Meta-analysis of the diagnostic efficacy of heparin binding protein in adult sepsis

Abstract

To evaluate the accuracy of heparin binding protein (HBP) in the diagnosis of adult sepsis by meta-analysis. The retrospective, randomized controlled and prospective researches on the diagnosis of adult sepsis using HBP were reviewed by searching VIP network, China National Knowledge Infrastructure (CNKI), Wanfang database, PubMed, Embase and Web of Science from foundation to August 2018. The researches were selected according to inclusion and exclusion criteria, and QUADAS was used to evaluate the quality of eligible studies. Metadisc 14.0 was used to analyze the threshold effect, pool the estimation and do Meta regression analysis. Publication bias was assessed using Stata 12.0 with Deeks funnel plot. A total of 728 related literatures were searched and 14 studies were enrolled in this study with a total of 2 023 subjects, with 1 120 in sepsis group and 903 in non-sepsis control group (795 patients with non-sepsis and 108 healthy volunteers). The total score of literature quality was 14-23, indicating that the overall quality was good. Meta-analysis showed that Spearman correlation coefficient between sensitivity logorithm and 1-specificity logorithm of 14 groups of data was 0.106 (P = 0.719), indicating no threshold effect. The pooled sensitivity was 0.74 [95% confidence interval (95%CI) was 0.72-0.77], the pooled specificity was 0.73 (95%CI was 0.70-0.76), the diagnostic odds ratio (DOR) was 14.15 (95%CI was 7.77-25.78), and the area under summary receiver operating characteristic curve (AUC) was 0.865. Deeks funnel plot was asymmetrical on the left and right (P = 0.60), indicating that there was no publication bias of the eligible studies. HBP has good accuracy for the diagnosis of adult sepsis and it can be used as an auxiliary index for adult sepsis diagnosis.