Validating chest MRI to detect and monitor cystic fibrosis lung disease in a pediatric cohort

Abstract

RESULTS: Fifty-seven patients who had an MRI, CT and CFQ-R during the study period were included. CF-MRI bronchiectasis correlated with FEV1 , Pseudomonas aeruginosa, pulmonary exacerbations and patient-reported respiratory symptoms. CF-MRI trapped air only correlated with FEV1 and Pseudomonas aeruginosa. ICCs between MRI and CT bronchiectasis and trapped air were 0.41 and 0.35 respectively. MRI tended to overestimate bronchiectasis compared to CT. CONCLUSION: The associations between CF-MRI scores and several important clinical parameters further contributes to the validation of MRI. MRI provides different information than CT. BACKGROUND: Computed Tomography (CT) is the gold standard to assess bronchiectasis and trapped air in cystic fibrosis (CF) lung disease, but has the disadvantage of radiation exposure. Magnetic Resonance Imaging (MRI) is a radiation free alternative. OBJECTIVE: To validate MRI as outcome measure by: correlating MRI scores for bronchiectasis and trapped air with clinical parameters, and by comparing those MRI scores with CT scores. METHODS: In patients with CF (aged 5.6-17.4 years), MRI and CT were alternated annually during routine annual check-ups between July 2007 and January 2010. Twenty-three children had an MRI performed 1 year prior to CT, 34 children had a CT 1 year prior to MRI. Bronchiectasis and trapped air were scored using the CF-MRI and CF-CT scoring system. CF-MRI scores were correlated with clinical parameters: FEV1 , Pseudomonas aeruginosa, pulmonary exacerbations and patient-reported respiratory symptoms measured on the Cystic Fibrosis Questionnaire-Revised (CFQ-R), using Spearman's correlation coefficient. MRI and CT scores were compared using intra-class correlation coefficients (ICC) and Bland-Altman plots.