Contribution to optimising the use of two promising preterm labour screening markers

Prematurity is a serious medical and socio-economic burden for the community. Every effort to improve its consequences is worthwhile. Preterm delivery is considered as a syndrome that can use different pathways. These allow different approaches for risk screening. In this work, we have studied the clinical value of 2 promising markers of preterm delivery. The first marker studied is based on the early separation of the chorion from the decidua. This is observed when preterm contractions occur, with release of a glycoprotein, called foetal fibronectin (fFn), in the cervico–vaginal secretions. The place of the test based on its dosage in our clinical arsenal remains unclear. We demonstrate here that, to be interpretable, the test should never be done on cervical samples, and never within the 24 h following coitus. Furthermore, considering likelihood ratios as the best objective parameters to evaluate the performance of a test, our two metaanalyses clearly showed that this test was disappointing as a screening method as well as a diagnostic tool whatever the clinical settings considered. The second marker studied concerns Ureaplasma spp., which is involved as pathogenic agent leading to preterm delivery. Thanks to two prospective studies, we have been able to demonstrate that carriage of U. spp. is constant during pregnancy and that it is present at a significantly higher level (both qualitatively and quantitatively) in vaginal secretions than in those from the cervix. We have also pointed out that bacterial vaginosis is more often associated with U. urealyticum than with U. parvum, indicating a greater potential deleterious effect from the first. This thesis provides new important information for the understanding of parts of the preterm delivery process and should be a step toward other clinical studies.