Computerized single-breath nitrogen washout in children: variability and reproducibility.

The variability, short-term (1 hour) and long-term (2 weeks) reproducibility of the variables derived from the single-breath N2 test were evaluated in a group of 56 healthy children aged 10 to 16 years with a computerized system. The variability was low for vital capacity and total lung capacity (coefficient of variation less than 3%), moderate for residual volume and phase III N2 slope (9 and 13%) and very high for closing volume (more than 50%). A closing volume could not be identified by the computer algorithm in 18.3% of the trials. The reproducibility of the variables was satisfactory both at 1 hour and at 2 weeks, with the exception of the closing volume, whose increase at 2 weeks seemed to be due to methodological factors. We conclude that reliable results of static lung volumes and phase III N2 slope, but not of closing volume, may be obtained in children by a computerized system in epidemiological studies.