studies may be due to glycogen stores being reduced through feeding participants low-carbohydrate, high-fat diets versus low-energy diets.

TABLE 1.

Influence of glycogen status or daily carbohydrate intake on exogenous carbohydrate oxidation rates during aerobic exercise in healthy active adults¹

Author (ref)	Article	Population	Study design	Carbohydrate supplement	Protocol
Margolis et al. (19)	Randomized, crossover design	12 men	Glycogen depletion followed by 24 h LG (1.5 g carbohydrate/kg, 3 g fat/kg) or HG (6.0 g carbohydrate/kg, 1g fat/kg)	146 g CHO (95 g GLU, 51 g FRU)	Cycling at 64% V̇O2 peak for 80 min
Ravussin et al. (18)	Randomized parallel design (n = 6), crossover (n = 2)	8 healthy individuals	Habitual diets consumed to elicit HG and 5 d consuming <5% total kcal from carbohydrate to elicit LG	100 g GLU	Cycling at 40% V̇O2max for 120 min
Péronnet et al. (15)	Randomized, crossover design	6 active males	90 min cycling 70% V̇O2max followed by 48 h LC (3000 kcal/d, 200 g carbohydrate) or HC (3500 kcal/d with 700 g carbohydrate)	200 g GLU	Cycling at 64% V̇O2max for 120 min
Jeukendrup et al. (16)	Randomized, crossover design	7 well-trained male cyclist and triathletes	Glycogen depletion followed by LG (14 g carbohydrate, 4 g fat, 6 g protein) or HG (250–300 g carbohydrate, 15 g fat, 20 g protein) the night before trials. Both consumed standardized breakfast (14 g carbohydrate, 4 g fat, 6 g protein) morning of testing	127 g GLU	Cycling at 57% V̇O2max for 120 min
Rowlands et al. (17)	Randomized, crossover design	7 active individuals (2 women, 5 men)	60 min cycling at 65% V̇O2max followed by 67 h LG (fasted, water only) or HG (1.5 × energy needs as 50% carbohydrate, 35% fat, 15% protein)	96 g GLU	Cycling at 50% V̇O2max for 80 min

¹CHO, carbohydrate; FRU, fructose; GLU, glucose; HG, high glycogen; LC, low carbohydrate; LG, low glycogen; V̇O_2peak, peak oxygen uptake; V̇O_2max, maximal oxygen uptake.

As carbohydrate intake is an important component to sustaining aerobic performance during prolonged physical activity, understanding how low glycogen resulting from carbohydrate restriction versus energy restriction impacts exogenous carbohydrate oxidation is integral to make appropriate fueling recommendations. The objective of this narrative review is to characterize the influence of low glycogen status resulting from carbohydrate restriction or energy restriction on exogenous carbohydrate oxidation. This review will also discuss the potential variables and mechanisms resulting in variant findings between carbohydrate- and energy-restricting conditions.

