Mauck, Christine K. MD1; Freziers, Ron G. MSPH2; Walsh, Terri L. MPH2; Peacock, Karen MA2; Schwartz, Jill L. MD1; Callahan, Marianne M. MEd1
Sodium cellulose sulfate is a noncytotoxic antifertility agent that has been studied as a vaginal gel.1–7 A gel containing cellulose sulfate has been shown to stimulate acrosomal loss, inhibit hyaluronidase, and impede sperm penetration into cervical mucus in vitro and is contraceptive in rabbits. Cellulose sulfate is not toxic to sperm but acts by inhibiting sperm function. Cellulose sulfate demonstrates in vitro activity against various sexually transmitted pathogens, organisms associated with bacterial vaginosis, and other organisms without inhibiting the proliferation of lactobacillus strains.
The primary objectives of this clinical trial were to estimate the cumulative probability of pregnancy for 6 months and six cycles of typical cellulose sulfate gel use and six cycles of consistent use, and discontinuation through 6 months of gel use. In addition, the cumulative probability of pregnancy for six cycles of correct and consistent “perfect” use was calculated. Secondary objectives of the trial were to estimate the acceptability of cellulose sulfate gel, the consistency of use among the study participants, and the frequency of abnormal cytology, candidiasis, urinary tract infections (UTIs), bacterial vaginosis, and genital irritation during cellulose sulfate gel use.
MATERIALS AND METHODS
This was a phase II noncomparative contraceptive effectiveness trial that enrolled 200 couples to use the study product as their primary means of contraception for 6 months and six cycles. The study was conducted between January 2004 and September 2005 by the California Family Health Council, in Los Angeles, California, after approval by the Health Council Institutional Review Board and in accordance with the Declaration of Helsinki. It was registered on www.clinicaltrials.gov as #NCT 002 55 294.
The decision to make the study noncomparative was based on the fact that the only marketed spermicide that could be used as an active control was nonoxynol-9. Nonoxynol-9 has been associated with an increased risk of human immunodeficiency virus (HIV) compared with placebo when used frequently by women at risk for HIV.8 In a phase I comparative study, nonoxynol-9 appeared to be more irritating that cellulose sulfate,9 and although the participants in this study were monogamous and at low risk for HIV and STDs, we anticipated some reluctance on the part of both the institutional review board and potential volunteers to permit randomization to an nonoxynol-9 arm. The study was, therefore, designed to be noncomparative and of an adequate size to give a reasonably precise estimate of the contraceptive effectiveness of cellulose sulfate gel, given the resources available. If the 6-month typical-use pregnancy probability was 18% or less and the probability of discontinuing before the end of month 6, for reasons other than pregnancy, was 32% or less, the half-width of the confidence interval for the 6-month typical-use pregnancy probability was expected to be no more than ±6%.
Couples were recruited from the family planning clients of the site and through advertisements in local newspapers, local Internet sites, and radio. Volunteers eligible for the study were couples in a stable, mutually monogamous relationship for at least 3 months, willing to have vaginal intercourse at least four times per month, not actively seeking pregnancy, and not known to be HIV-infected or allergic to study products. Females were eligible if they were between the ages of 18 and 40 years, had a negative pregnancy test at pre-enrollment and enrollment, had regular menstrual cycles (between 21 and 35 days in length) for the last two cycles, had no history of infertility or sterilization, and did not test positive for gonorrhea, chlamydia, or Trichomonas. Men were eligible if they were at least 18 years old, had no history of infertility or vasectomy, and had not had sex with a man in the last six months. Volunteers received compensation for time and travel depending on the length of their study participation. Participants who experienced any of the following during the study were discontinued: pregnancy, sexual inactivity for two cycles or longer, no menses for 90 days or longer, and symptomatic bacterial vaginosis, candidiasis, or UTI at enrollment that did not resolve within 14 days. Participants who stopped considering cellulose sulfate to be their primary contraceptive were also discontinued.
Participants were scheduled for four visits: pre-enrollment, enrollment, after one menstrual cycle, and after 6 months or six menstrual cycles, whichever came later. At the pre-enrollment visit, the study was explained, consent was obtained from both partners, and the woman underwent a urine pregnancy test. Medical and sexual histories were obtained from both partners, and instructions on gel use and diary completion were given. At the enrollment visit, vital signs were measured, and the woman underwent a pelvic examination with specimens taken for Pap test, wet prep, vaginal pH, and gonorrhea and chlamydia tests. When it was confirmed that the female participant met the entry criteria, she and her partner were enrolled and study drug, diaries, and a home pregnancy test were dispensed. The female participant was instructed to call the study staff at the onset of her first menses and to take a home pregnancy test 7–10 days later and call the study staff with the result.
