Abstract
BACKGROUND: Total lung capacity (TLC) measured with single-breath gas diffusion (TLCsb) is systematically lower than TLC measured with whole-body plethysmography (TLCwbp) especially in patients with obstructive defects. We aimed to develop and validate a regression correction equation to reduce the discrepancy between the two measurements of TLC. Second, we compared the ability to detect restriction (reduced TLC) from adjusted TLC measured by single-breath (TLCsb(adj)) with gold standard TLCwbp. METHODS: Lung function data from 800 consecutive patients were analysed with multivariable linear regression. A group of 530 were included for model development, and 270 were used for model validation. RESULTS: TLCsb was found to be on average 1.1 L lower than TLCwbp (p < 0.001). This difference increased with degree of airway obstruction. After adjustment TLCsb(adj) did not significantly differ from TLCwbp in obstructive and mixed obstructive-restrictive subjects. TLCsb(adj) had a sensitivity of 70% and a specificity of 99% to predict restriction on an individual basis, with a 95% confidence interval (CI) of [-19.6%; 17.7%] percentage when comparing adjusted values of TLCsb with the true TLCwbp value. CONCLUSIONS: After adjustment TLCsb was no longer significantly underestimated in obstructive and mixed restrictive-obstructive groups compared to TLCwbp. The adjustment can be used on individual subjects to estimate restriction via the TLCsb, thereby making the single-breath gas diffusion method a more valid alternative than without adjustment, when compared with the gold standard whole-body plethysmography to measure TLC.