Previously we established that an acute inflammatory response in the upper respiratory tract of humans could be studied by analyses of nasal lavages (NL). The relationship of these cellular responses to responses in the lower lung has not been thoroughly investigated in humans. In this study we have compared the cellular changes detected in NL with those detected in the bronchoalveolar lavage (BAL) taken from the same individual. A group of 10 subjects was exposed to either filtered air or 0.4 ppm ozone (O3), with exercise, for 2 h. The NL was done prior to, immediately following, and 18 h postexposure; the BAL was done only at 18 h postexposure. A significant increase in PMN was detected in the NL immediately postexposure to O3 (7.7-fold increase; p = 0.003) and remained elevated in the 18 h post-O3 NL (6.1-fold increase; p less than 0.001). A similar increase in PMN was detected in the BAL 18 h after exposure to O3 (6.0-fold increase; p less than 0.001). The albumin levels in the NL and BAL were also similarly increased 18 h after O3 (3.9-fold and 2.2-fold, respectively). Although a qualitative correlation in the mean number of PMN existed between the upper and lower respiratory tract after O3, comparison of the NL and BAL PMN from each individual showed a significant quantitative correlation for the air data (r = 0.741; p = 0.014) but not for the O3 data (r = 0.408; p = 0.243). This study demonstrates that PMN counts in the NL can be a useful, inexpensive means of studying the acute inflammatory effect of ozone and monitoring those effects in the lower lung.