Specific approach on dose-response curves to inhaled carbachol assessed by the interruption technique in children.

To determine significant thresholds obtained by bronchial challenge tests in children for assessment of bronchial reactivity, changes of respiratory resistance (Rint) induced by inhaled carbachol were evaluated in 20 asthmatic children (10 boys and 10 girls, mean age 9.1 +/- 2.8 y, range 5 to 15 y) and 20 healthy control children (10 boys and 10 girls, mean age 12.1 +/- 2.6 y, range 8 to 16 y). Baseline lung function was assessed by whole body plethysmography. The carbachol challenge test was performed by the interruption technique. The objectives were to search for the most predictable provocation dose (PD) according to statistical and practical considerations. The statistical argument is given by the condition that the target PD must definitively be out of the measurement error of baseline values. From the practical point of view, the object is to provoke the subjects the least amount possible. As thresholds of bronchial reactivity, the provocation doses inducing a 50%, 65%, 75%, or 100% increase in baseline Rint (PD50, PD65, PD75, PD100) were computed by a 4th power polynomial function analysis including data points of the entire dose-response curve. In comparison, a new threshold, the "variance based" provocation dose (PDvb), was defined as threshold to be definitively above the range of the baseline Rint variation before initiating challenge [PDvb > mean + 2 SD range of Rint(0)]. Although the PD50, PD65, PD75, and PD100 can be adequately computed by the 4th power function from the whole dose-response curve, comparison between the different provocation thresholds revealed that PDvb gives the best values to differentiate between healthy and asthmatic children (p < 0.002).(ABSTRACT TRUNCATED AT 250 WORDS)