| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Departments of Obstetrics and Gynecology, and Medicine, Division of Infectious Diseases, University of Washington, Seattle, Washington.
Address reprint requests to: David A. Eschenbach, MD University of Washington Department of Obstetrics and Gynecology Box 356460 Seattle, WA 98195-6460 E-mail: eschen{at}u.washington.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Women who desired DMPA for contraception were evaluated before and at 3 and 6 months after initiation of 150-mg DMPA injections every 3 months. At each visit, we assessed genital symptoms, vaginal signs, vaginal micro-flora, and histopathology by vaginal biopsies.
Results: Among 38 women observed for 6 months, there was significant reduction in mean serum estradiol level (99.9 ± 9.3 pg/mL to 26.6 ± 1.6 pg/mL, P < .001). The number of subjects with any Lactobacillus did not change, but the number with hydrogen peroxide (H2O2)–positive Lactobacillus decreased from 20% before to 12% after 6 months of DMPA (P = .005). The log concentration in colony-forming units per milliliter of vaginal fluid of H2O2–positive Lactobacillus decreased in a linear manner from 4.0 ± 0.6 at baseline to 2.5 ± 0.6 after 6 months of DMPA use (P = .006). The mean number of cell layers in the epithelium was reduced slightly from 28.1 ± 0.7 to 25.9 ± 0.9 (P = .05), epithelial thickness decreased from 1.02 ± 0.04 mm to 0.89 ± 0.05 mm (P = .005), and the glycogen-positive thickness decreased from 0.81 ± 0.04 mm at baseline to 0.66 ± 0.05 after 6 months of DMPA use (P = .005).
Conclusion: Depomedroxyprogesterone acetate produced a systemic hypoestrogenic state associated with decreased H2O2–positive Lactobacillus colonization and slight thinning of the glycogen vaginal epithelial layer. Such changes possibly compromise the vaginal barrier to infection.
Depomedroxyprogesterone acetate (DMPA) was approved by the Food and Drug Administration as a contraceptive in 1992 and is currently used by over 3 million women, including 7.9% of adolescents in the United States who use contraception.1 Medroxyprogesterone acetate inhibits gonadotropin and ovarian hormone synthesis, induces anovulation, enhances the cervical mucous barrier to sperm and causes loss of the endometrial glycogen needed for blastocyst support in humans.2 Amenorrhea occurs in over 90% of women who use DMPA for more than 2 years, and endometrial biopsies invariably show atrophic changes.2,3 Serum estradiol (E2) levels are markedly reduced in human DMPA users, and evidence of decreased bone density and high-density lipoprotein levels occurs among long-term users.4
Although a shift in vaginal cytologic maturation was documented with DMPA,5 its effects on vaginal epithelium have received little study until recently. In Macaca mulatta monkeys, profound vaginal atrophy resulted when the vaginal epithelial thickness went from greater than 25 cell layers before to less than ten cell layers after progesterone treatment.6 Experimental simian immunodeficiency virus transmission increased almost eightfold in the progesterone-treated animals.6 Although the dose of progesterone in those primates was not analogous to the DMPA dose used in humans, in a later primate study with medroxyprogesterone acetate in which progesterone levels were comparable with those in humans using DMPA for contraception, significant thinning of the vaginal epithelium still occurred and was reversible after cessation of progesterone.7
In humans, DMPA use has been associated with increased acquisition of human immunodeficiency virus (HIV) in sex workers8,9 and increased shedding of HIV among seropositive women,10 although DMPA has not been associated with HIV in large cross-sectional studies of family planning clinic attendees.11,12 The mechanism by which DMPA users might be at risk for HIV is not known. However, hydrogen peroxide (H2O2)–producing Lactobacillus inhibits HIV in vitro13 and in a recent report, the acquisition of HIV was significantly higher in women without Lactobacillus or with abnormal vaginal flora compared with those with H2O2–positive Lactobacillus in the vagina.14
In this study, we evaluated women before and during DMPA use for 6 months. We were particularly interested in whether DMPA use over that time produced changes in genital symptoms and signs; vaginal flora; visual signs of vaginal epithelial disruption; cytologic maturation index; or a reduction in the number of vaginal epithelial cell layers, glycogen content, and epithelial thickness.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Demographic, sexual, contraceptive, and gynecologic histories, including current genital symptoms, were collected on standardized forms at enrollment. Initial pelvic examinations were done 19–24 days from the last menstrual period. Depomedroxyprogesterone acetate (DepoProvera, Upjohn, Kalamazoo, MI) 150 mg was administered by gluteal intramuscular injection the next week between menstrual cycle days 1 and 5. Subsequent examinations were at 10- to 13-week intervals, and subsequent DMPA injections were given on examination days. At follow-up visits, we obtained interval sexual and gynecologic histories, and the vaginal and cervical examinations and specimens listed. A midstream clean catch urine sample was collected for culture before each examination. Serum was collected for E2 testing using the Immulite solid phase, chemiluminescent enzyme immunoassay (Diagnostic Products Corporation, Los Angeles, CA).
