Proceedings of the British Thoracic Society. 1987 winter meeting. 8-9 December 1987, London. Abstracts.

In assessment of bronchial responsiveness to inhaled pharmacological agents it is customary to "normal- ise" the concentration-response curve (CRC) so that the baseline measurement of airway calibre is expressed as 100% and responsiveness given as the concentration of agonist causing a given percentage change. This is valid only if the slope of the CRC relates to starting airway calibre in a consistent manner. Increasing use is being made of partial flow-volume loops in bronchial challenge testing, although the influence of starting flow rates on the slope of the CRC so obtained has not been studied. We have investigated this relationship in 20 CRCs from normal subjects and 19 CRCs from asthmatic subjects. Methacholine was administered by the method of Juniper et al (Thorax 1978;33:705-10). Max- imal and partial flow-volume loops were obtained according to the method of Zamel (Clin Respir Physiol 1984;20:471-5). The concentration of methacholine producing a 40% fall in flow at a lung volume 40% above residual volume on the partial loop (V40p) was determined from the log concen- tration-response curve and the slope of the curve determined at that point. Similar assessments were made for a 20% fall in FEV, derived from the maximal loops. In all subjects there was a significant negative correlation between the starting V,P and the slope of the CRC (r = -0 73, p < 0 001). In subgroups of eight asthmatic and six normal subjects whose baseline V40p values overlapped each other's range, there was no significant difference in mean slope between the normal subjects (-16 (055) 1 s-'/mg ml-') and the asthmatic subjects (-199 (1 2) 1 s-'/mg ml-'). In contrast, the correla- tion between starting FEV, and slope was much weaker (r = 0-38, p < 0 05). We conclude that "normalisation" of the CRC to methacholine using partial flow-volume loops is a valid procedure. Normal and asthmatic subjects cannot be distinguished on the basis of the slope of such curves. The practice of "normalising" the FEV, concentration-response curve is less justifiable.