THE RELIABILITY AND REPEATABILITY OF THE BLOOD-GLUCOSE RESPONSE TO PROLONGED EXERCISE IN ADOLESCENT BOYS WITH IDDM

OBJECTIVE - To determine the intrasubject reliability and repeatability of the blood glucose response to prolonged exercise in adolescents with insulin-dependent diabetes mellitus (IDDM) when pre-exercise meal, exercise, and insulin regimens are kept constant. RESEARCH DESIGN AND METHODS - Nine IDDM adolescent boys with diabetes duration of 9.7 +/- 4.8 years participated in two testing sessions 5-17 days apart. Carbohydrate intake, subcutaneous insulin injections, exercise bouts, and their timing were identical in both sessions. Exercise started 1 h after breakfast and consisted of six 10-min cycling bouts at moderate intensity (heart rate 145-150 beats/min), separated by 5-min rest periods. During rest periods, blood samples for glucose were taken and a supplemental carbohydrate beverage was consumed. Subjects were asked periodically to guess their own blood glucose levels. RESULTS - Intersession plasma glucose levels for each time period from the start of exercise to the end of the final recovery period were unchanged between sessions (P > 0.05) and highly correlated (r = 0.88 to r = 0.96, P < 0.01). The intraclass correlation for plasma glucose was 0.95. Decreases in plasma glucose from the start of exercise to the end of exercise were, al most, 5 mmol/l, and intersession decreases did not differ significantly (P > 0.05). One subject experienced hypoglycemia. The decrease in plasma glucose did not differ significantly between fair and poor glycemic control (glycosylated hemoglobin less than or equal to 10% and >10%, respectively). Subjects markedly underestimated their blood glucose levels when they were high, but their guesses were closer to the measured values at lower levels. CONCLUSIONS - Intersession observations demonstrated that the intrasubject blood glucose responses to prolonged moderate-intensity exercise were reliable and repeatable when pre-exercise meal, exercise, and insulin regimens were kept constant. This is an important finding for implementing and evaluating educational strategies for improving the metabolic control associated with prolonged exercise, as well as for future research on means for preventing exercise-induced hypoglycemia.