User Option

Specific Topics
Protein Nutrition and Endurance Exercise: What
Does Science Say?
Continuing Education ACSM, NATA & The USA
Cycling Coaching Assn
Meet our GSSI team
SSE #93: Highs and Lows of Carbohydrate Diets

Sports Science Exchange  93
VOLUME 17 (2004) Number 2


Edward F. Coyle, Ph.D., FACSM
Department of Kinesiology and Health Education
The University of Texas at Austin
Austin, TX


  • A low-carbohydrate diet in athletes impairs their exercise tolerance and their ability to beneficially adapt to long-term physical training.
  • Physical performance and mood state seem better maintained with a high vs. moderate carbohydrate diet, thus reducing the symptoms of overreaching and possibly overtraining.
  • Adequate dietary carbohydrate is critical to raise muscle glycogen to high levels in preparation for the next day?s endurance competition or hard training session.  Accordingly, during the 24 h prior to a hard training session or endurance competition, athletes should consume 7-12 g of carbohydrate per kilogram of body weight.  However, during the 24 h prior to a moderate or easy day of training, athletes need to consume only 5-7 g of carbohydrate per kilogram of body weight.
  • Just as not every day of training should be intense or prolonged, not every day of training requires a high intake of carbohydrate.
  • Obese people consuming a very low-carbohydrate diet for 3-6 months can lose about 8% of their body weight compared to a 4% loss in those who eat a conventional diet that stresses reduced calories and fat.  Approximately one-half of the weight loss may be derived from body fat on each diet.  However, after 12 months on either diet, body weight reductions are likely to average only 2-4%, with little difference between diets.


Athletes and non-athletes alike seek simple, practical, and achievable dietary advice to help them reach their physical goals.  Non-athletes, because their daily activities require minimal physical exertion, often try to prevent weight gain and obesity by restricting energy intake.  On the other hand, many athletes choose high-energy diets of varying composition in hopes of improving performance.  Popular books and articles urge the weight-conscious to speed fat loss, whereas athletes are told to 'carbo-load ' all the time.  It is no wonder that people are confused about carbohydrate nutrition.

The scientific truth is that the amount and type of dietary carbohydrate should vary directly with the intensity and volume of exercise. For example, the amount of high-glycemic and refined carbohydrate consumed daily should be related to the extent to which the individual depletes muscle glycogen in training and competition.  In applying this concept, it is necessary to focus on the timing of carbohydrate ingestion and on daily energy balance.  It is also necessary to appreciate the importance of muscle glycogen for exercising intensely and the need for periodic, very intense training that specifically mimics the stress of competitions.  Therefore, daily, weekly or monthly variations in exercise training intensity and duration should be accompanied by appropriate variations in carbohydrate intake. 

This brief review attempts to place the confusion about dietary carbohydrate into a logical and, when possible, scientific perspective, with an emphasis on the needs of people who are physically active.  


The Obesity Epidemic

The worldwide obesity epidemic appears to be the result of energy imbalance arising from too little physical activity relative to energy (food) consumption (World Health Organization, 1998).  Public health messages and programs that are effective at increasing energy expenditure while reducing intake are urgently needed.  One or two decades ago, popular advice focused simply on reducing intake of all dietary fat, but that advice did not lessen the obesity epidemic.  Now, we are bombarded with advertisements reinforcing the notion that restricting carbohydrate intake is the ideal approach to losing weight.

Excess dietary fat indeed contributes to excess energy (calorie) intake and obesity, justifying the general recommendation that most people should reduce total dietary fat (Astrup et al., 2000).  Yet it is recognized that low-fat diets will not be effective in reducing body weight without also producing a reduction in total energy intake (Willet, 1998).  There is virtual agreement that the incidence of obesity could be reduced if people were to dramatically increase their physical activity to regularly exceed their energy intake. 

Low-Carbohydrate Diets for Body Weight Loss in Obese Sedentary People

Obviously, the only means by which physically inactive persons can prevent obesity is to restrain themselves from eating more energy than they expend over a period of weeks and months.  This goal of restraining energy intake might be one of the foundations underlying the vague hypotheses surrounding most of the currently popular low-carbohydrate diets (Atkins, 1998).   The bottom line is that it is more difficult to overeat when carbohydrates are eliminated from the diet because they are the major food source and typically a main component of mixed meals.

