Lung Clearance Index: understanding variability from a paediatric perspective

Background: Cystic fibrosis (CF) is a recessive autosomal disorder caused by mutations on the CFTR gene, a membrane protein responsible for the transport of chloride. Alterations of ion content lead to thickened secretions in the airways, preventing clearance and encouraging infection and inflammation. In the management of CF, the most widespread lung function test is spirometry, but this is not sensitive in early disease and not feasible in infants. Multiple Breath Washout (MBW) measures the ventilation homogeneity by breathing of an inert gas and only requires passive cooperation, making it ideal in young children. Lung clearance index (LCI) is derived from the MBW and is the ratio of the cumulative expiratory volume (CEV) and the functional residual capacity (FRC). The LCI isn’t utilised routinely in clinical practice because of the relative lack of data, including the factors that influence its variability. Aims: To study LCI variability by: 1) the creation of z-scores for the reference population to aid in comparison of results 2) the exploration of the influence of the number of trials on the final LCI 3) the effect of time of the day on LCI. The results will then be used to determine the influence of age on LCI and LCI variability. Methods: MBW was performed using InnocorTM (inert gas: sulphur hexafluoride, SF6) and Exhalyzer D (inert gas: nitrogen, N2). Each test was performed in duplicate as a minimum, and analysed to obtain MBW derived indices. Data from healthy controls were used to create a z-score calculator to convert LCI raw values and to obtain reference values; these were then compared to those from the Exhalyzer D’s analysis software (Spiroware). For aim 2 the average LCI from 2 trials and 3 trials were compared with paired t-tests. For aim 3, one healthy control performed multiple MBW measurements in the morning and in the afternoon, and the mean LCI were compared with a paired t-test. To compare LCI and LCI variability in different age groups, HC and CF were divided into children and adults (≥18 years), and their LCI and coefficient of variation (CoV) analysed with t-tests. They were then further divided in children (<12 years), teenagers (between 12 and 18) and adults (≥18) and analysed with a one-way ANOVA test. Age and LCI were correlated with airway obstruction expressed in percent predicted FEV1 as per GLI reference equations. Results: LCI reference values from healthy controls are 5.35 to 7.31 on the Innocor and 5.91 to 8.66 on the Exhalyzer D (higher and more spread than Spiroware’s reference population). The LCI obtained from 2 trials is not significantly different from the LCI from 3 trials in HC and CF on Innocor and Exhalyzer D. Results from diurnal variation are contradicting. There is a significant difference in LCI between children and adults and between children and teenagers in HC and CF patients, but no difference in CoV was found. The CoV also remained similar in health and disease. When correlating percent predicted FEV1, LCI and age, children mainly have high FEV1 and low LCI, adults have low FEV1 and high LCI and teenagers are more uniformly distributed, with relatively more subjects in the mild obstruction group. Conclusion: Reference values for the RBHT population are now available, though the difference with Spiroware’s population (population size, ethnicity, baseline lung function) needs further investigation. The use of only 2 trials instead of 3 (as suggested by the guidelines) still provides reliable results and can potentially save time in the clinic. The contradicting results of diurnal variation suggest that studies with a larger population are needed. The similarity in CoV suggests that MBW is equally reliable in different age groups. Differences in LCI between age groups were seen in this cohort even in health; this might imply that different reference ranges could be useful. This confirms the potential value of LCI in early life