Evaluation of C-Reactive Protein and High Resolution Computed Tomography Scan Score to Initiate Early Steroid Therapy in Patients with COVID-19 Infection

Abstract

Background: The severity of COVID pneumonia depends on the degree of lung parenchyma inflammation. This study aimed to identify prognostic factors for COVID-19 and examines the relationship between CRP levels and computed tomography Severity Scores (CTSS).

Methods: An observational study was conducted at Dr. Ziauddin Hospital, Clifton campus, Karachi, Pakistan, from June to November 2021. Purposive sampling enrolled 209 patients, aged 18 years and older, PCR-confirmed cases, requiring both hospitalization and HRCT. Age, sex, admission CRP, symptom duration, disease severity at presentation, CTSS, length of hospital stays, and outcomes (death or discharge) were captured. STATA-12 was used for analysis. Chi-square assessed categorical variables' association with outcomes. Pearson/Spearman-rank test evaluated the correlation between CRP and CTSS. Binary logistic regression identified significant prognostic factors, and linear regression determined the association coefficient between CTSS and CRP (3.00; p=0.00). The coefficient established CRP cutoff for Moderate disease starting >= 27mg/L to initiate steroid therapy.

Results: Disease severity on presentation (p < 0.001), HRCT score (p < 0.001), hospital stay length (p = 0.002), prior illness duration (p = 0.011), and elevated CRP levels (p = 0.012) were associated with outcomes. A Spearman’s rho of 0.336 indicated a positive correlation between HRCT and CRP levels. Logistic regression showed CRP (OR=1.0051, p=0.003) and CT-severity-score (OR=2.1155, p=0.002) as significant. Linear regression yielded a coefficient of 3.0094 (p < 0.001) for CTSS and CRP levels. The transition from mild to moderate disease occurred at a CRP level of establishing ≥27 mg/L

Conclusion: A CRP cutoff of 27 mg/L can guide early steroid therapy to prevent disease progression and monitor therapeutic response.