In two separate but similar studies lasting 6 months (Brehm et al., 2003) or 12 months (Foster et al., 2003), the effectiveness of a low-carbohydrate  diet for weight loss was evaluated under  proper experimental conditions.  Obese people (35-50% body fat with average body weight of 95 kg or 209 lbs) were randomly assigned to a dietary plan, and real-world conditions were simulated by limiting professional contact regarding dietary monitoring after literature was distributed and questions answered.  The popular Atkins diet (Atkins, 1998) served as the low-carbohydrate diet, which is unrestricted in total calories and high in fat and protein.  With this diet, carbohydrate is restricted to 20 grams/day for a minimum of two weeks during an induction phase.  During phases thereafter, carbohydrate foods are added by 5-10 grams per week until weight loss stabilizes at approximately 2 lbs (0.9 kg) of body weight per week.  The amount of carbohydrate eaten in this phase is supposedly 40-60 grams/day (Atkins, 1998).  In actuality, inactive women reportedly eat approximately 100 grams of carbohydrate per day after 6 months on the low-carbohydrate diet (Brehm et al., 2003).

The low-carbohydrate diets in the studies of Brehm et al. (2003) and Foster et al. (2003) were compared to conventional diets that were low in fat and calories (approximately 1200-1500 kcal/day for women and 1500-1800 kcal/day for men, with about 55-60% of total energy from carbohydrate, 25-30% from fat, and 15% from protein).  These daily diets provided approximately 225 grams of carbohydrate, 42 grams of fat, and 56 grams of protein. 

The findings of Brehm et al. (2003) and Foster et al. (2003) were remarkably similar; thus, their results are combined and displayed in Figure 1.  After six months, the conventional diets reduced body weight by 4%, whereas the low-carbohydrate diets resulted in an 8% weight reduction, significantly greater than that elicited by the conventional diet.  Therefore, the loss of body weight produced during the first six months of the low-carbohydrate diet appeared to be roughly twice as large as that produced by the conventional diet, i.e.,  about 17 lbs (7.7 kg) vs. 8 lbs (3.6 kg).  It should also be noted that almost all of the reductions in body weight occurred during the first three months on either diet and that body weight did not change significantly during the 3-6 month period.  Furthermore, Foster et al. (2003) compared the two diets after 12 months and found that the 4.4% reduction in weight in those following the low-carbohydrate diet was not statistically different (P=0.26) from the 2.5% loss experienced by those on the conventional diet (Figure 1). 

In 2004 two similar studies, one lasting 6 months (Yancy et al., 2004) and the other 12 months (Stern et al., 2004), produced substantially similar results to those of Brehm et al. (2003) and Foster et al. (2003), respectively.  Therefore, the take-home message appears to be that diet-only approaches to weight reduction appear most effective during the first three months, with weight losses maintained for about six months and gradually diminishing over the next six months.  During only the first 3-6 months, low-carbohydrate diets in obese people seem to elicit twice as much body weight loss as do conventional dietary plans that stress reducing calories and fat.  After one year, weight changes on the two diet plans are likely to be similar and small.

Composition of body weight loss.  On both the low-carbohydrate as well as the conventional diet, about 50-60% of the reduction in body weight was due specifically to reductions in body fat.  The remaining 40-50% of weight loss was ascribed to lean tissue mass comprised mostly of the water (50-80% of lean tissue weight) and protein found in skeletal muscle and organs such as liver, intestines, heart, and skin.  From this perspective, it appears that during the 3-6 month period of a low-carbohydrate vs. conventional diet, people lose about twice as much body mass from fat, water and protein stores.  Therefore, as a function of greater reductions in body weight and lean tissue mass, the low-carbohydrate diet would be expected to elicit a slightly greater loss of total body water.

