Study population and specimen sampling

From October 2005 to August 2006, the samples of a 100 Japanese women from our university hospital were collected and a total of 31 patients were eligible with all data to be enrolled in this study. The ages of the cases were 20-53 years old and they did not receive any antibiotic treatment or other medication before sample collection that could have affected the study results. The sampling timing during menstrual cycle was not standardized in this study. Written informed consent was obtained from all participants in this prospective study, which was approved by the institutional review board at the University of Occupational and Environmental Health.

An unmoistened speculum was inserted before any other vaginal examination and vaginal pH was measured by use of color strips with a pH range of 3.6–6.1 (pH-Fix 3.6–6.1; Macherey-Nagel, Duren, Germany). The pH strip was placed on the lateral vaginal wall until wet, the color change of the strip was immediately compared with the colorimetric scale and the measurement was recorded. After making saline wet mount slides of vaginal fluid, specimens were collected by first inserting a sterile speculum into the vagina and wiping the vaginal discharge from the posterior fornix with sterile cotton swabs. The swabs were placed into an aerobic transport gel (Transwab, Medical Wire & Equipment Co. Ltd., Corsham, Wilts, England) and an anaerobic transport vial for up to 6 hours before assessment of microbial activity. The swabs were then rolled onto a glass slide for Gram staining to evaluate the Nugent score and to detect clue cells.

Lactobacillary grade (LAC grade)

All wet mount slides were evaluated for LAC grade and were classified according to Donders’ criteria. LAC grade I (normal flora) corresponds predominantly to Lactobacillary morphotypes, with very few coccoid bacteria present. LAC grade II (intermediate flora) corresponds to a diminished lactobacillary flora, which is mixed with other bacteria. LAC grade II was subdivided into 2 categories: one in which there was still a large amount of lactobacilli present (IIa), and a second group in which the coccoid flora outnumbered the lactobacillary flora (IIb). Finally, LAC grade III (BV) flora consists of numerous other bacteria, with no lactobacilli present. This grading system was chosen because of its simple and quick response, and its usefulness.

Grading of Nugent score and culture-based bacterial identification

Gram stained slides were examined for the presence of BV in accordance with standardized scoring criteria (Nugent score) by 1 trained technician who was blind to the origin of the slides and to the culture results, and reviewed by the author. BV was diagnosed if the score was 7-10; a score of 4-6 indicated intermediate vaginal flora; and a score of 0-3 indicated normal vaginal flora. Swabs of vaginal fluid were inoculated onto Drigalski agar medium, egg yolk salt agar medium, sheep blood agar medium (Nissui Pharmaceutical Co., Ltd., Tokyo), Desoxycholate agar (Kyokuto Pharmaceutical Industrial Co., Ltd., Tokyo, Japan), CHROM agar Candida for aerobic culture (Kanto Chemical Co., Inc., Tokyo, Japan), Thayer Martin agar (Becton, Dickinson and Company, Franklin Lakes, NJ), Gardnerella agar for CO ₂ culture (self-made), and Brucella HK agar for anaerobic culture (Kyokuto Pharmaceutical Industrial Co., Ltd., Tokyo, Japan). Semiquantitation of 1+, 2+, and 3+ means the colonies occupied one-third, one-half, and the entire area of the agar plate, respectively. The 3+ growth of Lactobacilli indicates that it is a 10 ⁷ colony forming unit.

Clone library method of broad-range 16S rRNA gene sequencing

For the clone library method, vaginal fluid was collected by the same method and was placed in a transport vial. Vaginal swabs were vigorously agitated in 2.0 mL Tris-HCl buffer (100 mM Tris-HCl, 50 mM EDTA-2Na, [pH 8.0]) to dislodge the cells. After treatment with a vortex mixer for 2 minutes, 900 μL aliquot of the cell suspension was transferred into a 2.5 mL polypropylene-tube. One hundred μL of 30% sodium dodecylsulfate (SDS, final concentration 3.0%) solution was added to the cell solution and approximately 0.3 g of a mixture of glass beads that consisted of equal weights of (a) 0.1 mm- and (b) 1 mm-diameter beads. The mixture was then vigorously shaken at 4500 rpm for 5 minutes on a MicroSmash MS-100 (Tomy Seiko Co., Ltd., Tokyo, Japan) and the supernatant was collected by centrifugation at 20,000 × g for 5 minutes at room temperature. This DNA extraction was carried out 3 times to improve a yield of DNA. The 3 supernatants were combined and treated with an equal volume of phenol-chloroform-isoamyl alcohol (25:24:1, vol/vol). The DNA in aqueous phase was concentrated and replaced by about 30 μL TE buffer using Montage PCR Centrifugal Filter Devices (Millipore, Bedford, MA). Using the extracted DNA as a template, the partial 16S rRNA gene fragments (approximately 550 base pair) were amplified by PCR method with a pair of universal primers (341F; 5′-CCTACGGGAGGCAGCAG-3′ and 907R; 5′-CCGTCAATTCMTTTRAGTTT-3′). Cycling conditions were as follows: 96C for 5 minutes, followed by 30 cycles of 96C for 30 seconds, 53C for 30 seconds and 72C for 1 minute, with final elongation at 72C for 7 minutes with a Gene Amp PCR system 9700 thermocycler (Applied Biosystems, Foster City, CA). The amplified products were cloned into Escherichia coli using a TOPO TA cloning kit (Invitrogen, Carlsbad, CA). Nucleotide sequences of the randomly chosen 96 clones were determined using the Big Dye Terminator v3.1 cycle sequencing kit with ABI3130 sequencer (Applied Biosystems). Homology searches were performed with the basic local alignment search tool (BLAST) using an inhouse software system and database, which collected type strains from Ribosome Database Project II ( http://rdp.cme.msu.edu/ ). The species was identified with more than 95% similarity to the nucleotide sequence of the type strain.

Principal component analysis (PCA)

PCA analysis was performed by using the proportions of bacterial species level obtained by the clone library method to characterize intravaginal microbial flora. R software version 2.9.1. ( http://cran.r-project.org/ ) was used for the PCA software.