Carbohydrate Restriction

Exogenous carbohydrate oxidation

To examine the influence of glycogen availability on exogenous carbohydrate oxidation, our laboratory (19) had participants perform a bout of glycogen-depletion cycling followed by 24 h of eucaloric refeeding, manipulating carbohydrate and fat intake, to elicit low (1.5 g/kg carbohydrate, 3.0 g/kg fat) or high (6.0 g/kg carbohydrate, 1.0 g/kg fat) glycogen stores. Glycogen content was determined from freeze-dried muscle samples obtained from biopsies of the vastus lateralis. At the onset of aerobic exercise, per protocol, glycogen content was lower in the low-glycogen group at 217 mmol/kg dry muscle wt compared with 396 mmol/kg dry muscle wt in the high-glycogen group (19). Despite a difference in glycogen content, there was no difference in exogenous carbohydrate oxidation, with rates of 0.84 in low- and 0.87 g/min in high-glycogen treatments, when 146 g carbohydrate (95 g glucose, 51 g fructose) was consumed while cycling at 65% maximal oxygen uptake (V̇O_2max) (Table 2). The primary metabolic adaptation to initiating exercise with low glycogen content following carbohydrate restriction was that fat oxidation increased by 0.18 g/min, whereas endogenous carbohydrate oxidation decreased by 0.42 g/min compared with high glycogen. In agreement with our work, Ravussin et al. (18) reported when participants consumed 100 g of glucose while cycling at 40% V̇O_2max, exogenous carbohydrate oxidation rates were 0.32 g/min for low and 0.34 g/min for high glycogen. In this study (18), glycogen stores were manipulated by having participants consume their habitual diet or a very low-carbohydrate (<5% total daily energy intake), high-fat diet for 3 d. However, no direct measurement of muscle glycogen content was conducted in this study. Similar to results from our laboratory, the main shifts in substrate oxidation were an increase in fat oxidation of 0.41 g/min, and a decrease in endogenous carbohydrate oxidation of 0.64 g/min in low-glycogen compared with high-glycogen treatments. After the carbohydrate ingestion, there were no significant differences in circulating insulin or glucose concentrations during the bout of cycling between the low- and high-glycogen treatments. In agreement with these differences in substrate oxidation, circulating free fatty acid concentrations were significantly higher when aerobic exercise was initiated with low compared to high glycogen content.

TABLE 2.

Substrate oxidation with low or high muscle glycogen status during aerobic exercise in healthy adults

	Total carbohydrate oxidation (g/min)		Endogenous carbohydrate oxidation (g/min)		Exogenous carbohydrate oxidation (g/min)		Fat oxidation (g/min)
Author (ref)	Low glycogen	High glycogen	Low glycogen	High glycogen	Low glycogen	High glycogen	Low glycogen	High glycogen
Carbohydrate restriction
1Margolis et al. (19)	1.59	2.033	0.75	1.173	0.84	0.87	0.55	0.383
2Peronnet et al. (15)	1.86	2.923	1.20	2.313	0.66	0.613	0.58	0.183
2Ravussin et al. (18)	0.51	1.23	0.19	0.833	0.32	0.34	0.61	0.23
Energy restriction
2Jeukendrup et al. (16)	1.40	1.933	0.81	1.123	0.60	0.823	1.52	1.12
2Rowlands et al. (17)	1.1	2.03	0.79	1.303	0.32	0.693	0.76	0.44

Data are presented as mean values.

¹Direct measurement of glycogen was conducted using a colorimetric assay in freeze-dried skeletal muscle.

²No direct measurement of skeletal muscle glycogen content.

³Reported as significantly different than low glucose in the primary manuscript.

Contrary to the work from our laboratory (19) and Ravussin et al. (18), Péronnet et al. (15) reported that exogenous carbohydrate oxidation was significantly higher at 0.66 g/min in low- compared with 0.61 g/min high-glycogen treatments when 200 g of glucose was consumed during a 120 min cycle at 64% V̇O_2max. Glycogen content was manipulated by having participants cycle for 90 min at 70% V̇O_2max followed by consuming a eucaloric diet either low (27% total daily energy needs) or high (80% total daily energy needs) in carbohydrate. No direct measurement of skeletal muscle glycogen content was conducted in this study. However, methods used by Péronnet et al. (15) to manipulate glycogen content were similar to those used in our laboratory's previous investigation (19). As such, it may be assumed that differences in glycogen content between treatments were similar to those reported by our laboratory (19). Though the rate of exogenous carbohydrate oxidation was reported to be higher with low compared to high glycogen, differences in exogenous carbohydrate oxidation were driven primarily by differences observed between 40 and 80 min, with no effect of glycogen status during 80–120 min of the aerobic exercise bout (15). Furthermore, assessing the percent contribution of exogenous carbohydrate oxidation to the total energy expended during exercise, there was no difference between treatments, with exogenous carbohydrate constituting 19% in low and 18% in high glycogen (15). Overall, there was no difference in circulating concentrations of glucose and insulin in responses to carbohydrate consumption during aerobic exercise between treatments. These data would indicate that overall, there was no difference in exogenous carbohydrate use for fuel during the 120 min bout of aerobic exercise following consumption of a carbohydrate-restrictive or carbohydrate-adequate diet. Again, the main adaptation to initiating aerobic exercise following carbohydrate restriction is an increase in fat oxidation and a decrease in endogenous carbohydrate oxidation (15). These shifts in substrate oxidation were reflected in higher circulating free fatty acid concentrations and lower circulating lactate concentrations in low- compared with high-glycogen treatments (15).