Unresolved mechanisms.  It remains unresolved how a low-carbohydrate diet might elicit a greater loss of body fat (roughly 2 kg) after three months compared to a conventional diet.  Diets that are controlled to contain the same number of calories produce similar reductions in body weight, independent of whether they are low or high in carbohydrate (Golay et al., 1996).  Therefore, the greater amount of fat loss observed with a low-carbohydrate diet that allows voluntary selection of foods suggests that low-carbohydrate dieters may consume about 300 fewer kcal/day or expend 300 more kcal/day (or some combination thereof) compared to those consuming a conventional diet (Brehm et al., 2003).  Foster et al. (2003) speculated that a low-carbohydrate diet with unrestricted protein and fat intake may reduce energy intake because of the monotony or simplicity of the diet or because some factor(s) associated with the low-carbohydrate diet produces greater satiety, other effects on appetite, or better dietary adherence.  

Because the efficacy of weight-reducing diets is lessened over a 12-month period, there does not appear to be a significant long-term advantage of a low-carbohydrate versus a conventional diet.  Although this eventual loss of efficacy in dieting regimens is common, the reasons for it are unclear.  Do people begin to eat more after several months on a diet?  Do they gradually expend less energy as the diet is extended?  Attrition rates with either conventional diets or low-carbohydrate Atkins-type diets are high, with an approximately 40% dropout rate before 12 months (Foster et al., 2003).

Risk factors for heart disease.  One potential, but apparently unfounded, concern about low-carbohydrate diets centers around the heart disease risk factors associated with a high intake of saturated fat.  In the study reported by Brehm et al. (2003), women who reduced energy intake from 1600-1700 kcal/day to 1150-1300 kcal/day using a low-carbohydrate diet, increased their intake of saturated dietary fat by only 8-15 grams per day.  Protein intake increased from 63 to 78 grams per day.  In this study, the low-carbohydrate diet did not negatively influence the profile of plasma lipids related to heart disease. 

Different results were seen after three months of dieting in the study by Foster et al. (2003), in which the low-carbohydrate diet tended to increase LDL-cholesterol, whereas LDL-cholesterol  decreased on the conventional diet; the differences between diets were significant.  However, the authors stated that the increased weight loss associated with the low-carbohydrate diet might have offset the adverse effect of saturated dietary fat on LDL-cholesterol.  On the other hand, the low-carbohydrate diet produced some beneficial changes in the risk of heart disease because it produced greater increases in HDL-cholesterol and decreases in plasma triglyceride compared to the conventional diet.  Stern et al. (2004) and Yancy et al. (2004) reported similar beneficial results for atherogenic markers for low-carbohydrate diets compared to low-fat diets.

It should also be recognized that the low amount of vegetables, fruit and fiber consumed on a low-carbohydrate diet has the potential to increase the risk of coronary heart disease, cancer and diabetes (Schaefer, 2002).  Therefore at present, it is not clear if the benefits of the Atkins-type low-carbohydrate diet in promoting greater body weight and fat loss over a 3-6 month period compared to a conventional diet outweigh the potentially increased long-term risk of coronary heart disease.  As discussed below, exercise tolerance is reduced on a low-carbohydrate diet, and physical inactivity also increases the risk of coronary heart disease.

Low-Carbohydrate Diets In Body-Weight-Stable People Who Are Physically Active

A premise of the low-carbohydrate diet proposed by Atkins (1998) is that once a person achieves the desired amount of body weight loss, carbohydrate is added back to the diet to levels that allow that individual to maintain body weight.   However, studies that monitored body weight during a full year on Atkins-like diets reported that body weight begins to increase after six months on the diet (Foster et al., 2003; Stern et al., 2004).  Therefore, the long-term effectiveness of a low-carbohydrate diet for maintaining body weight at desired levels has not been established.  Dieters may be unable to sustain a low daily intake of carbohydrate or they may be unable to incorporate sufficient daily exercise into their lifestyles, possibly because they lack the carbohydrate energy necessary to make exercise tolerable.  Programs that add exercise to long-term weight-control regimens using conventional diets in order to increase energy expenditure are effective at maintaining weight loss and preventing weight gain (World Health Organization, 1998).  However, whether a program combining diet and exercise can be accomplished with very-low-carbohydrate diets is uncertain.

