How should we estimate driving pressure to measure interrupter resistance in children?

Interrupter resistance (R(int)) is a widely used measure of airway caliber, but concerns remain about repeatability and sensitivity. Some R(int) variability may derive from the linear back-extrapolation algorithm (LBE 30/70) usually used to estimate driving pressure. To investigate whether other methods of estimating driving pressure could improve repeatability and sensitivity, we studied 39 children with asthma. Two measurements of R(int)-each the median of 10 interruptions-were made 5 min apart, and 14 children had a third measurement after bronchodilator (R(int)BD). Mouth pressure transients were analyzed using several algorithms, to compare the magnitude, repeatability, and sensitivity to bronchodilator change of R(int) values yielded. Algorithms taking driving pressure from later in the transient, predictably, yielded higher values of R(int) than those which back-extrapolated to time of valve closure. Algorithms which did not rely on back-extrapolation, including mean oscillation pressure (MOP) and mean plateau pressure (MP 30/70) had better repeatability. Sensitivity to detect change, calculated as ratio of bronchodilator response to repeatability coefficient (DeltaR(int)/CR), was also better for non-extrapolating algorithms: MP 30/70 1.67, LBE 30/70 1.28 (P = 0.0004). Measuring R(int) using techniques other than conventional back-extrapolation may give more consistent and clinically useful results, and these approaches merit further exploration.