|
|
Background: Differentiating Mycoplasma pneumoniae (MP) from Bordetella pertussis (BP) infections in children at primary care hospitals is challenging due to similar symptoms and overlapping epidemic seasons. This study evaluated the use of C-reactive protein (CRP) levels and lymphocyte counts from a single blood test to quickly distinguish between MP and BP infections.
Methods: The study included a derivation cohort (n = 301) and a validation cohort (n = 87). Logistic regression analyses identified predictors for differentiating MP and BP infections, leading to the development of a prognostic scoring model and nomogram. The model's effectiveness was assessed using the area under the receiver operating characteristic curve (AUC) and calibration analysis.
Results: CRP and lymphocyte count were key factors in distinguishing MP from BP infections (p < 0.001). A straightforward scoring model was developed utilizing these predictors. The score was significantly lower in the BP group compared to the MP group (-0.36 ± 0.17 vs. 2.09 ± 0.13; p < 0.001). The model exhibited excellent dis-criminative capability (AUC = 0.866), surpassing the performance of the neutrophil-to-lymphocyte ratio (NLR) (AUC = 0.704), CRP (AUC = 0.783), lymphocyte count (AUC = 0.838), and neutrophil count (AUC = 0.593). These findings were also validated in a separate cohort. Subsequently, a nomogram was constructed for clinical application.
Conclusions: The proposed predictive scoring model and nomogram, which incorporate CRP and lymphocyte count, offer an accessible and practical tool for differentiating between MP and BP infections in pediatric patients with respiratory tract infections in primary care outpatient settings.
DOI: 10.7754/Clin.Lab.2025.250836
|
