OBJECTIVES: To describe the short- and medium-term repeatability of lung clearance index at 2.5% (LCI2.5 ) in infants and calculate the number of patients needed to enroll in a study (N) using LCI2.5 as a primary outcome. METHODS: An 8-month follow-up observational study was employed for assessing short-term [coefficient of repeatability (CR) and intraclass correlation (ICC)] and medium-term repeatability (Bland-Altman method) of LCI2.5 in infants with cystic fibrosis (CF) or recurrent wheeze (RW) measured by the nitrogen multiple-breath washout test (N2 -MBW). Using these variability data, the N to reach 90% test power at the level of statistical significance (0.05) was calculated. RESULTS: Forty infants with CF and 21 with RW were enrolled. Initial N2 -MBW testing was successful in 33 and 17 patients, respectively. Follow-up data were available for 23 and 11 infants, respectively. Short-term repeatability of LCI2.5 was high (CR = 1.10 and 1.04 in CF and RW patients, respectively; ICC = 0.88 and 0.83 in CF and RW patients, respectively). The between-subject standard deviation was <13% of the actual LCI2.5 value. In clinically stable patients, LCI2.5 did not significantly change during the 8-month follow-up. Mean LCI2.5 change was -0.08 (1% of baseline) in CF and -0.05 (0.6%) in RW, with 95% limits of agreement being (-1.70; 1.53) in CF and (-1.51; 1.40) in RW patients. N = 23 infants if both intragroup differences of LCI2.5 and minimal difference to be detected would be 2.0. CONCLUSION: N2 -MBW may be a reproducible tool with reasonable test power to detect differences in infant studies.

Department of Pediatrics 2nd Faculty of Medicine Charles University and University Hospital Motol Prague Czech Republic

Department of Probability and Mathematical Statistics Faculty of Mathematics and Physics Charles University Prague Czech Republic

Citace poskytuje Crossref.org