Journal 2026 Vol.27 No.1
Procalcitonin for Differentiating Gram-Negative From Gram-Positive Sepsis in Children: A Retrospective Diagnostic Accuracy Study
Bianca Ana Maria R. Agbulos-Calupitan, M.D., Mary Antonette C. Madrid, M.D.

Abstract
Background : Pediatric sepsis is a leading cause of morbidity and mortality, and early appropriate antibiotic treatment is crucial in improving outcomes. Procalcitonin is a biomarker that increases in bacterial infections. Some adult studies have reported higher levels in Gram-negative compared to Gram-positive sepsis, suggesting a potential role in pathogen differentiation; however, findings have been inconsistent.
Objectives: To determine the diagnostic accuracy of procalcitonin levels in differentiating Gram-negative from Gram-positive sepsis in children aged one month and above admitted in Philippine Children’s Medical Center (PCMC) from January 2022 to December 2024.
Methodology: This analytic, cross-sectional study examined patient, parental, socioeconomic, cultural, and health system factors influencing delayed consultation among parents of patients 0 – 18 years at Batangas Medical Center and Lipa Medix Medical Center. A total of 668 parents were enrolled. Descriptive statistics and frequency tables summarized the key characteristics. Test of proportions assessed differences between groups. Univariate logistic regression screened possible predictors, followed This is a retrospective diagnostic accuracy study on pediatric patients aged one month and above admitted for sepsis in PCMC from January 2022 to December 2024. Inclusion required a positive blood culture for bacterial growth obtained on the same calendar day as procalcitonin level. Patients with non-infectious conditions that could influence procalcitonin level and those with polymicrobial or fungal growth on blood culture were excluded. Descriptive statistics were used to summarize the demographics and laboratory results and the diagnostic accuracy of rocalcitonin in differentiating Gram-negative from Grampositive sepsis were evaluated using the receiver operating characteristic (ROC) curve analysis.
Results: A total of 300 children with sepsis were included, of whom 73% had Gram-negative sepsis and 27% had Gram-positive sepsis. Median procalcitonin levels were not significantly different between Gram-negative and Gram-positive sepsis (p=0.452), although values were higher in Gram-negative cases (7.4 ng/mL [IQR 1.0–29.1] vs 2.2 ng/mL [IQR 0.3–33.0]). ROC analysis identified an optimal cutoff of 0.86 ng/mL, with a sensitivity of 77.3% (95% CI: 71.2% to 82.6%), specificity of 42.5% (95% CI: 31.5%-54.1%) and area under the curve (AUC) of 0.6 indicating poor discriminatory ability. Specific pathogens with the highest observed procalcitonin levels were E. coli (29.6 ng/mL, IQR 10.7-83.8), S. pneumoniae (21.6 ng/mL, IQR 2.1-87.2), and Klebsiella spp.(13.3 ng/mL, IQR 3.4-40.2). Subgroup analysis excluding prior antibiotic exposure showed similar findings, with poor overall discriminatory ability.
Conclusion: Procalcitonin alone cannot reliably differentiate between Gram-negative and Gram-positive sepsis in children. Unlike findings in adult studies, procalcitonin levels in Gram-negative sepsis were not significantly different from those in Grampositive sepsis. These findings may suggest possible age-related differences in host response, pathogen distribution, and disease severity.

KEYWORDS: procalcitonin, sepsis, children, Gram-negative sepsis, Gram-positive sepsis
https://doi.org/10.56964/pidspj20262701005
View Full Article in PDF format