How can the Treatment of Bacterial Vaginosis be Improved to Reduce the Risk of Preterm Delivery?

“… bacterial vaginosis is associated with a risk of obstetric complications … however, the benefits of screening and treatment of bacterial vaginosis during pregnancy remain unclear.”

Preterm delivery, defined as birth before 37 weeks of gestation, has an incidence of approximately 13% in the USA and between 5 and 9% in other developed countries [1]. It is a major public health problem incurring high monetary costs. The survival of children born prematurely has been made possible by advances in modern obstetrics (e.g., tocolytic agents and regionalization of perinatal care for high-risk mothers and their infants) and the development of neonatal resuscitation (e.g., corticosteroids for fetal maturation and newborn intensive care unit). While these measures have reduced perinatal mortality, the surviving children remain at risk of developing a wide array of complications in the short and long term. Premature birth can be divided into spontaneous preterm birth occurring after spontaneous labor with intact membranes or following preterm premature rupture of the membranes, and induced preterm birth occurring after labor induction or cesarean delivery for maternal or fetal indications. While the majority of premature births occur spontaneously, the rate of induced preterm births has been steadily increasing since 1990 [1]. The known causes of spontaneous preterm labor are multiple and complex and are mainly infection and/or inflammation related; however, vascular disease and uterine overdistension have also been linked to its occurrence. In this context, the control of infection and/or inflammation during pregnancy represents a major goal in the prevention of prematurity.

Bacterial vaginosis is a bacterial imbalance of normal vaginal flora that was first described in 1955 by Gardner and Dukes [2] and is characterized by a malodorous vaginal discharge. However, many affected women are asymptomatic. In the past 20 years, several investigations have shown that bacterial vaginosis is associated with a risk of obstetric complications (e.g., premature rupture of membranes, prematurity, chorioamnionitis and low birth weight). According to the meta-analysis carried out by Leitich and colleagues, pregnant women with bacterial vaginosis have an over twofold increase in risk of preterm delivery than women without bacterial vaginosis (odds ratio: 2.16; 95% CI: 1.56–3.00) [3]. This risk was shown to be even greater among women in whom bacterial vaginosis was diagnosed early during pregnancy [4]. Among women under going preterm labor, bacterial vaginosis doubles the risk of delivering before 37 weeks of gestation (odds ratio: 2.38; 95% CI: 1.02–5.58) [3]. In addition, bacterial vaginosis increases the risk of late miscarriage and that of maternal infection postpartum [3]. While bacterial vaginosis has been found associated with preterm birth, no link has been found with neonatal infection or perinatal mortality.

Other factors such as host local immune responses and genetic determinism should be considered to explain why not all bacterial vaginosis-positive pregnant women give birth prematurely [5,6]. Nevertheless, the association of bacterial vaginosis with preterm delivery would reasonably lead to the hypothesis that vaginal bacteria ascend into the amniotic cavity and there incite an inflammatory response thus inducing preterm labor [1]. Accordingly, bacterial vaginosis seems to be a favored target to prevent preterm delivery. However, the benefits of screening and treatment of bacterial vaginosis during pregnancy remain unclear. According to a recent meta-analysis, there are no benefits to screening for bacterial vaginosis in the low-risk population and arguments in favor of screening and treatment of bacterial vaginosis in women at high risk of premature delivery remain insufficient [7,8]. The limitations of existing studies concern the significant heterogeneity among ethnic groups, the prematurity rate and the methods of screening bacterial vaginosis and evaluating treatment efficacy. These last ones are mainly due to the lack of reliable standardized and objective diagnostic tools.

