Effect of serum inflammatory factors in predicting co-infection with influenza viruses and Omicron

To identify the key differences in laboratory indicators between mono-infection and co-infection by influenza viruses and Omicron to facilitate timely adjustments in patient treatment strategies.

Prealbumin and C-reactive protein (CRP) levels were analyzed in 161 COVID-19 cases infected by SARS-CoV-2 (wild type), 299 cases infected by Omicron, 95 cases infected by influenza virus A/B (Flu A/B) and 133 co-infection cases infected with Flu A/B and Omicron. The receiver operating characteristic (ROC) curve and logistic regression equation were used to analyze the clinical predictive capacity of prealbumin and CRP in coinfected patients.

The co-infected and wild-type infected patients had significantly different CRP and prealbumin levels compared to mono-infected patients with Omicron or Flu A/B (p < .001). The ROC curve results indicated that prealbumin was more efficient than CRP in identifying co-infection from Omicron (AUC: 0.867 vs. 0.724) or Flu A/B (AUC: 0.797 vs. 0.730), and joint prediction significantly improved the diagnostic ability to discriminate co-infection from mono-infection (AUC: 0.934 and 0.887).

The findings suggest that prealbumin is a valuable indicator that can warn of co-infection and guide timely treatment decisions. Joint prediction may offer an even more effective diagnostic tool for discriminating co-infection from mono-infection.