Association of inflammatory markers, neutrophil-lymphocyte ratio, and D-dimer with mortality in COVID-19 infection: A hospital-based retrospective analysis

Abstract

Introduction: To assess the association of blood biomarkers such as C-reactive protein (CRP), serum ferritin, lactate dehydrogenase (LDH), neutrophil-lymphocyte ratio, and D-dimer at admission with in-hospital mortality in COVID-19 and to determine best predictive cut-offs for them. Materials and Methods: This cross-sectional study included 984 confirmed cases of COVID-19 admitted in months of July and August 2020. The optimal biomarker cut-off points for mortality were defined by a receiver operating characteristic curve. Survival probabilities were estimated by the Kaplan–Meier method and compared with the log-rank test. Results: The overall mortality rate among the hospitalized cases was 254 (25.81%). All the markers were found to be significantly higher (P < 0.001) in nonsurvivor group as compared to the survivors at the time of admission. Serum CRP, ferritin, D-dimer and LDH were found to be elevated, i.e., higher than the upper limit of normal range in 426 (83%), 469 (68.37%), 449 (67.9%), and 380 (93.1%) respectively overall. However, these markers were significantly more elevated in nonsurvivor compared to survivors. A significant increasing trend of elevated level of all biomarkers was observed with increase of severity level (P < 0.0001). It was found that CRP ≥82 mg/L had sensitivity of 63.58% and specificity of 68.38% for predicting the mortality. Similarly, serum ferritin ≥475.6 mg/ml had sensitivity of 68.09% and specificity of 65.26%, D-dimer ≥0.65 had sensitivity of 90.71% and specificity of 55.45%, and LDH ≥915 U/L had sensitivity of 69.34% and specificity of 73.8% to predict the mortality. Furthermore, neutrophil and lymphocyte count ratio (NLR) ≥8.86 had sensitivity of 65.61% and specificity of 79.7% to predict the mortality. Conclusion: Levels of the blood biomarker such as CRP, serum ferritin, LDH, NLR, and D-dimer at admission can predict mortality in COVID-19 infection.