Association of Inflammatory Markers and CT Scoring as Severity Predictors in COVID-19 Patients

Abstract

Background: COVID-19 disease caused by SARS CoV-2 has rapidly spread worldwide and became a global concern for human health. Computed Tomography (CT) scan facility is not readily available in remote areas in developing nations like India. Previous studies have revealed the utility of CRP, LDH and other biomarkers in predicting the severity of COVID-19 pneumonia in resource limited settings and thus can help in stratification and early critical care transfer. Objective: The current study aimed to determine correlation between inflammatory markers with CT severity score to predict extent of COVID-19 pneumonitis.  Materials And Methods: A total of 240 patients more than 18 years of age hospitalized with COVID-19 confirmed diagnosis either by RT-PCR test or by Rapid Antigen Test were assessed in this monocentric prospective observational study at our center, Pune, India. CT severity score and blood parameters such hemogram, hsCRP, LDH, Ferritin, IL-6 and D-dimer levels documented on the day of admission. The total CT score was the sum of five individual lobar scores and defined as: 0, none; 1–5, minimal; 6–10, mild; 11–15, moderate; and 16–20, severe lung involvement. Results- Mean age of patient in clinically critical cases (61.40 years) was higher than those with moderate (55 years) and non-severe clinical presentation cases (50.51years). 180 (75%) subjects were having non-severe clinical presentation followed by 31 (12.9%) subjects with severe and 29 (12.1%) subjects were having critical clinical presentation on admission. hsCRP & LDH were having more correlation coefficient with CT severity score (0.599 & 0.570 respectively). The ROC analysis revealed that area under curve for hsCRP and  LDH on the day of hospitalization for predicting severe lung involvement was 0.796 at cut-off of 8.65 mg/dl (Sensitivity 76.9%, Specificity 74.9%) and 0.919 at cut-off of 425 U/L (Sensitivity 84.6%, Specificity 84.1%). Conclusion: On the basis of strong significant association with CT severity; hsCRP and LDH levels can be used to predict extent of lung involvement in chest CT in COVID-19 disease, thereby it will help in early triage and critical care transfer of patients to ensure optimal resource allocation in peripheral areas with limited facilities.