This study examined the effect of 6 weeks of prescribed sprint training on the human growth hormone (hGH) response to cycle ergometer sprinting. Sixteen male subjects were randomly assigned to a training (n=8) or a control (n=8) group. Each subject completed two main trials, consisting of two all-out 30-s cycle-ergometer sprints separated by 60 min of passive recovery, once before, and once after a 6-week training period. The training group completed three supervised sprint-training sessions per week in addition to their normal activity, whilst control subjects continued with their normal activity. In the training group, peak and mean power increased post-training by 6% (P<0.05) and 5% (P<0.05), respectively. Post-exercise blood pH did not change following training, but the highest post-exercise blood lactate concentrations were greater [highest measured value: 13.3 (1.0) vs 15.0 (1.1) mmol l(-1)], with lower blood lactate concentrations for the remainder of the recovery period (P<0.05). Post-exercise plasma ammonia concentrations were lower after training [mean highest measured value: 184.1 (9.8) vs 139.0 (11.7) micromol l(-1), P<0.05]. Resting serum hGH concentrations did not change following training, but the peak values measured post-exercise decreased by over 40% in the training group [10.3 (3.1) vs 5.8 (2.5) microg l(-1), P<0.05], and mean integrated serum hGH concentrations were 55% lower after training [567 (158) vs 256 (121) min microg l(-1), P<0.05]. The hGH response to the second sprint was attenuated similarly before and after training. This study showed that 6 weeks of combined speed- and speed-endurance training blunted the human growth hormone response to sprint exercise, despite an improvement in sprint performance.