Non-contrast enhanced magnetic resonance imaging detects mosaic signal intensity in early cystic fibrosis lung disease

Abstract

ObjectivesTo determine if morphological non-contrast enhanced magnetic resonance imaging (MRI) of the lung is sensitive to detect mosaic signal intensity in infants and preschool children with cystic fibrosis (CF). Materials and methods50 infant and preschool CF patients (mean age 3.5 ± 1.4y, range 0–6y) routinely underwent morphological (T2-weighted turbo-spin echo sequence with half-Fourier acquisition, HASTE) and contrast-enhanced 4D perfusion MRI (gradient echo sequence with parallel imaging and echo sharing, TWIST). MRI studies were independently scored by two readers blinded for patient age and clinical data (experienced Reader 1 = R1, inexperienced Reader 2 = R2). The extent of lung parenchyma signal abnormalities on HASTE was rated for each lobe from 0 (normal), 1 (<50% of lobe affected) to 2 (≥50% of lobe affected). Perfusion MRI was rated according to the previously established MRI score, and served as the standard of reference. ResultsInter-method agreement between MRI mosaic score and perfusion score was moderate with κ = 0.58 (confidence interval 0.45–0.71) for R1, and with κ = 0.59 (0.46–0.72) for R2. Bland-Altman analysis revealed a slight tendency of the mosaic score to underestimate perfusion abnormalities with a score bias of 0.48 for R1 and 0.46 for R2. Inter-reader agreement for mosaic score was substantial with κ = 0.71 (0.62–0.79), and a low bias of 0.02. ConclusionsThis study demonstrates that non-contrast enhanced MRI reliably detects mosaic signal intensity in infants and preschool children with CF, reflecting pulmonary blood volume distribution. It may thus be used as a surrogate for perfusion MRI if contrast material is contra-indicated or alternative techniques are not available.