The Predictive Role of Inflammatory Markers and Generating Cut-off Values in Determining the Mortality of COVID-19 Patients Admitted to Intensive Care Units of University Hospital, KDU

Abstract

Background: Covid-19 disease caused by SARS-CoV-2 had a rapid global spread leading to a pandemic with high mortality. Therefore, it is imperative to predict the mortality of the patients beforehand and generate cut-off values that support the prediction. Objectives: To study the predictive role of inflammatory markers and to generate optimal cutoff values of Covid-19 patients admitted to Intensive Care Units (ICUs) Method: A retrospective cohort study was conducted with 219 Covid-confirmed patients admitted to ICUs of the University Hospital KDU from June to December 2021. Patients without a confirmatory diagnosis for Covid-19 infection (by PCR or RAT), with missing records and patients with malignancy were excluded. Bed headnotes were traced from medical records to extract the peak values of inflammatory markers [C-Reactive Protein (CRP), Lactate Dehydrogenase (LDH), procalcitonin, serum ferritin, neutrophil-lymphocyte ratio (N/L ratio)] and patient outcomes (deceased or discharged). Correlation between inflammatory markers and patient outcomes was assessed using Point Biserial Correlation. Optimal cut-off values of the significantly correlated inflammatory markers were predicted by Receiver Operating Characteristics (ROC) curve and Area Under the ROC curve (AUC). p<0.05 was considered as statistically significant. Results: Among 219 patients, 119 patients were deceased, and 100 patients were discharged. CRP (r=0.432), LDH (r=0.194), procalcitonin (r=0.177) and N/L ratio (r=0.371) showed statistically significant positive correlations with poor patient outcomes while serum ferritin depicted a negative correlation. The ROC analysis showed that AUC for CRP was 0.762 (sensitivity=0.731 and specificity=0.75), which is higher than the AUCs of N/L ratio (0.755), LDH (0.697) and procalcitonin (0.685). The significant optimal cut-off values of CRP, LDH, procalcitonin, and N/L ratio predicting mortality were 158.43 mg/L, 749.50 U/L, 0.345 ng/mL and 14.465, respectively. Conclusions: Study suggests that CRP, LDH, procalcitonin and N/L ratio along with their cut-off values possess a potential predictive role in determining the mortality of Covid-19 patients. Less number of laboratory investigations for serum ferritin, missing values of laboratory parameters and conducting a retrospective study are the limitations of the study. Keywords: Covid-19, Cutoff values, Inflammatory markers, Patient outcomes 42