PURPOSE: The purpose of this study was to examine whether a swimming session performed in a pool sanitized with chlorine-based agents induces lung inflammation, modifies lung epithelium permeability, and alters lung function. METHODS: Eleven volunteers performed two standardized swimming sessions: one in a nonchlorinated indoor swimming pool and the other one in a chlorinated indoor pool. Lung inflammation was assessed by fractional exhaled nitric oxide (FE(NO)). Changes in lung epithelium permeability were estimated by measuring the surfactant-associated proteins Type A and Type B (SP-A and SP-B), the Clara cell protein (CC16), and the Krebs von den Lungen-6 protein (KL-6). Lung function tests were also performed. All measurements were carried out in basal conditions, after training completion and 3 h postexercise. Nitrogen trichloride (NCl3), the most concentrated gas derived from pool water chlorination, was measured in each pool during the swimming sessions. RESULTS: NCl3 ranged from 160 to 280 microg x m(-3) in the air of the chlorinated pool and was undetectable in the nonchlorinated one. Lung function was affected neither by the exercise session nor by the type of sanitation. Serum pneumoproteins were unchanged excepted SP-A which decreased by 8% after exercise in the chlorinated pool (P < 0.05). FE(NO) increased by 34% (P < 0.05) after exercise in the nonchlorinated pool, whereas it was unaffected in the chlorinated one. CONCLUSIONS: At concentrations lower than 300 microg x m(-3), NCl3 in an indoor chlorinated pool, does not produce short-term changes in lung function or in epithelial permeability. The unchanged FE(NO) found in the chlorinated pool after exercise suggests that chlorination might inhibit NO-induced vasodilation observed during exercise.