# Comparison of Pediatric Risk of Mortality-III, Phoenix Sepsis, and pediatric Sequential Organ Failure Assessment scores for predicting septic shock in Vietnamese children with sepsis

**Authors:** Khai Quang Tran, Ngan Tuong Thien Pham, Tri Duc Nguyen, Quan Minh Pham

PMC · DOI: 10.1016/j.bjid.2026.104612 · The Brazilian Journal of Infectious Diseases · 2026-01-20

## TL;DR

This study compares three scoring systems for predicting septic shock in Vietnamese children with sepsis, finding that the Phoenix Sepsis Score has the best sensitivity for early detection.

## Contribution

The study evaluates the performance of PRISM-III, Phoenix Sepsis Score, and pSOFA in a Vietnamese pediatric population for septic shock prediction.

## Key findings

- Phoenix Sepsis Score (PSS) showed the highest AUROC (0.867) for predicting septic shock.
- PSS had the highest sensitivity (95.7%) and negative predictive value (97.6%).
- PRISM-III had the highest specificity (90.5%) and positive predictive value (70.0%).

## Abstract

Early recognition of septic shock is crucial for improving outcomes in children with sepsis. This study aimed to compare the predictive performance of the Pediatric Risk of Mortality-III (PRISM-III), Phoenix Sepsis Score (PSS), and pediatric Sequential Organ Failure Assessment (pSOFA) scores for septic shock in Vietnamese children.

A cross-sectional study was conducted on 86 children aged 2-months to 15-years with sepsis (including 23 with septic shock) admitted to a pediatric intensive care unit. Septic shock classification was performed independently and single ‒ blinded to score calculations to minimize assessment bias. The PSS and pSOFA were calculated using the worst parameters within the first 6-hours, and PRISM-III within the first 24 hours of admission. Discriminatory ability was assessed by the Area Under the Receiver Operating Characteristic Curve (AUROC). Multivariable logistic regression and calibration analyses were performed. Calibration results should be interpreted cautiously due to the small sample size.

The PSS showed the highest AUROC (0.867, 95 % CI: 0.777–0.931), followed by PRISM-III (0.826, 95 % CI: 0.729–0.899) and pSOFA (0.791, 95 % CI: 0.690–0.871); pairwise comparisons were not statistically significant. The PSS demonstrated the highest sensitivity (95.7 %) and negative predictive value (97.6 %), while PRISM-III had the highest specificity (90.5 %) and positive predictive value (70.0 %). In multivariable analysis, both PSS (Odds Ratio, OR = 2.78) and PRISM-III (OR = 1.23) were independent predictors of septic shock.

The PSS and PRISM-III provide complementary value. A two-step approach using the sensitive PSS for initial screening and the specific PRISM III for confirmation may enhance early septic shock recognition in resource-limited settings.

## Full-text entities

- **Genes:** PSS (Potocki-Shaffer syndrome) [NCBI Gene 780904]
- **Diseases:** infection (MESH:D007239), leukocytosis (MESH:D007964), Obesity (MESH:D009765), tachycardia (MESH:D013610), infectious (MESH:D003141), Organ Failure (MESH:D009102), respiratory, cardiovascular, coagulation, and neurologic dysfunction (MESH:D001778), Sepsis (MESH:D018805), cancer (MESH:D009369), critically ill (MESH:D016638), inflammatory response (MESH:D018746), Tachypnea (MESH:D059246), hypotension (MESH:D007022), acute malnutrition (MESH:D000067011), septic (MESH:D001170), thalassemia (MESH:D013789), Respiratory infections (MESH:D012141), -III (MESH:C537189), PRISM-III (MESH:D003643), Cardiovascular dysfunction (MESH:D002318), Malnutrition (MESH:D044342), acquired immunodeficiency (MESH:D000163), Overweight (MESH:D050177), Fever (MESH:D005334), congenital heart disease (MESH:D006330), leukopenia (MESH:D007970), nephrotic syndrome (MESH:D009404), hypothermia (MESH:D007035), Septic shock (MESH:D012772), immunodeficiencies (MESH:D007153), shock (MESH:D012769)
- **Chemicals:** phoenix (-), lactate (MESH:D019344)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12857169/full.md

## Figures

4 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12857169/full.md

## References

30 references — full list in the complete paper: https://tomesphere.com/paper/PMC12857169/full.md

---
Source: https://tomesphere.com/paper/PMC12857169