Glucose turnover

Preservation in the rates of exogenous carbohydrate oxidation with low glycogen status following carbohydrate restriction is likely due to maintenance of glucose uptake in peripheral tissue. Several studies (20–24) have examined the impact of initiating aerobic exercise with low compared to adequate glycogen content on glucose rates of disappearance (e.g., glucose uptake) and plasma glucose oxidation in healthy participants using primed/continuous infusions of 6,6-²H₂-glucose. Two separate investigations by Weltan et al. (20, 21) manipulated glycogen content by having participants complete a bout of glycogen-depleting cycling and then consuming a carbohydrate-restriction or -adequate diet for 48 h. Participants then returned to their laboratory to perform 145 min of steady-state cycling at 70% V̇O_2max while being infused with a glucose solution. Both studies concluded that there was no difference in plasma glucose uptake or oxidation when exercise was initiated with low or high glycogen stores (20, 21). Furthermore, Chokkalingam et al. (25) observed that following 6 d of consuming a carbohydrate-restrictive diet, the glucose rate of disappearance was slightly higher during the last 30 min of a 4-h hyperinsulinemic euglycemic clamp compared with a control mixed-macronutrient diet. The increase in glucose rate of disappearance appears to be the result of higher rates of nonoxidative glucose disposal (e.g., glycogen synthesis) in response to the hyperinsulinemic euglycemic clamp following carbohydrate restriction. Overall, these data indicate that the rate of glucose uptake is maintained following carbohydrate restriction, and that there is a shift to increase glycogen synthesis to replenish endogenous carbohydrate stores.

Thus, the primary differences observed between glycogen conditions are that with low glycogen a higher rate of fat oxidation, circulating free fatty acid, norepinephrine, and lactate concentrations, and lower rates of total carbohydrate and muscle glycogen oxidation are elicited compared with high glycogen (20, 21). Several other investigations (22–24, 26) similarly reported that there was no effect of glycogen content on glucose uptake or plasma glucose oxidation. In agreement with Weltan et al. (20, 21), these studies (22–24, 26) reported that the primary adaptation to low compared with high glycogen content was increased fat oxidation and decreased total and endogenous carbohydrate oxidation. To the best of our knowledge, no study has used a duel tracer technique, combining oral ¹³C-isotopes with primed/continuous 6,6-²H₂-glucose to assess exogenous carbohydrate oxidation and glucose turnover, nevertheless the combined results of these investigations and the ones discussed in the above section indicate that maintained uptake of circulating glucose by peripheral tissue is the mechanism resulting in no difference in exogenous carbohydrate oxidation when aerobic exercise is initiated with low or adequate glycogen stores.