As discussed below, it is clear that a person's ability to tolerate and recover from intense exercise lasting several minutes or longer is directly related to the daily carbohydrate intake.  Therefore, people eating a low-carbohydrate diet (20-100 grams per day) should have a reduced exercise tolerance and thus an impaired ability to improve their physical endurance through training.  This is evidenced in a report by Helge et al. (1996), who studied young men who attempted to perform endurance training 3-4 times per week for seven weeks.  A low-carbohydrate group ate 15% of total calories as carbohydrate, whereas a high-carbohydrate group ate 65% or their calories as carbohydrate.  The authors concluded that  "a low-carbohydrate diet during an endurance training program is detrimental to improvement in endurance.?   Table 1 outlines some of the advantages and disadvantages of low- and high-carbohydrate diets for various populations and purposes.

Table 1.  Advantages and disadvantages of diets low and high in carbohydrate.

  Low-carbohydrate Diet High-carbohydrate Diet








Inactive obese


Twice the weight loss and fat loss after 3-6 months

May elicit satiety

More effective at increasing HDL-cholesterol and lowering plasma triglycerides

Part of the increased weight loss is derived from lean tissue.  Fat loss not maintained after 1 year. 

Produces ketosis.

May decrease ability to concentrate

Low in healthy fruits, vegetables, and fiber 

Does not lower LDL-cholesterol and total cholesterol in proportion to weight loss.


Low-calorie diets are not ketogenic.

May contain healthy fruits & vegetables

Low-calorie diets minimize loss of lean body mass.

Less loss of body weight

May not elicit satiety

May not decrease plasma triglycerides

Recreationally active, healthy body weight


Fewer food choices may reduce appetite

Difficult to exercise intensely; low in healthy fruits, vegetables, fiber;  

fewer food choices; boring


Intense exercise is less difficult to maintain.

May increase plasma triglycerides

Competitive athlete, lean


May help lose body fat.

May cause loss of muscle mass

May lead to overtraining


Increases Performance

Decreases Over-training

May increase plasma triglycerides

Carbohydrate loading every day may increase lipogenesis and/or body weight & fat

Note: Another classification of people is the recreationally active, but obese, individual.  These individuals might first focus on their obesity by adopting a diet that leads to negative energy balance.

Benefits of a High-Carbohydrate Diet in Athletes

Recovery of muscle glycogen stores between training sessions is critical for an athlete to train at the intensity of competition for prolonged periods in practice.  At the very least, about 24 hours are needed to replenish muscle glycogen after very hard exercise such as encountered in playing soccer, basketball or tennis intensely for 30-90 minutes.  Muscle glycogen can also be depleted after only 10-20 minutes of interval training in sports such as swimming, running, and cycling, during which exercise at intensities eliciting maximal oxygen uptake is performed for 1-5 minutes periods interspersed with 1-5 minutes of active recovery.  To replenish muscle glycogen in 20-24 hours, the diet must contain the proper amount and type of carbohydrate, and the carbohydrate consumption should be timed properly.  It obviously is not possible to fully restore muscle glycogen when athletes perform "two-a-days," i.e., exercising a given muscle group at moderate or high intensity twice in one day, typically with 4-12 hours between sessions.

Periodization of diet and emphasis on carbohydrate.   Just as not every day of training should be intense or prolonged, not every day of training requires a high intake of carbohydrate.  Unfortunately, there has been little investigation of how best to vary carbohydrate intake on a day-to-day basis to match the typical alteration of hard, easy, and moderate days of training performed during a week by well-coached competitive athletes.  It is assumed that the most important objective of periodization of daily carbohydrate intake would be to ensure high muscle glycogen levels at the start of the hard training sessions.  Athletes typically perform 2-4 'hard' training sessions per week.  To raise muscle glycogen to high levels, athletes should eat a total of 7-12 grams of carbohydrate/kg body weight during recovery from the last training session.  The recovery period should be not be less than 24 hours (Burke et al., 2004).   However, during the 24 hours prior to a moderate or easy day of training, it may be satisfactory for athletes to eat 5-7 grams of carbohydrate/kg.  If muscle glycogen is not fully recovered and the athletes sense this as a feeling of slight residual fatigue, they may refrain from exercising too intensely.  

An athlete's daily energy intake should generally match energy expenditure to minimize hunger and stress.  Fluctuations in carbohydrate intake can be matched by inverse fluctuations in calories from fat and or protein.  Thus, on the day before an easy day of training, if athletes choose to eat a moderate amount of carbohydrate (5 grams/kg), they can appropriately increase their intake of fat and protein.  In addition to providing them with a varied diet to satisfy taste, the extra dietary fat has the potential to raise the concentration of intramuscular triglyceride (Coyle et al., 2001),  a source of muscle fuel; extra protein may also be beneficial on a periodic basis. 

