The purpose of this study was to compare chemoresponses following two different intermittent hypoxia (IH) protocols in humans. Ten men underwent two 7-day courses of poikilocapnic IH. The long-duration IH (LDIH) protocol consisted of daily 60-min exposures to normobaric 12% O(2). The short-duration IH (SDIH) protocol comprised twelve 5-min bouts of 12% O(2), separated by 5-min bouts of room air, daily. Isocapnic hypoxic ventilatory response (HVR) was measured daily during the protocol and 1 and 7 days following. Hypercapnic ventilatory response (HCVR) and CO(2) threshold and sensitivity (by the modified Read rebreathing technique) were measured on days 1, 8, and 14. Following 7 days of IH, the mean HVR was significantly increased from 0.47 +/- 0.07 and 0.47 +/- 0.08 to 0.70 +/- 0.06 and 0.79 +/- 0.06 l.min(-1).%Sa(O(2))(-1) (LDIH and SDIH, respectively), where %Sa(O(2)) is percent arterial oxygen saturation. The increase in HVR reached a plateau after the third day. One week post-IH, HVR values were unchanged from baseline. HCVR increased from 3.0 +/- 0.4 to 4.0 +/- 0.5 l.min(-1).mmHg(-1). In both the hyperoxic and hypoxic modified Read rebreathing tests, the slope of the CO(2)/ventilation plot was unchanged by either intervention, but the CO(2)/ventilation curve shifted to the left following IH. There were no correlations between the changes in response to hypoxia and hypercapnia. There were no significant differences between the two IH protocols for any measures, indicating that comparable changes in chemoreflex control occur with either protocol. These results also suggest that the two methods of measuring CO(2) response are not completely concordant and that the changes in CO(2) control do not correlate with the increase in the HVR.