Infection is the principal cause of preterm births. Most (90%) women with preterm deliveries have no abnormal history. It is widely agreed that preterm delivery is often associated with bacterial vaginosis. One of the major difficulties at this time is that the diagnosis of bacterial vaginosis is based on heterogeneous criteria. The technique currently used is the Nugent score, but it lacks the characteristics necessary for widespread use in the general population. It must be performed on a fresh swab, and any delay in transporting it can cause drying that makes the test difficult to perform.

The investigators have developed a rapid diagnostic tool for bacterial vaginosis using molecular biology based on a point of care model and obtained a patent (European Patent Office N° 2087134). In comparison with the reference techniques, our tool's performance has been excellent, in terms of specificity, sensitivity, and positive and negative predictive values. In particular, our work showed that 57% of the flora samples rated as intermediate on the Nugent score were in reality true bacterial vaginosis. Molecular biology therefore identifies a homogeneous population of women with vaginal flora anomalies. The investigators recently showed that the carriage of Atopobium vaginae and/or Gardnerella vaginalis >105/mL shortens the time to delivery in a population at risk of preterm delivery (PHRC 2006). Vaginal flora anomalies are therefore an important target for preventing preterm delivery.

The principal objective is to use a cost-effectiveness study to assess the medical and economic impact of a new strategy for the screening and subsequent treatment of vaginal flora anomalies before the 20th week of gestation- molecular biology techniques (PCR for A. vaginae and G. vaginalis by point-of-care testing) will test self-collected vaginal samples in a population of pregnant women at low risk of preterm delivery. There are multiple secondary objectives, including assessment of the effect of the strategy on the following indicators- delivery before 26, 32, and 37 weeks. Methods-An open-label prospective randomized study will compare 2 groups with different management of their pregnancies. Pregnant women at low risk, regardless of any symptoms, will be randomized into 2 groups. The study will not include women at high risk of preterm delivery. Innovative strategy (group A)- these women will undergo routine screening for bacterial vaginosis by analysis of their self-collected vaginal samples with this innovative technique.

Inclusion Criteria:

• Women over 18 years with a pregnancy before 20 weeks are some parity and gravidity;
• Woman who understood the process and the objectives of the study and who agreed to sign an informed consent;
• Without a history of premature birth or late abortion (population at low risk of preterm birth);
• Having no major risk factors for prematurity: insulin-dependent diabetes, systemic lupus erythematosus, hypertension, uterine malformation, cone biopsy, multiple pregnancy;
• No pre-existing hypertension;
• Asymptomatic or symptomatic regarding the diagnosis of bacterial vaginosis.

Exclusion Criteria:

• Woman withdrawing her consent during the study