The aim of this study was to evaluate the precision and variability with which resting pulmonary function tests (resting PFTs) can be used to predict the capacity in exercise, and the usefulness of the different measurements of pulmonary function in the evaluation of impairment/disability in patients with chronic obstructive pulmonary disease (COPD). We studied 78 patients with stabilized COPD (FEV1 45.1 +/- 17.1%). Of these, 39 suffered severe impairment according to the resting PFTs. Both the group with severe impairment (maximal oxygen consumption [VO2,max]: 16.22 +/- 5 ml/kg/min; maximal minute ventilation [VEmax]: 31.87 +/- 7.1 L/min; maximal heart rate [HRmax]: 133.8 +/- 10.9 beats/min) and the group with nonsevere impairment according to the resting PFTs (VO2max: 22.55 +/- 7.9 ml/kg/min; VEmax: 42.11 +/- 10.9 L/min; HRmax: 138 +/- 13.7 beats/min) showed ventilatory limitation during exercise. FEV1 was the most prevalent criterion for the determination of severe impairment (FEV1 < 40%), and was the variable best correlated to VO2max (r = 0.52 with VO2max as absolute value and r = 0.54 with VO2max expressed as a percentage of the reference value), but the resting PFTs were not predictive of exercise performance including patients with severe COPD. Evaluation of working capacity (based on VO2max) revealed a significant number of inaccurate predictions among the results of the resting PFTs. Sensitivity and specificity analysis were used to compare the different criteria used to evaluate the severity of disability. They reveal that the classification will be different according to the criteria used.(ABSTRACT TRUNCATED AT 250 WORDS)