Porras, Carolina MS*; Safaeian, Mahboohbeh PhD, MPH†; González, Paula MD*; Hildesheim, Allan PhD†; Silva, Sandra MS‡; Schiffman, Mark MD, PhD†; Rodríguez, Ana Cecilia MD†; Wacholder, Sholom PhD†; Freer, Enrique MD, PhD‡; Quint, Koen BSc§; Bratti, Concepción MD, MPH*; Espinoza, Albert MBA*; Cortes, Bernal PharmD*; Herrero, Rolando MD, PhD*; for the Costa Rica HPV Vaccine Trial (CVT) Group
CHLAMYDIA TRACHOMATIS (CT) IS THE MOST common bacterial STI. Worldwide, this intracellular, Gram-negative bacterium produces around 92 million new genital infections.1
Ct infects endocervical epithelium and produces urogenital infections, which are asymptomatic in 70% to 90% of women and 50% of men, and persist months to years. Manifestations in women include dysuria, discharge, bleeding, abdominal discomfort, and cervical ectopy or friability; in men, urethral discharge or pruritus and dysuria.2,3
Serious consequences include pelvic inflammatory disease when infection ascends to the uterus, tubes, or ovaries. Resulting tissue scarring causes infertility, pelvic pain, and risk of ectopic pregnancy. Ct during pregnancy causes chorioamnionitis, premature membrane rupture and delivery, and early pregnancy loss. Neonatal infection includes neonatal conjunctivitis and pneumonia.4 Genital infection is a possible cervical cancer cofactor,5 and can facilitate transmission of human immunodeficiency virus.6
Ct prevalence is highest in women aged ≤24 years.7,8 Risk factors for infection include having new or multiple sexual partners, history of previous STIs and being unmarried.4,9,10 Some studies have suggested that cervical ectopy, oral contraceptive use, or intrauterine devices (IUD) may also play a role.11–14 Consistent and correct use of condoms is at least partially protective,15,16 and male circumcision could play a role in reduction of risk.17 However, few studies have investigated in detail the role of male sexual behavior in the risk of female acquisition of Ct infection.
Public health measures for prevention, early diagnosis, and treatment to prevent complications require epidemiologic data. However, there is very limited information about the epidemiology of Ct infection in developing countries. A study performed in 10 areas including countries from Latin America,8 reported the highest prevalence among women 15 to 24 years old in Colombia (7.2%), while the third was in Argentina (5.4%). In the US-Mexico border, Baldwin et al.18 reported a prevalence of 14% (Hybrid capture 2, HC2), while in Honduras, among women 18 to 35 years old; the prevalence was 6% (ELISA-immunofluorescence).19
In Costa Rica (CR), a population-based study evaluated Ct seroepidemiology among women 25 to 59 years, seroprevalence was 56.1%, and increased with sexual partners.20 However, serology represents past exposure and cross-reacts with C. pneumoniae.
In the United States, annual screening is recommended for sexually active women ≤24 years.10 In CR, despite universal health care, no screening has been implemented, partly as a result of lack of knowledge about the epidemiology of this asymptomatic but potentially serious infection.
We are conducting a community based trial of an human papillomavirus (HPV) vaccine in CR, among 7466 young women. At enrollment, we tested all sexually experienced women for Ct using the HC2 test (Digene Corp) to provide a benefit to participants and to study in detail the epidemiology and some clinical manifestations of Ct. Additionally we determined serovars using a polymerase chain reaction (PCR)-based detection and genotyping assay (Ct-DT),21 providing unique opportunities to describe genotypes in this population and evaluate whether genotype-specific associations exist which may provide guidance for future prevention strategies (e.g., screening or vaccination).
Materials and Methods
We included women 18 to 25 years who had initiated sexual activity at enrollment into a community-based trial investigating efficacy of an HPV16/18 vaccine to prevent cervical neoplasia in Guanacaste and Puntarenas, CR. Women were identified through a population census performed for the study. Potential participants were invited via personally delivered appointments to study clinics.22
Participants presented their ID, signed informed consent and were interviewed. A medical history and physical examination were performed, including pelvic examination for sexually experienced women (see below). To be eligible, physicians verified that they were not hysterectomized, pregnant or lactating, were mentally competent, in good general health, willing to use contraception and not moving for 6 months. Women excluded had histories of chronic or immunodeficiency conditions and history of vaccination against or known history of hepatitis A. A total of 7466 women, approximately 30% of the census agreed and fulfilled inclusion criteria. Women enrolled randomly received candidate vaccine against HPV 16/18 or hepatitis A control vaccine. Protocols were approved by the institutional review boards of INCIENSA and the US NCI.
