Abstract

BACKGROUND--Exercise testing has become an important tool in the diagnosis and treatment of restrictive lung disease. The reproducibility of variables measured during exercise testing was examined in subjects with stable restrictive lung disease. METHODS--Six subjects, who had never previously undergone exercise testing, each underwent three maximal incremental exercise studies on a bicycle ergometer conducted during a 28 day period. RESULTS--Data collected at rest, before exercise, were not significantly different during the three study days. Comparison of results at the end of the exercise tests from the three studies also revealed no evidence of a significant learning effect. Reproducibility of exercise performance by subjects was assessed by the coefficient of variation. The mean within subject coefficient of variation at the end of the exercise tests was 5.6% for work rate, 7.9% for exercise duration, and 9.5% for dyspnoea. The mean within subject coefficient of variation obtained at the end of the exercise tests was 5.3% for oxygen uptake (VO2), 2.5% for oxygen saturation (SaO2), 4.0% for heart rate (HR), 5.5% for minute ventilation (VE), 5.8% for respiratory frequency (f), and 4.6% for tidal volume (VT). The mean within subject coefficient of variation at 40% and 70% of maximal work rates for VO2 was 5.7% and 5.6% respectively, for SaO2 1.3% and 1.5%, for HR 4.8% and 4.0%, for VE 6.3% and 6.6%, for f 10.1% and 7.8%, and for VT 6.0% and 4.5%. CONCLUSIONS--Variables measured during clinical exercise testing in subjects with restrictive lung disease are highly reproducible. No significant learning effect was found on repeated testing in subjects who had never previously undergone exercise testing.