{"paper_id":"499cc29f-5c8b-4156-8aae-22d6dc98c85c","body_text":"Chlamydia trachomatis  is the most common bacterial sexually transmitted infection worldwide, especially among young adults [ 1 ]. Chlamydia infections remain often undiagnosed, as they are asymptomatic in the majority of patients. Undiagnosed and untreated chlamydia infections can ascend to the upper genital tract, where they colonize the endometrial mucosa and the fallopian tubes, leading to pelvic inflammatory disease (PID). Chlamydial PID can cause tubal occlusion and subsequent infertility, or partial occlusion with an increased risk for ectopic pregnancy [ 2 ]. However, most women who have tubal infertility or ectopic pregnancy have never been diagnosed with C. trachomatis  PID because their infections have been asymptomatic or subclinical.\nRisk factors frequently associated with chlamydial PID and its sequelae are young age, sexual intercourse at an early age, a large number of sexual partners,\ninconsistent condom use, and the presence of chlamydia antibodies [ 3 ].\nSeroepidemiological studies have indicated that chlamydia infections account for a large proportion of asymptomatic genital tract infections by demonstrating a strong link between tubal pathology and the presence of chlamydia antibodies [ 4 , 5 ]. Thus, chlamydia IgG antibodies are associated with the development of late sequelae and are markers for previous exposure or endogenous reactivation of a previous chlamydia infection. In\nchronically infected patients negative for endocervical C. trachomatis , a positive serological test may be the only indication of chlamydia involvement [ 6 ].\nIn the present study, we evaluated whether prior exposure to C. trachomatis , as determined by IgG chlamydia antibody titers, was\nassociated with infertility due to tubal occlusion and ectopic pregnancy in\nBrazilian women.\n\nThe study was performed in subfertile and parous women who entered a clinic specialized in human reproduction in the municipality of Goiânia,\na city with 1 093 007 inhabitants in the central region of Brazil. During the period from\nMarch to December 2001, 110 women aged 18 to 38 were selected. Group I\nconsisted of 33 women with infertility due to unilateral or bilateral tubal\nocclusion, confirmed by laparoscopy, and 22 women that presented one or more\nprior episodes of ectopic pregnancy. The group II consisted of 55 parous women.\nWomen who had used oral or topical vaginal antimicrobial treatments 15days prior to sample collection were excluded from the study. After giving written consent, women underwent standardized interviews concerning demographic characteristics, gynecological and obstetrics antecedents, self behaviors, potential risk factors for cervical and vaginal infections and symptoms. Peripheral blood was collected and all sera were transported, under refrigerated conditions, to the Laboratory of Cellular Immunology, Institute of Tropical Pathology and Public Health of the Federal University of Goiás, where they were cryopreserved at −20°C.\nThe gynecological examination included a careful examination of the abdomen, external and internal genitalia, and a speculum examination was performed.\nAfter cleaning the ectocervix, collection proceeded with an appropriate swab,\nwhich was introduced into the endocervix and then rotated for five seconds and\nremoved carefully. It was immediately placed in a PCR transportation tube and\nagitated for five seconds. The samples were transported to the laboratory,\nwhere they were at 2–8°C until processing, which\noccurred within 7days.\nC. trachomatis  DNA was amplified using the\nAmplicor kit (Roche Molecular Systems, Branchburg, NJ, USA),\naccording to the manufacturer's instructions. \nThe internal control was used in each amplification reaction, such as positive and negative controls to C. trachomatis,  purchased by the kit.\nSerum samples were assayed for chlamydia IgG antibodies employing the Hemagen Virgo C. trachomatis  IgG test (Electronucleonics Incorporation, Columbia, Ill, USA), according to the manufacturer's instructions. This is a whole cell inclusion immunofluorescence assay (WIF) that uses L2 serotype of C. trachomatis . Positive reactions, inclusions presenting a brilliant apple green fluorescence, were identified with the aid of a fluorescence microscope (Olympus Vanox with a B2 filter) at 400× magnification. For a quantitative determination, serial dilutions in PBS were performed. Dilutions of sera were expressed as antibody titers from 1/16 to 1/4096, or negative (<1/16).\n\nData processing and analyses were realized using the software programs Epi-Info 6.0 (CDC, Atlanta, Ga, USA) and SPSS 8.0 (SPSS, Chicago, Ill, USA).