Presented at the Neonatal Society 2008 Autumn Meeting.
Tan S1,4, Nilsson C5, Tang C2, Li YW2, Sullivan M3, Lee C5, Lim YW5, Raun YJ5, Hibberd M5, Edwards AD1,4
1 Division of Clinical Sciences and MRC Clinical Sciences Centre, Imperial College London
2 Centre for Molecular Microbiology and Infection, Imperial College London
3 Division of Surgery Reproduction and Oncology, Imperial College London
4 Division of Neonatology, Imperial College Comprehensive Biomedical Research Centre, DuCane Road, London, W12 0HS, UK
5 Singapore Bioinformatics Institute, Singapore
Background: Although infection is recognised as the most important cause of very early preterm birth, the roles of specific bacterial species are not well characterized. Previous research has predominantly studied only preterm infants, or utilised culture techniques. Molecular investigation of intrauterine bacteria in both term and preterm subjects offers potentially new information on infection and preterm birth.
Aim: To survey bacterial species present in fetal membranes from preterm and term delivery, using 16Sr Polymerise Chain Reaction (PCR) amplification of bacterial DNA.
Methods: Ethical approval was obtained from the Riverside Ethics Committee. Fetal membrane samples were collected from 801 subjects at birth, using sterile techniques. All recruited infants of gestation 32 weeks and under were studied: n=90. A term cohort was randomly selected for comparison: n= 147. PCR was performed on purified DNA extracted from membranes, using established primers and optimised conditions. PCR eluates showing detectable levels of bacteria on agarose gel electrophoresis were extensively cloned. Sequences obtained were searched using Blastn for homologies in the Ribosomal Database Project.
Results: Bacterial DNA was detected in 30/247 fetal membranes studied. Bacterial DNA was only detected in 2/147 (1.4%) of term membranes, both vaginal delivery (VD) with prolonged membrane rupture. One demonstrated multiple colonisation with faecal flora and the other showed dual colonisation with Ureaplasma and Mycoplasma. 28/90 (31%) of preterm membranes had detectable bacterial DNA. A single species was detected in 17: Ureaplasma (9), Fusobacteria (4), Streptococcus (2) Mycoplasma (1) and Sneathia (1). Two species were detected in 6: Fusobacteria and Mycoplasma (3 each), Ureaplasma (2), Sneathia, Bacteroides, Peptoniphilus and Granulicatella (1 each). 5 cases had polymicrobial colonisation with a wide range of bacteria, but only one Lactobacillus and no organism classified as Escherichia Coli were detected. The monocolonised group included 12 VD and 5 Caesarian Sections (CS). Dual-colonised subjects included 5 with VD and 1 with CS. Multiple-colonised subjects included 4 with VD and 1 with CS.
Conclusion: This is a comparable detection rate with previous 16SrRNA studies on preterm subjects (1,2) Preterm birth appears to be associated with a restricted range of bacteria in the fetal membranes, in particular with Ureaplasma, Fusobacteria and Mycoplasma. This suggests that specific mechanisms may be involved in preterm delivery and that experimental studies using E Coli endotoxin are of limited value.
References
1. DiGiulio et al PLoS ONE 2008; 26;3(8):e3056
2. Miralles R, et al. Pediatr Res. 2005; 57(4):570-7