Sympathetic and respiratory motor activities are entrained centrally. We hypothesize that this coupling may partially underlie changes in sympathetic activity evoked by hypoxia due to activity-dependent changes in the respiratory pattern. Specifically, we tested the hypothesis that sympathetic nerve activity (SNA) expresses a short-term potentiation in activity after hypoxia similar to that expressed in phrenic nerve activity (PNA). Adult male, Sprague-Dawley (Zivic Miller) rats (n = 19) were anesthetized (Equithesin), vagotomized, paralyzed, ventilated, and pneumothoracotomized. We recorded PNA and splanchnic SNA (sSNA) and generated cycle-triggered averages (CTAs) of rectified and integrated sSNA before, during, and after exposures to hypoxia (8% O(2) and 92% N(2) for 45 s). Inspiration (I) and expiration (E) were divided in half, and the average and area of integrated sSNA were calculated and compared at the following time points: before hypoxia, at the peak breathing frequency during hypoxia, immediately before the end of hypoxia, immediately after hypoxia, and 60 s after hypoxia. In our animal model, sSNA bursts consistently followed the I-E phase transition. With hypoxia, sSNA increased in both halves of E, but preferentially in the second rather than the first half of E, and decreased in I. After hypoxia, sSNA decreased abruptly, but the coefficient of variation in respiratory modulation of sSNA was significantly less than that at baseline. The hypoxic-evoked changes in sympathetic activity and respiratory pattern resulted in sSNA in the first half of E being correlated negatively to that in the second half of E (r = -0.65, P < 0.05) and positively to Te (r = 0.40, P < 0.05). Short-term potentiation in sSNA appeared not as an increase in the magnitude of activity but as an increased consistency of its respiratory modulation. By 60 s after hypoxia, the variability in the entrainment pattern had returned to baseline. The preferential recruitment of late expiratory sSNA during hypoxia results from either activation by expiratory-modulated neurons or by non-modulated neurons whose excitatory drive is not gated during late E.