Related Experiment Video
Updated: Mar 12, 2026

Evaluation of a Reliable Biomarker in a Cecal Ligation and Puncture-Induced Mouse Model of Sepsis
Published on: December 9, 2022
Comparing procalcitonin and C-reactive protein levels in very low birth weight infants with sepsis
Tess Stopczynski1, Samantha Eschborn2, Dorianna C Dobson2
1Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA.
Background:
While C-reactive protein (CRP) is a commonly used sepsis biomarker in neonates, studies in older patients propose procalcitonin (PCT) as potentially superior due to earlier rise in response to bacterial infection and shorter half-life.
Methods:
We conducted a prospective cohort study of VLBW infants undergoing sepsis evaluations at a level IV neonatal intensive care unit (NICU) and combined the analysis with a retrospective data review from a different level III NICU. We adjusted all analyses for study site to account for potential heterogeneity. Two blood samples for each biomarker obtained within a 24-h period were analyzed. Diagnostic accuracy was assessed by sensitivity, specificity, predictive values, likelihood ratios, and the area under the receiver operating characteristic (ROC) curve (AUC).
Results:
At literature-defined cut-offs, PCT demonstrated higher sensitivity, but lower specificity for sepsis compared to CRP. Using ROC-optimized cut-offs, CRP showed increased specificity (85%) but lower sensitivity (48%), whereas PCT maintained relatively high sensitivity (73%) but even lower specificity (60%). In multivariable modeling, only CRP was significantly associated with sepsis. Correlation between the biomarkers was weak.
Conclusions:
PCT did not outperform CRP as a sepsis biomarker. CRP was more specific and remained the stronger independent predictor of sepsis.
Impact:
Procalcitonin (PCT) shows higher sensitivity, but lower specificity compared to C-reactive protein (CRP) in very low birth weight infants with sepsis. CRP emerges as the stronger independent predictor of sepsis in multivariable models. This study constitutes one of the largest United States cohorts directly comparing CRP and PCT in this high-risk group.

