Title : Hematologic and inflammatory profiles across different clinical presentations of pediatric brucellosis: A retrospective cohort study
Abstract:
Background: Pediatric brucellosis has heterogeneous clinical presentations, while laboratory features of relapse and chronic disease remain poorly characterized. We aimed to compare hematologic, inflammatory, serological, and microbiological findings across presentations.
Methods: This retrospective cohort included 97 children with acute (n=52), subacute (n=14), chronic (n=14), relapse (n=9), or asymptomatic brucellosis identified through household screening (n=8), diagnosed at a center in eastern Türkiye between April 2021 and January 2023. Hematologic indices, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), serological tests, and blood cultures were compared using non-parametric methods. Exploratory receiver operating characteristic (ROC) analysis evaluated total lymphocyte count (TLC) for distinguishing relapse from chronic disease.
Results: Significant differences were observed for TLC (p=0.016), ESR (p<0.001), CRP (p<0.001), neutrophil-to-lymphocyte ratio (NLR; p=0.030), and platelet-to-lymphocyte ratio (PLR; p=0.023). Acute disease showed the highest ESR and CRP levels, whereas chronic disease had higher NLR and PLR values. TLC was highest in relapse (3480/µL) compared with chronic disease (2250/µL). Brucella IgM concentrations (p=0.016) and blood culture positivity (p=0.018) differed, whereas SAT, Coombs agglutination, Rose Bengal, and Brucella IgG did not. Exploratory ROC analysis suggested a discriminatory signal of TLC for relapse versus chronic disease (AUC=0.857; 95% CI, 0.691–1.000; p=0.005).
Conclusion: Hematologic and inflammatory parameters show different patterns across pediatric brucellosis presentations. Higher TLC in relapse is hypothesis-generating and should not be considered a relapse marker. Prospective multicenter validation is needed.
Keywords: pediatric brucellosis; relapse; total lymphocyte count; inflammatory markers; NLR; PLR