Molecular adaptations

Alterations in molecular processes within skeletal muscle are probably the primary mechanism regulating shifts in whole-body substrate oxidation when aerobic exercise is performed with low glycogen (27–29). Recent work from our laboratory (19) suggests that molecular pathways that regulate carbohydrate uptake are maintained when aerobic exercise is initiated with low glycogen content following 24 h of carbohydrate restriction. The maintenance of these molecular pathways likely accounts for the lack of effect of glycogen content on rates of exogenous carbohydrate oxidation and glucose uptake. Primary molecular alterations following short-term (≤5 d) carbohydrate restriction are a decreased reliance on endogenous carbohydrate stores for fuel, and increased fatty acid uptake, transport, and oxidation (19, 30). Specifically, when muscle glycogen is low, pyruvate dehydrogenase (PDH) activity (19, 31, 32) is suppressed by increased pyruvate dehydrogenase kinase 4 (PDK4) (25, 33), reducing glycolytic flux into the tricarboxylic acid cycle, accounting for the reduction in carbohydrate oxidation. To offset lower rates of total and endogenous carbohydrate oxidation, transcription factors peroxisome proliferator-activated receptors (PPARs) and PPAR-γ coactivator 1α (PGC-1α), which are central regulators of mitochondrial function and oxidative capacity (34, 35), are upregulated. These factors increase the transcription of fatty acid translocase (FAT), carnitine palmitoyl transferase 1a (CPT1a), and hydroxyacyl-CoA dehydrogenase (HADHA) when muscle glycogen content is low (33, 36–38). The increase in transcriptional regulation of fatty acid uptake, transport, and oxidation contribute to greater whole-body fat oxidation (36). Importantly, when aerobic exercise is initiated with low glycogen increased FAT, FABP3, CPT1a, HADHA, and PPARδ expression is maintained even when carbohydrate was consumed during exercise compared with high glycogen (19). Sustained increases in transcriptional regulation of fatty acid uptake, transport, and oxidation when carbohydrate was consumed during exercise was likely the result of glycogen content remaining postaerobic exercise in low- compared with high-glycogen treatments (19). Together, these results indicate consuming carbohydrate may be a viable fueling strategy to prolong submaximal exercise without impairing adaptations to enhance whole-body and skeletal muscle fat metabolism when glycogen stores are reduced following dietary carbohydrate restriction.

Energy Restriction

Exogenous carbohydrate oxidation

In contrast to studies showing no effect of carbohydrate restriction on exogenous carbohydrate oxidation (15, 18, 19), studies by Jeukendrup et al. (16) and Rowland et al. (17) reported exogenous carbohydrate oxidation rates were higher when aerobic exercise was initiated with high compared to low glycogen stores (Table 2). Discordant results in these 2 studies compared with the 3 discussed in the previous section can likely be attributed to methodological differences in how glycogen stores were manipulated. Juekendrup et al. (16) had participants complete a bout of glycogen-depletion cycling, followed by feeding either a lower energy and lower carbohydrate meal (116 kcals, 14 g carbohydrate) to achieve low glycogen content or a higher energy and higher carbohydrate meal (1200–1400 kcal, 250–300 g carbohydrate) to achieve high glycogen content the evening prior to completing a substrate oxidation protocol. No direct measure of muscle glycogen content was conducted in this study, however, this method of reducing energy and carbohydrate intake has been shown to alter muscle glycogen content (6). The following day participants performed 120 min of cycling at 57% V̇O_2max. Exogenous carbohydrate oxidation was 0.60 g/min with low compared to 0.82 g/min with high glycogen stores, when 127 g glucose was consumed during cycling. Fat oxidation was 0.40 g/min higher and endogenous carbohydrate oxidation was 0.31 g/min lower in the low- compared with high-glycogen treatments. There was no difference in circulating glucose concentrations between treatments, whereas insulin concentrations were significantly lower in low- compared with high-glycogen treatments. Rowlands et al. (17) similarly reported reduced exogenous carbohydrate oxidation in low- compared to high-glycogen treatments, with rates of 0.32 g/min and 0.69 g/min, when 96 g glucose was consumed while cycling at 56% V̇O_2max for 80 min. Low-glycogen treatment was achieved by having participants consume only water for 67 h before the steady-state aerobic exercise test, whereas high glycogen was achieved by participants consuming their habitual diet. No direct measurement of muscle glycogen content was assessed in this investigation, however, muscle glycogen content may be depleted within 24 h of starvation (39). Along with lower rates of exogenous carbohydrate oxidation, endogenous carbohydrate oxidation was 0.51 g/min lower, whereas fat oxidation was 0.32 g/min higher in low- compared to high-glycogen treatments. Interestingly, in contrast to Jeukendrup et al. (16), Rowlands et al. (17) reported higher concentrations of circulating insulin in the low- compared with high-glycogen treatments in response to carbohydrate consumption during aerobic exercise. There was no difference in circulating blood glucose concentrations between treatments. Discrepancies between insulin responses in these studies may be due to the severity of energy restriction. As we will discuss in detail in the next section, starvation reduces insulin sensitivity and glucose uptake (40), which may explain lower exogenous carbohydrate oxidation rates following periods of energy restriction. Reductions in exogenous carbohydrate oxidation during exercise may indicate carbohydrate supplements may not be effective in improving physical performance when glycogen stores are low due to energy restriction.