Data and Specimen Collection
Extensively trained female interviewed participants in private settings about sociodemographic, sexual and reproductive history, contraceptives, and smoking. Women reporting one lifetime sexual partner were asked about their partner's characteristics because in this group the evaluation of male characteristics that may determine Ct infection is more feasible, given the complexity of analyzing behavior of multiple partners.
The pelvic examination included evaluation of the external genitalia, vagina, and cervix. After visualizing the cervix, the transformation zone was described and discharge, warts, and any abnormalities were documented. Exfoliated cervical cells were collected using a Cervex brush (Rovers Medical Devices BV, Netherlands). The long bristles were inserted in the os, the brush was rotated 5 times and introduced into vials containing 20 mL of PreservCyt solution (Cytyc, Marlborough), where cells were rinsed vigorously. Samples were transported at approximately 20°C to the laboratory, where two 0.5 mL aliquots were drawn using PCR safe procedures. ThinPrep slides were then prepared for cytology, and the remaining solution was sent for HPV, Ct and Neisseria gonorrhoeae (GC) testing by HC2 (University of CR).
The cytology slides were screened sequentially by 2 cytotechnologists and interpreted by 1 cytopathologist who also reported the presence of infections.
Participants with Ct received counseling and treatment with 1 g Azithromycin for them and their partner(s). Treatment efficacy was confirmed at the next visit.
All study procedures were conducted by staff specifically hired and extensively trained for the trial, according to Good Clinical Practices and using detailed standard operating procedures in accordance with the study protocol.
Ct Testing by HC2-CT/GC and HC2-CT
Both HC2-CT/GC and HC2-CT are FDA-approved tests. It is a nucleic acid hybridization assay with signal amplification that combines antibody capture of DNA and RNA probe hybrids and chemiluminescent signal detection. Testing was done according to manufacturer's instructions. All samples were first tested by a combined HC2 CT/GC DNA test, which contains a probe cocktail mixture complementary to approximately 39,300 base-pair (4%) of the Ct genomic DNA and 7500 bp or 100% of the cryptic plasmid; and 9700 base-pair (0.5%) of the GC genomic DNA and 4200 bp or 100% of the cryptic plasmid. This is a qualitative test, specimens with relative light units/cutoff (RLU/CO) value <1.00 indicate the absence of Ct and GC DNA or their presence in levels below the detection limit of the assay. Specimens with an RLU/CO value ≥1.00 were further tested by HC2 CT and GC specific tests to identify the organism(s) present in each sample.
Serovars were determined on all HC2-CT positive specimens and an 8% random sample of HC2-CT/GC negatives, using Ct-DT genotyping.21 Briefly, total DNA was isolated from PreservCyt aliquots, and eluted in water. Extraction runs contained positive and negative controls. The Ct PCR primer set was used on extracted DNA to amplify all serovars available in GeneBank. This multiplex primer set amplifies a fragment of 241 bp from the cryptic plasmid and a fragment of 160/157 bp from the Variable Region 2 of the omp1 gene.
Reverse primers containing a biotin label at the 5′ end enabled capture of the reverse strand onto streptavidin coated plates. Captured amplimers were denatured by alkaline treatment, and detected by a defined cocktail of digoxigenin-labeled probes. This Ct DNA enzyme immunoassay provides optical density values. Each DNA enzyme immunoassay experiment contains positive, borderline-positive and negative controls and a PCR-positive control. It detects all serovars, subserovars, and genovariants in GeneBank but cannot differentiate between serovars. Borderline samples were considered positive.
The same amplimers used to detect Ct-positive samples were analyzed to differentiate Ct serovars by reverse hybridization (RHA) on a line probe assay, with probes for the cryptic plasmid, 3 different Ct serogroups (B, C, and Intermediate), for the 14 probes that hybridizes 19 serovars (A, B/Ba, C, D/Da, E, F, G/Ga, H, I/Ia, J, K, L1, L2/L2a, and L3). One extra probe was added to detect a genovariant of serovar J that otherwise remains undetected. Each RHA-run contained a negative and a positive control.