\nInitially, a descriptive analysis of the main sociodemographic characteristics,\nsexual behavior of the participants, and its related risk factors was\nperformed. The prevalence of C. trachomatis  antibodies was calculated with corresponding 95% confidence interval (CI) and compared between groups by the χ \n 2  test or the Fischer exact test. For\ncomparison of chlamydia IgG antibody titers, the Mann-Whitney test was used.\nUnivariate and multivariate logistic regression analyses were performed to\ndetermine the risk factors associated with chlamydial antibodies. P  < .05 were considered statistically significant.\n\nThe study protocol was approved by the Ethics Committee on Human and Animal Medical Research of the University Hospital, Federal University of Goiás (Protocol no. 047/2001).\n\nThe demographic characteristics and sexual history of the study population are\noutlined in Table 1 . The mean age ± the standard deviation for group I was 30.7 ± 4.3 years, while for group\nII it was 34.0 ± 4.6 years. Among the 55 women from group I, the great majority were married/cohabiting (87.3%); the same was observed in group II (81.8%). More than 2/3 of the population of groups I (91.0%) and II (76.3%) were educated to high school or university level, either complete or incomplete, which indicates a good level\nof education in this population. In relation to sexual behavior, the mean age\nfor initiating sexual intercourse in group I was 19.4 ± 3.7 years old and for group\nII, 18.4 ± 2.7 years old. Seventeen women from group I (30.9%) and 11 from group II (20.0%) reported having four or more partners in life.\nThe prevalence of chlamydia IgG antibodies was significantly higher\n( P  < .01) in the group I (31/55–56.4%) when compared to group II (17/55–31.0%). In the women with tubal occlusion this\nvalue was 54.5% and in those with previous ectopic pregnancy it was 59.1% ( P  > .05). Nine women in the group I (16.4%) reported a previous PID episode. The clinical diagnosis for PID was based on the following criteria: acute pelvic pain, especially in the postmenstrual period, fever, abnormal cervical discharge, uterine/adnexial tenderness, and cervical motion tenderness. Eight of these women (88.9%) presented with chlamydia IgG antibodies.\nAmong the 31 positive samples from group I, 23 (74.2%) presented titers greater than 1/64, while in group II this occurred in only two (11.8%) of the 17 positive\nsamples ( P  < .01) ( Figure 1 ). Moreover, titers equal to or greater than 1/128 were found in 42.4% of the samples of women with tubal occlusion and in 40.9% of those with previous ectopic pregnancy ( P  < .05). Titers ≥1/1024 were found in six patients (21.4%) and two of them were positive to C. trachomatis  plasmid DNA.\nC. trachomatis  DNA was only detected in two\nendocervical samples from group I (3.6%; IC 95%) and in none of the 55 samples from group II.\nUnivariate analysis was performed to determine the degree of association between sociodemographic findings and sexual behavior and the presence of chlamydia antibodies ( Table 2 ). The odds ratios (OR), with their respective 95% confidence intervals (CI) showed statistical significance in group I for the variables: number of sexual partners (2 to 3 partners compared to 1; ≥4\ncompared to 1) and previous PID. In group II, only the previous STD variable\nshowed significant association.\nAfter adjustment to a logistic regression model, chlamydia IgG antibodies were significantly associated with a greater number of lifetime sexual partners\n(estimated OR 3.06; CI 95% 1.7–5.5; P  < .05) and with tubal pathology (estimated OR 2.93; CI 95% 1.2–6.9; P  < .05).\n\nIn the current study, we found a high prevalence rate and titers of chlamydia IgG antibody among women with tubal occlusion or previous ectopic pregnancy from the central part of Brazil and an association between chlamydial antibodies and a higher number of lifetime sexual partners.\nThe prevalence rate of chlamydia IgG antibodies was significantly higher (56.4%) in the subfertility group than in parous women (31.0%). These results are in agreement with other studies that used immunofluorescence, like that of\nKihlström et al. [ 7 ], who found\nchlamydia IgG antibodies in 56% of women with previous PID, tubal factor\ninfertility, or prior ectopic pregnancy. Other authors have also reported high\nlevels of chlamydia antibodies in women with tubal pathology, using other\ndiagnostic tests [ 8 ].\nChlamydia antibody titers (CATs) have been shown to be of predictive value in the detection of tubal damage and increased risk of ectopic pregnancy \n[ 4 , 9 , 10 ] and are quantitatively related to the severity of damage [ 6 ]. These observations are in agreement with our results. Titers equal to or greater than 1/128 were found in around 40% of the women with tubal occlusion or previous ectopic pregnancy and in six patients the titers were ≥1/1024 (21.4%).