Procalcitonin in preterm rupture of membranes: a systematic review and meta-analysis

摘要

Purpose Early detection of infection is of supreme importance in obstetrics; however, during pregnancy it is not reliably predicted by standard laboratory tests. We aimed to determine if procalcitonin (PCT) is a reliable predictor of chorioamnionitis (CA) in women with premature rupture of membranes (PPROM). Methods An electronic search of Scopus, ISI, Medline, Embase, ClinicalTrials.gov and the Cochrane Library databases was performed using specified key words. We examined all English and French reports on PCT measurement after admission for PPROM and considered: human studies published between 1990 and 2019; observational studies; and randomized controlled trials. A protocol was determined previously, registered at PROSPERO as CRD42019145464. The eligibility was independently assessed by two researchers and literature search yielded 590 studies; after revision of the titles and abstracts, 46 articles were identified as potentially eligible; eight studies were included in the meta-analysis. Primary data synthesis was performed in Review Manager Version 5.3 and average sensitivity and specificity was calculated using Midas, Stata. Results From the eight studies included, 335 participants with PPROM were enrolled. Our meta-analysis disclosed that PCT has a poor sensitivity (0.50; 95 CI 0.28-0.73) and a modest specificity (0.72; 95 CI 0.51-0.87) in diagnosing CA. C-reactive protein (CRP) not only has better sensitivity (0.71; 95 CI 0.53-0.84), but also better specificity (0.75; 95 CI 0.55-0.88), compared with the other inflammatory parameters analyzed. Procalcitonin does not seems to be better than CRP in preterm rupture of membranes for chorioamnionitis diagnosis.