Effect of sample volume size on single-breath transfer factor for the lung

Background: ATS/ERS 2005 guidelines advise a standardised protocol for measuring Transfer factor for the lung (TL CO ) via a single breath-hold method. The volume of exhaled gas sampled limits the test in subjects with a vital capacity (VC) under 1.0L. A measure of TL CO in this group would be clinically useful. This study aimed to see how reducing sample volume (V samp ) influenced TL CO . Methods: We randomly reviewed 125 (87 males, mean age 49-±8.8 years (± sd)) reproducible TL CO tests from our database. Five subgroups (n=25) were included: 1) healthy 2) restrictive lung function, and 3) mild 4) moderate 5) severe obstructive lung disease. For each patient two real-time gas traces were analysed. Washout volume was maintained, whilst the V samp was varied to give TL CO readings for V samp of 0.2L, 0.4L, 0.6L and 0.8L and 1.0L. The repeatability of TL CO at each V samp was investigated with repeatability coefficient. TL CO at each V samp was compared to the TL CO at 1.0L V samp and the difference illustrated using Bland & Altman plots. Results: Repeatability was not affected by changing V samp in any of the subgroups (repeatability coefficient: 0.42-0.64). Mean TL CO increased as little as 3-5% with increasing V samp, with the largest difference between a 0.2L and 1.0L V samp. However the imprecision of the mean (limits of agreement up to 0.65mmol/min/kPa) indicated a fall in TL CO equal to 15% was possible. Conclusion: In this study changing V samp does not compromise repeatability. A clinically significant variation in TL CO of up to 15% is possible with a reduced V samp . Further study is needed to confirm this observation and investigate the effect of changing V samp on transfer coefficient and alveolar volume.