RATIONALE AND OBJECTIVES: Hyperpolarized (HP) gas magnetic resonance imaging (MRI) is an advanced imaging technique that provides high-resolution regional information on lung function without using ionizing radiation. Before this modality can be considered for assessing clinical or investigational interventions, baseline repeatability needs to be established. We assessed repeatability of lung function measurement using HP helium-3 MRI (HP (3)He MRI) in a small cohort of patients with cystic fibrosis (CF). MATERIALS AND METHODS: We examined repeatability of HP (3)He MR images of five patients with CF in four scanning sessions over a 4-week period. We acquired images on a Philips 3.0 Tesla Achieva MRI scanner using a quadrature, flexible, wrap-around, (3)He radiofrequency coil with a fast gradient-echo pulse sequence. We determined ventilation volume and ventilation defect volume using an advanced semiautomatic segmentation algorithm and also quantified ventilation heterogeneity. RESULTS: There were no significant differences in total ventilation volume, ventilation defect volume, ventilation defect percentage, or mean ventilation heterogeneity (repeated-measures analysis of variance, P = .2116, P = .2825, P = .2871, and P = .7265, respectively) in the patients across the four scanning sessions. CONCLUSIONS: Our results indicate that total ventilation volume, ventilation defect volume, ventilation defect percentage, and mean ventilation heterogeneity as assessed by HP gas MRI in CF patients with stable health are reproducible over time. This repeatability and the technique's capability to provide noninvasive high-resolution data on regional lung function without ionizing radiation make (3)He MRI a potentially useful outcome measure for CF-related clinical trials.