Treatment for exercise induced asthma (EIA) in sporting competition is controlled to prevent the use of agents which might enhance physical performance. There is little information concerning the effects of the long-acting inhaled, and oral, sustained release type bronchodilators on the cardiorespiratory effects of submaximal exercise. The aim of this study was to compare the cardiorespiratory effects of submaximal exercise in patients with EIA before and after pretreatment with high-dose inhaled salmeterol xinafoate (SX) and controlled release oral salbutamol (CR). Patients were treated with SX (100 micrograms b.d.) and CR (8 mg b.d.) for > or = 3 days in a double-blind randomized cross-over design, with a 5-14 day washout period between treatments. A submaximal exercise test (total exercise time 6 min, final 3 min at 60% of VEpeak) was performed prior to each treatment period, and repeated at 1, 6, and 12 h postdose at the end of the treatment period. Two subjects were withdrawn from the study. Three subjects required relief medication after 1 h (CR) and one subject after 6 h (SX) and they did not perform further exercise tests. Both treatments increased baseline FEV1, with SX producing significantly greater pre-exercise bronchodilation than CR (P = 0.04). Following CR, there were no significant differences from the pretreatment values for VO2, VE, respiratory exchange ratio, heart rate, ventilatory equivalents for VO2, and oxygen pulse during the submaximal exercise challenge. Following SX, there were no significant differences for any of the exercise variables except for VE at 6 and 12 h (mean increase 4.27 l min-1 at 6 h, P < 0.01 and 4.69 l min-1 at 12 h, P = 0.05). The changes in ventilation following SX did not have an effect on oxygen consumption, and the ventilatory efficiency (VE/VO2) remained unchanged. The findings from this study demonstrate that, despite exercising from a higher baseline FEV1, short pretreatment periods with controlled release oral salbutamol and with inhaled salmeterol do not confer any cardiorespiratory advantage during submaximal exercise in subjects with EIA.