Association between Prior Chlamydia trachomatis Infection and Ectopic Pregnancy at a Tertiary Care Hospital in South Western Uganda

Background Increase in the number of ectopic pregnancy is attributed to increase in the incidence of pelvic infections. Chlamydia trachomatis is responsible for most of the sexually transmitted bacterial infections. If undetected and untreated, the infection can ascend to the upper genital tract and cause pelvic inflammatory disease (PID) and related sequelae (ectopic pregnancy and tubal factor infertility). Objective To determine the association between prior Chlamydia trachomatis infection and ectopic pregnancy at Mbarara Regional Referral Hospital (MRRH). Methods This was an unmatched case-control study carried out at MRRH involving 25 cases and 76 controls. Serological evidence of prior chlamydial infection was determined by testing for the presence of Chlamydia immunoglobulin G antibodies in their blood. Logistic regression was used to determine the association between prior Chlamydia trachomatis infection and also the factors associated with ectopic pregnancy. The significant level of <0.05 was used. Results. Chlamydia antibodies were found in 60% of patients with ectopic pregnancy and 26.3% of the controls (p=0.002). The presence of Chlamydia antibodies was associated with a fourfold risk of ectopic pregnancy. Conclusion There was a strong association between prior Chlamydia trachomatis infection and ectopic pregnancy.


Introduction
In Uganda, in the last three decades, ectopic pregnancy has assumed epidemic proportion with signi cant maternal morbidity, mortality, and fetal wastage (Aliya (2010)). e incidence of ectopic pregnancy is directly related to the prevalence of salpingitis. Endosalpingeal damage secondary to pelvic in ammatory disease caused by Chlamydia trachomatis infection has been found to be a major risk factor for the development of ectopic pregnancy [1]. Studies from northern and southwestern Nigeria showed macroscopic evidence of pelvic in ammatory disease at laparotomy in over 60% of patients with ectopic pregnancy [2]. Both increased incidence of C. trachomatis infection and the e cacy of antibiotic therapy in preventing total tubal occlusion, after an episode of salpingitis, are related to increased incidence of ectopic pregnancy [3].
In females, up to 40% of chlamydial cervicitis may ascend to the endometrium and is responsible for the etiology of endometritis and salpingitis [4]. Fallopian tube samples of patients with ectopic pregnancy have also been found positive for C. trachomatis deoxyribonucleic acid using the polymerase chain reaction. e frequent association between chlamydial cervicitis and the presence of vaginal clue cells or Gram stain abnormalities indicative of an overgrowth of anaerobic bacteria has led to the speculation that C. trachomatis alters the normal vaginal ecology, thereby setting the stage for a complex polymicrobial upper genital tract infection. Untreated or poorly treated chlamydial infection of the genital tract can therefore have serious long-term reproductive consequences. Late sequelae of the disease include chronic pelvic in ammatory disease, tubal blockage and infertility, chronic pelvic pain, and ectopic pregnancy [5,6]. Seroepidemiological studies have indicated that chlamydial infections account for a large proportion of asymptomatic genital tract infection, by demonstrating a strong link between tubal pathology and the presence of chlamydial antibodies. e chlamydial immunoglobulin G antibodies are associated with the development of late sequelae and are markers for previous exposure. In chronically infected patients negative for endocervical C. trachomatis, a positive serological test may be the only indication of chlamydial involvement. Serum immunoglobulin G antibodies to Chlamydia have also been associated with tubal factor infertility and ectopic pregnancy [7].
is study therefore evaluates prior exposure to C. trachomatis in patients treated for ectopic pregnancy by determining the presence of immunoglobulin G antibodies in their sera. By comparing the same with women carrying normal intrauterine pregnancy, it seeks to determine the risk association between prior C. trachomatis infection and ectopic pregnancy at Mbarara Regional Referral Hospital.

