Transit time indices derived from forced expiratory spirograms: repeatability and criteria for curve selection and truncation.

Transit time analysis applied to spirograms truncated with respect to volume, not time, could be informative in population surveys. The present paper asks the questions: How does averaging several forced expirations affect the within subject variability at different levels of truncation? If only one spirogram is analysed, how should it be chosen? Are the results likely to be influenced by prior bronchodilation? The subjects were 46 men, selected by stratified random sampling from a shipyard population. Spirograms were obtained in quintuplicate, on two occasions, an average of nine days apart. Using the three "best" spirograms, transit time indices were calculated from average moments and from single curves selected in eight different ways; the analyses were performed on complete expirations, and after truncation by volume at levels between 75-99% of forced vital capacity (FVC), and by time (at 6 s). Indices from averaged moments were found to be more reproducible and yielded higher intra-class correlations than indices from single curves. Of the latter, the curve selected as having the largest forced vital capacity also yielded acceptable results. Prior inhalation of a bronchodilator aerosol did not reduce the variability. The averaging procedure is recommended for respiratory surveys of working men. For this application, the optimal level of truncation appeared to be by volume at 90% FVC, but on theoretical grounds truncation at 75 and 99% of FVC could also be informative.