In this cross-sectional study, we enrolled women attending the Gynecology Out Patient Department (OPD) of Seth G.S. Medical College and King Edward Memorial (KEM) Hospital, Parel, Mumbai, between 2003 to 2009. The group comprised of women with histories of recurrent spontaneous abortion (RSA, n=143), infertility (n=264), symptoms and signs of lower genital tract infections (LGTI, n=213), pregnant women (n=174) attending the antenatal care (ANC) unit, as well as those who had no symptoms and signs of any infection (asymptomatic controls, n=102) but came for family planning advice.

Ethics Committees of the institute, as well as of KEM Hospital approved the study. Each woman was informed about the study and written consent was obtained from all the women before enrollment.

The clinician team did a routine gynecological per speculum examination to record signs of infection and collected endocervical and vaginal swab specimens. First, endocervical swab (Hi-media Laboratories Pvt. Limited, Mumbai, India) specimens were collected in a sterile container with 1ml PBS (pH=7.5) for immediate processing to detect C. trachomatis infection, while second specimens were stored in dry sterile vials at -20°C for confirmatory tests, if required. Vaginal specimens were also collected from the posterior fornix using a wooden spatula for the diagnosis of bacterial vaginosis (BV) using Nugent's scoring system for Gram stain smears (11) and for the detection of trichomonas and candida by wet mount. Blood specimens collected from the women were used for antibody test using commercially available ELISA kit (Novatech Immuno-diagnostica, GMBH).

In order to check specimen adequacy, each cervical specimen in PBS was vortexed for 30 s, the swab was squeezed and 10 μl of specimen was examined under microscope to see the presence of epithelial cells. Four to five epithelial cells per high power field was considered as an adequately collected specimen for further processing.

Severely eroded cervix with hypertrophic cervical erosions and a mucopurulent endocervical discharge or leucorrhoea were recorded as signs while burning micturation and pain in the abdomen reported by the women were recorded as symptoms.

DNA was isolated from cervical specimen using a rapid non-enzymatic method. The cells were pelleted and resuspended in Tris-MgCl₂-KCl buffer (pH=7.4) and treated with 10% sodium dodecyl sulphate at 55 °C for 10 min to lyse the cells. The proteins were precipitated using saturated sodium chloride solution. DNA was precipitated by 100% ethanol and eluted in Tris EDTA buffer (12). The quantity and quality of DNA was estimated spectrophotometrically and by loading an aliquot of DNA on 0.8% agarose gel. As an internal control PCR for beta-globin gene was also performed for each sample to rule out the presence of inhibitory factors in the extracted specimens.

PCR was performed on extracted DNA using primers designed from the conserved region of MOMP gene of C. trachomatis with sense primer: 5’ GCC GCT TTG AGT TCT GCT TCC 3’ and anti-sense primer: 5’ GTC GAA AAC AAA GTC ACC ATA GTA 3’ to amplify a 180 bp DNA fragment common to all serotypes (13). The reaction was carried out in a volume of 50 µl. It contained primers (0.5 µm each), 0.2 mM dNTP's, PCR buffer (10 mM Tris buffer; pH=9), 1.25 units of Taq polymerase, 10 µl of DNA specimen and the volume was adjusted with sterile distilled water. Positive and negative controls were also run in each experiment. Reaction was performed in a thermal cycler (Perkin Elmer 2400) as per the following protocol: initial denaturation was done for 5 min at 94 °C. This was followed by 35 cycles of 30 s each of denaturation at 94 °C, annealing at 55 °C and extension at 72 °C for 1 min. The final extension step was carried out at 72 °C for 5 min. The amplified products were run on 2% Agarose gel, observed under a UV transilluminator while the results were being documented. Presence of 180 bp repeat sequences in positive control specimen and its absence in the negative control indicated reaction had been completed satisfactorily. Presence of 180 bp repeat sequences in other clinical specimens indicated presence of C. trachomatis infection. Further confirmation of these amplified products was carried out using specific C. trachomatis probe in Southern hybridization (14). Probe was prepared using PCR dig-labeling kit (Roche diagnostics). Standard protocol for Southern blotting was followed for transfer of PCR products to a nylon membrane, which was then processed for hybridization using a generic probe. Instruction manual was followed to detect the probe complex using Dig-luminescence detection kit (Roche diagnostics).

Commercially available enzyme-linked immunosorbent assay (ELISA) kit was used to detect C. trachomatis specific IgG antibody (NovaTec Immunodiagnostica, GMBH). In brief, microtitre wells precoated with C. trachomatis antigens were incubated with serum specimen at a 1:100 dilution so that any corresponding antibodies present in the serum would bind to the antigen to form complexes. After washing the wells to remove all unbound sample material, horseradish peroxidase (HRP) labeled anti-human IgG conjugate was added which would bind to captured Chlamydia specific antibodies. The immune complex formed by the bound conjugate was visualized by adding tetramethylbenzidine (TMB) substrate, which gives a blue colored reaction product.