Of the 7466 women, we excluded 1595 without pelvic examination (1592 virgins, 3 examinations not possible). We further excluded 42 (0.7%) with insufficient sample. Of the 5829 Ct results, 140 were tested using samples collected a few weeks after enrollment, given insufficient volume. These had similar Ct positivity and were included.
We calculated prevalence of Ct by HC2 and 95% confidence intervals (CI) using standard methods. To study women's characteristics associated with Ct infection we calculated age—and lifetime number of sexual partner—adjusted odds ratios (ORs) and 95% CI using unconditional logistic regression. A multivariate model was developed including the variables associated with Ct in this study (age, education, marital status, lifetime sexual partners, IUD, history of miscarriage, smoking, and cervical ectopy-cervicitis). We also included use of condoms (never, past, and in last 30 days) given its anticipated reduction in Ct infection. Another multivariate model restricted to women reporting 1 lifetime sexual partner was prepared including the same variables mentioned above plus the lifetime partners of the partner.
We considered ectopy as a possible risk factor because it can facilitate Ct infection. As we evaluated ectopy by visual inspection, it can be confounded by cervicitis (which can be a Ct manifestation), therefore we labeled it as ectopy-cervicitis, this is important for the interpretation of the results.
We assessed dose response associations by treating ordinal variables as continuous assuming a linear trend in the models (Ptrend).
Clinical findings found at the pelvic examination (vaginal discharge, location of transformation zone) or in cytology (squamous diagnosis and degree of inflammation) are presented as a comparison of the percent of women with the characteristic among Ct positive and Ct negative women; a Pearson χ2 test was done to define statistical significance of the difference between positive and negative women.
To evaluate the association of other cervico-vaginal infections with Ct infection, crude ORs and 95% CI were calculated using unconditional logistic regression.
We also considered genotype-specific associations. Since agreement between HC2 Ct and Ct-DT test was excellent,23 all HC2 Ct negative specimens, not tested by the Ct-DT assay were considered negative. We reanalyzed data at the level of the most common genotypes (serovar D/Da, E, F, and I/Ia) using multivariate logistic regression to identify factors associated with specific Ct genotype infection.
All analyses were carried out using Stata/SE version 8.1.
Characteristics of Women
Median age was 21 years old, 68% had over 7 school years and 15% some university education, 52% were married, 44% single, and 3% divorced, separated, or widowed. Median age of sexual debut was 17 years old and 42% reported one lifetime sexual partner. Sixty percent had been pregnant and 8.8% reported 1 or more miscarriages, 53% currently used oral contraceptives, 20% condoms, and 2.6% an IUD in the last month. Only 16% reported ever having smoked. Among women reporting one lifetime partner, 22% reported their partner had never had other partners.
Prevalence of Ct Infection
Overall prevalence was 14.2% (95% CI, 13.3–15.1), peaked at age 20 to 21 (15.5%) and declined slightly with age (Table 1).
Women's Characteristics and Ct Infection
Women with ≥12 years of education had a slightly higher adjusted OR for Ct detection (OR, 1.4, 95% CI, 1.1–1.7). Compared with married women; divorced, separated, widowed, and single women were more likely to have Ct infection. Age at sexual debut, years of sexual activity, and frequency of intercourse were not associated with Ct after adjustment for sexual partners.
The strongest association was with increasing number of lifetime sexual partners. Women reporting 2 sexual partners had a 2-fold increase and those with 3 or more had 3-fold increase in Ct detection compared with women with only 1 partner (P for trend <0.001). Age at first menses, phase of menstrual cycle, frequency of menses, and days since last sexual intercourse were not associated with Ct detection (data not shown).
Ct detection was not associated with oral contraceptives or injectable contraceptive use (Table 2). Current IUD users had a 60% increase in Ct detection. Ever use of condoms, use in the last month, and consistency of use were not associated with infection.
Pregnancies did not increase Ct detection, but women with miscarriages had a 30% lower detection. Self-reported histories of STIs did not increase Ct detection. Current smokers had a higher Ct detection than never smokers, particularly if they smoked regularly. Compared with women who had no visible transformation zone, an increase in Ct detection was seen with increasing degree of cervical ectopy-cervicitis, with a 2-fold detection among those with extensive ectopy-cervicitis.
Multivariate Models of Female Characteristics
In the multivariate model including all the variables associated in the age- and sexual partners-adjusted model, all the variables that were significant remained associated with Ct detection (Table 3).
