Several studies have examined the reproducibility over time of standard clinical measurements of pulmonary airways parameters, such as FEV1 and FVC. These studies indicate fairly consistent reproducibility of such measurements over periods of months and years. Although such traditional pulmonary function tests can provide quantitative measures of lung function, they do not provide any insight into the local or regional variation of individual airways. Therefore, in a longitudinal study in dogs, we directly measured the baseline size (cross-sectional area) of the same individual canine airways multiple times over a 1-yr period, using high-resolution computed tomography (HRCT). We compared the variability of individual airway size both within and between subjects to the various sizes of airways and to seasonal differences in size. Five anesthetized, intubated, and ventilated dogs were studied between five and eight times over a 15-mo period. HRCT scans were acquired at end expiration with the dogs at FRC on all occasions. All individual airways showed considerable variability in size over time as measured by the coefficient of variation (CV) (range: 9.2 to 69.3). The sizes of individual airways within a dog were seen to change by as much as twofold over time. Some dogs also showed significantly greater overall variability in airway size than did other dogs (p < 0.01), but this variability was related neither to the size of the airway (p = 0.34) nor to the season. These findings indicate a substantial local and regional variability in airway caliber that would not be detectable with conventional pulmonary function tests.(ABSTRACT TRUNCATED AT 250 WORDS)