Abstract
RATIONALE: Factors associated with long-term sequelae emerging after the acute phase of COVID-19 (so called "long COVID") are unclear. Here, we aimed to identify risk factors for the development of COVID-19 sequelae in a prospective cohort of subjects hospitalized for SARS-CoV-2 infection and followed up one year after discharge. METHODS: A total of 324 subjects underwent a comprehensive and multidisciplinary evaluation one year after hospital discharge for COVID-19. A subgroup of 247/324 who consented to donate a blood sample were tested for a panel of circulating cytokines. RESULTS: In 122 patients (37.8%) there was evidence of at least one persisting physical symptom. After correcting for comorbidities and COVID-19 severity, the risk of developing long COVID was lower in the 109 subjects admitted to the hospital in the third wave of the pandemic than in the 215 admitted during the first wave, (OR 0.69, 95%CI 0.51-0.93, p=0.01). Univariable analysis revealed female sex, diffusing capacity of the lungs for carbon monoxide (DLCO) value, body mass index, anxiety and depressive symptoms to be positively associated with COVID-19 sequelae at 1 year. Following logistic regression analysis, DLCO was the only independent predictor of residual symptoms (OR 0.98 CI 95% (0.96-0.99), p=0.01). In the subgroup of subjects with normal DLCO (> 80%), for whom residual lung damage was an unlikely explanation for long COVID, the presence of anxiety and depressive symptoms was significantly associated to persistent symptoms, together with increased levels of a set of pro-inflammatory cytokines: interferon-gamma, tumor necrosis factor-alpha, interleukin (IL)-2, IL-12, IL-1β, IL-17. In logistic regression analysis, depressive symptoms (p=0.02, OR 4.57 [1.21-17.21]) and IL-12 levels (p=0.03, OR 1.06 [1.00-1.11]) 1-year after hospital discharge were independently associated with persistence of symptoms. CONCLUSIONS: Long COVID appears mainly related to respiratory sequelae, prevalently observed during the first pandemic wave. Among patients with little or no residual lung damage, a cytokine pattern consistent with systemic inflammation is in place.