In analyses restricted to women reporting one lifetime sexual partner, including all female variables of the multivariate model, with additional adjustment for sexual partners of the male, the associations with marital status, IUD use, miscarriage, and smoking were not present (Table 4). Women with a partner with 4 to 6 partners had a 2-fold increase of Ct infection compared with women whose partners had only one lifetime partner, but those with partners with 7 or more partners had an OR of 1.0. After adjustment for number of partners of the partner, other male characteristics evaluated were not associated with Ct detection (age, education, living with the woman, and smoking).
Clinical Findings Among Women With and Without Ct Infection
Table 5 presents the prevalence of clinical and cytologic findings among Ct positive and Ct negative women. Vaginal discharge was significantly more common among Ct positive women (20.7% vs. 15.2%, P <0.001). Minor grade cytologic abnormalities were much more common among Ct positive women (P <0.001), particularly reactive changes (74.2% vs. 63.0%), as was moderate or severe inflammation (61.7% vs. 41.4%, P <0.001).
STIs Correlated With Ct Infection
Detection of high-risk HPV by HC2 was associated with 90% increase in Ct (Table 6). Prevalence of GC infection detected by HC2 in this population was 0.8% (95% CI, 0.6–1.1) and 78% of infected women also had Ct infection. Detection of GC was associated with a 21-fold increase in Ct detection.
Although cytology is not accurate for cervical infections, evidence of herpes, trichomonas, cocobacilli, and gardnerella were associated with Ct detection, but candida was not.
Analysis of Ct Genotypes
Our ancillary analysis of Ct genotypes showed serovar E to be the most common type (4.3%, 30.1% of Ct positive), followed by serovar F (3.0%, 20.6% of Ct positive), serovar D/Da (2.9%, 20.4% of Ct positive), and serovar I/Ia (2.1%,15.0% of Ct positive). Serovars B/Ba, G/Ga, H, J, and K were found in less than 1% of the samples. Serovars rarely found to infect the cervix (A, C, L1, L2/L2a, and L3) were not detected. When we examined associations of the most common serovars (E, F, D/Da, and I/Ia) with risk factors, we did not observe any statistically significant effect modification by serovar, in other words, each of the Ct genotypes displayed the same associations with the variables under study (data not shown).
To our knowledge, this is the largest study in young women in a developing country and the first reporting community-based data on prevalence of Ct using HC2.
Our results indicate that this is a very high-risk area for Ct infection (14.2%), higher than other population-based studies8 and similar to that observed in high-risk populations or sexually transmitted disease clinics.18,24,25 The high prevalence in our study is consistent with high prevalence of HPV (approximately 50%) and other STIs. Different Ct detection assays may partially explain prevalence differences between studies, but we confirmed our results using the new Ct-DT assay.23 High prevalence could also be partially explained by selection bias resulting in a sample at higher risk of STIs than the general population (see below).
We confirmed strong associations with sexual behavior of women, both through variables that assess it indirectly (e.g., marital status) and directly (e.g., lifetime number of sexual partners), with a significant trend of increasing Ct detection with increasing number of sexual partners.
We found a nearly 2-fold increase in Ct detection among current users of IUD; but it was not evident among women reporting 1 lifetime partner after adjusting for male sexual behavior. IUD use has been associated with pelvic inflammatory disease, through contamination of the uterine cavity during insertion.26–28 Several authors have reported increases in genital infections and bacterial vaginosis,14,29,30 but there is controversy over whether IUDs lower host resistance to infections.31,32
Condoms were not protective, contradicting some reports mainly from developed countries, although lack of association has been also reported [reviewed in Ref. 15]. This is possibly related to nonuse or inconsistent use of condoms during each sexual intercourse, or improper ascertainment. This finding might imply that in high-prevalence areas, condoms are less effective unless used perfectly with all partners, pointing to the need of education about the correct use of condoms for STI prevention.
Current but not former smoking was associated with Ct detection, but the association disappeared among women reporting 1 lifetime partner adjusting for male sexual behavior. An effect of smoking is not entirely implausible, and some studies have demonstrated immunosuppressive effects in the cervix of smokers,33,34 but our findings seem to be the result of residual confounding.