The participant was instructed to call the site at the beginning of each menses, at which time the diaries were reviewed and recorded. A follow-up visit was scheduled after the beginning of the second menstrual period after enrollment. Diaries were reviewed again, completed diaries were collected, and an acceptability questionnaire was administered. The participant’s weight and blood pressure were measured and a pelvic examination and urine pregnancy test were done. The final visit was scheduled after the first menses, following 6 months of product use and six complete menstrual cycles. The procedures for the final visit were the same as the interim follow-up visit, except that a Pap test was done and unused study supplies were collected.
The 6% cellulose sulfate gel was manufactured and packaged under Good Manufacturing Practice conditions in plain, white, single-use applicators by Patheon Inc. (Mississauga, Ontario, Canada). Six percent cellulose sulfate gel contains 60 mg of the active ingredient, sodium cellulose sulfate, per gram of gel. Each 3.5-mL application of 6% gel, therefore, contained 225 mg of the active ingredient. The vehicle gel contained 5% glycerine and 5% sorbitol as humectants, up to 0.25% carbomer as a thickener, 1% benzyl alcohol as a preservative, and water. The pH of 6% cellulose sulfate gel is 7 (neutral), with low buffering capacity.
Participants were instructed to insert one applicator of the study gel into the vagina before each act of intercourse. If they did not have intercourse within 1 hour of inserting the gel, they were to insert another applicator of gel before having intercourse. Participant diaries were used to collect information on acts of intercourse, uses of cellulose sulfate gel, uses of other methods of birth control, incorrect use of cellulose sulfate gel, unprotected intercourse, and transient adverse events related to cellulose sulfate gel use. Cycles were classified into correct and consistent, consistent, and inconsistent use cycles.
The cumulative probability of pregnancy for 6 months of typical use was estimated by conventional life-table analysis. The life-table analysis used the following: date of method acceptance (onset date of menses following study enrollment), date of last use of study method for women withdrawing early and end of last cycle for women completing the study, and date of estimated conception if pregnant (estimated by California Family Health Council research medical director using ultrasound results). Data from participants who exited the study early were censored after the date of their last use of cellulose sulfate gel. Cycles with no acts of sexual intercourse and cycles in which emergency contraception was used were excluded from the six-cycle analysis. Cycles with unprotected intercourse and/or nonstudy method use alone were excluded from the consistent-use cycle analysis.
The cumulative probability of pregnancy for six cycles of typical use and consistent use was estimated by a life-table analysis program devised to handle cycle data. The typical-use pregnancy rate was the cumulative risk of pregnancy among participants who used cellulose sulfate gel as their primary method of birth control, even if they used cellulose sulfate gel incorrectly or inconsistently. The consistent-use pregnancy rate was the cumulative risk of pregnancy when cellulose sulfate gel was used for every act of intercourse, even if the cellulose sulfate gel was used incorrectly. The perfect use pregnancy rate was the cumulative risk of pregnancy when cellulose sulfate gel was used correctly for every act of intercourse. Six-month and six-cycle cumulative probabilities, 95% confidence intervals, and standard errors were calculated.
Information on adverse events was collected in two ways. Participants recorded on their study diaries any genital tract problems that were mild in severity, not serious, and transient, that occurred in conjunction with gel use and that did not require treatment or interruption of study gel use. In addition, participants recorded on their Adverse Health Event/Medication Log all adverse events experienced by the female participant that did not meet the criteria for recording on the diary and all adverse events experienced by the male partner that were serious, severe, possibly related to the female’s use of the gel, or involving the genital tract. The California Family Health Council medical director classified each adverse event as definitely related, probably related, possibly related, or unrelated to gel use. Relatedness was based on the biological plausibility of a causal relationship between gel use and the adverse event and the strength of the temporal relationship between gel use and the onset of the event. Data on all gel-related adverse events of the genital tract were tabulated and summarized by gender with frequencies and percentages. Adverse events recorded on the diary were summed on the cycle interview and tabulated separately, by gender, from other adverse events. It was not possible to combine figures for adverse events reported on Adverse Health Event/Medication Logs and genital tract problems reported on diaries because the diary collected the number of days on which the various types of discomfort occurred, while the adverse event log collected single instances of discomfort or patterns of intermittent discomfort.
