C-reactive protein and serum amyloid A protein as complementary biomarkers in differentiating viral and bacterial community-acquired pneumonia in children.

Abstract

Objective: Distinguishing between viral and bacterial community-acquired pneumonia (CAP) in children is crucial for guiding targeted management and antibiotic use. This study aimed to evaluate the efficacy of serum amyloid A protein(SAA) and C-reactive protein (CRP) as biomarkers to differentiate viral from bacterial CAP in children.

Methods: A total of 441 hospitalized children with a confirmed diagnosis of CAP were initially selected. Of these, 206 met the inclusion and exclusion criteria, including 132 cases of viral pneumonia and 74 cases of bacterial pneumonia. Baseline data and clinical characteristics were collected. Respiratory pathogen detection and blood biomarker measurements, including SAA and CRP levels, were completed within 24 h of admission. Receiver operating characteristic (ROC) curves were plotted to evaluate the effectiveness of SAA, CRP, and their combination in differentiating between viral and bacterial CAP.

Results: Fever (body temperature ≥ 38 °C) was more frequently observed in the bacterial CAP group (95.9%) compared to the viral CAP group (75%, P = 0.000). Wheezing was more prevalent in the viral CAP group (40.2% vs. 24.3%, P = 0.022). CRP and SAA levels were significantly higher in the bacterial CAP group (CRP: 27.6 (6.5, 49.4) mg/L vs. 3 (0.7, 8.4) mg/L, P = 0.000; SAA: 190.1 (70, 297.4) mg/L vs. 13.5 (1.4, 48.2) mg/L, P = 0.000). The area under the ROC curve for CRP and SAA was 0.84 (0.78 ∼ 0.90) and 0.85 (0.79 ∼ 0.91), respectively. The cutoff points were 86.55 mg/L for SAA and 19.65 mg/L for CRP, with sensitivities of 86.9% and 94.6%, and specificities of 73.0% and 63.5%, respectively. Combining SAA and CRP detection with clinical symptoms increases specificity to 93.2% and 97.3% but reduces sensitivity to 31.3% and 22.7% in distinguishing viral from bacterial pneumonia. Multivariate regression analysis confirmed that CRP was an independent predictor of bacterial pneumonia (OR = 1.098, P < 0.001) and was strongly correlated with SAA (Pearson r = 0.816, P < 0.001).

Conclusion: CRP and SAA are effective biomarkers for distinguishing between viral and bacterial CAP in children, with CRP demonstrating independent predictive value. The combined detection of CRP and SAA, along with clinical symptoms (such as fever or wheezing), significantly enhances diagnostic specificity but requires a trade-off with reduced sensitivity. This finding provides important evidence for the early precise classification of pediatric CAP and the rational use of antibiotics, but its clinical value still needs to be validated through larger-scale multicenter studies.