· Index 
Pre-Event Meals:  High or Low Glycemic Index Foods? 

Louise Burke, Australian Institute of Sport, Canberra, Australia 

Athletes have been cautioned that eating carbohydrate foods in the hour before exercise may alter exercise metabolism by stimulating insulin production, which in turn increases the rate at which the muscles burn carbohydrate.  As a result of this faster rate of carbohydrate oxidation, blood glucose levels may actually fall (a condition known as hypoglycemia) shortly after exercise begins.  In most cases this effect is short term, and metabolism corrects itself as exercise continues.  The occasional athlete experiences impaired performance, and it has been found in one study. But mostly there are no problems, and in some situations pre-exercise carbohydrate can even improve exercise performance.  Nevertheless, the stigma about athletes eating carbohydrate before exercise persists, and many athletes and coaches talk about "rebound hypoglycemia" with fear, even though eating carbohydrate could potentially provide extra fuel during a prolonged training session or race.

But what if you could have your cake and eat it too?  What if there were carbohydrate foods that didn't cause such a large insulin response?   Since the 1980s nutritionists have recognized that carbohydrate-rich foods differ in their effects on blood glucose and insulin.  We now even have tables listing the carbohydrate-rich foods that promote a high glucose/insulin response (called high glycemic-index foods or high GI) and those that produce lower responses (low GI).  See my earlier article in Sportscience News on the glycemic index.

Some Australian researchers, led by Dr Diana Thomas, first raised interest in the role of the glycemic index in sports nutrition by manipulating the glycemic index of pre-exercise meals eaten by endurance athletes.  They reported a low GI carbohydrate meal (lentils) eaten one hour prior to exercise increased the time that cyclists could ride before fatiguing compared with an equal amount of carbohydrate eaten in the form of a high GI carbohydrate food (potatoes).  These findings were attributed to lower glucose and insulin responses to the low GI trial, which better stabilized blood glucose and fat levels throughout the cycling bout.  Although muscle glycogen was not measured,  the researchers suggested that glycogen sparing (slower use of muscle glycogen) may have occurred with the low GI carbohydrate trial, promoting better endurance (Thomas, Brotherhood, and Brand, 1991).

This is an example when publicity from a single study has lead to well-intentioned but premature nutrition advice.  Many sports nutrition articles and guidelines for athletes now recommend that endurance athletes choose low glycemic carbohydrate foods for their pre-event or pre-training meals.  Unfortunately, other investigations, including further studies from Dr Thomas' group, have not shown a clear benefit resulting from eating low GI carbohydrate foods before exercise.  The majority of studies show that even if  low GI pre-exercise meals produce better metabolic conditions during exercise compared with high GI meals, the differences are small and short lived.  The bottom line is that athletes probably perform the same on both pre-race menus.

Confused? Well, one important ingredient has been left out of the planning of all these studies.  In endurance exercise events the most common and effective strategy used by athletes to promote fuel availability is to consume carbohydrate-rich drinks or foods during the event.  The impact of consuming carbohydrate before and during exercise has not been well investigated.  So, while on sabbatical at the Sports Science Institute of South Africa in Cape Town,  I recently conducted a study, along with co-researchers Amanda Claassen,  John Hawley and Tim Noakes, to see whether the GI of pre-exercise meals has any effect on exercise metabolism and performance when large amounts of carbohydrate are consumed during the session (Burke et al., 1998).  Our study design attempted to blend sports nutrition guidelines with the real-life practices of competitive athletes.

We recruited six well-trained cyclists from the Cape Town area. These cyclists had an average maximum oxygen uptake of 68 ml/kg/min, which classifies them as good cyclists.  Each of them performed three trials in which they consumed a different pre-race meal two hours before undertaking an exercise test.  The three test meals consisted of a high GI carbohydrate meal (mashed potatoes topped with pasta sauce), a low GI carbohydrate meal (pasta topped with the pasta sauce), and a placebo or control meal (subjects ate low-calorie jelly, believing it to be a new "sports jelly").  The cyclists rode for two hours at 70% of their maximum oxygen uptake, equivalent to marathon pace or about 80% of maximum heart rate.  During this time we took blood and breath samples to determine which fuels they were burning.  At the end of the two hours, the cyclists did a time trial to complete a set amount of work, lasting around 15-17 minutes.

Fifteen minutes before starting their time trial, the cyclists consumed about 300 ml of a sports drink.  Then throughout the two hours of steady riding they continued to take regular drinks of this carbohydrate mixture.  In total, they drank about 700 ml per hour of the sports drink, taking in the recommended carbohydrate intake of about 60 g each hour.

We found that the placebo (jelly) meal did not change blood profiles before the start of exercise: blood glucose and insulin levels remained stable, and blood fats were at the high levels found after an overnight fast.  The low GI meal caused a small rise in blood glucose and insulin, while the high GI meal caused a much larger and sustained rise in glucose and insulin.  Both carbohydrate meals caused fatty-acid concentration in the blood to drop.

With the start of exercise and the intake of the sports drink, blood glucose increased in all trials.  In fact, blood glucose and the rate of carbohydrate burning were sustained throughout the two hours of riding and did not differ between trials.  There was no sign of the feared hypoglycemia, and subjects showed no sign of running out of carbohydrate fuel.  The contribution of the sports drink to the fuel mix was not changed by the choice of pre-event meal.

In the time trial that followed, there were no differences between the three test meals in the time to complete the performance ride. Five out of six subjects did slightly better after eating one of the carbohydrate meals, compared to the jelly (placebo), but there was a negligible difference in performance between eating pasta or potato before the trial.

In conclusion, our study showed that the intake of large amounts of carbohydrate during prolonged moderate-intensity exercise, according to current sports nutrition guidelines, maintained the fuel needs of our athletes in our endurance test.  Furthermore, it overrode any metabolic or performance effects arising from the type of pre-event meal.  If these results transfer to real endurance events, athletes don't need to worry about the glycemic index of the foods they eat before competing.  Rather, athletes should feel confident that when they consume adequate amounts of carbohydrate drinks or foods during endurance exercise sessions, they can choose their pre-exercise menu according to their personal preferences and previous experiences.


Burke, L.M., Claassen, A., Hawley, J.A., Noakes, T.D. (1998). No effect of glycemic index of pre-exercise meals with carbohydrate intake during exercise. Medicine & Science in Sports & Exercise, 30, S82 (Abstract 471).

Thomas, D.E., Brotherhood, J.R., Brand, J.C. (1991).  Carbohydrate feeding before exercise:  effect of glycemic index.  International Journal of Sports Medicine, 12, 180-186.

newseditor=AT=sportsci.org · webmaster=AT=sportsci.org · Homepage · Copyright ©1998
Edited by Mary Ann Wallace
· Webmastered by Jason Nugent · Last updated 15 May 98