At each visit, the external genitalia were viewed. A nonlubricated speculum was inserted for visual inspection and colposcopic examination of vaginal walls. The amount of vaginal discharge was qualitatively assessed and recorded. A Color pHast indicator strip (EM Science, Gibbstown, NJ) was applied directly to the vaginal sidewall to determine pH. A spatula was used to scrape the right vaginal wall for Gram stain15 and a separate cytologic specimen for Papanicolaou staining and calculation of the maturation index. Maturation index was calculated by counting 100 cells in five nonadjacent areas of each vaginal cytology sample and converting the numbers into percentages of superficial, intermediate, and parabasal cells.16 For example, a maturation index of 100/0/0 is interpreted as 100% superficial cells consistent with maximal estrogen effect.
The left vaginal wall was sampled for Gram stain and wet-mount analysis. The vaginal posterior fornix was swabbed, and the swabs were placed in transport media (Port-A-Cul, Becton Dickinson, Cockeysville, MD) and cultured for aerobic and anaerobic bacteria within 12 hours as reported.17 Hydrogen peroxide–producing Lactobacillus were identified by a blue pigment formed when H2O2 oxidizes tetramethylbenzidine in brucella agar base.18 Cervical samples for cytology and for Chlamydia trachomatis culture19 were collected at enrollment, and an endocervical Gram stain was collected at all visits.
A full-thickness biopsy of the vaginal epithelium was collected from the upper one third of the vagina of each subject at each visit. In ten subjects, at baseline and at 3 and 6 months, second biopsies were taken from the lower one third of the vaginal wall for comparison. Biopsies were done by applying a small amount of topical 5% lidocaine ointment (Fougera, E. Fougera and Co., a division of Atlanta, Inc., Melville, NY) followed by a 1-mL injection of 2% lidocaine hydrochloride with 1:100,000 epinephrine (Abbott Laboratories, North Chicago, IL) using a 30-gauge needle. Approximately a 2 x 4-mm full thickness of vaginal epithelium was removed with Mini-Townsend (Cooper Surgical, Shelton, CT) cervical biopsy forceps. Biopsy tissue was pinned flat on styrofoam to minimize distortion during formalin fixation. The number of epithelial layers were counted on hemotoxylin and eosin–stained samples in three separate fields at 40x magnification using an ocular micrometer grid by an observer (DP) who had no knowledge of other patient data or the visit number. Data are presented as the mean of the three fields. Periodic acid-Schiff reaction stain was used to evaluate the epithelial glycogen content by percentage and intensity of stained layers with a three-point scale (0 = absent, 1 = present, 2 = intense). Epithelial thickness was measured at 40x magnification using the grid with each grid equivalent to 0.1 mm at 40x magnification. Thus, a 0.05-mm difference could be measured accurately.
Statistical analyses were based on an examination of a trend over time in the same individual. Logistic regression was used for correlated data in the contingency tables, and linear regression for correlated data20 was used for continuous variables with significance set at P < .05 (Generalized Estimating Equations function in Splus, Mathsoft, Inc., Cambridge, MA, using an Auto-regressive-1 correlation structure).
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The 38 subjects remained monogamous, without new sexual partners or a changes in frequency of intercourse while using DMPA (Table 1
). However, there was a marked decrease in condom use after starting DMPA (P < .001).
|
. Coital frequency per week was zero for 10% of subjects throughout the study and under two times per week for approximately 30%, without significant change throughout the study. External genital pruritus and dyspareunia were infrequent and unchanged (Table 1) as was history of dysuria, urinary urgency, and external genitalia burning or pain (data not shown). The subjective amount of vaginal discharge reported also did not change. All subjects had regular menses at baseline, but only 12 (42%) of 38 had them after 6 months of DMPA. Intermenstrual bleeding was present in 20 subjects (all with no regular menses) at 6 months compared with none at baseline (Table 1).