The importance of periodization of both training intensity (i.e., easy days to prepare for hard days) and amount of dietary carbohydrate is not typically addressed in scientific studies, in which training tends to be uniform to reduce experimental variability.  Furthermore, periodization is not specifically or adequately addressed in popular magazines for runners, cyclists or triathletes, possibly because the focus of the week then becomes the 1-2 hardest sessions performed at racing pace.  The vast majority of the readers of popular magazines are recreational athletes whose training and coaching is not specifically geared to eliciting peak performance, i.e., not hard-easy but instead uniformly moderate training.  Therefore, it is understandable how athletes might be exposed only to the simplistic message that they need to eat very-high carbohydrate diets.  

There are as many approaches to varying dietary carbohydrate as there are to weekly and monthly periodization of training intensity.  However, the most important aspect is that endurance athletes should not exercise for 20-24 hours prior to a hard training session, and they should eat 7-12 grams of carbohydrate/kg of body weight, as discussed below.

High-Carbohydrate Diets in the Daily Diets of Athletes.  Athletes in many sports attempt to reduce body fat as much as is appropriate for their circumstances.  Therefore, the simple advice to eat a high-carbohydrate diet is met with concern that it may lead to a positive energy balance and a gain in body fat.  In a 65-kg (143 lb) athlete, a daily intake of 7-12 grams of carbohydrate/kg of body weight would be 455-780 grams, amounting to 1820-3120 kcal.  This is the amount of carbohydrate needed to fully recover muscle glycogen.  However, this amount of carbohydrate can represent either a relatively large or small portion of an athlete's daily energy needs.  For example, for athletes who have depleted their muscle glycogen stores with brief, high-intensity interval training, a positive energy balance during recovery may be elicited with 7-12 grams of carbohydrate/kg of body weight.  On the other hand, in cyclists training for 4-6 hours per day, this amount of carbohydrate, while sufficient to replenish glycogen stores, may represent only one-half of the total energy intake needed for energy balance.  For these reasons, it is better to express an individual's carbohydrate requirements in grams/day rather than expressing it as a percentage of total energy, which varies greatly.  Besides, from the practical perspective of devising  diets for athletes, the number of grams of carbohydrate is easier to calculate than percentage of energy from carbohydrate, as the latter requires an accurate measure of total energy expenditure.  In other words, it is most effective to simply advise an athlete to eat a certain number of grams of carbohydrate per day. 

Gender considerations.  Given that female athletes are often concerned with minimizing body fat, they often are more reluctant to eat the large amounts of carbohydrate needed to fully restore muscle glycogen.  Female endurance athletes can replenish muscle glycogen to levels similar to those in males, but to do so, they must be willing to increase total energy (Tarnopolsky et al., 2001).  In other words, females are typically restrained eaters compared to males; because of their concern with body weight, they are normally reluctant to eat the large absolute amounts of carbohydrate, along with protein and some fat, that are needed to fully recover muscle glycogen prior to a hard bout of training or competition.

Practical Recommendations for Optimal Recovery of Muscle Glycogen

The supplement to this article presents recently published recommendations from world experts (Burke et al., 2004) aimed at athletes who need to quickly resynthesize muscle glycogen.   Given that complete muscle glycogen restoration takes at least 20-24 hours, athletes should not waste time.  They should ingest approximately 1 gram of carbohydrate/kg each hour after exercise until they eat their next large meal.  As discussed by Burke et al. (2004), attention has recently been focused on whether there is a benefit of adding protein (20-25% of energy intake) to a carbohydrate recovery drink.  It seems that adding protein to a carbohydrate recovery drink speeds glycogen recovery over the first 40 minutes, yet this benefit is lost by 60 and 120 min.  However, there may be an additional surge in glycogen resynthesis when protein is added to a carbohydrate feeding taken after 2 hours of recovery (Ivy et al., 2002).  This potential benefit of adding protein to recovery drinks may be negated by giving larger and more frequent carbohydrate feedings (Burke et al., 2004), but this seems to require a large positive energy balance and thus may not be practical. 

