<![CDATA[Introduction: Bacterial vaginosis (BV) represents the most prevalent vaginal dysbiosis among reproductive-age women, characterized by depletion of hydrogen peroxide-producing Lactobacillus species and overgrowth of anaerobes including Gardnerella vaginalis and Atopobium vaginae. During gestation, BV has been mechanistically linked to premature rupture of membranes (PROM) through enzymatic degradation of chorioamniotic membranes, upregulation of matrix metalloproteinases (MMPs), and pro-inflammatory cytokine cascade activation. This systematic review aimed to synthesize evidence from randomized controlled trials (RCTs) and primary studies examining the association between BV infection and PROM risk. Methods: A comprehensive systematic review was conducted following PRISMA 2020 guidelines. Included studies comprised pregnant women with confirmed BV diagnosis (Nugent score ≥7, Amsel criteria, or molecular methods) reporting PROM as an outcome. Risk of bias was assessed using Cochrane RoB 2 for RCTs and Newcastle-Ottawa Scale for observational studies. Fifteen primary studies encompassing 42,318 pregnant women met full inclusion criteria. Results: BV-positive pregnant women demonstrated significantly elevated PROM risk compared to BV-negative controls (pooled OR 2.19, 95% CI 1.54–3.12). Preterm PROM (PPROM) risk was more pronounced (OR 2.73, 95% CI 1.84–4.05). The PREMEVA trial (n=3,344) showed no significant PROM reduction with early clindamycin in low-risk populations, whereas high-risk subgroups demonstrated benefit (OR 0.14–0.56). BV-associated organisms produce phospholipase A2, sialidase, and collagenase, directly degrading fetal membrane integrity. Secondary outcomes included chorioamnionitis (OR ~3.0), neonatal sepsis (OR 1.60), and low birth weight (OR 1.73). Discussion: Consistent evidence supports a biologically plausible BV-PROM association through enzymatic and inflammatory pathways. Treatment trial heterogeneity arises from differences in diagnostic criteria, gestational age at intervention, population risk profiles, and antibiotic regimens. Microbiome-based community state type IV (high diversity, Lactobacillus-depleted) confers highest PPROM risk. Conclusion: Bacterial vaginosis constitutes a significant, modifiable risk factor for PROM/PPROM. Universal early-pregnancy BV screening with targeted antibiotic treatment for high-risk populations is warranted. Microbiome-guided therapeutic strategies require further RCT investigation.]]>
