Glycemic response to exercise in the subarctic depends on time of day, season and sex

Between December 1992 and September 1996, a total of 62 healthy athletes (32 males and 30 females) exercised for 30 minutes, with an intensity of 60% VO2, in one or more morning and afternoon sessions in the months December, February, April, June and September. Blood samples obtained for determinations of glucose were drawn at 0, 1, 5, 10 and 30 minutes of each exercise span. The total area under the glucose-time curves was calculated both in original units (mmol/l) and after normalization to percent of starting value for 263 exercise spans. These calculated areas, representing the glucose ''response'' during exercise, were tested for the effect of sex, time of day and season by ANOVA and by the least-squares fit of a 1-year cosine. During exercise a significant effect was found by ANOVA for sex (p<0.001), time of exercise (p=0.006) and month (p=0.007). No significant interactions were found. Exercise in the morning beginning at 11:30h produced a smaller glucose response for both sexes, when compared with exercise in the afternoon beginning at 16:30h. With regard to sex, a smaller integrated glucose response to exercise occurred in females. The response was lowest for both sexes in Dec (winter) compared to other months, while the response was greatest for females in September, and for males in April and June. A circannual rhythm was found both for baseline glucose (p=0.005, acrophase=Feb 20) and glucose-time response areas in mmol/l (p<0.001, acrophase=Feb 4), and normalized values (p,=0.05, acrophase=Dec 19). This finding is in accordance with previous reports that, at least in the subarctic area, the glycemic response to a standardized exercise is influenced by season and timing of exercise. This observation may be of importance for athletes involved in vigorous training and patients with Diabetes Mellitus, among others.