The Value of HALP Score, SII, and SIRI in Predicting the Need for Intensive Care and Assessing Disease Severity in Pediatric Diabetic Ketoacidosis Cases.

Objective: The aim of this study was to evaluate whether Hemoglobin, Albumin, Lymphocyte, and Platelet (HALP) scores, the Systemic Immune-Inflammation Index (SII), and the Systemic Inflammatory Response Index (SIRI) can predict intensive care unit (ICU) or inpatient admissions in pediatric diabetic ketoacidosis (DKA) cases and to determine their sensitivity and specificity. Methods: This retrospective study included 39 pediatric patients (<18 years) diagnosed with DKA (pH < 7.3, HCO3 < 15). HALP, SII, SIRI, and urine ketone values were collected from medical records. Statistical analyses included ROC curve analysis, correlation tests, and appropriate parametric or non-parametric comparisons to evaluate associations with 30-day outcomes. Results: The median age was 13 years (IQR: 8-15), 56.4% were male, and 64.1% required ICU monitoring. ICU patients had significantly lower pH (p = 0.005) and HCO3 (p = 0.012) and significantly higher monocyte, SII, and SIRI values (all p ≤ 0.018). ROC analysis showed SIRI had the highest predictive power for ICU admission (cut-off: 3911; sensitivity: 76%; specificity: 85.7%), followed by SII. HALP scores did not demonstrate any value in assessingdisease severity for predicting ICU admission (AUC = 0.25). Conclusion: SIRI and SII are reliable predictors of ICU admission in pediatric DKA. HALP scores do not predict ICU admission and should not be considered a marker of disease severity. Incorporating SIRI and SII into clinical decision-making may improve early risk stratification. Prospective multicenter studies are warranted to validate these results.