Training Benefits of a High-Carbohydrate Diet

It is difficult to conduct long-term training studies that compare performance improvements in athletes training on diets moderate in carbohydrate (5 grams/kg daily) and high in carbohydrate (7-12 grams kg daily).  Arguably the best-controlled training study, conducted with competitive rowers, demonstrated that a high-carbohydrate diet (10 grams/kg daily) usually, but not always, elicited superior rowing performance compared to a moderate-carbohydrate diet (5 grams/kg daily) over the course of four weeks of very intense training (Simonsen et al., 1991).  At no time was performance significantly better on the moderate-carbohydrate diet.

In another well-controlled study, runners were followed during 11 days of intensified training on either a moderate (5.4 grams/kg daily) or high (8.5 grams/kg daily) carbohydrate diet.  Physical performance was better maintained with the high- versus the moderate-carbohydrate diet as was mood state, thus reducing the symptoms of overreaching and possibly overtraining (Achten et al., 2004).


Achten, J., S.L. Halson, L. Moseley, M.P. Rayson, A. Casey, and A.E. Jeukendrup (2004). Higher dietary carbohydrate content during intensified running training results in better maintenance of performance and mood state.  J. Appl. Physiol. 96:1331-1340.

Astrup, A., G.K. Grunwald, E.L. Melanson, W.H. Saris, and J.O. Hill (2000). The role of low-fat diets in body weight control: a meta-analysis of ad libitum dietary intervention studies. Int. J. Obes. Relat .Metab. Disord. 24:1545-52.

Atkins, R. (1998). Dr. Atkins' new diet revolution.  New York: Avon Books.

Brehm, B.J., R.J. Seeley, S.R. Daniels, and D.A. D'Alessio (2003).  A randomized trial comparing a very low carbohydrate diet and a calorie-restricted low fat diet on body weight and cardiovascular risk factors in healthy women.  J. Clin. Endocrinol. Metab. 88: 1617-1623.

Burke, L.M., B. Kiens, and J.L. Ivy (2004). Carbohydrates and fat for training and recovery. J. Sports Sci. 22:15-30.

Coyle, E.F., A.E. Jeukendrup, M.C. Oseto, B.J. Hodgkinson, and T.W. Zderic (2001). Low-fat diet alters intramuscular substrates and reduces lipolysis and fat oxidation during exercise. Am. J. Physiol. Endocrinol. Metab. 280:E391-398.

Foster, G.D., H.R. Wyatt, J.O. Hill, B.G. McGuckin, C. Brill, C.S. Mohammed, P.O. Szapary, D.J. Rader, J.S. Edman, and S. Klein (2003). A randomized trial of a low-carbohydrate diet for obesity. N. Engl. J. Med. 348:2082-2090.

Golay, A., A. Allaz, Y. Morel, N. de Tonnac, S. Tankova, and G. Reaven (1996). Similar weight loss with low- or high-carbohydrate diets. Am. J. Clinical Nutrition 63:174-178.

Helge, J.W., E.A. Richter, and B. Kiens (1996).  Interaction of training and diet on metabolism and endurance during exercise in man.  J Physiol (Lond) 492:293-306.

Ivy, J.L., H. W. Goforth, Jr., B.M. Damon, T.R. McCauley, E.C. Parsons, and T.B. Price (2002).  Early postexercise muscle glycogen recovery is enhanced with a carbohydrate-protein supplement.  J. Appl. Physiol. 93:1337-1344.

Schaefer, E. J. (2002).  Lipoproteins, nutrition, and heart disease.  Am. J. Clin. Nutr. 75:191-212.

Simonsen, J.C., W.M. Sherman, D.R. Lamb, A.R. Dernbach, J.A. Doyle, and R. Strauss (1991). Dietary carbohydrate, muscle glycogen, and power output during rowing training.  J. Appl. Physiol. 70:1500-1505.

Stern, L., N. Igbal, P. Seshadri, K.L. Chicano, D.A. Daily, J. McGrory, M. Williams, E.J. Gracely, and F.F. Samaha (2004).  The effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-ujp of a randomized trial.  Ann. Intern. Med. 140:778-785.

