Longitudinal assessment of patients with cystic fibrosis lung disease with multivolume noncontrast MRI and spirometry

Background: MRI has been suggested as a radiation-free imaging modality to investigate early structural alterations and regional functional impairment in cystic fibrosis (CF) lung disease. Purpose/Hypothesis: To compare functional and morphological MRI changes over the course of the disease to changes in spirometry. Study Type: Longitudinal retrospective study. Population: Twenty patients with CF lung disease (at baseline, age = 16.5 (13.3–20.6) years, forced expiratory volume in 1 second (as % of predicted [%pred]) FEV1 = 71 (59–87) %pred, forced expiratory flow at 25–75% of forced vital capacity FEF25-75 = 39 (25–63) %pred. Field Strength/Sequence: 1.5T / T2-weighted HASTE; T2-weighted TSE-PROPELLER; T2-weighted bSSFP; T1-weighted 3D GRE. Assessment: Nonenhanced chest MRI and spirometry were retrospectively collected over a 3-year period from the initial recruitment visit. Images acquired at end-inspiration and end-expiration were registered by software using the optical flow method to measure expiratory-inspiratory differences in MR signal-intensity (Δ1H-MRI). Measures of CF functional impairment were defined from Δ1H-MRI: Δ1H-MRI median, Δ1H-MRI quartile coefficient of variation (QCV), and percent low-signal-variation volume (LVV). MR images were also evaluated by three readers using a CF-specific scoring system. Statistical Tests: Spearman correlation analysis, Spearman rank correlation analysis, linear mixed-effect model analysis, intraclass correlation coefficient. Results: Functional imaging parameters and total morphological score correlated with all spirometric measures, as did subscores of bronchial wall thickening/bronchiectasis, mucus plugging, and consolidation. Overall, the percent change of Δ1H-MRI median correlated with the percent change of FEV1 (ΔFEV1, r = 0.41, P < 0.01) and the percent change of FEF25-75 (ΔFEF25-75%, r = 0.38, P < 0.01). The percent change of LVV correlated with ΔFEV1 (r = –0.47, P < 0.001) and ΔFEF25-75 (r = –0.50, P < 0.001). Data Conclusion: These preliminary results suggest that nonenhanced multivolume MRI may provide a feasible tool to regionally map early pulmonary alterations for longitudinal evaluation of CF lung disease, without exposing the patients to ionizing radiation. Level of Evidence: 3T. Technical Efficacy Stage: 5.