Post-exercise ketosis in non-diabetic subjects

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

The effect of exercise on the total ketone body (acetoacetate + D-3-hydroxybutyrate) concentrations in the blood was studied to find out whether the susceptibility of non-athletes, compared with athletes, to develop post-exercise ketosis is the result of the former's increased reliance on glycolysis during exercise. In the first experiments, use was made of the diving reflex to induce peripheral vasoconstriction during exercise in both physically trained and untrained subjects. It was hoped that under these circumstances athletes and non-athletes would utilize similar amounts of muscle glycogen during exercise, and therefore develop similar degrees of ketosis after exercise, if the glycogen content of the muscles was in fact the factor which determined post-exercise ketosis. Ten non-athletic subjects, six long-distance runners, and three competitive swimmers were therefore studied before, and for 9½ hours after swimming in the early morning. The last meal was eaten during the evening before the swim. On the first test day the subjects swam underwater for as far as they could go three times in succession. A week later the same distance was swum on the surface without breath-holding. There was no increase in the post-exercise blood ketone body concentrations in any of the subjects after either form of the exercise, compared with control day values (when the subjects fasted, but did not swim at 07h30). Similar results were obtained when healthy young medical students (aged 18 - 23 years; trained and untrained) performed maximal exercise for 15 minutes, or moderate exercise for up to 90 minutes, on a bicycle ergometer. When six older subjects (aged 30 - 51 years) exercised at 75 W for 90 minutes, three of them developed ketonaemia, which reached its maximum intensity about three hours after exercise. The exercising heart rates of these older subjects were similar to those of the younger non-athletic subjects who had performed the same exercise, but had not developed post-exercise ketosis., An extra 60 - 90 g sucrose in the diet of the subject who had developed the most marked post-exercise ketonaemia, abolished the response, whereas carbohydrate restriction intensified it. A protein-fat diet caused two well trained marathon runners to develop the highest post-exercise blood ketone body levels yet recorded (3,88 mmoles/l). Free fatty acid, glucose, growth hormone and insulin concentrations in the serum followed patterns different from the ketone body levels during, and for 7 ½ hours after exercise, but were also affected more by diet than by training. Post-exercise ketosis, previously ascribed to a lack of athletic training, could equally well be ascribed to the lower carbohydrate intake of sedentary subjects compared with athletes: the two marathon runners were estimated to eat about twice as much carbohydrate in their regular diet than the sedentary subject who had developed post-exercise ketonaemia without carbohydrate restriction. The final experiments were designed to find out whether post-exercise ketosis was the result of the low levels of glycogen in the body, or of the gluconeogenesis which occurs after exercise to replenish the carbohydrate stores. Twenty-four highly trained athletes were therefore studied after prolonged exercise following a protein-fat diet to induce post-exercise ketosis. Six of them were then given 100 g alanine to take by mouth, six ingested 100 g glucose, six ingested 100 g starch, and the remaining six acted as controls. It was found that both alanine and glucose ingestion reduced the blood ketone body concentration from about 2 mmoles/l to less than 0,4 mmoles/l in 3 hours. Starch had a minimal effect on the blood ketone body levels during the 5-hour observation period. Alanine and glucose exerted their antiketogenic effects in the context of widely different serum insulin, glucagon and growth hormone concentrations. Similar results were obtained in starvational ketosis, and even in normoketonaemic subjects. The results indicate that ketogenesis is not the result of gluconeogenesis, nor of a low insulin/glucagon (+ growth hormone) ratio in the blood. It is concluded that low levels of glycogen, or of a metabolic intermediary of glycogen metabolism (such as glucose-1-phosphate, or glucose-6-phosphate) in the liver is probably the single most important stimulus for ketogenesis after exercise and starvation.