BACKGROUND: It is possible to measure nitric oxide (NO) levels in exhaled air. The absolute concentrations of exhaled NO obtained by separate workers in similar patient groups and normal subjects with apparently similar techniques have been very different. A study was undertaken to determine whether changes in measurement conditions alter the concentration of exhaled NO. METHOD: NO concentrations measured by a chemiluminescence analyser (Dasibi Environmental Corporation) and carbon dioxide (CO2) measured by a Morgan capnograph were analysed in single exhalations from total lung capacity in healthy volunteers (mean age 35.9 years). Ten subjects performed five exhalations at four different expiratory flow rates, at four different expiratory mouth pressures, and before and after drinking hot (n = 5) or cold (n = 5) water. Three subjects performed five exhalations on a day of high background NO (mean NO level 134 ppb) before and after a set of five exhalations made while both the subject and analysers were sampling from a low NO/NO-free reservoir system. RESULTS: The mean peak concentration of NO decreased by 35 ppb (95% CI 25.7 to 43.4) from a mean (SE) of 79.0 (15.5) ppb at an expiratory flow rate of 250 ml/min to 54.1 (10.7) ppb at 1100 ml/min. The mean peak concentration of NO did not change significantly with change in mouth pressure. The mean (SE) peak NO concentration decreased from 94.4 (20.8) ppb to 70.8 (16.5) ppb (p = 0.002, 95% CI 12.9 to 33.1) with water consumption. The mean NO concentration with machine and subject sampling from the low NO reservoir was 123.1 (19.4) ppb, an increase from results obtained before (81.9 (10.2) ppb, p = 0.001, 95% CI -19.9 to -62.7) and after (94.2 (18.3) ppb, p = 0.017, 95% CI 6.0 to 51.8) sampling with high ambient NO. CONCLUSIONS: The measurement of exhaled NO must be performed in a carefully standardised manner to enable different teams of investigators to compare results.