Gursahaney, Priya R. BS*; Meyn, Leslie A. MS*; Hillier, Sharon L. PHD*; Sweet, Richard L. MD†; Wiesenfeld, Harold C. MD, CM*‡
To date, the transmission dynamics of Neisseria gonorrhoeae are poorly understood. Host susceptibility to gonorrhea plays a critical role in determining who becomes infected with this pathogen, as exposure alone does not ensure transmission. Estimates of N. gonorrhoeae infection rates in female partners of men with gonococcal urethritis are 60% to 80%.1–4 Transmission is affected by behavioral factors, including number of sex acts and condom use.5,6 Biologic factors, such as anatomical changes and the hormonal and microbiologic environment of the lower genital tract further affect sexually transmitted disease (STD) transmission.7,8 The role of hormonal contraception on acquisition of gonorrhea has not been well-characterized. Results of previous studies have been inconsistent, with some reporting enhanced susceptibility to gonorrhea while others did not find this association.8–13 This study was designed to further investigate the effect of hormonal contraception on the transmission of N. gonorrhoeae.
MATERIALS AND METHODS
This study is a subanalysis of a longitudinal study examining the association between lower genital tract infection and infertility, as previously described.14 Briefly, females seeking care between 1998 and 2005 at the following sites in Pittsburgh, PA, were eligible for participation: the STD Clinic at the Allegheny County Health Department, the ambulatory clinics of the Family Health Council Inc, and the outpatient clinic at Magee-Womens Hospital. Women aged 15 to 35 years were included in this analysis if they reported sexual contact to a male diagnosed with N. gonorrhoeae. Length of time since sexual encounter with an infected male partner did not affect eligibility. Women were excluded if they were pregnant, had taken antibiotics during the previous 2 weeks, or had a hysterectomy. The Institutional Review Board at the University of Pittsburgh granted human subject approval.
The participants underwent a standardized interview obtaining medical, sexual, social, and demographic information. They also provided information on current contraceptive use at the time of enrollment. A comprehensive gynecologic exam was performed. Vaginal fluid was collected for microscopy, Gram stain, and Trichomonas vaginalis culture. Endocervical samples were obtained for N. gonorrhoeae and Chlamydia trachomatis testing.
Cervical swabs were cultured for N. gonorrhoeae by streaking endocervical samples on modified Thayer-Martin and chocolate media (confirmed by Gram stain, oxidase testing and Gonochek II analysis [EY Laboratories, San Mateo, CA]), while cervical swabs were tested for C. trachomatis by polymerase chain reaction (Roche Diagnostics, Branchburg, NJ) between 1998 and 2003. Cervical samples were tested for both N. gonorrhoeae and C. trachomatis using strand displacement amplification (Becton Dickinson, Franklin Lakes, NJ) between 2003 and 2005. For T. vaginalis culture, vaginal swabs were placed in modified Diamond's media, incubated at 37°C, and examined every other day for 7 days for motile trichomonads (InPouch TV, Biomed Diagnostics, White City, OR).
Data were analyzed with SPSS statistical software, release 17.0.0 (SPSS, Inc., Chicago, IL). All statistical tests were evaluated at the 2-sided 0.05 level of significance. Pearson χ2 test was used to evaluate associations between self-reported condom use, sexual activity, and hormonal contraception use. Prevalence ratios (PR) were used to evaluate the association of enrollment characteristics, exam findings, and laboratory findings with testing positive for N. gonorrhoeae. Poisson regression with robust variance was used to calculate PRs and their respective 95% confidence intervals (CI).15 Multivariable Poisson regression was used to calculate PRs for hormonal contraception use, and was adjusted for self-reported condom use and sexual behaviors, which were determined a priori to be potential confounding factors.
We enrolled 107 women reporting sexual contact with a male infected with N. gonorrhoeae. The mean age of the participants was 21.2 ± 2.8 years (range: 15–35 years). Sixty-five percent of the participants were black and 23% were white. Most women were single (97%) and 55% had completed high school. Only 21% of the women had private health insurance; the remaining participants were uninsured or received public health care funding. The overall rate of gonorrhea in the cohort was 63.6% (68/107). Thirty-seven (34.6%) women tested positive for C. trachomatis, and T. vaginalis was identified in 24 women (22%). Seventy-two percent of women reported the number of days since their last sexual encounter with the suspected source partner. The mean time between suspected exposure and study enrollment was 16.5 days. A smaller percentage of female participants (41%) were able to provide information about partner treatment before enrollment.