Materials and Methods
is was an unmatched case-control study that is (1 : 3 ratio of cases to controls) carried out at Mbarara Regional Referral Hospital between September 2016 and January 2017. Consecutive sampling method was used to enrol all mothers who met the inclusion criteria until the sample size was achieved. e cases were 25 consecutive women with a diagnosis of ectopic pregnancy during the study period. e control group was made up of women with con rmed uncomplicated intrauterine pregnancy, attending the antenatal clinic of MRRH. Each case of ectopic pregnancy was followed by pregnant controls attending the antenatal clinic. e sample size for this study was calculated using the formula provided in Kelsey et al. (1996) for calculating the sample size for unmatched case-control studies. After computing for the expected 10% attrition, the sample size was calculated to be 25 cases and 76 controls giving a total sample size of 102. All patients who had laparotomy for tubal ectopic pregnancy, who consented to this study and satis ed the inclusion and exclusion criteria, were recruited until the minimum sample size was obtained. e approval to conduct the study was obtained from the Department of Obstetrics and Gynecology, Faculty of Medicine Research Committee (DMS 6), MUST Research Ethics Committee (Number 21/7-16), and the Uganda National Council of Science and Technology (HS 2146). Informed consent was obtained from all respondents, and con dentiality was ensured. Study participants were identi ed by study codes and not their names, for issues of con dentiality. In addition, authority was sought from the o ce of the hospital director of MRRH to conduct the study in this institution.
Serological evidence of prior chlamydial infection was determined in both groups by examining for the presence of Chlamydia immunoglobulin G antibodies in their blood. Five milliliters (5 ml) of blood was collected from the volar surface of the forearm of all the participants and emptied into clean, gold coated vacutainer. e specimen was collected from patients with ectopic pregnancy before blood transfusion and surgery and also from the control group. e blood was taken to the research laboratory (lancet laboratory) where the specimen was allowed to clot, centrifuged, and sera obtained. e sera were analyzed in batches for Chlamydia antibodies. e serological assay was done using the ImmunoComb Chlamydia trachomatis Immunoglobulin G kit (Orgenics, France), an indirect solid-phase enzyme immunoassay (EIA) test that quantitatively measures antibodies to Chlamydia trachomatis in human serum. e reagent test kit was brought to room temperature, and then, 10 μL (microlitre) pipetted serum was assayed with reagent control samples for Chlamydia trachomatis antibodies. e collected data were entered into a password protected computer using Microsoft excel, and data were imported into STATA 11 software for analysis. Results were presented as means with standard deviations, frequency distribution tables, and cross tables. Descriptive analysis was done for sociodemographic, sexual, and reproductive characteristic data, and tests of signi cance were done using chi-square test. e level of statistical signi cance was set at p value < 0.05. e regression logistic analysis was used to determine the association between Chlamydia trachomatis and ectopic pregnancy and also found the other factors associated with ectopic pregnancy.

Results
Majority of the participants were Banyankole, had at least primary level of education, and were residents of Mbarara. Mean age of the cases was 27.8 ± 4.3 and 25.2 ± 5.9 for controls. ere were signi cantly more (12 (48.0%)) single women among the cases population with ectopic pregnancy than the controls (5 (6.58%)) (p < 0.01) ( Table 1). Table 2 shows sexual behaviour of the participants. e cases engaged in sexual intercourse at a signi cantly younger age than the controls (p < 0.01). ere were more cases with multiple sexual partners compared to the control group. Previous history of pelvic in ammatory disease was obtained in 24 (96.00%) among the cases. is was signi cantly more than 8 (10.53%) among the controls (p � 0.02). e cases group had signi cantly more surgeries 13 (52.00%) compared to only 6 (7.89%) among the control group.
Chlamydia antibodies were found in 15 (60%) and 20 (26.32%) of the cases and controls, respectively (p � 0.02). e presence of Chlamydia antibodies is associated with a fourfold risk of ectopic pregnancy (OR 4.4; CI 1.70-11.48) ( Table 3). Table 4 shows that prior Chlamydia trachomatis infection, single women, and a history of previous pelvic/ abdominal surgery are independently associated with ectopic pregnancy at Mbarara Regional Referral Hospital.
Logistic regression model analysis of risk factors for ectopic pregnancy showed a fourfold increase in the risk of ectopic pregnancy in those with C. trachomatis antibodies when controlling for previous history of pelvic in ammatory disease (OR 4.9; 95% CI 1.15-21.29).
Controlling for the e ects of sociodemographic characteristics, previous history of pelvic in ammatory disease, marital status, and number of sexual partners showed that single women (OR 12.0; 95% CI 2.21-65.32) and having a history of previous abdominal/pelvic surgery (OR 6.9; 95% CI 1.47-32.98) were shown to be more likely to be associated with ectopic pregnancy as well.