Glucose turnover

Lower rates of exogenous carbohydrate oxidation with low glycogen content following energy restrictions may be due to decreased glucose uptake by peripheral tissue. Periods of starvation result in reductions in whole-body glucose tolerance and insulin sensitivity (40, 41). Mansell and MacDonald (40) reported following 48 h of starvation (water only), rates of glucose disposal are lowered due to reductions in glucose oxidation compared with the consumption of an energy-adequate diet. Furthermore, the rate of glucose uptake was significantly reduced during starvation due to impaired glucose tolerance and insulin sensitivity (40). Corroborating these data, 3 separate investigations assessing glucose kinetics in response to a hyperinsulinemic euglycemic clamp reported the glucose rate of disappearance and nonoxidative glucose disposal were reduced following 60–67 h of starvation in healthy lean adults (42–44). When nonoxidative glucose disposal was assessed as a percentage of the glucose rate of disappearance, there was no difference following starvation (42). These data further indicate that starvation-induced insulin resistance results in decreased uptake into peripheral tissue, while the intracellular mechanisms regulating glucose storage are maintained.

Starvation-induced insulin resistance appears to be independent of carbohydrate restriction. Johnson et al. (45) compared the effects of 67 h of starvation to an energy-adequate low-carbohydrate, high-fat diet on glucose tolerance and insulin sensitivity, using an intravenous glucose tolerance test. Although both groups were reported to have lower glucose tolerance and insulin sensitivity compared with a control mixed-macronutrient diet, insulin sensitivity was lower with starvation compared to a low-carbohydrate, high-fat diet (45). Using similar methodologies, Green et al. (46) observed reductions in glucose tolerance and insulin sensitivity following 72 h of starvation compared with consuming an energy-adequate low-carbohydrate, high-protein diet. Conversely, Areta et al. (47) recently reported an impaired glycemic response to consuming 1.2 g/kg fat-free mass carbohydrate plus 0.38 g/kg fat-free mass protein postaerobic exercise following 24 h of a high-fat, low-carbohydrate diet compared with an energy-restrictive diet. Specifically, the AUC for circulating concentrations of glucose and insulin were higher in response to oral carbohydrate consumption following carbohydrate restriction compared with energy restriction. Discrepancies between this and the previously discussed work may be due to methodological differences. Results may differ when a bolus of carbohydrate is consumed compared with intravenous glucose infusion, and if the glycemic response is measured during compared with postaerobic exercise. Though there are some discordant results in the literature regarding carbohydrate versus energy restriction, the majority of investigations suggest that starvation results in reduced glucose tolerance and insulin sensitivity, which likely accounts for subsequent lower exogenous carbohydrate oxidation rates during exercise compared with an energy-adequate diet.

