Effects of a Low-Carbohydrate-High-Protein Pre-Exercise Meal in Type 1 Diabetes-a Randomized Crossover Trial

Context Current guidelines for exercise-related glucose management focus on reducing bolus and/or basal insulin doses and considering carbohydrate intake. Yet far less attention has been paid to the potential role of other macronutrients alongside carbohydrates on glucose dynamics around exercise. Objective To investigate the effects of a low-carbohydrate-high-protein (LCHP) compared with a high-carbohydrate-low-protein (HCLP) pre-exercise meal on the metabolic, hormonal, and physiological responses to exercise in adults with insulin pump-treated type 1 diabetes. Methods Fourteen adults (11 women, 3 men) with insulin pump-treated type 1 diabetes (median [range] HbA(1c) of 50 [43-59] mmol/mol (6.7% [6.1%-7.5%]), age of 49 [25-65] years, and body mass index of 24.0 [19.3-27.1] kg/m(2)) completed an unblinded, 2-arm, randomized, crossover study. Participants ingested isocaloric meals that were either LCHP (carbohydrate 21%, protein 52%, fat 27%) or HCLP (carbohydrate 52%, protein 21%, fat 27%) 90 minutes prior to undertaking 45 minutes of cycling at moderate intensity. Meal insulin bolus was dosed according to meal carbohydrate content but reduced by 25%. Basal insulin rates were reduced by 35% from meal ingestion to end of exercise. Results Around exercise the coefficient of variability was lower during LCHP (LCHP: 14.5 & PLUSMN; 5.3 vs HCLP: 24.9 & PLUSMN; 7.7%, P = .001). Over exercise, LCHP was associated with a lesser drop (LCHP: & UDelta;-1.49 & PLUSMN; 1.89 vs HCLP: & UDelta;-3.78 & PLUSMN; 1.95 mmol/L, P = .001). Mean insulin concentration was 30% lower during exercise for LCHP compared with HCLP (LCHP: 25.5 & PLUSMN; 11.0 vs HCLP: 36.5 & PLUSMN; 15.9 mU/L, P < .001). Conclusion Ingesting a LCHP pre-exercise meal lowered plasma glucose variability around exercise and diminished the drop in plasma glucose over exercise.