The few external genital or vaginal abnormalities observed throughout the study occurred randomly and included a superficial vulvar abrasion, a superficial vulvar hemorrhage, localized erythema of the vagina in one quadrant only (which spontaneously cleared in all three subjects at follow-up), diffuse erythema of the entire vagina at 3 and 6 months, and vaginal warts in one subject (Table 1). More subjects reported scant vaginal discharge at 6 months compared with baseline. The viscosity of vaginal discharge was unchanged. Cervical bleeding upon contact, cervical ectopy (one subject at each visit had over 75% ectopy), and a semiquantitative estimate of the amount of cervical mucous were not changed after 6 months of DMPA (Table 1).
No change occurred in the number of subjects with any Lactobacillus recovered at baseline compared with the 3 and 6 month visits (Table 2). However, the proportion of women with H2O2-positive Lactobacillus decreased significantly in a linear manner during the 6 months of DMPA use (P = .005, Table 2). A significant linear decrease also occurred in high concentrations (at least 105 colony forming units/mL) of H2O2-positive Lactobacillus over 6 months of DMPA and in the total log concentration of H2O2-positive Lactobacillus, from 4.0 ± 0.6 to 2.5 ± 0.6. The decrease in H2O2-positive Lactobacillus was not related to condom use at baseline (too few condoms were used after DMPA for logistic regression analysis). Conversely, there was a slight increase in high concentrations (at least 105 colony forming units/mL) of H2O2-negative Lactobacillus (P = .09) and a significant increase in the log concentration of H2O2-negative Lactobacillus after 6 months of DMPA, from 0.5 ± 0.3 to 1.4 ± 0.5. The few other Lactobacillus isolates (that did not grow in the tetramethyphenzidine-brucella medium and could not be tested for H2O2 production) did not change in frequency or concentration with DMPA use. A significant linear decrease also occurred in Candida albicans over 6 months of DMPA use.
|
). The number of subjects with Ureaplasma urealyticum increased during the study, but the number with high concentrations (at least 105 colony forming units/mL) of U urealyticum did not change. Interpretation of the vaginal Gram stain by the Nugent criteria15 showed no change in the prevalence of either intermediate flora or bacterial vaginosis during the 6 months of DMPA use.
Three women (8%) had more than 105 bacteria/mL of a single uropathogen on urine culture at enrollment, and that number decreased slightly with DMPA use. Vaginal neutrophils (at least 1 per 400x high-powered field) on Gram stain did not change (more than ten neutrophils per high-powered field were present at baseline in three subjects and at 6 months in six subjects). Cervical neutrophils on Gram stain did not change with DMPA use (one subject had more than 30 neutrophils at 3 months). No subject had Chlamydia trachomatis or trichomoniasis, or yeast by wet-mount testing at any point in the study. The number of subjects with vaginal pH over 4.7 and the mean pH measurements were similar at enrollment and after 6 months of DMPA use (Table 2).
The mean serum E2 level decreased significantly from 99.9 ± 9.3 pg/mL at baseline to 26.6 ± 1.6 pg/mL after 6 months of DMPA use (Table 3, P < .001). There was one subject at baseline and 13 after 6 months of DMPA use who had serum E2 levels below 20 pg/mL (a level lower than early follicular E2 values).21
|
). As indicated in the figure, most subjects with high or low values at baseline tended to have values return to the mean at follow-up visits, an indication that the decrease in the mean number of layers occurred for the whole group and not a subset of subjects.
|
). The thickness of glycogen-positive vaginal epithelium decreased between baseline (0.81 ± 0.04 mm) and 6 months of DMPA use (0.66 ± 0.05, P = .005). The thickness of intensely periodic acid-Schiff–stained vaginal epithelium did not decrease with 6 months of DMPA use. Ten subjects had upper and lower vaginal biopsies taken at baseline and after 3 and 6 months of DMPA use. In a comparison of 27 specimens by linear regression for correlated data over three visits, the mean difference in epithelial thickness of the upper compared with the lower vaginal biopsy site was 0.006 ± 0.05 mm (P = .3), and the mean difference in glycogen thickness between the two sites was 0.03 ± 0.05 mm (P = .8). Thus, upper and lower vaginal epithelial thickness was virtually identical in the subjects.
