Clinical Feasibility of Structural and Functional MRI in Free-Breathing Neonates and Infants

Background: Evaluation of structural lung abnormalities with magnetic resonance imaging (MRI) has previously been shown to be predictive of clinical neonatal outcomes in preterm birth. MRI during free-breathing with phase-resolved functional lung (PREFUL) may allow for complimentary functional information without exogenous contrast. Purpose: To investigate the feasibility of structural and functional pulmonary MRI in a cohort of neonates and infants with no cardiorespiratory disease. Macrovascular pulmonary blood flows were also evaluated. Study Type: Prospective. Population: Ten term infants with no clinically defined cardiorespiratory disease were imaged. Infants recruited from the general population and neonatal intensive care unit (NICU) were studied. Field Strength/Sequence: T1-weighted VIBE, T2-weighted BLADE uncorrected for motion. Ultrashort echo time (UTE) and 3D-flow data were acquired during free-breathing with self-navigation and retrospective reconstruction. Single slice 2D-gradient echo (GRE) images were acquired during free-breathing for PREFUL analysis. Imaging was performed at 3 T. Assessment: T1, T2, and UTE images were scored according to the modified Ochiai scheme by three pediatric body radiologists. Ventilation/perfusion-weighted maps were extracted from free-breathing GRE images using PREFUL analysis. Ventilation and perfusion defect percent (VDP, QDP) were calculated from the segmented ventilation and perfusion-weighted maps. Time-averaged cardiac blood velocities from three-dimensional-flow were evaluated in major pulmonary arteries and veins. Statistical Test: Intraclass correlation coefficient (ICC). Results: The ICC of replicate structural scores was 0.81 (95% CI: 0.45–0.95) across three observers. Elevated Ochiai scores, VDP, and QDP were observed in two NICU participants. Excluding these participants, mean ± standard deviation structural scores were 1.2 ± 0.8, while VDP and QDP were 1.0% ± 1.1% and 0.4% ± 0.5%, respectively. Main pulmonary arterial blood flows normalized to body surface area were 3.15 ± 0.78 L/min/m2. Data Conclusion: Structural and functional pulmonary imaging is feasible using standard clinical MRI hardware (commercial whole-body 3 T scanner, table spine array, and flexible thoracic array) in free-breathing infants. Evidence Level: 2. Technical Efficacy: Stage 1.