P-472. Evaluating and Refining the Candida Score for Invasive Candidiasis in Critically Ill Non-Neutropenic Children

TL;DR

This study developed a new Pediatric Candida score to better diagnose invasive candidiasis in critically ill children, improving accuracy over the adult version.

Contribution

The study introduces and validates a Pediatric Candida score tailored for non-neutropenic critically ill children.

Findings

The Pediatric Candida score showed higher sensitivity (87.9%) and specificity (86.4%) compared to the original score.

Key risk factors included bowel disease, colonization sites, and thrombocytopenia.

The new score could improve early diagnosis and antifungal treatment timing in children.

Abstract

Delayed antifungal treatment of invasive candidiasis (IC) in critically ill children is associated with increased morbidity and mortality. While the Candida score is widely used in critically ill non-neutropenic adults, its applicability in pediatric patients remains unvalidated. This study aimed to assess the performance of the original Candida score in critically ill non-neutropenic pediatric patients and to develop and internally validate a Pediatric Candida score.Table 1.Demographic Characteristics and Underlying Diseases of Critically Ill Non-Neutropenic Patients Stratified by Invasive candidiasis StatusTable 2.Invasive Candidiasis-Associated Predictor Variables Identified Through Multivariate Analysis and Stepwise Regression

Demographic Characteristics and Underlying Diseases of Critically Ill Non-Neutropenic Patients Stratified by Invasive candidiasis Status

Invasive Candidiasis-Associated Predictor Variables Identified Through Multivariate Analysis and Stepwise Regression

A retrospective matched case-control study was conducted at Songklanagarind Hospital, Thailand (2009–2023). Sixty-six critically ill non-neutropenic children with proven IC were matched by age, sex, and time at risk with 66 controls. Risk factors were identified using univariate analysis and parsimonious logistic regression to formulate the Pediatric Candida score.Table 3.Pediatric Candida Score and Original Candida Score Cutoffs With Diagnostic ParametersFigure 1.Receiver operating characteristic curves of the predictive scores using K-fold cross-validation.

Pediatric Candida Score and Original Candida Score Cutoffs With Diagnostic Parameters

Receiver operating characteristic curves of the predictive scores using K-fold cross-validation.

From the parsimonious model, the Pediatric Candida score included the following risk factors: bowel disease (odds ratio [OR], 33.51; 95% confidence interval [CI]: 4.22–266.17, P< 0.001), number of Candida colonization sites (OR: 26.81; 95%CI: 6.17–116.56, P< 0.001), thrombocytopenia (OR: 23.43; 95%CI: 5.31–103.44, P< 0.001), duration of mechanical ventilation (OR: 1.05; 95%CI: 1.01–1.08, P = 0.003), length of ICU stay (OR: 0.96; 95%CI: 0.93–0.99, P = 0.026), peritoneal catheter use (OR: 7.85; 95%CI: 0.68–90.42, P = 0.073), and prior surgery (OR: 3.6; 95% CI: 0.88–14.69, P = 0.064). The formula:

Pediatric Candida Score = 3.29×(Colonization sites) + 3.15×(Thrombocytopenia) – 0.04×(ICU days) + 3.51×(Bowel disease) + 2.06×(Peritoneal catheter) + 0.04×(Ventilation days) + 1.28×(Surgery). Compared to the original Candida score (sensitivity 65.1%, specificity 69.6%), the Pediatric Candida score showed superior performance (sensitivity 87.9%, specificity 86.4%, accuracy 87.1%).

The Pediatric Candida score outperformed the original Candida score, demonstrating superior diagnostic accuracy for IC in critically ill, non-neutropenic children. This tailored tool supports early diagnosis and timely antifungal therapy, potentially improving outcomes in vulnerable populations. Future multicenter validation is necessary to confirm its clinical utility and optimize pediatric antifungal stewardship.

All Authors: No reported disclosures

Figures