Discussion
Serum immunoglobulin G antibodies have been associated with endosalpingeal damage and ectopic pregnancy resulting from pelvic in ammatory disease [8,9]. In this study, the prevalence of chlamydial antibodies was signi cantly higher   (60.0%) in patients with ectopic pregnancy than in women with intrauterine pregnancy (26.32%). is nding con rms the report of other researchers that patients with ectopic pregnancy are more likely to have immunoglobulin G antibodies against C. trachomatis when compared with women with intrauterine pregnancy [4,9]. is suggested relationship was further strengthened by the nding in this study of a fourfold increase in the risk of ectopic pregnancy in women with chlamydial antibodies. e prevalence of prior C. trachomatis infection, as evidenced by the presence of chlamydial antibodies in this and other studies, underscores the reason why ectopic pregnancy has remained a major complication of reproductive health of women and why the incidence is increasing in parallel with the rise in the rate of chlamydial infection [10]. is study also found that women with ectopic pregnancy were more likely to be single, corroborating the ndings in another Nigerian study [11]. is can be explained by the fact that single women are more likely to have multiple sexual partners than married women. In examining the risk factors for Chlamydia infection among patients with ectopic pregnancy and normal pregnancy, this study looked at the sexual and reproductive characteristics of both groups. e results of univariate analysis showed that women with ectopic pregnancy engaged in sexual intercourse at a signi cantly younger age than women with intrauterine pregnancy. is nding is in line with previous studies which linked early coitarche with Chlamydia trachomatis infection [12]. Adolescents are prone to sexually transmitted disease (STD) because of high risk sexual behaviour. Biologically, the adolescents are particularly at risk of STD because the columnar epithelium, which is susceptible to Chlamydia and gonococcal organisms, extends from the endocervical canal to the ectocervix making it fully exposed to pathogens. ey also have di culties using barrier methods of contraception and may have less access to STD care because of limited facilities, negative peer pressure, concealment, and restrictive policies especially in the developing countries [13]. Previous history of pelvic inammatory disease was signi cantly higher in patients with ectopic pregnancy when compared with women with  [8]. In contrast to other studies [8], this study found that induced abortions were relatively low in women with ectopic pregnancy and did not have any association with ectopic pregnancy. is could be explained by the small number of participants in this study who had a history of induced abortions, and therefore it was di cult to assess statistical di erence with such a small number. In our study, previous history of abdominal/pelvic surgery was associated with ectopic pregnancy and this compares with studies done in other areas [14]. is could be explained by the fact that previous surgery in the pelvic area or on the tubes can cause adhesions and impede the egg's movement causing implantation in the tubes and other sites.
However, when these were subjected to logistic regression model (multivariate) analysis, patients with Chlamydia antibodies showed a fourfold increase in the risk of ectopic pregnancy when controlling for previous history of pelvic in ammatory disease. In addition, the multivariate analysis revealed that women with a history of previous abdominal/ pelvic surgery and those who were single were signi cantly associated with ectopic pregnancy.

Conclusion
is study has shown that a greater proportion of women with ectopic pregnancy had serological evidence of prior C. trachomatis infection than women with intrauterine pregnancy. It also demonstrated a fourfold risk association between prior C. trachomatis infection and ectopic pregnancy. Furthermore, being single and having a history of previous abdominal/pelvic surgery were positively associated with ectopic pregnancy.
ere is therefore a need for the sexual reproductive health division in the MoH to strengthen/ improve sex education and for prompt and e ective treatment of sexually transmitted infections.

Conflicts of Interest
e authors declare that they have no con icts of interest.