Molecular adaptations

Alterations in molecular adaptations resulting in reduced glucose uptake and oxidation during periods of starvation appear to result downstream of the insulin receptor substrate 1 (IRS1) (48). Under normal physiological conditions IRS1 activates downstream intracellular signaling proteins to stimulate translocation of glucose transporter 4 (GLUT4) in response to increased insulin concentrations and binding to the insulin receptor (49). Translocation of GLUT4 via insulin stimulation occurs through increased phosphorylation of protein kinase B (AKT) and AKT substrate of 160 kDa (AS160) (49). Following 62 h of fasting, the change in the phosphorylation status of AKT and AS160 is blunted in response to hyperinsulinemic euglycemic clamp (42). Furthermore, Tsintzas et al. (50) observed that following 48 h of fasting, concurrent with reductions in the glucose rate of disappearance and insulin sensitivity, the mRNA expression of hexokinase II (HKII) was decreased, whereas mRNA expression and total protein content of PDK4 were increased. Interestingly, in contrast to carbohydrate-restriction studies, Tsintzas et al. (50) reported no change in transcriptional regulators of fatty acid uptake, transport, and oxidation. This lack of transcription modification occurred despite higher circulating concentrations of free fatty acids and glycerol (50), and others have observed increased rates of fat oxidation following starvation (17). Molecular findings from these studies suggest that lower rates of exogenous carbohydrate oxidation following energy restriction are the result of decreased GLUT4 translocation due to blunted insulin-stimulated activation of AKT and AS160.

Contrary to results from human studies, several studies in rats (51–55) have reported that low glycogen content following 24 h of starvation increases glucose uptake into the muscle compared with high glycogen content. Increased glucose uptake in rat models has been attributed to increased AKT signaling activity, as well as increased activation of AMP-activated protein kinase (AMPK) (51–55). Similar to AKT, AMPK phosphorylates AS160 to stimulate GLUT4 translocation and glucose uptake into the cell, however, this occurs independent of insulin (56, 57). AMPK is activated in response to muscle contraction and low energy availability, and is a primary regulator of cellular energy metabolism (58–60). Lai et al. (55) reported in response to contraction, AMPKα2 activity and phosphorylation was highest in rats with low glycogen following 24 h of starvation compared to rats with high glycogen. Increased AMPK activity in the low glycogen rats resulted in increased glucose uptake, determined using a 2-deoxy-D-[³H] glucose tracer methodology in dissected muscle (55). Increased glucose uptake with contraction in these rat models appears to drive increased rates of glycogen synthesis when glycogen content is low (53). Increased rates of glucose uptake appear absent in human models, as exogenous carbohydrate oxidation, glucose rate of disappearance, and metabolic clearance rate are maintained or lowered when glycogen stores are reduced following carbohydrate or energy restriction. Discordant results between rat and human studies are not easily resolved but may be the result of model and methodological differences. Most noteworthy, rat studies assessed glucose uptake on isolated tissue that was dissected from the animal and incubated in buffers while being contracted. Human studies must be whole-body experiments relying on multiple biological processes, such as the endocrine and circulatory system, to interact with skeletal muscle to take up and use glucose as fuel. Though differences may be explained by use of isolated tissue versus whole-body experiments, these discrepancies highlight the need for future human investigations to couple methods that measure substrate metabolism at the whole-body level with direct measurements within skeletal muscle to characterize changes in molecular pathways regulating glucose uptake and oxidation following carbohydrate and/or energy restriction.

