Contribution of Erythrocyte Sedimentation Rate to Predict Disease Severity and Outcome in COVID-19 Patients.

Abstract

Aim: The use of erythrocyte sedimentation rate (ESR) in coronavirus disease 2019 (COVID-19) to determine disease severity and prognosis is limited. This study aimed to interrogate the diagnostic and prognostic role of ESR compared to other acute-phase reactants.

Method: This retrospective cross-sectional study included 493 confirmed and hospitalized adult COVID-19 patients. Pneumonia, radiological severity, oxygen, intensive care requirements, mortality, ESR, and other acute-phase reactant values were recorded. Logistic regression and ROC analysis identified the effect of ESR on mortality and the sensitivity and specificity of the optimal cutoff values of ESR for the prediction of pneumonia, intensive care needs, and mortality and compared these with values for CRP.

Results: Of patients, 346 (70.2%) had pneumonia, 98 (19.9%) required intensive care, 183 (37.1%) required oxygen support, and 62 (12.6%) died. ESR data were obtained for 278 patients. Among patients, 80.2% had ESR above 20 mm/h, with a median value of 53 (interquartile range: 49). ESR was higher among those with pneumonia (p < 0.001), requiring oxygen (p < 0.001), and requiring intensive care (p=0.003) compared to those without these, and in exitus patients (p=0.043) compared to survivors. Logistic regression analysis identified that ESR did not impact mortality. ROC analysis found the AUC, cutoff, sensitivity, and specificity results of ESR for pneumonia were 0.827, 37 mm/h, 77%, and 78%; for intensive care were 0.625, 50 mm/h, 74%, and 52; and for mortality were 0.606, 51 mm/h, 71%, and 49%, respectively. However, ROC analysis values for CRP were superior to ESR for all these categories.

Conclusion: ESR increased in COVID-19 patients in the presence of pneumonia and severe disease; however, it was not prognostic. Sensitivity and specificity values for pneumonia, intensive care requirements, and mortality were lower than those for CRP.