\nThe immunofluorescence test employed in the present study is highly sensitive, as shown by a blinded comparative study of other serological tests for C. trachomatis  antibody carried out in two international centres [ 11 , 12 ]. The test detects both group-specific lipopolysaccharide and species-specific antibodies [ 13 ]. Therefore, women with a positive serology but with a normal pelvis may have cross-reactive responses to past infection with other species of chlamydia such as Chlamydia pneumoniae  [ 14 ] or Chlamydia psittaci  [ 12 , 15 ]. In our study we did not evaluate C. pneumoniae  antibodies. However, den Hartog et al. [ 6 ] evaluated serum of women with distal tubal pathology (DTP) and found that the presence of C. trachomatis  antibodies was the\nonly independent predictor for DTP. The predictive value of C. trachomatis  antibodies for DTP could not be improved by adding test results of \n C. pneumoniae  or lipopolysaccharide antibody testing.\nBased on the correlation between chlamydia IgG antibody titers and the presence of tubal sequelae, some authors suggest that testing for these antibodies should be part of the basic routine investigation in infertility clinics [ 5 , 16 ]. Land et al. [ 10 ] have incorporated this procedure into the routine of their services since 1992. The predictive value of chlamydia IgG antibody titers detected by indirect immunofluorescence in the diagnosis of tubal pathology is considered by some to be comparable, or even superior to, that of histerosalpingography, according to some authors [ 17 , 18 ].\nAn association between the presence of chlamydia IgG antibodies and the number of sexual partners occurred in this study. This finding is in agreement with\nprevious studies which show that this is one of the most relevant risk factors\nin the acquisition of PID and its sequelae [ 19 , 20 ].\nIn the current study, chlamydia IgG antibodies were also an important risk factor for tubal occlusion and ectopic pregnancy. After adjustment to a multivariate model, a frequency of 2.9-fold greater exposure to C. trachomatis  in\nsubfertile women was found in comparison with parous women. These data suggest\na positive association between chlamydia infection and the risk of developing\nectopic pregnancy and tubal infertility. In a previous study, it was shown that\nexposure to this pathogen is approximately three-fold greater in women with\ntubal infertility and ectopic pregnancy in comparison with a control group of\nfertile women [ 21 ].\nOnly two women in the present study were positive for\n C. trachomatis  in their endocervix by PCR. Other authors have also\nfound low prevalence rates of 1.3–8.3% in similar\npopulations, using nucleic acid amplification techniques [ 11 , 22 , 23 ]. One possible explanation for this low chlamydia detection rate could be that in\nwomen with tubal infertility or with a previous ectopic pregnancy the bacterium has ascended through the cervix and endometrium to the fallopian tubes and is no longer present in the endocervix. This hypothesis is supported by the data of Patton et al. \n[ 24 ]. The authors found C. trachomatis  DNA and/or antigens in fallopian tubes from 19 of 24 women (79.2%) with tubal\nfactor infertility suggesting a persistent upper genital tract chlamydial\ninfection. Furthermore, evidence exists that C. trachomatis  may persist in a viable and metabolically active state in the upper genital tract, despite\nnegative PCR results from the cervix. Possible reactivation of the\nmicroorganism, for example by uterine instrumentation, may result in a renewed\nupper genital tract infection [ 25 ].\nIn conclusion, we demonstrated a high prevalence and titers of chlamydia IgG antibody in Brazilian women with tubal occlusion or prior ectopic pregnancy.\nChlamydia antibodies were associated with sexual behavior. Thus, an inapparent\nchlamydia infection could have triggered these sequelae in the studied\npopulation confirming the importance of C. trachomatis  as a cause of tubal dysfunction in this population. For practical clinical purposes, chlamydia serology is useful mainly as a screening test for the likelihood of tubal damage in infertile women and may facilitate decisions on which women should proceed with further more invasive investigations. Our data reinforce the need for public health policies in developing countries, which promise triage and eventual treatment for young women with C. trachomatis  genital infections, thus avoiding the serious sequelae to women's reproductive health and a reduction in the financial burden hospital commitment.","source_license":"CC-BY-4.0","license_restricted":false}