Ct increased with increasing cervical ectopy-cervicitis, as reported by others.13 Ct infects the endocervical tissue, which is more exposed when there is ectopy. Hence ectopy can facilitate infection but can also make Ct more detectable. Alternatively, Ct can induce cervicitis that can be interpreted as ectopy. We evaluated ectopy-cervicitis only by visual inspection, which may be inaccurate.35
As expected, Ct infection was associated with detection of other sexually transmitted infections. In particular, there was a strong correlation between detection of gonorrhea and Ct, with an OR of 21. The prevalence of GC (0.8%) is very similar to that reported in high-risk populations.8 The high frequency of coinfection with Ct may be explained by the sexual cotransmission of these organisms, and it has been suggested that some infections can facilitate acquisition of other infections36,37
Among monogamous women, number of sexual partners of the partners, as reported by the women, was strongly associated with Ct. Interestingly, the association went back to unity for women reporting partners with 6 or more female partners. Whether this is a spurious finding or is related to some unknown immunologic aspect of the infection in men is unknown.
In this analysis we had the unique opportunity to examine the epidemiology of Ct serovars and whether particular serovar(s) may be more “aggressive” than others or associated with a specific risk factor, but risk factors did not differ across the most prevalent genotypes examined.
This study was very large and community based, but participants were volunteers that had to meet several eligibility criteria including not being pregnant, using an acceptable contraceptive method, and being in good health. Therefore, our group may not be free of selection bias, and we may have selected a group at higher risk than the general population, although it is difficult to speculate about the possible direction of bias without additional information. Another limitation of this study is that it is cross-sectional and hence causality or temporality with certain characteristics cannot be established. We had the opportunity to evaluate male sexual behavior in relation to risk of Ct infection. However, this analysis was restricted to women with 1 lifetime partner and therefore it cannot be generalized to other groups.
In summary, our study of sexually transmitted Ct infection in a rural population in Costa Rica has uncovered one of the highest reported prevalences in the world and suggests the need to establish prevention, screening, and treatment programs in the study area.
1. World Health Organization. Global prevalence and incidence of selected curable sexually transmitted infections: Overview and estimates. Geneva, Switzerland: World Health Organization, 2001.
2. Brunham RC, Rey-Ladino J. Immunology of Chlamydia infection: implications for a Chlamydia trachomatis
vaccine. Nat Rev Immunol 2005; 5:149–161. Review.
3. Paavonen J, Eggert-Kruse W. Chlamydia trachomatis
: Impact on human reproduction. Hum Reprod Update 19995:433–447.
4. Peipert JF. Clinical practice. Genital Chlamydia infections. N Engl J Med 2003; 349:2424–2430. Review.
5. Smith JS, Bosetti C, Munoz N, et al. Chlamydia trachomatis
and invasive cervical cancer: A pooled analysis of the IARC multicentric case-control study. Int J Cancer 2004; 111:431–439.
6. Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: The contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Transm Infect 1999; 75:3–17. Review.
7. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2004. Atlanta, GA: U S. Department of Health and Human Services, 2005.
8. Franceschi S, Smith JS, van den Brule A, et al. Cervical infection with Chlamydia trachomatis
and Neisseria gonorrhoeae
in women from ten areas in four continents. A cross-sectional study. Sex Transm Dis 2007; 34:563–569.
9. Fenton KA, Korovessis C, Johnson AM, et al. Sexual behaviour in Britain: Reported sexually transmitted infections and prevalent genital Chlamydia trachomatis
infection. Lancet 2001; 358:1851–1854.
10. U.S. Preventive Services Task Force. Screening for chlamydial infection: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2007; 147:128–134.
11. Cottingham J, Hunter D. Chlamydia trachomatis
and oral contraceptive use: A quantitative review. Genitourin Med 1992; 68:209–216.
12. Baeten JM, Nyange PM, Richardson BA, et al. Hormonal contraception and risk of sexually transmitted disease acquisition: Results from a prospective study. Am J Obstet Gynecol 2001; 185:380–385.
13. Lee V, Tobin JM, Foley E. Relationship of cervical ectopy to chlamydia infection in young women. J Fam Plann Reprod Health Care 2006; 32:104–106.
14. Hiltunen-Back E, Haikala O, Kautiainen H, et al. A nationwide sentinel clinic survey of Chlamydia trachomatis
infection in Finland. Sex Transm Dis 2001; 28:252–258.
15. Warner L, Stone KM, Macaluso M, et al. Condom use and risk of gonorrhea and Chlamydia: A systematic review of design and measurement factors assessed in epidemiologic studies. Sex Transm Dis 2006; 33:36–51 Review.