Women less than 21 years of age were more likely to test positive for N. gonorrhoeae than older women, although this trend was not statistically significant (Table 1). Race was not associated with infection, with gonorrhea detected in 63% of blacks and 65% of women from other racial backgrounds (P = 0.8). Education, insurance, and substance abuse were not associated with testing positive for N. gonorrhoeae. Similarly, history of gonorrhea, chlamydia, genital herpes, or bacterial vaginosis was not related to current gonococcal infection.
The effect of sexual behaviors on gonorrhea infection rates was examined (Table 2). A new sexual partner in the month before enrollment was more common among women testing positive for N. gonorrhoeae compared to uninfected women (35% vs. 13%; PR: 1.47; 95% CI: 1.13, 1.90; P = 0.004). Gonorrhea rates were not different among women reporting multiple (≥2) sexual partners in the preceding 3 months than among women with one sexual partner (70% vs. 58%). On univariate analysis, sex during menses was associated with gonorrhea. Ninety percent of women reported sexual activity within 14 days of enrollment, and while gonorrhea rates were higher among women with recent sexual activity, this difference did not attain statistical significance (67% vs. 36%, P = 0.137). Frequency of sexual activity and douching were unrelated to gonorrhea infection. Sex with an uncircumcised partner was uncommon (n = 6) in our cohort.
Combined hormonal contraception (estrogen with progestin-containing oral contraceptives, n = 19; vaginal ring, n = 1; or transdermal patch, n = 2) was reported by 21% of women in the cohort, and 8% used depomedroxyprogesterone acetate (DMPA; Depo-Provera, Pfizer Inc, NY). There was no difference in the rates of any hormonal contraceptive use over the course of the study, and nearly all (94%) of the women reported using their method of hormonal contraception for at least 2 months before enrollment. Length of time since exposure to N. gonorrhoeae was similar in women using and not using hormonal contraception. Furthermore, the proportion of male partners treated before subject enrollment as well as the time interval from partner treatment to subject enrollment was similar between these 2 groups. Women using combined hormonal contraception were significantly less likely than women not using hormonal contraception to test positive for N. gonorrhoeae (32% vs. 76%; PR: 0.42; 95% CI: 0.22, 0.78; P = 0.006). An inverse relationship between DMPA use and gonorrhea approached statistical significance on univariate analysis (PR: 0.44; 95% CI: 0.17, 1.11; P = 0.083). Condom use, which was reported for the last 10 coital episodes, was not associated with a lower rate of N. gonorrhoeae infection.
We examined sexual behaviors among participants to determine whether women using hormonal contraception (combined hormonal contraception or DMPA) participate in less risky sex, and therefore are at lower risk for STDs. Frequent sexual activity, defined as 4 or more sexual episodes per month, was reported by 65% of hormonal contraceptive users compared to 72% of nonusers (P = 0.420), while 90% of users and nonusers reported recent sexual activity (preceding 14 days). Likewise, having a new sexual partner (23% vs. 29%, P = 0.502) or multiple sexual partners (32% vs. 49%, P = 0.120) was not different among women using and not using hormonal contraception. Finally, male condom use was similar between the 2 groups, as none of the women on hormonal contraception reported perfect condom use and 55% reported any condom use by their sexual partners, compared to 10% and 67% of women not using hormonal contraception, respectively.
Multivariable Poisson regression (Table 3) confirmed that combined hormonal contraception remained associated with lower rates of N. gonorrhoeae infection (PR: 0.43; 95% CI: 0.23, 0.81) after adjusting for condom use and sexual behaviors. After adjustment, DMPA use was independently associated with lower N. gonorrhoeae rates as well (PR: 0.39; 95% CI: 0.15, 0.99). Sex during menses was excluded from this analysis to avoid bias since women on hormonal contraception are less likely to have menstrual flow. A new sexual partner during the previous month remained independently associated with testing positive for gonorrhea.
