Lung function distinguishes preschool children with CF from healthy controls in a multi-center setting.

RATIONALE: Conducting clinical trials in cystic fibrosis (CF) preschoolers has been limited by lack of sensitive lung function measures performed across sites.
OBJECTIVES: (1) Assess feasibility and short-term reproducibility of spirometry, forced oscillometry (FO), and inductance plethysmography (IP) in a multi-center preschool population; (2) compare ability of each technique to differentiate lung function of CF preschoolers and controls; (3) evaluate longitudinal changes in lung function; (4) estimate sample sizes for future trials.
METHODS: A longitudinal, multi-center study of CF preschoolers was conducted utilizing standardized equipment, rigorous site training, and centralized lung function data review. CF subjects participated in up to four study visits 6 months apart, plus a 2-week reproducibility visit. Controls had one study visit.
RESULTS: Ninety-three CF subjects and 87 controls participated. Acceptability rates were lowest for spirometry (55%) and highest for IP (77%). Spirometry success increased with age and having a prior acceptable measurement. FEV(1) , FEV(0.5) , and FEF(25-75) were lower for CF subjects than for controls; spirometric z-scores declined with age. IP measures of thoracoabdominal asynchrony were greater for CF subjects than for controls. FO indices did not distinguish CF from controls. FEV(1) and FEV(0.5) are able to detect the smallest treatment effect for a given sample size.
CONCLUSIONS: Spirometry appears more sensitive than IP or FO for detecting lung disease in CF preschoolers; spirometric indices decline with age. Future trials using spirometry should include a run-in period for training and require acceptable data prior to enrollment. However, near-normal spirometric measurements in CF preschoolers may lead to difficulty detecting a treatment effect.