The Prognostic Value of the Lactate-to-Albumin Ratio in Predicting Mortality in the Pediatric Intensive Care Unit

Abstract Purpose To evaluate the prognostic value of admission lactate-to-albumin ratio (LAR) for mortality in a large pediatric intensive care unit (PICU) cohort, compare it with lactate, albumin, Pediatric risk of mortality-3 (PRISM-3), and Pediatric logistic organ dysfunction-2 (PELOD-2) scores, and assess its performance across diagnostic subgroups and association with intubation. Methods This retrospective cohort included 7,317 patients aged 1 month–18 years admitted to a tertiary PICU between September 2019 and April 2025 with admission lactate and albumin measurements. ROC analysis, logistic regression, DeLong tests, and analyses of 10 diagnostic subgroups with Bonferroni correction were performed. Results Mortality was 10.39% (760/7,317). LAR was higher in non-survivors than survivors (median 0.90 vs. 0.47; p < 0.001) and predicted mortality with an AUC of 0.743 (cutoff 0.70; sensitivity 63.3%; specificity 74.9%). It outperformed lactate (AUC 0.691) but not albumin (AUC 0.748). Each 1-unit increase in LAR increased mortality odds 2.86-fold (95% CI 2.60–3.14); this association remained after adjustment for age, sex, and PRISM-3 (adjusted OR 2.07) or PELOD-2 (adjusted OR 2.12). LAR was associated with mortality in all subgroups (OR 1.96–5.67), remaining significant after Bonferroni correction in nine, and was modestly higher in intubated survivors (p < 0.001). PRISM-3 and PELOD-2 outperformed LAR (AUC 0.897 and 0.889; both p < 0.001), although adding LAR increased their AUCs to 0.910 and 0.906. Conclusion Admission LAR is a rapid, independent predictor of PICU mortality that outperforms lactate and remains informative across diagnoses. Although not a substitute for PRISM-3 or PELOD-2, it may complement these scores in early risk stratification.