The mean number of neutrophils in five fields decreased from 4.3 ± 0.5 at baseline to 2.8 ± 0.3 after 6 months of DMPA, as did the number of subjects with five or more neutrophils in all fields. The number of subjects with chronic inflammation, mitotic division, or plasma cells did not change.
The change in E2 levels from baseline to 6 months of DMPA use ranged from an increase of 11 pg/mL to a decrease of 231 pg/mL and was linear without an apparent natural break. We therefore divided E2 levels by tertials and medians and found that there was no correlation between serum E2 level at 6 months and amount of vaginal discharge, menses, intermenstrual bleeding, mean number of epithelial layers, epithelial thickness, glycogen layer thickness, neutrophils in the subepithelium, or concentration of H2O2-positive Lactobacillus. Smoking and BMI had no effect on the serum E2 levels at 6 months in that group or on any of the variables in the tables that were changed significantly with 6 months of DMPA use.
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
), findings that were documented previously.2,3,5 The percentage of subjects with very low E2 levels (under 20 pg/mL)21 increased from 3% at baseline to 34% at 6 months. Over one third of subjects were hypoestrogenic after 6 months of DMPA. A single E2 measurement was shown to be a valid measure of the estrogen milieu of DMPA users because the measurement of daily E2 levels yielded the same results as levels taken at greater intervals.22 Estradiol levels measured at the time of DMPA injection, as in our study, resulted in higher estrogen levels than when E2 levels were measured 1 week after injection,3 indicating that E2 levels reported here might be even lower for much of the 3 months between DMPA injections. The persistent low estrogen levels in these reproductive-age, previously menstruating women resulted in changes in the vaginal flora and a demonstrable, albeit surprisingly small, effect on vaginal epithelium. The subjects are to be observed for 2 years to evaluate these changes over a longer period. We expected DMPA to reduce thickness of the vaginal epithelium because in a primate model it markedly reduced the number of vaginal epithelial layers.6,7 When simian immunodeficiency virus was inoculated into the vagina, there was was significantly more simian immunodeficiency virus infection in progesterone-treated macaques than in controls,6 suggesting that thinning of the vaginal epithelium in progesterone-treated animals largely explained the increased acquisition of simian immunodeficiency virus by the macaque.6 It is reassuring that after 6 months of DMPA use, our subjects did not show marked loss of vaginal epithelial cell layers as with progesterone treatment of macaques. The reduction of E2 appeared only to reduce the mean number of cell layers in the vaginal epithelium from 28 to 26 after 6 months of DMPA. Whereas the 10% reduction found in epithelial thickness was statistically significant, that small decrease should not markedly reduce the mechanical barrier effectiveness against microorganisms traversing the vaginal epithelium.
The lack of a profound effect on vaginal epithelial cell layers with short-term DMPA was consistent with findings from other studies in humans. In 1936, Zondek and Friedman23 took biopsies from women at different points in the menstrual cycle and found no cyclic changes in number of vaginal epithelial layers among normally menstruating women compared with those with primary or secondary amenorrhea, a finding confirmed in menstruating women.24 A recent report in which the same 16 women had biopsies before and 1 and 3 months after DMPA injections also found slightly fewer epithelial cells25 of the magnitude found in the present study (Table 3).
The change in total epithelial thickness was confined primarily to the glycogen-stained portion. Although the exact amount of reduced thickness is difficult to measure because fixation of tissue in formalin causes extensive dehydration, our quantitative measurement of thickness at different time points in the same person is likely to assess accurately the relative amount of glycogen in the epithelium. In animal models, glycogen is produced in estrogen-stimulated vaginal epithelial cells,26 and oophorectomy reduces their glycogen content.27 Premenarchal and postmenopausal women have virtually no glycogen in the vaginal epithelium. Glycogen is a very hydrophilic molecule and attracts water into the cell. Intracellular glycogen correlates with water content, and a reduction of glycogen could cause some degree of cytoplasmic dehydration. Whether a reduction of water-enriched cells could reduce the integrity of the mechanical barrier of vaginal epithelium is not known.
