The clinical value of cord-blood procalcitonin for the diagnosis of neonatal intrauterine bacterial infection: a meta-analysis

Objective To study the clinical value of cord-blood procalcitonin (PCT) for the diagnosis of neonatal intrauterine bacterial infection. Method Embase, Cochrane library, PubMed, Wanfang Database, CNKI Database and Chinese Biomedical Literature Database were searched from the establishment date of the databases to the end of 2018. Articles involving PCT and intrauterine infection were enrolled. The research background, design information and diagnostic parameters were extracted and analysed using Meta-Disc 1.4 software. Heterogeneity of the data were studied and corresponding effect models were selected according to the heterogeneity results. Data from all studies were weighted and quantitatively combined. Summary receiver operating curve (SROC) were drawn and the area under the curve (AUC) were calculated. Finally, sensitivity analysis were carried out. Result A total of 318 articles, including 242 in foreign languages and 66 in Chinese, were retrieved. According to the inclusion criteria, 9 articles were finally obtained. A total of 3 571 newborns from five countries (France, Poland, Australia, Czech and China) were included. In the 9 articles, cord-blood PCT was used to diagnose intrauterine infection in six articles. The pooled sensitivity, specificity and SROC AUC were 73.7%, 82.6% and 0.839. For the rest three articles in which cord-blood PCT was used to diagnose early-onset sepsis, the pooled sensitivity, specificity and SROC AUC were 92.3%, 97.2% and 0.987, respectively. The sensitivity analysis of cord-blood PCT detection showed that disease severities and the differences of PCT detection threshold caused the heterogeneity. Conclusion Cord-blood PCT has diagnostic value for intrauterine infection of neonates and for early-onset sepsis of neonates. Key words: Meta-analysis; Infection; Cord blood; Procalcitonin; Infant, newborn