We investigated the effect of oral creatine supplementation (20 g d(-1) for 7 days) on metabolism during a 1-h cycling performance trial. Twenty endurance-trained cyclists participated in this double-blind placebo controlled study. Five days after familiarization with the exercise test, the subjects underwent a baseline muscle biopsy. Thereafter, a cannula was inserted into a forearm vein before performing the baseline maximal 1-h cycle (test 1 (T1)). Blood samples were drawn at regular intervals during exercise and recovery. After creatine (Cr) loading, the muscle biopsy, 1-h cycling test (test 2 (T2)) and blood sampling were repeated. Resting muscle total creatine (TCr), measured by high performance liquid chromatography, was increased (P < 0.001) in the creatine group from 123.0 +/- 3.8 - 159.8 +/- 7.9 mmol kg(-1) dry wt, but was unchanged in the placebo group (126.7 +/- 4.7 - 127.5 +/- 3.6 mmol kg(-1) dry wt). The extent of Cr loading was unrelated to baseline Cr levels (r=0.33, not significant). Supplementation did not significantly improve exercise performance (Cr group: 39.1 +/- 0.9 vs. 39.8 +/- 0.8 km and placebo group: 39.3 +/- 0.8 vs. 39.2 +/- 1.1 km) or change plasma lactate concentrations. Plasma concentrations of ammonia (NH(3)) (P < 0.05) and hypoxanthine (Hx) (P < 0.01) were lower in the Cr group from T1 to T2. Our results indicate that Cr supplementation alters the metabolic response during sustained high-intensity submaximal exercise. Plasma data suggest that nett intramuscular adenine nucleotide degradation may be decreased in the presence of enhanced intramuscular TCr concentration even during submaximal exercise.