Limitations of the Current Literature and Future Investigations

Although each individual study powered their investigations to adequately assess the impact of muscle glycogen on exogenous carbohydrate oxidation rates, the overall number of studies and thus sample size for this body of research is relatively small. There is likely interindividual variance that may result in different individuals, based on sex, training status, or genetic predisposition, exhibiting an altered response of exogenous carbohydrate oxidation following periods of energy or carbohydrate restriction during aerobic exercise. More research is required to better understand potential variances across individuals. Additionally, it should be noted that of the 5 studies that have assessed exogenous carbohydrate oxidation following carbohydrate restriction or energy restriction, only the study from our laboratory (19) directly measured muscle glycogen. Although the manipulation of study diets and/or exercise protocols were appropriate in all investigations, and these methods have been reported to be effective in reducing muscle glycogen content in other investigations (6, 19, 23, 24), the lack of a direct measurement makes the severity of muscle glycogen depletion, and thus its contribution to changes in substrate oxidation, unknown. To better understand the influence of the severity of glycogen depletion, future investigations seeking to understand the impact of carbohydrate restriction or energy restriction on the use of exogenous carbohydrate oxidation and physical performance should include the direct measurement of muscle glycogen and the molecular pathways that regulate glucose uptake and glycogen synthesis.

Combined results across studies suggest that exogenous carbohydrate supplementation is a viable nutrition intervention to sustain physical performance following acute carbohydrate restriction but may not be an effective ergogenic aid following energy restriction. Though energy restriction appears to reduce exogenous carbohydrate oxidation due to lower rates of glucose uptake resulting from insulin resistance, it is important to note these data primarily result from complete starvation (i.e., 0 kcals consumed). Energy restriction is common for athletes and military personnel, however, the severity is more likely to range from 20 to 60% energy deficit (13, 14), as opposed to full starvation. Presently, it is unclear if reductions in exogenous carbohydrate oxidation and insulin sensitivity will occur with less severe energy restrictions. Additionally, the severity of starvation-induced insulin resistance worsens with more prolonged starvation (42, 43). Potential time course shifts in exogenous carbohydrate oxidation and physical performance during energy restriction are currently unknown. Finally, consuming multiple transportable carbohydrates, such as combined glucose and fructose, has been well-established to increase exogenous and decrease endogenous carbohydrate oxidation rates compared with glucose alone during aerobic exercise under energy-/carbohydrate-adequate conditions (61–65). Whether consuming multiple transportable carbohydrates compared with glucose alone increases exogenous carbohydrate oxidation and improves physical performance during periods of energy restriction has not been determined.

Conclusion

In conclusion, the results from these individual studies suggest when exercise is initiated with low or high muscle glycogen content following an energy-adequate carbohydrate-restricted feeding there is no difference in exogenous carbohydrate oxidation during aerobic exercise compared with a mixed-macronutrient control diet. Conversely, if glycogen stores are reduced due to low energy availability exogenous carbohydrate oxidation is impaired compared with an energy-adequate diet. The difference between these 2 interventions is likely due to reduced glucose uptake into skeletal muscle as a result of starvation-induced insulin resistance. The molecular adaptations resulting in reduced glucose turnover and insulin sensitivity appear due to reductions in the signaling cascade that stimulates GLUT4 translocation. However, studies thus far are limited to severe energy restriction (e.g., starvation). The impact of more moderate energy restriction on exogenous carbohydrate oxidation remains unclear. Overall, it appears that exogenous carbohydrate may be an effective intervention to sustain physical performance following short-term energy-adequate carbohydrate-restriction diets. However, reductions in exogenous carbohydrate oxidation due to lower glucose uptake indicates carbohydrate supplements may not be an effective ergogenic aid when glycogen stores are low due to energy restriction.

Notes

This work was supported by the US Army Medical Research and Development Command.

Author disclosures: The authors report no conflicts of interest.

The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Army or the Department of Defense. Any citations of commercial organizations and trade names in this article do not constitute an official Department of the Army endorsement of approval of the products or services of these organizations.

Abbreviations used: AKT, protein kinase B; AMPK, AMP-activated protein kinase; AS160, AKT substrate of 160 kD; CPT1a, carnitine palmitoyl transferase 1a; FAT, fatty acid translocase; GLUT4, glucose transporter 4; HADHA, hydroxyacyl-CoA dehydrogenase; IRS1, insulin receptor substrate 1; PPAR, peroxisome proliferator-activated receptor; V̇O_2max, maximal oxygen uptake.