No association was observed between presence of N. gonorrhoeae and symptoms, and overall, 50% (34/68) of women with gonorrhea were asymptomatic. Similarly, physical findings did not correlate with infection, and only 1 woman had cervical motion tenderness. Gonorrhea rates were similar among women with and without cervical ectopy (63% in each group).
This study demonstrates that combined hormonal contraceptive use may be protective against acquisition of gonorrhea. Among women reporting sexual contact to an infected male partner, combined hormonal contraception use was associated with a significantly lower likelihood of testing positive for N. gonorrhoeae. After controlling for factors associated with N. gonorrhoeae infection, women who were using combined hormonal contraception were 56% less likely to test positive for gonorrhea than nonusers (adjusted PR: 0.43; 95% CI: 0.23, 0.81). Gonococcal infection was also less common among women using DMPA, although only 9 women reported using this method of contraception (adjusted PR: 0.39; 95% CI: 0.15, 0.99).
Prior studies evaluating the association between hormonal contraception and gonorrhea are inconsistent.8–13,16 Louv et al reported that compared to nonusers, the relative rate of gonorrhea in oral contraceptive users was increased by approximately 70%.10 An earlier study examining the prevalence of gonorrhea in a predominantly black cohort found that the highest rate of gonorrhea was associated with oral contraceptive use.11 Gonorrhea rates in women using oral contraception were greater than those in women using barrier methods of contraception, but they were not increased compared with women using no contraception, suggesting that lack of barrier contraception, rather than oral contraceptive use, is responsible for higher risk of gonorrhea. Other studies have not observed an association between combined hormonal contraception and N. gonorrhoeae infection. In a prospective cohort assessing risk of HIV acquisition in 4200 women in South Africa, gonorrhea was not associated with oral contraceptives or DMPA use.12 Similarly, the incidence of gonococcal infection was not increased with combined hormonal contraception in another prospective study of Kenyan commercial sex workers.8,13 Finally, a systematic review concluded that the majority of studies failed to identify an association between oral contraceptive use and gonorrhea.9 Our study provides further evidence against an association between combined hormonal contraception and increased N. gonorrhoeae susceptibility; rather our results suggest that combined hormonal contraception use may reduce the risk of N. gonorrhoeae acquisition.
The influence of steroid hormones on genital tract infection has been under considerable investigation. Clinical observations that gonococcal pelvic inflammatory disease and disseminated gonococcal infection occur more frequently during menses suggest an interaction between sex hormones and the pathogenesis of gonorrhea.17,18 In mice, estrogen increases susceptibility to disseminated gonococcal infection.19 Sex steroids (estrogens and progestins) can affect immune function, including estrogen-related decrease in cell-mediated immune response, interactions with mucosal and nonmucosal humoral immunity, and cytokine production.20 Progesterone suppresses the growth of N. gonorrhoeae, whereas estrogen slightly enhances gonococcal growth or has no effect in human and mouse models.20–22 Progesterone-induced thickening of cervical mucus may also limit STD acquisition. Cervical mucus contains antimicrobial agents, such as secretory leukocyte protease inhibitor (SLPI). SLPI concentrations are highest in the ovulatory phase of the menstrual cycle, implicating SLPI regulation by ovarian steroids.23,24 The viscosity of cervical mucus progressively increases after ovulation in response to progesterone.25 The rate of infection of N. gonorrhoeae may vary by oral contraceptive formulation, with higher risk of gonorrhea seen with more androgenic progestins.10 We do not have information on the progestin formulation of oral contraceptives used by our study participants and therefore cannot comment on the influence of the types of progestins on N. gonorrhoeae infection. Additional studies in humans are needed to further characterize the influence of steroid hormones on the pathogenesis of STDs.
