Contrast enhanced MRI does not detect differences in pulmonary perfusion between patients with long COVID and controls

<bold>Aim:</bold> To evaluate whether lung perfusion MRI detects lung perfusion differences in patients with long COVID (never hospitalised) both with and without dyspnoea, and controls. <bold>Methods:</bold> Non-hospitalised long COVID (NHLC) patients and controls who recovered fully from COVID-19 were recruited from post-COVID clinics, underwent MRI including contrast enhanced lung perfusion at either 1.5 or 3T. Variable flip angle images were acquired to correct for lung T1 and proton density. Participants with significant smoking history, cardiorespiratory diseases or abnormal CT were excluded. A general linear model was constructed to evaluate differences between participant groups (controls, NHLC with dyspnoea and NHLC without dyspnoea) adjusting for age, height and field strength using SPSS. <bold>Results:</bold> 94 patients with NHLC and dyspnoea (61% female, median(IQR) age=45(15) years), 35 patients with NHLC without dyspnoea (54% female, 47(19) years) and 31 controls (77% female, 46(21) years) underwent successful lung perfusion MRI. PBV was lower at 3T (p=<0.001). There were no significant differences in PBV between groups (p=0.746). PBF decreased with age (p<0.001) and height (p=0.010). There were no significant differences in PBF between groups (p=0.568). MTT increased with age (p=0.011), height (p=0.019) and was shorter at 3T (p=0.022). There were no significant differences in MTT between groups (p=0.346). Lung perfusion heterogeneity was assessed using the interquartile range of PBV, PBF and MTT. No significant differences were found in any heterogeneity metrics between groups. <bold>Conclusion:</bold> No differences in pulmonary perfusion were found between controls and NHLC patients with and without dyspnoea.