Journal Article

BS26 Neuroendocrine activation and cardio-renal imaging post-COVID-19

Agtmael Tom Van, Ang, Daniel T, Friday, Jocelyn, Morphett, James, McGuinness, David, Lee, Jiyoung, McConnachie, Alex, Berry, Colin, Cole, John, Kamdar, Anna, Welsh, Paul, Jamieson, Nigel, Mangion Kenneth, Thomson, Cameron R

2025

Overview

IntroductionPost-COVID-19 syndromes remain a medical enigma. A disconnect between objective clinical markers and subjective symptoms complicates diagnosis and management for these patients in clinical practice. Catecholamines mediate the host stress response to illness. We therefore hypothesised dysregulation of catecholamine pathways influences the illness trajectory following COVID-19 infection.MethodsThis was a prospective, multicentre cohort study in the West of Scotland, involving post-COVID-19 patients enrolled during their hospitalisation or shortly after discharge (CISCO-19; NCT04403607). Participants underwent serial assessments, including during the index admission and then again 28–60 days after discharge. The design involved a self-controlled time-series of multisystem evaluations. We conducted a plasma proteomics analysis using the SOMAscan® assay v4.1, measuring 7288 proteins in a cohort of 155 participants who had undergone prospective, multisystem phenotyping (MRI, Siemens 3.0T PRISMA; CT chest and pulmonary angiography, Canon Aquilion ONE), incorporating clinical data gathered during the original study period. The statistical analysis of SOMAscan probe-sets, referred to as ‘proteins,’ was performed using R (v4.4.1) and Python (v3.11.5) using the Searchlight2 workflow. Differences in protein expression were assessed using the Wilcoxon rank-sum test. Over-representation analysis was conducted using the MSigDB 2024.1.Hs release for Gene Ontology (GO ) biological processes (GO-BP), GO cellular components (GO-CC), Hallmark, Reactome and STRING.A threshold of Benjamini-Hochberg adjusted p-value (p.adj) < 0.05 was established to determine statistical significance.ResultsAmong the participants (mean age: 55 years; 43% female), 72 (47%) had a history of cardiovascular disease, and 57 (37%) belonged to the highest quintile of social deprivation. Principal component analysis of 401 catecholamine proteins in GO-BP catecholamine pathways identified a distinct participant cluster (n=25, figure 1) characterised by protein expression indicative of heightened catecholamine activity, oxidative stress (Hallmark Reactive Oxygen Species, p.adj=0.03,figure 2) and mitochondrial dysfunction (Reactome, Mitochondrial Protein Degradation, p.adj<0.001) independent of disease severity. This group was characterised by heightened illness perception and reduced cardiorespiratory function, as measured by validated patient reported outcome measures, despite the absence of objective evidence of more severe disease on multisystem, cardio-renal and respiratory imaging in comparison to the remaining cohort (n=130).Abstract BS26 Figure 1Principal component analysis[Image Omitted. See PDF.]Abstract BS26 Figure 2Reactive oxygen species proteins. Blue/Turquoise=Group 1. Brown/Orange=Group 2[Image Omitted. See PDF.]ConclusionTargeted therapies aimed at reducing reactive oxygen species while also improving nitric oxide utilisation and addressing mitochondrial dysfunction within the vasculature could be areas warranting attention in Long-COVID.

• Publisher: BMJ Publishing Group LTD
• Subject: Catecholamines / COVID-19 / Illnesses / Principal components analysis / Protein expression / Proteins / Reactive oxygen species