RESUMO
Accumulating evidence supports that the viral-induced hyper-inflammatory immune response plays a central role in COVID-19 pathogenesis. It might be involved in the progression to acute respiratory distress syndrome (ARDS), multi-organ failure leading to death. In this study, we aimed to evaluate the prognostic value of the immune-inflammatory biomarkers in COVID-19, then determine optimal thresholds for assessing severe and fatal forms of this disease.153 patients with confirmed COVID-19 were included in this study, and classified into non-severe and severe groups. Plasmatic levels of interleukin 6 (IL6), C-reactive protein (CRP), soluble-IL2 receptor (IL2Rα), procalcitonin (PCT) and ferritin were measured using chemiluminescence assay. Complete blood count was performed by Convergys 3X® hematology analyzer. Our results demonstrated that the peripheral blood levels of IL6, PCT, CRP, ferritin, IL2Rα, white blood cell count (WBC), neutrophil count (NEU), neutrophil-to-lymphocyte ratio (NLR), derived neutrophil-to-lymphocyte ratio (d-NLR) were significantly higher in severe forms of COVID-19. The ROC curve analysis showed that IL6 was the most accurate inflammatory biomarker. The calculated cutoff of IL6 (42 pg/ml) could correctly classify > 90% of patients regarding their risk of severity (area under ROC curve (AUROC) = 0.972) and the threshold value of 83 pg/ml was highly predictive of the progression to death (AUROC = 0.94, OR = 184) after a median of 3 days. Besides, IL-6 was positively correlated with other inflammatory markers and the kinetic analysis highlighted its value for monitoring COVID-19 patients. PCT and NLR had also a high prognostic relevance to assess severe forms of COVID-19 with corresponding AUROC of 0.856, 0.831 respectively. Furthermore the cut-off values of PCT (0.16 ng/ml) and NLR (7.4) allowed to predict mortality with high accuracy (se = 96.3%, sp = 70.5%,OR = 61.2)' (se = 75%, sp = 84%, OR = 14.6).The levels of these parameters were not influenced by corticosteroid treatment, which make them potential prognostic markers when patients are already undergoing steroid therapy.
