Correlation of Patient Profiles and Biomarkers with Outcomes in Covid-19 Icu Patients: A Retrospective Analysis

Background: COVID-19 is a novel disease with a highly variable and unpredictable clinical course. Various clinicodemographic factors and numerous biomarkers have been identified in studies from the West and marked as possible predictors of severe illness and mortality which may be used to triage patients for early aggressive care. This triaging becomes even more significant in resource-limited critical care settings of the Indian subcontinent.

Methods: This retrospective observational study recruited 99 cases of COVID-19 admitted to intensive care from 1 May to 1 August 2020. Demographic, clinical and baseline laboratory data were collected and analysed for association with clinical outcomes, including survival and need for mechanical ventilatory support.

Results: Male gender (p=0.044) and diabetes mellitus (p=0.042) were associated with increased mortality. Binomial logistic regression analysis revealed Interleukin-6 (IL6) (p=0.024), D-dimer (p=0.025) and CRP (p<0.001) as significant predictors of need of ventilatory support and IL6 (p=0.036), CRP (p=0.041), D-dimer (p=0.006) and PaO2FiO2 ratio (p=0.019) as significant predictors of mortality. CRP >40 mg/L predicted mortality with sensitivity of 93.3% and specificity of 88.9% (AUC 0.933) and IL6> 32.5 pg/ml with a sensitivity of 82.2% and specificity of 70.4% (AUC 0.821).

Conclusion: Our results suggest that a baseline CRP >40 mg/L, IL6 >32.5 pg/ml or D-dimer >810 ng/ml are early accurate predictors of severe illness and adverse outcomes and may be used to triage patients for early intensive care.