Tarnopolsky, M.A., C. Zawada, L.B. Richmond, S. Carter, J. Shearer, T. Graham, and S.M. Phillips (2001).  Gender differences in carbohydrate loading are related to energy intake.  J. Appl. Physiol. 91:225-230.

World Health Organization (1998).  Obesity: preventing and managing the global epidemic.  WHO Technical Report Series, No. 916.  Geneva: World Health Organization.

Willet, W. (1998).  Is dietary fat a major determinant of body fat?  Am. J. Clin. Nutr. 67:S565-625.

Yancy, W.S. Jr., M.K. Olsen, J.R. Guyton, R.P. Bakst, and E.C. Westman (2004).  A low-carbohydrate, ketogenic diet versus a low-fat diet to treat obesity and hyperlipidemia: a randomized, controlled trial.  Ann. Intern. Med. 140:769-777.

Sports Science Exchange 93
VOLUME 17  (2004) NUMBER 2


It is widely known that athletes who train at high levels and/or for long durations must begin their competitions or hard training sessions with their muscles and livers well stocked with glycogen, the storage form of carbohydrate.  Without adequate stores of glycogen, maximal performance cannot be achieved.

Here are some practical dietary recommendations to help ensure that your glycogen stores are sufficiently replenished before your next training session or competition.  These recommendations were developed on behalf of the International Olympic Committee by a group of experts in sports nutrition and were modified from an original article cited at the end of this supplement.

  • You should aim to eat enough carbohydrate to meet the fuel requirements of your training program and to optimize restoration of muscle glycogen stores between workouts.  The following general recommendations are provided, but they should be fine-tuned with consideration of your individual total energy needs, your specific training objectives, and how well you perform after adjusting your diet
    • For fast recovery after a hard workout or competition, eat 1.0-1.2 grams of carbohydrate/kg body weight (0.45-0.55 g/lb) each hour for the first 4 hours of recovery.
    • In preparation for an easy day of moderate-duration, low-intensity training, your 24-hour recovery diet should include 5-7 grams of carbohydrate/kg (2.3-3.2 g/lb).
    • In preparation for a session of moderate or heavy endurance training or competition, your 24-hour recovery diet should include 7-12 grams of carbohydrate/kg (3.2-5.5 g/lb).
    • If you are participating in extreme exercise training (4-6 hours per day or more), your daily diet should include at least 10-12 grams of carbohydrate/kg (4.5-5.5 g/lb).
  • Choose nutrient-rich carbohydrate foods like fruits and vegetables and add other foods to recovery meals and snacks to provide a good source of protein and other nutrients.  These nutrients may assist in other recovery processes and, in the case of protein, may promote additional glycogen recovery when your carbohydrate intake is sub-optimal or when frequent snacking is not possible.
  • When the period between exercise sessions is less than 8 hours, you should begin eating carbohydrate as soon as practical after each workout to maximize recovery between sessions.  It may be advantageous to eat your carbohydrates as a series of snacks during the early recovery phase.
  • During longer recovery periods (24 h), you should organize the pattern and timing of carbohydrate-rich meals and snacks according to what is practical and comfortable for your individual situation.  Liquid and solid forms of carbohydrate are equally effective in replenishing glycogen.
  • Carbohydrate-rich foods like potatoes, pasta, oatmeal, and sports drinks that have a moderate-to-high glycemic index are good sources of carbohydrate for muscle glycogen synthesis and should be the major carbohydrate choices in recovery meals.
  • Adequate energy (calorie) intake is also important for optimal glycogen recovery; if you deliberately restrict your energy intake to lose weight, you may find it difficult to eat enough carbohydrate to optimize glycogen storage.
  • Don't base your intake of carbohydrate, fat, or protein on a percentage of your total energy intake because such guidelines are difficult to follow and are not strongly related to your muscles' absolute need for fuel.
  • Avoid drinking excessive amounts of alcohol during the recovery period because it is likely to interfere with your ability or motivation to follow guidelines for recovery eating.  Follow sensible drinking practices at all times, but particularly in the recovery period after exercise.


L.M. Burke, B. Kiens & J.L. Ivy, Carbohydrates and fat for training and recovery.  J. Sports Sci. 22:15-30, 2004.