Hyperpolarized 3He and 129Xe MR Imaging in Healthy Volunteers and Patients with Chronic Obstructive Pulmonary Disease

Purpose: To quantitatively compare hyperpolarized helium 3 (He-3) and xenon 129 (Xe-129) magnetic resonance (MR) images obtained within 5 minutes in healthy volunteers and patients with chronic obstructive pulmonary disease (COPD) and to evaluate the correlations between He-3 and Xe-129 MR imaging measurements and those from spirometry and plethysmography. Materials and Methods: This study was approved by an ethics board and compliant with HIPAA. Written informed consent was obtained from all subjects. Eight healthy volunteers and 10 patients with COPD underwent MR imaging, spirometry, and plethysmography. Ventilation defect percentages (VDPs) at He-3 and Xe-129 imaging were obtained by using semiautomated segmentation. Apparent diffusion coefficients (ADCs) were calculated from He-3 (b = 1.6 sec/cm(2)) and Xe-129 (b = 12 sec/cm(2)) diffusion-weighted images. VDPs at hyperpolarized He-3 and Xe-129 imaging were compared with a two-tailed Wilcoxon signed rank test and analysis of variance; Pearson correlation coefficients were used to evaluate the relationships among measurements. Results: Xe-129 VDP was significantly greater than He-3 VDP for patients with COPD (P < .0001) but not for healthy volunteers (P = .35), although 3He and Xe-129 VDPs showed a significant correlation for all subjects (r = 0.91, P < .0001). The forced expiratory volume in 1 second (FEV1) showed a similar and significant correlation with He-3 VDP (r = 20.84, P < .0001) and Xe-129 VDP (r = -0.89, P < .0001), although the correlation between the FEV1/forced vital capacity (FVC) ratio and Xe-129 VDP (r = -0.95, P < .0001) was significantly greater (P = .01) than that for FEV1/FVC and He-3 VDP (r = 20.84, P < .0001). A significant correlation was also observed for He-3 and Xe-129 ADC (r = 0.97, P < .0001); Xe-129 ADC was significantly correlated with diffusing capacity of lung for carbon monoxide (r = 20.79, P = .03) and computed tomographic emphysema measurements (areas with attenuation values in the 15th percentile: r = 20.91, P = .0003; relative areas with attenuation values of less than 2950 HU: r = 0.87, P = .001). Conclusion: In patients with COPD, the VDP obtained with hyperpolarized Xe-29 MR imaging was significantly greater than that with He-3 MR imaging, suggesting incomplete or delayed filling of lung regions that may be related to the different properties of Xe-129 gas and physiologic and/or anatomic abnormalities in COPD.