Meta-analysis of the accuracy of CD64, HBP, and PCT in differential diagnosis of sepsis patients

Abstract

Introduction: The aim of this study was to systematically evaluate the accuracy of the cluster of differentiation 64 (CD64) molecule, heparin binding protein (HBP), and procalcitonin (PCT) in the differential diagnosis of sepsis patients.

Methodology: Literature in multiple Chinese and English databases were searched to screen for CD64, PCT, and HBP related studies that focused on the differential diagnosis of sepsis. The literature was reviewed to extract the true positive, false positive, true negative, and false negative data; and STATA 17.0 software was used to combine the sensitivity, specificity, and area under the sensitivity receiver operating characteristic curve (SROC) of CD64, PCT, and HBP.

Results: This study included 17 articles. The combined sensitivity of CD64, HBP, and PCT were 0.87 [0.73~0.94], 0.85 [0.71~0.93], and 0.86 [0.64~0.96], respectively. The combined specificity of CD64, HBP, and PCT were 0.87 [0.78~0.93], 0.80 [0.06~1.00], and 0.63 [0.23~0.91], respectively; and all showed significant heterogeneity. There was a significant change in diagnostic odds ratio (DOR) values after excluding individual studies. The DOR value was overestimated when the sample size was small. The specificity of research in China was relatively low.

Conclusions: CD64, HBP, and PCT are all useful in the diagnosis of sepsis; and further optimization of diagnostic thresholds is needed before clinical application to improve the quality of testing. A prospective study with larger sample size is needed to improve the reliability of meta-analysis results.