CT Pulmonary Vessels and MRI Ventilation in Chronic Obstructive Pulmonary Disease: Relationship with worsening FEV1 in the TINCan cohort study

Abstract

The relationships between computed tomography (CT) pulmonary vascularity and MRI ventilation is not well-understood in chronic obstructive pulmonary disease (COPD) patients. Our objective was to evaluate CT pulmonary vascular and MRI ventilation measurements in ex-smokers and to investigate their associations and how such measurements change over time. Ninety ex-smokers (n = 41 without COPD 71 ± 10 years and n = 49 COPD 71 ± 8 years) provided written informed-consent to an ethics-board approved protocol and underwent imaging and pulmonary-function-tests twice, 31 ± 7 months apart. 3He MRI was acquired to generate ventilation-defect-percent (VDP). CT measurements of the relative area-of-the-lung with attenuation <-950 Hounsfield units (RA950), pulmonary vascular total-blood-volume (TBV) and percent of vessels with radius < one voxel (PV1) were evaluated. At baseline, there were significant differences in RA950 (p = 0.0001), VDP (p = 0.0001), total-blood-volume (p = 0.0001) and PV1 (p = 0.01) between ex-smokers and COPD participants as well as for VDP (p = 0.0001) in COPD participants with and without emphysema. The annual FEV1 change (-40 ± 93 mL/year) was not different among participant subgroups (p = 0.87), but the annual RA950 (p = 0.01) and PV1 (p = 0.007) changes were significantly different in participants with an accelerated annual FEV1 decline as compared to participants with a diminished annual FEV1 decline. There were significant but weak relationships for PV1 with FEV1%pred (p = 0.02), FEV1/FVC (p = 0.001), and log RA950 (p = 0.0001), but not VDP (p=0.20). The mean change in PV1 was also weakly but significantly related to the change in RA950 (p = 0.02). CT pulmonary vascular measurements were significantly different in ex-smokers and participants with COPD and related to RA950 but not VDP worsening over 2.5 years.