Among menstruating women with normal E2 levels, low vaginal pH is believed to be caused by glycogen. Lactobacilli normally colonize the vaginas of menstruating women at concentrations exceeding 105 organisms per milliliter of vaginal fluid.17,18 Vaginal lactobacilli use glycogen to produce lactic acid that helps maintain low pH and dominance of lactobacilli and other acidophilic bacteria in the vagina. Our subjects had only a small change in vaginal pH and in the proportion with any Lactobacillus, despite very low E2 levels in one third of subjects. These results raised the possibilities that mature vaginal epithelium can maintain itself despite low estrogen levels for at least 6 months or that the effect of estrogen is not necessarily dose-related but perhaps analogous to the all-or-nothing effect proposed for bone physiology.28 Large individual2 variations in medroxyprogesterone acetate serum levels also occurred, so perhaps our study was too small to detect the effect.
The low E2 levels in DMPA users affected two important vaginal microorganisms in our study, with potential impact on HIV acquisition. Although the total number of subjects with any Lactobacillus did not change, the number of subjects with H2O2-positive Lactobacillus decreased significantly from 53% to 32% after 6 months of DMPA use. That reduction in H2O2-producing Lactobacillus was not likely caused by chance, not attributed to changes in new partners or intercourse frequency, and not apparently related to decreased condom use. In a separate study, a decrease in H2O2-producing Lactobacillus was not noted in OC users in whom H2O2-producing Lactobacillus was present in half of subjects at baseline and 46% of 28 subjects after 2 months of OC use.29 The production of H2O2 by Lactobacillus appears to represent an important mechanism by which it maintains its dominance over other vaginal flora.18 The H2O2 formed might inhibit or kill other vaginal flora, particularly flora that lack or have low levels of H2O2-scavenging enzymes, such as catalase.18 The combination of H2O2, peroxidase, and a halide ion greatly increases the inhibition of many vaginal microbes compared with H2O2 alone.30 The halide ions and peroxidase necessary to augment the inhibitory potency of that system are present in the vagina31 and can interact with the H2O2 generated by Lactobacillus. For example, H2O2-positive Lactobacillus at a concentration of 107 colony forming units/mL kills HIV virus in vitro in half of the samples, but HIV was totally eradicated by H2O2-positive Lactobacillus at only 106 colony forming units/mL when myeloperoxidase and chloride were added to the medium.13 Recently, compared with women with H2O2-positive Lactobacillus, a twofold increased rate of HIV acquisition was found for women without Lactobacillus or with intermediate or bacterial vaginosis flora on vaginal Gram stain.14 Thus, it is possible that DMPA increases the risk of HIV acquisition through two mechanisms that involve Lactobacillus. Hydrogen peroxide–positive Lactobacillus in the vagina might directly kill free virus in the vagina at intercourse,13 or H2O2-positive Lactobacillus might prevent bacterial vaginosis,31 which has been associated with an increased risk of HIV acquisition.9,32
Depomedroxyprogesterone acetate also decreased vaginal Candida colonization. A cross-sectional study of symptomatic yeast vulvovaginitis found that only 3.2% of women used DMPA, whereas 52% used OC pills.33 Our data confirmed a report of decreased positive Candida cultures (from 32% to 8%) after only 3 months of DMPA, and the lack of symptomatic Candida infection among women with more than 1 year of DMPA use.34 Decreased symptomatic candidiasis also might reduce the risk of acquiring HIV.9
Langerhans cells, which express CD4 receptors, are the probable target cell in the vagina and cervix for heterosexual transmission of HIV.35 It is not known whether the vagina is a more important portal of HIV entry in women than the cervix. We did not determine the effect of DMPA on the number of Langerhans cells in the vagina, but DMPA did not appear to influence the number of cervical neutrophils or cervical friability, and appeared to reduce the number of vaginal subepithelial neutrophils, a finding not previously reported.
Depomedroxyprogesterone acetate use has been associated with the acquisition of HIV in two prospective studies of high-risk sex workers.8,9 In the largest cohort, DMPA use remained associated with HIV acquisition in a multivariate model after controlling for many other risk factors associated with HIV.9 However, DMPA use has not been consistently associated with HIV infection in cross-sectional studies.11,12 Many factors might confound the relationship between HIV and DMPA. Human immunodeficiency virus seroconversion or seroprevalence has been linked to cervical infection, inflammation, and ectopy9; genital ulcers,9 Candida,9 and bacterial vaginosis9,14,32; and to inconsistent condom use.36,37 Most reports have not provided extensive contraceptive data or information on duration of DMPA use. Even a multivariate adjustment of multiple factors9 might not adequately control for confounding variables. For example, amenorrhea among sex workers using DMPA might allow for more customers, the reliable contraceptive feature might reduce consistent condom use, and the prevention of pregnancy could continue the length of sex work, thereby increasing exposure to HIV. Users of DMPA also might be more likely to remain in studies to receive DMPA injections and hence be more likely to have repeat testing for HIV.
