D-Dimer, ferritin, and lactate dehydrogenase (LDH) as predictors of mortality in hospitalized COVID-19 patients

Abstract

Introduction: Several laboratory parameters may be predictors of coronavirus disease 2019 (COVID-19) mortality. This study aimed to analyze the potential of D-dimer, ferritin, and lactate dehydrogenase (LDH) to predict mortality in severe COVID-19 patients.

Methodology: A retrospective cohort study, including 147 patients, was examined using secondary data from medical records of hospitalized COVID-19 patients. D-dimer, ferritin, and LDH levels were obtained from the patients’ blood analysis on first hospitalization. Patients were then categorized into a survival group (97 patients) and a non-survival group (50 patients) based on final outcome. Proportions and means were analyzed using Chi square and Mann-Whitney tests. Further, the correlation and accuracy were analyzed using partial correlations test and receiver operating characteristic curve analysis. The combination of multiple predictors was also analyzed.

Results: The non-survival group had significantly higher levels of D-dimer (32.11 ± 13.05 vs. 9.57 ± 16.65; p < 0.001), ferritin (1719.84 ± 539.52 vs. 808.83 ± 664.81; p < 0.001), and LDH (1782.92 ± 1537.92 vs. 622.848 ± 274.79; p < 0.001) than the survival group. These parameters also had a moderate correlation with mortality (r > 0.500) and robust sensitivity and specificity for predicting mortality, especially ferritin (AUC = 0.906; sensitivity = 92.3%; specificity = 87.5%; p < 0.001), and the combination of ferritin and LDH with or without D-dimer (AUC = 0.959; sensitivity = 100%; specificity = 87.5%; p < 0.001).

Conclusions: The levels of these parameters are significantly higher, have robust sensitivity and specificity, and can be used as predictors of mortality.