Abstract Rationale The Lung Health Cohort (LHC) is examining determinants of lung health in 4000 young adults (aged 25-35). In this ancillary study, we are collecting xenon MRI (Xe-MRI) in 260 LHC participants. Xe-MRI enables measurements of regional ventilation, airspace size, and gas exchange. These measures augment the data collected in the LHC and provide additional understanding of structure and function in lung health. Our hypothesis is that impaired lung health is associated with abnormal alveolar structure and function assessed using Xe-MRI. We present a preliminary analysis of Xe-MRI features, lung function, and respiratory symptoms from the first 130 participants enrolled in XeLHC. Methods 130 participants (80 female) have been enrolled across 3 sites (KUMC-83, Duke-27, Iowa-20). Images were analyzed to generate quantitative image metrics of ventilation, alveolar size, and gas exchange including ventilation defect percent (VDP), apparent diffusion coefficient (ADC), membrane uptake, red blood cell (RBC) transfer, RBC/membrane, and RBC oscillation amplitude. Spirometry, respiratory symptoms, and demographic information were extracted from the baseline visit from the parent LHC. Linear models were used to examine the association of Xe-MRI measures of lung function with clinical measures of lung function (specifically forced expiratory volume in 1 second FEV1) and respiratory symptoms, controlling for sex, age, and the site of data acquisition as fixed effects. Results Xenon MRI was collected at a median of 131 days (interquartile range: 37-383 days) from the baseline visit. N = 15 patients had forced expiratory volume in 1 s (FEV1) ≤85%; N = 15 patients exhibited respiratory symptoms; N = 2 were in both groups. ADC (primary endpoint for comparison with lung function) had no association with FEV1 as a continuous variable (p = 0.6; Figure). However, ADC approached significance in participants with FEV1≤85% compared to those with FEV185% (p = 0.12; Figure). Neither VDP (p = 0.12), ADC (p = 0.39), nor RBC/Membrane (p = 0.77) (co-primary endpoints for respiratory symptoms) were significant predictors for respiratory symptoms (Figure). Secondary analyses suggest that VDP (p 0.001), ADC (p = 0.006), and RBC/Membrane (p = 0.007) are associated with FEV1/Forced Vital Capacity (FVC; Figure). Conclusions In this preliminary analysis for the Xenon MRI ancillary study to the lung health cohort, there is a trend between ADC and FEV1, but additional participants are needed to interrogate this fully. Secondary analyses demonstrate a strong association of Xe-MRI features with FEV1/FVC. Longitudinal follow-up of these participants will be critical for assessing which patients are at greatest risk for developing chronic lung disease. This abstract is funded by: R01HL168446
Niedbalski et al. (Fri,) studied this question.