ECTOPIC FAT DEPOSITION IN THE LIVER AND PANCREAS AND THE PRESENCE OF EXCESS VISCERAL ADIPOSE TISSUE SIGNIFICANTLY INCREASE HOSPITALISATION RISK FOLLOWING COVID-19

ABSTRACT

Background and Aims: In patients with COVID-19, obesity may increase risk of hospitalisation, use of mechanical ventilation and patient mortality. High liver fat, body mass index (BMI) and male sex are significant predictors of hospitalisation risk following COVID-19. However, BMI is a poor indicator of body fat distribution. Here, we studied ectopic fat accumulation within the liver and pancreas and body composition through multiparametric magnetic resonance (mpMR) and compared participants with and without hospitalisation for COVID-19. Method: Participants with laboratory-confirmed or clinically suspected SARSCoV- 2 infection were recruited to the COVERSCAN study (NCT04369807;median time from initial symptoms = 177 days) and underwent a multi-organ mpMR scan (CoverScan®, Perspectum Ltd). Measures of liver and pancreatic fat (PDFF), liver fibroinflammation (cT1) and body composition [visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), skeletal muscle index (SMI)] were analysed. Differences between participants hospitalised (n = 59) and not hospitalised (n = 348) for COVID-19 were assessed using Wilcoxon signedrank tests. Univariate and multivariate analyses were performed on all biomarkers to assess the hospitalisation risk. Data presented are median values. Results: Approximately 6-months after initial symptoms, participants hospitalised following COVID-19 had significantly elevated pancreatic fat (3.8 % vs 2.8 %, p < 0.01), liver fat (3.8 % vs 2.4 %, p < 0.01) and liver cT1 (735ms vs 706ms, p < 0.01) compared to those who convalesced at home. Though hospitalised participants had a significantly elevated BMI (27 kg/m2 vs 25 kg/m2, p = 0.014), it was VAT, but not SAT, that was significantly elevated (132 cm2 vs 86 cm2, p < 0.01). Univariate analysis revealed that male sex, advanced age and elevated BMI, VAT, pancreatic fat, liver fat, and liver cT1 were all significantly predictive of hospitalisation following COVID- 19. In multivariate analysis, only age remained significantly predictive of hospitalisation. In hospitalised people with obesity (³ 30 kg/m2), VAT, liver cT1 and liver fat, but not BMI nor pancreatic fat, remained significantly elevated [VAT 220 cm2 vs 152cm2, p = 0.01 (Figure 1);liver fat 9.9 % vs 4.2 %, p = 0.003;liver cT1 782ms vs 742ms, p = 0.012]. Conclusion: mpMR revealed significantly elevated visceral and ectopic fat deposition within the liver and pancreas in hospitalised participants following COVID-19. In obese participants, BMI was not significantly different in hospitalised, and non-hospitalised patients, whereas visceral fat, liver fibroinflammation and liver fat were significantly elevated. Our work highlights body fat distribution an important consideration for COVID-19 risk profiling, which cannot be sufficiently evaluated based on BMI alone. (Figure Presented) Figure 1. Comparison of liver fat (left), pancreatic fat (middle) and visceral adipose tissue (right) between participants hospitalised and not hospitalised following COVID-19.