The relationships of metals exposure and disturbance of the vaginal microbiota with the risk of PROM: Results from a birth cohort study

The vaginal microbiota is proposed to be associated with reproductive health. Exposure to metals during pregnancy is a risk factor for premature rupture of membranes (PROM). PROM can lead to serious maternal complications, thus, identifying the cause and therapeutic targets for it is crucial. However, the role of vaginal microbiota in the association between metals exposure and the risk of PROM are not clear. Based on a prospective birth cohort study including 668 pregnant women, maternal blood levels of 15 metals in the first trimester (n=668) and microbiota of vaginal secretions in the third trimester (n=244) were assessed. The metals that significantly associated with the risk of PROM were screened out via four statistical models, the top three were barium (Ba), chromium (Cr) and thallium (Tl) according to their weight indices. The results from the BKMR model showed a positive association of the mixture (Ba, Cr and Tl) with the risk of PROM. PROM and non-PROM were characterised by different beta diversities, moreover, the relative abundances of Bifidobacterium, Corynebacterium and Collinsella were statistically and negatively related to the risk of PROM [the adjusted odds ratios (ORs) and 95 % confidence intervals (CIs) were 0.06 (0.00, 0.82), 0.32 (0.14, 0.74) and 0.50 (0.30, 0.84), respectively]. On the other hand, women with different levels of Ba exposure were also characterised by different beta diversities (p value = 0.047); and blood Ba levels were also negatively associated with the relative abundances of Collinsella; additionally, Cr levels were positively associated with alpha diversity indices [Shannon index: β (95 % CI) = 0.25 (0.01, 0.50); Simpson index: β (95 % CI) = 0.08 (0.00, 0.17), respectively]. The results from mediation analysis showed the proportion of the relationship between Ba exposure and PROM risk mediated by the relative abundance of Collinsella was 26.4 %. Further verification analysis exploring the potential cause of the above phenomenon indicated that the neutrophil count, one of blood inflammation indicators for PROM, was higher in women with the absence of Collinsella (p value = 0.039), moreover, the cumulative hazard of PROM for women with the presence of Collinsella was also significantly lower than that of those without Collinsella (p value = 0.007). Collectively, the changes in the diversity and composition of the bacterial community, especially the reduction in Collinsella abundance caused by metal exposure, may be related to the occurrence of PROM, which provides a new microbiota-based perspective for intervention in metal exposure-related PROM. Confirming these relationships and determining the possible processes at play will require more investigation.