Our data show significant reduction of serum E2 and changes in the vagina after 6 months of DMPA use. It is plausible that some of the changes, such as the reduction of H2O2-positive Lactobacillus or the reduction in vaginal glycogen, could enhance the acquisition of HIV. Conversely, DMPA use in humans does not produce the marked epithelial thinning observed in primates. The DMPA-associated reduction of neutrophils in the subepithelium and of Candida might provide some protection against the acquisition of HIV. Depomedroxyprogesterone acetate produces complex effects and further studies are warranted on H2O2-positive Lactobacillus and the risk of HIV infection in women.
|
Footnotes |
|---|
Received December 30, 1999. Received in revised form March 14, 2000. Accepted April 7, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Mishell DR. Pharmacokinetics of depot medroxyprogesterone acetate contraception. J Reprod Med 1996;41(Suppl):381–90.[Medline]
3. Jeppsson S, Johansson EDB, Ljungberg O, Sjoberg NO. Endometrial histology and circulating levels of medroxyprogesterone acetate, estradiol, FSH and LH in women with MPA induced amenorrhea compared with women with secondary amenorrhea. Acta Obstet Gynecol Scand 1977;56:43–8.[Medline]
4. Cundy T, Cornish J, Roberts H, Elder H, Reid IR. Spinal bone density in women using depot medroxyprogesterone contraception. Obstet Gynecol 1998;92:569–73.[Abstract]
5. Mishell DR, El-Habashy MA, Good RG, Moyer DL. Contraception with an injectable progestin; A study of its use in postpartum women. Am J Obstet Gynecol 1968;101:1046–53.[Medline]
6. Marx PA, Spira AI, Gettie A, Dailey PJ, Veazey RS, Lackner AA, et al. Progesterone implants enhance simian immunodeficiency virus vaginal transmission and early viral load. Nature Med 1996;2: 1084–9.[Medline]
7. Hild-Petito S, Veasey RS, Larner JM, Reel JR, Blye RP. Effects of two progestin only contraceptives, Depo-Provera, and Norplant II, on the vaginal epithelium of rhesus monkeys. AIDS Res Hum Retroviruses 1998;14:S125–30.
8. Ungchusak K, Rehle T, Thammapornpilap P, Spieglman D, Brinkmann U, Siraprapasiri T. Determinants of HIV infection among female commercial sex workers in northeastern Thailand: Results from a longitudinal study. J Acquir Immune Defic Syndr Hum Retrovirol 1996;12:500–7.[Medline]
9. Martin HL, Nyange PM, Richardson BA, Lavreys L, Mandaliya K, Jackson DJ, et al. Hormonal contraception, sexually transmitted diseases, and risk of heterosexual transmission of human immunodeficiency virus type 1. J Infect Dis 1998;178:1053–9.[Medline]
10. Mostad SB, Overbaugh J, DeVange DM, Welch MJ, Chohan B, Mandaliya K, et al. Hormonal contraception, vitamin A deficiency, and other risk factors for shedding of HIV-1 infected cells from the cervix and vagina. Lancet 1997;350:922–7.[Medline]
11. Mati JKG, Hunter DJ, Maggwa BN, Tukei PM. Contraceptive use and the risk of HIV infection in Nairobi, Kenya. Int J Gynaecol Obstet 1995;48:61–7.[Medline]
12. Kapiga SH, Lyamuya EF, Lwihula GK, Hunter DJ. The incidence of HIV infection among women using family planning methods in Dar es Sallam, Tanzania. AIDS 1998;12:75–84.[Medline]
13. Klebanoff SJ, Coombs RW. Viricidal effect of Lactobacillus acidophilus on human immunodeficiency virus type 1: Possible role in heterosexual transmission. J Exp Med 1991;174:282–92.
14. Martin HL, Richardson BA, Nyange PM, Lavreys L, Hillier SL, Chohan B, et al. Vaginal lactobacilli, microbial flora, and risk of human immunodeficiency virus type 1 and sexually transmitted disease acquisition. J Infect Dis 1999;180:1863–8.[Medline]
15. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of Gram stain interpretation. J Clin Microbiol 1991;29:297–301.
