Printed in UK -all rights reserved D Dy ys sp pn no oe ea a a an nd d f fl lo ow w--v vo ol lu um me e c cu ur rv ve e d du ur ri in ng g e ex xe er rc ci is se e i in n C CO OP PD D p pa at ti ie en nt ts s Copyright ERS Journals Ltd 1994 European Respiratory Journal ISSN 0903 -1936 A. Noseda*, J-P. Carpiaux*, J. Schmerber*, F. Valente**, J-C. Yernault*** ABSTRACT: The purpose of this study, in patients with chronic obstructive pulmonary disease (COPD), was to examine the relationship of dyspnoea, rated on a visual analogue scale (VAS), to 1) tidal ventilatory variables measured on exercise and 2) pre-exercise lung function. Twenty one patients (forced expiratory volume in one second (FEV 1 ) mean (SD) 1.19 (0.32) l) were studied. During a preliminary test, the maximal workload was assessed and the upper end of the VAS was anchored. On the study day, the tidal flow-volume curve on exercise was monitored and dyspnoea was assessed serially every minute. In each individual, the relationship of dyspnoea to various tidal ventilatory variables was studied using linear regression analysis; results were reported as squared correlation coefficients, slopes and dyspnoea thresholds. Subsequently, the relationship of slopes and thresholds to pre-exercise lung function was examined.In all patients, dyspnoea showed a close correlation with ventilation, tidal volume, breathing frequency and tidal flow. The tidal peak inspiratory flow was the best individual predictor of dyspnoea with a median r 2 of 0.91. Patients with the poorest pre-exercise lung function exhibited the highest rates of increase in dyspnoea and the lowest thresholds, the strongest correlation being observed between the dyspnoea/ventilation slope and pre-exercise maximal peak inspiratory flow (r 2 = 0.54).In conclusion, for individual COPD patients dyspnoea on exercise is closely related to inspiratory flow. The degree of pre-exercise ventilatory impairment accounts, at most, for only half of the variation in dyspnoea perception between subjects.