Faced with this paradox, various efforts are being made to improve the management of bacterial vaginosis during pregnancy to prevent preterm delivery. The first step has been to define the abnormal vaginal flora and since 2005, the application of molecular biological techniques have expanded our knowledge of the vaginal microbiome [9]. Not one bacterial species but numerous bacteria have thus been implicated in bacterial vaginosis and the recent technical advances have facilitated the detection and the identification of bacteria without the need for cultivation. Some of these bacteria had not previously been described or well characterized. It is now clear that the resident Lactobacillus species are replaced by an overgrowth of vaginal anaerobes or Gramnegative bacteria including Gardnerella vaginalis, Atopobium vaginae, bacterial vaginosis-associated bacteria, Megasphaera species, Mycoplasma hominis, Mobiluncus species, Ureaplasma urealyticum, Prevotella and Peptostreptococcus species [9]. Moreover, bacterial vaginosis-associated bacteria have been shown to form a prolific polymicrobial biofilm, the main component of which was found to be A. vaginae and G. vaginalis that adhere to the vaginal epithelium [10]. Such results have led some authors to propose new diagnostic tools based on molecular biological techniques [11]. Of these, one of the most interesting is the quantitative real-time PCR assay for A. vaginae as high vaginal quantification of A. vaginae is highly predictive of bacterial vaginosis [12]. A reliable method to objectively analyze the abnormal vaginal flora is critical to the development of urgently needed diagnostic tools.

“… molecular biological techniques have expanded our knowledge of the vaginal microbiome … [and] the link between abnormal vaginal flora and obstetrical complications.”

The second step has been establishing and defining a relationship between molecular abnormalities of vaginal flora and obstetrical complications. Such a link was recently investigated among women with preterm labor, among whom the risk of preterm delivery was significantly associated with high vaginal concentrations of A. vaginae and G. vaginalis [13]. In addition, high vaginal concentrations of these two microorganisms together were associated with a significantly shorter interval between preterm labor and delivery than were lower concentrations. This suggests an implication of the vaginal load of both bacteria in the occurrence of preterm delivery. Moreover, the recent identification of A. vaginae in the amniotic fluid in preterm prelabor rupture of membranes may suggest the microbial invasion of the amniotic cavity from the vagina [14]. Finally, a case report of septic abortion, involving A. vaginae, following trophoblast biopsy by transcervical chorionic villus sampling provided evidence of the pathogenicity of this organism during pregnancy [15]. Confirmation of these data and the identification of the pathophysiological mechanisms responsible are now required to enhance our understanding of the link between abnormal vaginal flora and obstetrical complications.

The third step has been the testing of new therapeutic options. Currently, oral metronidazole or oral clindamycin are recommended regimens for pregnant women [101]. However the treatment is suboptimal with frequent relapse after antibiotic treatment. The recurrence rate within 12 months of oral metronidazole can be as high as 58% [16]. High-level resistance to metronidazole of some strains of A. vaginae, survival of bacterial vaginosis-associated bacteria as a biofilm on the vaginal epithelium after therapy and possible reinfection are all possible reasons for treatment failure [17]. Nonantibiotic therapy, including probiotics and prebiotics, should be more thoroughly investigated. Probiotics are live microorganisms that can be found in Lactobacillus preparations and yogurts which, when administered in adequate amounts, confer a health benefit on the host. A Cochrane review supported the use of probiotics to treat vaginal infections among pregnant women, but was not able to assess any effect on preterm birth [18]. Interestingly, a recent study from Norway that evaluated 950 women with spontaneous preterm birth and 17,938 women with at-term delivery, showed that intake of probiotics may be associated with reduced risk of preterm birth [19]. Further investigations are needed to explain such observed benefits of probiotics administered during pregnancy. For prebiotics that stimulate the growth of lactobacilli to induce a vaginal flora shift from abnormal to normal flora, their efficacy in treating bacterial vaginosis and their benefits during pregnancy remain to be demonstrated. Another therapeutic option to consider is preconception screening. All studies to date have been conducted during early pregnancy; however, at this stage, inflammatory mechanisms and infections that are potentially involved in preterm delivery would already be installed. A healthy vaginal microbiome preconception may be of paramount interest to the primary and secondary prevention of bacterial vaginosis and preterm birth.