16. Holmes KK, Levine R, Weaver M. Effectiveness of condoms in preventing sexually transmitted infections.: Bull World Health Organ 2004; 82:454–461.
17. Castellsague X, Peeling RW, Franceschi S, et al. Chlamydia trachomatis
infection in female partners of circumcised and uncircumcised adult men. Am J Epidemiol 2005; 162:907–916.
18. Baldwin SB, Djambazov B, Papenfuss M, et al. Chlamydial infection in women along the US-Mexico border. Int J STD AIDS 2004; 15:815–821.
19. Tabora N, Zelaya A, Bakkers J, et al. Chlamydia trachomatis
and genital human papillomavirus infections in female university students in Honduras Am J Trop Med Hyg 2005; 73:50–53.
20. Vetter KM, Barnes RC, Oberle MW, et al. Seroepidemiology of chlamydia in Costa Rica. Genitourin Med 1990; 66:182–188.
21. Quint KD, van Doorn LJ, Kleter B, et al. A highly sensitive, multiplex broad-spectrum PCR-DNA-enzyme immuno assay and reverse hybridization assay for rapid detection and identification of Chlamydia trachomatis
serovars. J Mol Diagn 2007; 9:631–638.
22. Hildesheim A, Herrero R, Wacholder S, et al. Effect of human papillomavirus 16/18 L1 virus like particle vaccine among young women with preexisting infection: A randomized trial. JAMA 2007; 298:743–753.
23. Quint KD, Porras C, Safaeian M, et al. Evaluation of a novel PCR-based assay for the Detection and Identification of Chlamydia trachomatis
Serovars in Cervical Specimens. Submitted to J Clin Microbiol 2007; 45:3986–3991.
24. Poulin C, Alary M, Bernier F, et al. Prevalence of Chlamydia trachomatis
and Neisseria gonorrhoeae
among at-risk women, young sex workers, and street youth attending community organizations in Quebec City, Canada Sex Transm Dis 2001; 28:437–443.
25. Miller WC, Ford CA, Morris M, et al. Prevalence of chlamydial and gonococcal infections among young adults in the United States. JAMA 2004; 291:2229–2236.
26. Steen R, Shapiro K. Intrauterine contraceptive devices and risk of pelvic inflammatory disease: Standard of care in high STI prevalence settings Reprod Health Matters 2004; 12:136–143.
27. Arias RD. Compelling reasons for recommending IUDs to any woman of reproductive age. Int J Fertil Womens Med 2002; 47:87–95. Review.
28. Farley TM, Rosenberg MJ, Rowe PJ, et al. Intrauterine devices and pelvic inflammatory disease: An international perspective. Lancet 1992; 339:785–788.
29. Joesoef MR, Karundeng A, Runtupalit C, et al. High rate of bacterial vaginosis among women with intrauterine devices in Manado, Indonesia Contraception 2001; 64:169–172.
30. Hodoglugil NN, Aslan D, Bertan M. Intrauterine device use and some issues related to sexually transmitted disease screening and occurrence. Contraception 2000; 61:359–364.
31. Grimes DA. Intrauterine device and upper-genital-tract infection. Lancet 2000; 356:1013–1019. Review.
32. Westhoff CL. Current assessment of the use of intrauterine devices. J Nurse Midwifery 1996; 41:218–223. Review.
33. Poppe WA, Ide PS, Drijkoningen MP, et al. Tobacco smoking impairs the local immunosurveillance in the uterine cervix. An immunohistochemical study. Gynecol Obstet Invest 1995; 39:34–38.
34. Barton SE, Maddox PH, Jenkins D, et al. Effect of cigarette smoking on cervical epithelial immunity: A mechanism for neoplastic change? Lancet 1988; 2:652–654.
35. Jacobson DL, Peralta L, Farmer M, et al. Relationship of hormonal contraception and cervical ectopy as measured by computerized planimetry to Chlamydia infection in adolescents. Sex Transm Dis 2000; 27:313–319.
36. Ness RB, Kip KE, Soper DE, et al. Bacterial vaginosis (BV) and the risk of incident gonococcal or chlamydial genital infection in a predominantly black population Sex Transm Dis 2005; 32:413–417.
37. Wiesenfeld HC, Hillier SL, Krohn MA, et al. Bacterial vaginosis is a strong predictor of Neisseria gonorrhoeae
and Chlamydia trachomatis
infection. Clin Infect Dis 2003; 36:663–668.