After terminating the reaction using a stop solution (Sulphuric acid, 0.2 mol/l), the absorbance of the end product, which is yellow in color, was read at 450 nm using an ELISA plate reader (μ Quant, Bio-Tek Instruments Inc.). The intensity of this product is directly proportional to the amount of Chlamydia-specific IgG antibodies in the specimen. The specimens with O.D. higher than the cut-off value (0.250−0.900) were considered positive for Chlamydia-specific antibodies and used as an indicator of past Chlamydia infection. Each positive sample was again confirmed using another serum aliquot of the same participant. The results were found to be reproducible.

Counseled each enrolled women to come back to take the report. Those found to be infected with any of these infections were treated by the clinician.

Statistical analysis using Epi Info version 6 software for Chi-squares (χ ² ) test was applied to study the association between C. trachomatis infection with the clinical manifestations. The test of significance for proportion between different groups was carried out using “Z” test. A probability value of pࣘ0.05 was considered as significant.

Eight hundred and ninety-six women were tested for current C. trachomatis infection by PCR. The participants were between 16 to 45 years old with a median age of 29 yrs, and an interquartile range (IQR) value of 10. They belonged to middle socio-economic groups and their personal history did not reveal any high risk behavior. The number of women in asymptomatic control group, as well as those in groups with different clinical histories like RSA, infertility, with lower genital tract infection (LGTI), pregnant women from antenatal care (ANC) centers, their age and the infection rate in each group is presented in Table 1.

In the RSA group, there were 58 women with 2 pregnancy losses (2SA), 77 women with more than 3 pregnancy losses (>3SA) and 8 with ectopic pregnancy. In the ANC group, the gestational period of the pregnant women varied from 2 to 4 months. There were 108 (12.1%) women with current C. trachomatis, indicating the prevalence of this infection in the study population.

Of the 108 C. trachomatis infected women, 4 (3.7%) had concomitant BV while 1 (0.9%) had concomitant Candida albicans. In the rest of participants (n=788), C. trachomatis specific antibody was present in 14 women, one woman had both the antibody and the antigen. Eighty women had other infections such as BV, Candida or Trichomonas infections and the related infection rates were 14.5%, 4.3% and 0.9%, respectively. For further analysis, these women with C. trachomatis antibody (n=15), as well as those with other infections (n=80) were excluded. Hence, there were 693 women without any infection, who were taken into consideration for comparative analysis (Table 2).

Infection rate varied from 1.96% to 25.0% among the different groups of participants (Table 1). Among the RSA subgroup a significant (p<0.001) proportion of women with ectopic pregnancy (25%) and with more than 2 spontaneous abortions (10.4%) had this infection, compared to women with 2 spontaneous abortions (5.2%). C. trachomatis infection rate was significantly low in the group of women with children or expecting a child such as asymptomatic controls (n=102), LGTI (n=213) or in ANC (n=174) groups, {9.4% (46 of 489); odds ratio: 2.19, p<0.0005)} when compared to infected infertile women (18.6%, 49 of 264). Comparison of clinical manifestation of women with only current C. trachomatis infection (n=103) with that of uninfected women (n=693) revealed significant association of C. trachomatis infection with infertility (45.6% vs. 27.3%, p=0.0001; Table 2). Another significant observation was the absence of any symptoms or signs on per speculum examination in infertile (64% vs. 36%, p<0.01) and pregnant (79.2% vs. 22.8%, p ≤ 0.001) infected women, indicating asymptomatic nature of this infection (Figure 1).

C. trachomatis infection was highest (21.8%) among women 20 years old or younger, though not statistically significant, and lowest in 21−25 year old age group. The infection rate again showed an increasing trend in women above 26−40 years of age.

Among the C. trachomatis infected women, (20.2%) had high pH (≥5), (47.7%) had more than 10 polymorphonuclear leukocytes (PMNs) in their specimens, 2.7% had yellow/grayish coloured discharge and 7.3% had blood on swab or bled during collection of specimens.

Seventy-two women with C. trachomatis infection came for the report, which were subsequently treated. Follow up record on 60 women was available only for one year. Thereafter, they could not be traced due to several reasons. In ANC group, 8 of 24 C. trachomatis positive pregnant women came for follow up and were treated. No further testing was done in these pregnant women after completion of treatment in accordance with the clinician advice to avoid any risk during pregnancy. These eight women had live births. Result of follow up in other groups revealed that four of the 52 infected individuals were positive even after treatment. There were eight live births in these groups following treatment (Table 2). Women with other infections such as BV, Candida and Trichomonas were also treated as per hospital routine procedure.