Particulate air pollution is associated with asthma exacerbations and increased morbidity and mortality from respiratory causes. Ultrafine particles (particles less than 0.1 microm in diameter) may contribute to these adverse effects because they have a higher predicted pulmonary deposition, greater potential to induce pulmonary inflammation, larger surface area, and enhanced oxidant capacity when compared with larger particles on a mass basis. We hypothesized that ultrafine particle exposure would induce airway inflammation in susceptible humans. This hypothesis was tested in a series of randomized, double-blind studies by exposing healthy subjects and mild asthmatic subjects to carbon ultrafine particles versus filtered air. Both exposures were delivered via a mouthpiece system during rest and moderate exercise. Healthy subjects were exposed to particle concentrations of 10, 25, and 50 microg/m(3), while asthmatics were exposed to 10 microg/m(3). Lung function and airway inflammation were assessed by symptom scores, pulmonary function tests, and airway nitric oxide parameters. Airway inflammatory cells were measured via induced sputum analysis in several of the protocols. There were no differences in any of these measurements in normal or asthmatic subjects when exposed to ultrafine particles at concentrations of 10 or 25 microg/m(3). However, exposing 16 normal subjects to the higher concentration of 50 microg/m(3) caused a reduction in maximal midexpiratory flow rate (-4.34 +/- 1.78% [ultrafine particles] vs. +1.08 +/- 1.86% [air], p =.042) and carbon monoxide diffusing capacity (-1.76 +/- 0.66 ml/min/mm Hg [ultrafine particles] vs. -0.18 +/- 0.41 ml/min/mm Hg [air], p =.040) at 21 h after exposure. There were no consistent differences in symptoms, induced sputum, or exhaled nitric oxide parameters in any of these studies. These results suggest that exposure to carbon ultrafine particles results in mild small-airways dysfunction together with impaired alveolar gas exchange in normal subjects. These effects do not appear related to airway inflammation. Additional studies are required to confirm these findings in normal subjects, compare them with additional susceptible patient populations, and determine their pathophysiologic mechanisms.