•  
  •  
 

Abstract

Objective: Meteorin-like protein (METRNL), soluble CD14 subtype (presepsin), and bone morphogenetic protein 9 (BMP9) show promising for sepsis diagnosis and prognosis assessment in children. Their diagnostic and prognostic utility is comprehensively assessed in this study. Methods: A total of 360 children, aged 3-14 years, with sepsis, infection, and healthy groups were retrospectively analyzed. Receiver Operating Characteristic curve evaluated the diagnostic performance of METRNL, presepsin, and BMP9 for sepsis. The 28-day prognostic information was collected for the sepsis group. Risk factors for poor prognosis in the sepsis group were analyzed using logistic regression. A nomogram prediction model for prognostic prediction in the sepsis group was developed. Results: In differentiating healthy from infection, the combination diagnostic of METRNL, presepsin, and BMP9 showed excellent accuracy with an area under the curve (AUC) of 0.969 (95% CI: 0.952-0.987, P < 0.001), sensitivity of 0.925, and specificity of 0.908. In differentiating healthy from sepsis, the combination diagnostic showed enhanced performance with an AUC of 0.999 (95% CI: 0.996-1.000, P < 0.001), sensitivity of 0.983, and specificity of 1.000. In differentiating infection from sepsis, the combination diagnostic showed an AUC of 0.971 (95% CI: 0.950-0.993, P < 0.001), sensitivity of 0.883, and specificity of 1.000. In the sepsis group, 91 patients achieved a favorable prognosis, while 29 had a poor prognosis. The logistic regression analysis identified that METRNL, presepsin, and BMP9 were risk factors for poor prognosis. The nomogram prediction model incorporating these biomarkers showed good prognostic performance with an AUC of 0.823 (95% CI: 0.725-0.922). Conclusions: The METRNL, presepsin, and BMP9 panel offers high clinical value for sepsis diagnosis and prognosis in children. The developed nomogram prediction model provides a useful instrument for early risk assessment and personalized management.

Download

Share

COinS