Anatomical and physiological effects of combined hormonal contraceptives may modulate the pathogenesis of STDs. Cervical ectopy, frequently seen in oral contraceptive users, may enhance susceptibility to chlamydial infection.10 While N. gonorrhoeae infects columnar epithelial cells, which are increasingly exposed with hormonal contraception use, a relationship between cervical ectopy and N. gonorrhoeae infection has not been observed.10,26 Similarly, we did not identify an increase in gonorrhea rates in women with cervical ectopy, suggesting that cervical ectopy plays little, if any role in the acquisition of gonorrhea in women. Another effect of combined hormonal contraception is a reduction in menstrual blood flow. Gonococcal growth is augmented in iron-rich menstrual blood.27,28 Lighter menstrual flow resulting from hormonal contraceptive regulation reduces iron sources, potentially inhibiting growth of N. gonorrhoeae.
Sexual behaviors may also modify the risk of STD acquisition. We did not find that consistent condom was protective against gonococcal infection. Our findings may be explained by others who postulate that self-reported condom use is not associated with lower STD incidence because of reporting bias.6 Although we asked subjects about the frequency of condom use, we did not further explore the manner of proper condom practices. Having a new sexual partner in the preceding month increased the likelihood of testing positive for N. gonorrhoeae 4-fold; however, the frequency of sexual intercourse and recent sexual intercourse were not associated with an increased rate of gonorrhea. These findings are consistent with those of Klausner et al who did not identify an association between sexual risk behavior and repeated gonorrhea infections.29
Because of the cross-sectional study design, we are unable to determine whether contraceptive use preceded acquisition of gonorrhea, and whether the women were the “index” case or acquired the infection from their partner. Another limitation is that the infection status of the male partner relied on self-report by the female participants and was not confirmed; therefore, the transmission rate cannot be precisely determined. However, the observed transmission rate of N. gonorrhoeae in our cohort (64%) is consistent with other reports.2–4,26 Given this high prevalence of disease, we used the PR in our analyses because even though it may demonstrate a moderate preventive effect of hormonal contraception on gonorrhea, it is a more accurate measure of effect.
The small number of women using combined hormonal contraception (n = 22) and DMPA (n = 9) also limits the strength of the conclusions. Although our study group of 107 women is among the largest cohorts of women reporting sexual exposure to N. gonorrhoeae, the study may still be underpowered to determine other variables associated with gonorrhea acquisition. Previous studies have examined risk factors for N. gonorrhoeae infection in populations undergoing STD screening; our study is restricted to those women suspected of a recent sexual exposure to N. gonorrhoeae, enabling a stronger analysis of risk factors for infection given the higher probability of infection. Finally, a strength of our study is that it explores N. gonorrhoeae acquisition in users of modern hormonal contraceptives. Nonetheless we support recent appeals for a randomized controlled trial of hormonal contraception and the risk of STD acquisition.30
In summary, our study provides important information on the modulation of N. gonorrhoeae transmission by hormonal contraception. An increase in gonococcal infection rates among combined hormonal contraceptive or DMPA users following reported sexual exposure was not observed. Rather, hormonal contraceptive users in our cohort were less likely to test positive for gonorrhea after exposure. These data highlight the need for additional studies to determine whether, in addition to contraceptive benefits, modern hormonal contraception has a protective effect on the acquisition of N. gonorrhoeae.
1. McCormack WM, Stumacher RJ, Johnson K, et al. Clinical spectrum of gonococcal infection in women. Lancet 1977; 1:1182–1185.
2. McCormack WM, Reynolds GH; Cooperative Study Group. Effect of menstrual cycle and method of contraception on recovery of Neisseria gonorrhoeae
. JAMA 1982; 247:1292–1294.
3. Lin JL, Donegan SP, Heeren TC, et al. Transmission of Chlamydia trachomatis
and Neisseria gonorrhoeae
among men with urethritis and their female sex partners. J Infect Dis 1998; 178:1707–1712.
4. Platt R, Rice PA, McCormack WM. Risk of acquiring gonorrhea and prevalence of abnormal adnexal findings among women recently exposed to gonorrhea. JAMA 1983; 250:3205–3209.
5. 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.
6. Zenilman JM, Weisman CS, Rompalo AM, et al. Condom use to prevent incident STDs: The validity of self-reported condom use. Sex Transm Dis 1995; 22:15–21.
