Test-retest reliability of symptom-limited cycle ergometer tests in patients with chronic obstructive pulmonary disease

Background: Symptom-limited exercise tests are widely used to evaluate the effects of pulmonary rehabilitation in patients with chronic obstructive pu lmonary disease (COPD), but the reliability of these tests is not well esta blished in COPD patients. Objectives: We compared test-retest reliability of two repeated symptom-lim ited exercise tests between COPD patients and healthy elderly subjects and between male and female patients. Method: Fifty-six COPD patients (40 men, 16 women) and 16 healthy subjects (6 men, 10 women) performed two symptom-limited exercise tests approximatel y 2 weeks apart. Measures of oxygen uptake ((V) over dot O-2), minute venti lation ((V) over dot E), heart rate, and ratings of breathlessness and leg fatigue were obtained at peak exercise at each symptom-limited exercise tes t. Results: Repeated measures of peak exercise responses were stable for patie nts and healthy subjects and for male and female patients. Although mean pe rcent error (absolute difference divided by mean) for peak exercise respons es was low, some individuals' values exceeded 10%. There was no difference in the percent error between CORD patients and healthy subjects or between men and women with COPD. Test-retest reliability was lower for breathlessne ss ratings than for other peak exercise responses for all groups. Conclusions: Repeated symptom-limited exercise tests are reliable in COPD p atients and healthy subjects. However, some individuals are less reliable, and these patients may require more than one exercise test to establish rel iable performance.