16. Jones HW, Wentz AC, Burnett LS. Novak’s textbook of gynecology. 11th ed. Baltimore: Williams & Wilkins, 1988:85–6.
17. Hillier SL, Krohn MA, Rabe LK, Klebanoff SJ, Eschenbach DA. Normal vaginal flora, H2O2-producing lactobacilli and bacterial vaginosis in pregnant women. Clin Infect Dis 1993;16(Suppl 4): S273–81.
18. Eschenbach DA, Davick PR, Williams BL, Klebanoff SJ, Young-Smith K, Critchlow CM, et al. Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis. J Clin Microbiol 1989;27:251–6.
19. Stamm WE, Tam MR, Koester M, Cles L. Detection of Chlamydia trachomatis inclusions in McCoy cell cultures with fluorescein-conjugated monoclonal antibodies. J Clin Microbiol 1983;17:660–8.
20. Disser PJ, Liang KY, Zeger SL. Analysis of longitudinal data. Oxford, Great Britain: Oxford University Press, 1994:146–68.
21. Buster JE. Estrogen kinetics for clinicians. In: Sciarra JJ, ed. Gynecology and obstetrics. Philadelphia: Lippincott, Williams & Wilkins, 1999.
22. Mishell DR, Dharma KM, Thorneycroft IH, Nadamura RM. Estrogenic activity in women receiving an injectable progestogen for contraception. Am J Obstet Gynecol 1972;113:372–6.[Medline]
23. Zondek B, Friedmann M. Are there cyclic changes in the human vaginal mucosa? JAMA 1936;106:1051–4.
24. Traut HF, Block PW, Kuder A. Cyclic changes in the human vaginal mucosa. Surg Obstet Gynecol 1936;63:7–15.
25. Mauck CK, Callahan MM, Baker J, Arbogast K, Veazey R, Stock R, et al. The effect of one injection of Depo-Provera on the human vaginal epithelium and cervical ectopy. Contraception 1999;60:15–24.[Medline]
26. van Dyke HB, Chen G. Observations on the biochemistry of the genital tract of the female macaque particularly during the menstrual cycle. Am J Anat 1936;58:473–92.
27. Carrington LJ, Bailey CJ. Effects of natural and synthetic estrogens and progestins on glycogen deposition in female mice. Horm Res 1985;21:199–203.[Medline]
28. Raisz LG, Prestwood KM. Estrogen and the risk of fracture-new data, new questions. N Engl J Med 1998;339:767–8.
29. Patton DL, Thwin SS, Agnew K, Eschenbach DA. Influence of the normal menstrual cycle and combination oral contraceptives on vaginal epithelium and flora. Abstract. Infect Dis Obstet Gynecol 1997;4:356.
30. Klebanoff SJ, Hillier SL, Eschenbach DA, Waltersdorph AM. Control of the microbial flora of the vagina by H2O2-generating lactobacilli. J Infect Dis 1991;164:94–100.[Medline]
31. Hawes SE, Hillier SL, Benedetti J, Stevens CE, Koutsky LA, Wølner-Hanssen P, et al. Hydrogen peroxide producing lactobacilli and acquisition of vaginal infection. J Infect Dis 1996;174:1058–63.[Medline]
32. Taha TE, Hoover DR, Dallabetta GA, Kumwenda NI, Mtimavalye LAR, Yang LP, et al. Bacterial vaginosis and disturbances of vaginal flora: Association with increased acquisition of HIV. AIDS 1998;12:1699–706.[Medline]
33. Nelson AL. The impact of contraceptive methods on the onset of symptomatic vulvovaginal candidiasis within the menstrual cycle. Am J Obstet Gynecol 1997;176:1376–80.[Medline]
34. Toppozada M, Onsy FA, Fares E, Amir S, Shaala S. The protective influence of progestogen only contraception against vaginal moniliasis. Contraception 1979;20:99–103.[Medline]
35. Soto-Ramirez LE, Renjifo B, McLane MF, Marlink R, O’Hara C, Sutthent R, et al. HIV-1 Langerhans’ cell tropism associated with heterosexual transmission of HIV. Science 1996;271:1291–3.[Abstract]
36. Plourde PJ, Pepin J, Agoki E, Ronald AR, Ombette J, Tyndall M, et al. Human immunodeficiency virus type I seroconversion in women with genital ulcers. J Infect Dis 1994;170:313–7.[Medline]
37. Royce RH, Sena A, Cates W Jr, Cohen MS. Sexual transmission of HIV. N Engl J Med 1997;336:1072–8.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
