Could we predict the prognosis of the COVID-19 disease?

Coronavirus 2019 disease (COVID-19) lead to one of the pandemics of the last century. We aimed to predict poor prognosis among severe patients to lead early intervention.

The data of 534 hospitalized patients were assessed retrospectively. Risk factors and laboratory tests that might enable the prediction of prognosis defined as being transferred to the intensive care unit and/or exitus have been investigated.

At the admission, 398 of 534 patients (74.5%) were mild-moderate ill. It was determined that the male gender, advanced age, and comorbidity were risk factors for severity. To estimate the severity of the disease, receiver operating characteristic analysis revealed that the areas under the curve which were determined based on the optimal cut off values that were calculated for the variables of values of neutrophil to lymphocyte ratio (NLR > 3.69), C-reactive protein (CRP > 46 mg/L), troponin I ( > 5.3 ng/L), lactate dehydrogenase (LDH > 325 U/L), ferritin ( > 303 ug/L), d-dimer ( > 574 μg/L), neutrophil NE ( > 4.99 × 10⁹ /L), lymphocyte (LE < 1.04 × 10⁹ /L), SO₂ ( < %92) were 0.762, 0.757,0.742, 0.705, 0.698, 0.694,0.688, 0.678, and 0.66, respectively. To predict mortality, AUC of values for optimal cutoff troponin I ( > 7.4 ng/L), age ( > 62), SO₂ ( < %89), urea ( > 40 mg/dL), procalcitonin ( > 0.21 ug/L), CKMB ( > 2.6 ng/L) were 0.715, 0.685, 0.644, 0.632, 0.627, and 0.617, respectively.

The clinical progress could be severe if the baseline values of NLR, CRP, troponin I, LDH, are above, and LE is below the specified cut-off point. We found that the troponin I, elder age, and SO₂ values could predict mortality.