BACKGROUND: Sepsis in children is a significant cause of morbidity and mortality in the pediatric intensive care unit (PICU). Assessment of pediatric sepsis using serial Pediatric Logistic Organ Dysfunction (PELOD)-2 score can be used as a prognostic factor. The use of biomarkers of sepsis is also used for diagnosis and predicting outcomes. Many studies have suggested that C-reactive protein (CRP) and procalcitonin (PCT) can be used to predict mortality.METHODS: A prospective cohort study was conducted to evaluate CRP, PCT, PELOD-2 score and its combination as a predictive factor of mortality in sepsis. All patients admitted to PICU Haji Adam Malik General Hospital, Medan, from April to November 2019 with suspected sepsis were included in this study. Blood examination and PELOD-2 scores were examined in the first 24 hours.RESULTS: A total of 79 children were included with a mortality rate 55.7%. The CRP, PCT, and PELOD-2 score were higher in nonsurvivor (2.8 (0.5-22.4) mg/dL; 9.36 (0.13-79.8) ng/mL; 9 (3-21), respectively). In multivariate logistic regression analysis, neither CRP nor PCT values could be independent predictors of mortality. The PELOD-2 score can be an independent predictor for mortality at a cut-off score of 7 (OR: 3.47 (95% CI: 1.68-7.19)). The combination of PELOD-2 and CRP scores as predictors of mortality showed lower values than PELOD-2 and PCT scores (0.80 vs. 0.95). The combination of all parameters only adds 1% of the predicted mortality value.CONCLUSION: PELOD-2 score with PCT value are recommended to predict mortality children with sepsis.KEYWORDS: sepsis, mortality, C-reactive protein, procalcitonin, PELOD-2 score