Long-COVID Phenotypes Are Not Associated With Persistent Inflammation, Endothelial Dysfunction, or Complement Activation Over 3-years in COVID-19 ALI/ARDS Survivors

Abstract RATIONALE: Greater inflammation and endothelial dysfunction are associated with COVID-19 acute lung injury and acute respiratory distress syndrome (ALI/ARDS) and may persist among those who develop long-COVID. We tested associations of inflammation, endothelial dysfunction, and complement activation with reported long-COVID symptoms and physical function deficits in the first three years following COVID ALI/ARDS. METHODS: We conducted a single-center New York City-based prospective longitudinal cohort study of 150 adults hospitalized with severe or critical COVID-19 in 2020, with sampling weighted to include 50% invasive mechanical ventilation survivors. Serum samples were obtained at hospital discharge, 4-months, 15-months, and 3-years. Biomarkers of inflammation and aging included interleukin-6 (IL-6), tumor necrosis factor (sTNFRI), interleukin-8 (IL-8), and growth differentiation factor-15 (GDF-15). Biomarkers of endothelial activation included ADAMTS13, angiopoietin, intercellular adhesion molecule-1 (sICAM-1), and vascular cell adhesion molecule-1 (sVCAM-1). Biomarkers of complement activation included C2, C4b, and C5. Symptom questionnaires and frailty assessments were conducted at 15-months and 3-years. Using questionnaires about severity and frequency of 43 symptoms, we calculated and defined post-acute sequelae of SARS-CoV-2 (PASC) positivity based on the NIH Recover definition of a PASC score ≥12. In sensitivity analyses, we tested associations with severe symptom phenotypes, defined as patient-reported moderate or greater post-exertional malaise (PEM), brain fog, or fatigue. We tested unadjusted associations and used generalized additive models with covariate balancing propensity scores to test adjusted associations between biomarkers and the predicted risk of PASC positivity, severe symptom phenotypes, and frailty categories (robust, pre-frail, and frail) while controlling for demographics and comorbidities. Lastly, we examined the association between PASC and severe symptom phenotypes with physical function deficits. RESULTS: Participants had a mean (SD) age at hospital admission of 56 (13) years and 44% were women. At 15-months and 3-years, the prevalence of PASC positivity was 25% and 25%, post-exertional malaise was 46% and 43%, brain fog was 22% and 28%, and fatigue was 27% and 41%, respectively. Greater frailty phenotype category and lower short physical performance battery score were significantly associated with PASC positivity and severe symptom phenotypes at 15-months, but not at 3-years. In adjusted and unadjusted analyses, no biomarkers were consistently associated with frailty categories, PASC positivity, or severe symptom phenotypes. CONCLUSIONS: In the 3 years following hospital discharge, biomarkers of inflammation and endothelial dysfunction are not associated with long-COVID phenotypes that have greater physical frailty. Other pathobiology may underlie long-COVID phenotypes in COVID-19 ALI/ARDS survivors and warrants further investigation.