Slope of the dose-response curve: usefulness in assessing bronchial responses to inhaled histamine

The value of determining the slope of the histamine dose-response curve, in addition to the histamine provocation concentration producing a 20% reduction in FEV1 (PC20-FEV1), was assessed by analysis of histamine dose-response curves in 40 patients selected as having a wide range of increased non-specific bronchial responsiveness to inhaled histamine. The histamine dose-response curves were found to be fit the linear curve (dose v response, mean r2 = 0.97) better than the logarithmic curve (log dose v response, mean r2 = 0.93), the difference being significant (p less than 0.001). There was a strong negative correlation between the PC20-Fev1 and the slope (r = -0.98, p much less than 0.001) and a weak negative correlation between the PC20-FEV1 and the log-dose-response slope (r = -0.38, p greater than 0.05). Sixteen normal subjects and 16 asthmatic patients were compared on the basis of histamine dose-response curves measuring fal in sGaw. In this study there was no difference between r2 for the linear determination and for the logarithmic determination (0.91 v 0.90, p less than 0.05). The PC35-sGaw showed a strong negative correlation with the dose-response slope (r = -0.95, p much less than 0.01) and no correlation with the log-dose-response slope (r = 0.09, p greater than 0.05). In the two studies there appeared to be little information gained from the determination of either the dose-response slope or the log-dose-response slope. The slope and the PC20-FEV1 were equally reproducible, duplicate determinations showing less than a two-fold difference in 14 of 15 paired PC20 measurements and in 13 of 15 paired slope measurements. In summary, the slope of the histamine dose-response curve appears to fit the linear model better than the logarithmic model. It is feasible to calculate it from the results of a standardised histamine inhalation test; determination of either the slope or the log-dose-response slope, however, appears to add little useful information. It is recommended that bronchial provocation test results should be expressed in terms of a threshold concentration such as the PC20-FEV1 or the PC35-sGaw.