Changes in pulmonary microvascular perfusion after SARS-CoV-2 infection remains poorly characterized due to limitations in longitudinal imaging. To provide a longitudinal evaluation of lung perfusion, we used quantitative dynamic contrast enhanced MRI (DCE-MRI) on a 0.55T MRI system in patients following COVID-19. In this prospective study, we performed quantitative DCE-MRI to generate perfusion maps on patients with a history of COVID-19 between June 2020 and June 2023. Imaging studies were divided into acute, recovery, convalescent, and extended study phases, collected over 3 years. Median lung perfusion, perfusion heterogeneity, perfusion defect percent, pulmonary transit time, arterial transit time, and transit time defect percent were measured. Perfusion metrics were correlated with pulmonary function tests (PFT), disease severity, cardiorespiratory symptoms, vaccine status, viral variant, and chest CT findings. We included 84 post-COVID-19 patients and 156 separate DCE-MRI exams for analysis, and 10 healthy volunteers. Statistical significance between patients with COVID-19 and healthy volunteers was observed for perfusion defect percent in all study phases (p < 0.01). Five patients had visible perfusion defects. Lower median perfusion was found in patients with low diffusing capacity for carbon monoxide (DLCO) in the recovery phase (p = 0.02), and patients with heterogeneous perfusion maps in the acute phase were more likely to have low DLCO at their final PFT measurements (p = 0.01). During the convalescent phase, patients with residual symptoms had lower median perfusion (p = 0.01), unvaccinated patients had higher transit time defect percent (p = 0.0047), and ground glass opacities on CT were associated with lower median perfusion (p = 0.004). Pulmonary microvascular perfusion abnormalities are found months after COVID-19, with correlative findings in patients with persistence of pulmonary symptoms and impaired pulmonary function. Our findings further support the utilization of DCE-MRI in a broad range of pulmonary vascular disorders. clinicaltrials.gov NCT04401449 (registered 2020-05-22) and NCT03331380 (registered 2017-11-02).
Campbell-Washburn et al. (Thu,) studied this question.