1. ¹Perinatology Research Branch, NICHD/NIH/DHHS, Bethesda, MD, USA
2. ²Perinatology Research Branch, NICHD/NIH/DHHS, Detroit, MI, USA
3. ³Center for Molecular Medicine and Genetics, Wayne State University, Detroit, MI, USA
4. ⁴Department of Obstetrics and Gynecology, Wayne State University School of Medicine, Detroit, MI, USA
5. ⁵UO Ostetricia e Ginecologia, Azienda Ospedale/Universita, Padova, Italy
6. ⁶Department of Pathology, Wayne State University School of Medicine, Detroit, MI, USA
7. ⁷Translational Research and Clinical Epidemiology, Department of Internal Medicine, Wayne State University School of Medicine, Detroit, MI, USA

Correspondence: Dr CJ Kim, MD, PhD, Department of Pathology, Wayne State University School of Medicine, 4 Brush North, Room 4620, Hutzel Women's Hospital, 3990 John R, Detroit, MI 48201, USA. E-mail: cjkim@med.wayne.edu; Dr R Romero, MD, Perinatology Research Branch, 4 Brush, Hutzel Women's Hospital, 3990 John R, Detroit, MI 48201, USA. E-mail: prbchiefstaff@med.wayne.edu

Received 18 March 2009; Revised 18 April 2009; Accepted 29 April 2009; Published online 8 June 2009.

Abstract

Acute chorioamnionitis is a response to amniotic fluid (AF) infection. However, it remains unclear whether substantial bacterial propagation in the chorioamniotic membranes (CAMs) precedes microbial invasion of the amniotic cavity (MIAC), which is inconsistent with characteristic 'amniotropic neutrophil migration' in acute chorioamnionitis. This study was performed to determine whether CAMs have widespread bacterial infection during MIAC and whether bacteria normally colonize CAMs. AF pellets and CAMs from the following groups were studied: group 1, patients with positive (n=18) or negative (n=22) AF cultures; group 2, patients with or without acute chorioamnionitis in which the amnion and chorion were studied separately (n=60); and group 3, patients at term who underwent a cesarean delivery (n=30). SYTO 9/propidium iodide fluorescent staining and fluorescent in situ hybridization for 16S rRNA were performed. Real-time quantitative PCR for 16S rDNA and PCR for genital mycoplasmas were also conducted. Bacteria were more frequently detected in AF than in CAMs of patients with positive AF culture (100 vs. 33%; P<0.0001). Bacteria were detected more frequently in CAMs as the severity of chorioamnionitis increased (P<0.01). The median 16S rRNA gene copy number in the amnion was significantly greater than in the chorion (group 2; P<0.0001). Bacteria were not detected in CAMs or AF in women at term before labor (group 3). A fraction of patients with chorioamnionitis or MIAC did not have bacteria in CAMs. Collectively, the findings herein indicate that MIAC does not follow widespread infection of CAMs, but precedes it. We propose a model of MIAC: the initial stage is intra-amniotic bacterial invasion through a discrete region of the CAMs, followed by intra-amniotic proliferation, and bacterial invasion of CAMs primarily extends from the amniotic fluid. This study emphasizes the importance of assessing the intra-amniotic compartment for diagnosis and treatment of preterm birth.