Introduction:

Due to the clinical heterogeneity of COVID-19, biomarkers must be used to confirm the disease's diagnosis and determine its prognosis.

Objective:

To assess the performance of ferritin as a diagnostic biomarker in cases of suspected diagnosis of COVID-19 in inpatients and as a prognostic biomarker in those with a confirmed diagnosis.

Materials and methods:

Multicenter, prospective, observational study conducted in 711 adult patients hospitalized between April and November 2020 in three quaternary care hospitals in Bogotá D.C., Colombia, due to suspected COVID-19. Based on ferritin levels on hospital admission, ROC curves were created for three outcomes: diagnosis, admission to the intensive care unit (ICU), and death. The operating characteristics of this biomarker were calculated for each outcome, and a multivariate analysis was carried out using a linear regression model to evaluate the association between ferritin levels and each outcome.

Results:

COVID-19 diagnosis was confirmed in 592 patients, of whom 160 (27.02%) were admitted to the ICU and 107 (18.07%) died. The areas under the ROC curve (AUC) for diagnosis, ICU admission, and death were 0.67 (95%CI: 0.62-0.73), 0.58 (95%CI: 0.57-0.67), and 0.56 (95%CI: 0.50-0.63), respectively. In the bivariate analysis, ferritin levels were significantly associated with diagnosis (p=0.003) and admission to the ICU (p<0.001), but not with mortality (p=0.326). In the multivariate analysis, ferritin was only significantly associated with admission to the ICU (p=0.009).

Conclusions:

Ferritin showed poor and moderate performance as a prognostic and a diagnostic biomarker, respectively. Therefore, neither the diagnosis of patients with suspected COVID-19, nor the prognosis of those with a confirmed diagnosis can be determined based only on serum ferritin levels.

        · abstract in Spanish     · text in English     · English ( pdf )