Key points are not available for this paper at this time.
BACKGROUND: Pancreatic Stone Protein (PSP) has been proposed as a sepsis biomarker, but its role in severity stratification in pediatric intensive care and its relation to infectious source have not been explored. METHODS: We conducted a retrospective analysis of 97 pediatric patients admitted to the PICU with new-onset sepsis (within 24h from diagnosis). Blood samples were collected within 24h to measure PSP levels and compared with microbiological culture results. RESULTS: Among 97 patients, PSP levels were significantly higher in those with positive blood cultures (n = 24; median 108 ng/mL) compared with negatives (n = 73; 82 ng/mL, p = 0.008). When combining blood molecular testing with cultures, PSP remained higher in positives (n = 31; 111 ng/mL) than negatives (n = 66; 85 ng/mL, p = 0.026). PSP levels also correlated with sepsis severity: Non-septic (n = 55, median 72 ng/mL), sepsis (n = 22, median 238 ng/mL), septic shock (n = 20, 375 ng/mL) (p = 0.001). Receiver operating characteristic (ROC) curve analysis showed that PSP had superior accuracy in predicting sepsis severity (AUC 0.75, 95% CI 0.64-0.87) compared with C-reactive protein (C-RP, AUC 0.54, 95% 0.37-0.64) and procalcitonin (PCT, AUC 0.60, 95% CI 0.35-0.67). CONCLUSIONS: PSP is a promising biomarker for sepsis detection and severity prediction in critically ill pediatric patients. Further studies are required to validate its integration into diagnostic protocols.
Bottari et al. (Mon,) studied this question.