EVALUATING THE PREDICTIVE VALIDITY OF PLATELET COUNT, PLATELET-TO-LYMPHOCYTE RATIO, AND HEMATOCRIT FOR DENGUE SHOCK SYNDROME IN PEDIATRIC PATIENTS

Tóm tắt

Background: Dengue infection remains a major public health burden in tropical and subtropical regions, contributing significantly to pediatric hospitalizations, morbidity, and healthcare resource utilization. Early identification of patients at risk of developing dengue shock syndrome (DSS) is crucial to reduce complications and mortality. However, in resource-limited settings such as Vietnam, access to advanced diagnostic tools is often restricted, underscoring the need for simple, cost-effective, and widely available biomarkers. Despite growing interest in hematological indices, evidence on their predictive validity for DSS remains scarce and inconclusive, particularly in pediatric settings. Objectives: This study aimed to evaluate the prognostic value of platelet count, platelet-to-lymphocyte ratio (PLR), and hematocrit in predicting DSS in pediatric patients with dengue hemorrhagic fever (DHF), focusing on both predictive accuracy and model fit. Materials and methods: This analytical study included pediatric patients diagnosed with DHF at Can Tho Children’s Hospital between December 2022 and February 2025. Eligible participants were followed from hospital admission until either DSS development or discharge. Demographic, clinical, and laboratory data, including complete blood count parameters, were collected. The discriminatory ability of platelet count, PLR, and hematocrit for predicting DSS was assessed using receiver operating characteristic (ROC) curve analysis. Calibration was evaluated using the Hosmer-Lemeshow test. Results: A total of 318 children were included, with 22% progressing to DSS. Platelet count, PLR, and hematocrit significantly differed between the DSS and non-DSS groups (p<0.001). Platelet count demonstrated the highest predictive performance (AUC=0.847, 95% CI: 0.8–0.894), followed by hematocrit (AUC=0.772, 95% CI: 0.7–0.844) and PLR (AUC=0.727, 95% CI: 0.657–0.797). Calibration analysis showed good agreement between predicted and observed DSS cases (platelet: p=0.604; PLR: p=0.499; hematocrit: p=0.764). Conclusions: Among the hematological parameters studied, platelet count exhibited the strongest predictive value for DSS. Given its accessibility in resource-limited settings, platelet count can serve as a practical marker for early DSS risk stratification in pediatric patients.