7. Wiesenfeld HC, Hillier SL, Krohn MA, et al. Bacterial Vaginosis is a strong predictor of Neisseria gonorrhoeae
and Chlamydia trachomatis
infection. Clin Inf Dis 2003; 36:663–667.
8. 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.
9. Mohllajee AP, Curtis KM, Martins SL, et al. Hormonal contraceptive use and risk of sexually transmitted infections: A systematic review. Contraception 2006; 73:154–165.
10. Louv WC, Austin H, Perlman J, et al. Oral contraceptive use and the risk of chlamydial and gonococcal infections. Am J Obstet Gynecol 1989; 160:396–402.
11. Berger GS, Keith L, Moss W. Prevalence of gonorrhoea among women using various methods of contraception. Br J Vener Dis 1975; 51:307–309.
12. Myer L, Denny L, Wright TC, et al. Prospective study of hormonal contraception and women's risk of HIV infection in South Africa. Int J Epidemiol 2007; 36:166–174.
13. Lavreys L, Chohan V, Overbaugh J, et al. Hormonal contraception and risk of cervical infections among HIV-1-seropositive Kenyan women. AIDS 2004; 18:2179–2184.
14. Wiesenfeld HC, Hillier SL, Krohn MA, et al. Lower genital tract infection and endometritis: Insight into subclinical pelvic inflammatory disease. Obstet Gynecol 2002; 100:456–463.
15. Coutinho LM, Scazufca M, Menezes PR. Methods for estimating prevalence ratios in cross-sectional studies. Rev Saude Publica 2008; 42:992–998.
16. Austin H, Louv WC, Alexander WJ. A case-control study of spermicides and gonorrhea. JAMA 1984; 251:2822–2824.
17. Sweet RL, Blankfort-Doyle M, Robbie MO, et al. The occurrence of chlamydial and gonococcal salpingitis during the menstrual cycle. JAMA 1986; 255:2062–2064.
18. Britigan BE, Cohen MS, Sparling PF. Gonococcal infection: A model of molecular pathogenesis. N Engl J Med 1985; 312:1683–1694.
19. Kita E, Takahashi S, Yasui K, et al. Effect of estrogen (17 beta-estradiol) on the susceptibility of mice to disseminated gonococcal infection. Infect Immune 1985; 49:238–243.
20. Sonnex C. Influence of ovarian hormones on urogenital infection. Sex Transm Infect 1998; 74:11–19.
21. Morse SA, Fitzgerald TJ. Effect of progesterone on Neisseria gonorrhoeae
. Infect Immune 1974; 10:1370–1377.
22. Salit IE. The differential susceptibility of gonococcal opacity variants to sex hormones. Can J Microbiol 1982; 28:301–306.
23. Moriyama A, Shimoya K, Ogata I, et al. Secretory leukocyte protease inhibitor (SLPI) concentrations in cervical mucus of women with normal menstrual cycle. Mol Hum Reprod 1999; 5:656–661.
24. Ming L, Xiaoling P, Yan L, et al. Purification of antimicrobial factors from human cervical mucus. Hum Reprod 2007; 22:1810–1815.
25. Wolf DP, Blasco L, Khan MA, et al. Human cervical mucus: Part II. Changes in viscoelasticity during the ovulatory menstrual cycle. Fertil Steril 1977; 28:47–52.
26. Bhattacharyya MN, Jephcott AE. Diagnosis of gonorrhea in women–influence of the contraceptive pill. J Am Vener Dis Assoc 1976; 2:21–24.
27. McKenna WR, Mickelson PA, Sparling PF, et al. Iron uptake from lactoferrin and transferrin by Neisseria gonorrhoeae
. Infect Immune 1988; 56:785–791.
28. Serkin CD, Seifert HS. Iron availability regulates DNA recombination in Neisseria gonorrhoeae
. Mol Microbiol 2000; 37:1075–1086.
29. Klausner JD, Barrett DC, Dithmer D, et al. Risk factors for repeated gonococcal infections: San Francisco, 1990–1992. J Infect Dis 1998; 177:1766–1769.
30. Hubacher D, Raymond ER, Beksinska M, et al. Hormonal contraception and the risks of STI acquisition: Results of a feasibility study to plan a future randomized trial. Contraception 2008; 77:366–370.