Reproducibility of spirometric indices

The standard deviation of repeated measurements of the VC and the FEV1 in the same healthy adult is 90-200 mL, on average about 150 mL (ref. 1). The within subject reproducibility expressed as a coefficient of variation varies between 0.3 and 11.4%. Use of the coefficient of variation requires the spread in measurements to be proportional to the average value of the index, which is rarely the case in ventilatory indices.

The reproducibility of spirometric measurements performed on different days is better for the FVC and FEV1 than for forced expiratory flows; the above figure illustrates this in healthy adolescents investigated daily for one week.

The colored area represents one standard deviation above and below the mean (white line). Differences in the level of FEV1 and FVC between subjects have been eliminated; hence the figure shows exclusively the variability between subjects. The data derive from adults (figures modified from ref. 2).

If measurements are not performed at the same time of the day, then their reproducibility is affected by within day variations in ventilatory function. The highest values of FEV1 (figure above), FVC (figure below) and peak expiratory flow occur at about noon; the within day variability is greater in smokers and in subjects with respiratory symptoms, and is additionally age dependent (ref. 2).

Short term variability in FEV1 and FVC in patients with lung disease is larger than in healthy subjects. The 95th percentile of a chance change in FEV1 in adults is about 200 mL, in FVC about 340 mL (ref. 3).