Percentiles of inspiratory capacity in healthy nonsmokers: a pilot study.

Background: Over the past decade inspiratory capacity (IC) has received increasing attention, especially in studies on chronic obstructive pulmonary disease. Though a few recent studies provide reference values for IC, still little is known on the distribution of its values in healthy subjects.

Objective: To estimate percentiles of IC in a sample of healthy nonsmokers by applying quantile regression.

Methods: We applied quantile regression to estimate seven percentiles of IC from 5 to 95% for a given age, height and body mass index, separately in males and females, in a sample of 411 healthy nonsmokers, aged 8-73 years, drawn from the general population in Northern Italy.

Results: The magnitude and statistical significance of the effect of age, height and body mass on IC varied across percentiles and between genders. When the 50th percentile of IC was compared with the predicted values of several studies, including those published by the European Community for Coal and Steel, the differences were as large as 0.86 liters in males and 0.90 liters in females. The possible advantage of the 5th percentile predicted from quantile regression over the 'traditional 5th percentile' as lower limit of normal was also illustrated.

Conclusions: Inference on percentiles, rather than just the mean, of IC and other lung function measures may help in the future to better understand the effect of various risk factors, model growth curves and derive more accurate reference values.