HRCT and MRI of the lung in children with cystic fibrosis: comparison of different scoring systems

Background: Chest imaging is essential in the assessment of respiratory disease in cystic fibrosis (CF). High-resolution computed tomography (HRCT) can detect progressive lung disease but involves significant delivered dose of ionizing radiation. Magnetic resonance imaging (MRI) is radiation-free but is rarely used in CF. Based on the limited information on the potential interest of chest MRI in CF pediatric patients, the aims of our study were: 1) to evaluate and compare the reproducibility of HRCT and MRI scores; and 2) to evaluate the agreement between HRCT and MRI scores using both Helbich and Eichinger scores.

Methods: In this prospective study, CF children who were having a HRCT for their routine assessment were proposed to perform a chest MRI the same day. 17 patients were included (median age 12.7 years). Two radiologists scored independently HRCT (Helbich score) and MRI (Helbich and Eichinger scores); and established a consensus score. Concordance was assessed using the Intraclass Correlation Coefficient (ICC); and the inter-observer reproducibility between methods was compared using Fisher's Z test for dependent observations.

Results: Concordance between readers was almost perfect for HRCT score (ICC = 96%) and MRI-Eichinger score (84%), and substantial for MRI-Helbich score (68%). Correlation was strong between HRCT and MRI (r = 0.86 and 0.91 for HRCT and respectively MRI-Eichinger and MRI-Helbich scores) and the concordance almost perfect and substantial (ICC = 86% and 78% for HRCT and respectively MRI-Eichinger and MRI-Helbich scores).

Conclusions: We showed that, in CF children, MRI could adequately visualize lung morphologic changes when compared with the "gold-standard" HRCT. Regarding the potential cancer risks from associated ionizing radiation with HRCT, these results lead us to propose larger intervals of time between two lung HRCTs with realization of lung MRI in the meantime.