Two thousand twenty volunteers were screened by telephone, and 200 couples (400 participants) were enrolled. About two thirds of the callers declined to participate after learning about study requirements while the remaining third did not meet one or more of the study eligibility criteria. Eighty-two (41%) of those enrolled completed the study without becoming pregnant; two couples (1%) were lost to follow-up, and 18 (9%) of those enrolled discontinued without ever using the study gel and were not included in the efficacy analysis. The remaining 98 (49%) discontinued after using study gel at least once: 18 due to pregnancy, 14 for gel-related reasons, and 66 because they no longer met entry criteria, no longer wanted to participate, or for personal reasons. The most common specific reason for discontinuing was separation from study partner (24 discontinuations) (Fig. 1). The life-table overall continuation rate, excluding pregnancy, was 52.1% (95% confidence interval [CI] 44.5–59.6%).
Table 1 shows sociodemographic characteristics of the participants. Study gel use was recorded 8,067 times in the study, on average, 9.5 times per menstrual cycle, over a total of 813 full and partial cycles. Of 758 complete cycles, 343 (45.3%) were cycles of correct and consistent (“perfect”) gel use. The remaining cycles involved either incorrect use of cellulose sulfate gel, use of an additional contraceptive method, use of an alternative contraceptive method, and/or unprotected intercourse. Unprotected intercourse was reported in 125 cycles (16.5% of total cycles). There was an average of 11.5 coital acts per cycle. Over three quarters of coital acts involved the study gel used alone as instructed (“perfect use”). Four percent of acts involved the study gel used alone but incorrectly, 5% involved use of an additional method, 10% involved use of an alternative method, and 4% involved unprotected intercourse.
Life-table pregnancy probabilities were as follows: cumulative probability of pregnancy for 6 months and six cycles of typical use of the cellulose sulfate gel: 13.4% (95% CI 7.5–19.4%) and 13.9% (95% CI 7.7–20.2%), respectively. The cumulative probability of pregnancy for six cycles of consistent use was 3.0% (95% CI 0.0–7.2%) and for six cycles of correct and consistent (“perfect”) use was 3.9% (95% CI 0.0–9.2%). The Pearl index was 31.7 for typical use, 7.0 for consistent use, and 8.6 for correct and consistent (“perfect” use). Due to the small number of pregnancies in the study, no adjustments for covariates were done. Twenty-seven pregnancies were conceived among the study participants: 13 (48%) resulted in live births, 11 (41%) in induced abortions, and three (11%) in spontaneous abortions.
Of the 403 nondiary adverse events recorded over the course of the study, 96 adverse events reported by 57 (28.5%) of the female participants were at least possibly related to the study gel (Table 2).
Most (n=88 or 92%) of gel-related adverse events were only possibly related to gel and most (n=73 or 78%) were mild. Sixty-five (69%) were both mild and only possibly related. There were no serious adverse events related to the study gel.
Thirteen vaginal yeast infections were diagnosed among 12 women: seven were possibly gel-related and six unrelated. Eleven cases of bacterial vaginosis were diagnosed among 10 women: five possibly related and six unrelated. Twenty-three UTIs were diagnosed in 22 (9%) females and one male; 20 met the protocol definition for UTI or were diagnosed by the women’s private physician, two were self-diagnosed, and one woman received a prescription over the phone from her private physician. Seventeen UTIs among 14 (7%) women were considered possibly related to study gel.
Sixteen adverse events designated as “gel reaction” were recorded among 14 (7%) female participants and one male participant. All but one of the gel reactions were mild, four required temporary cessation of gel use, and two were self-treated with over-the-counter medication.
About one third of female participants reported at least one genital tract problem in their diaries, as did 18% of men. Female participants reported a problem in 13.5% of cycles, on average, on one day out of every 3.3 cycles. Male participants reported having problems in 6.0% of their partners’ cycles, on average, on one day out of every 10 cycles. The most common problems were irritation and itching, with vaginal discharge, painful urination, and “other discomfort” being reported less frequently. A total of 43.5% of female participants reported either discomfort on a cycle interview (diary) or a gel-related adverse event.
There was no clinically significant increase or change in physical examination findings seen in scheduled visits over the course of the study. The most common abnormal examination finding was vaginal discharge, seen in about 13% of women. Five Pap tests showing atypical cells of undetermined significance and one showing low-grade squamous intraepithelial lesions were observed during the study.
More than 80% of women and their partners reported that they liked the gel “very much” or “OK” at both the follow-up visit and study exit. More than half would purchase the gel to prevent STDs, and more than three quarters would buy it to prevent pregnancy.
Cellulose sulfate gel has been studied in 13 clinical trials: six safety studies in women,9–14 two safety studies in men,15,16 an magnetic resonance imaging study,17 two HIV prevention studies (van Damme L. Phase III trial of 6% cellulose sulfate [CS] gel for the prevention of HIV transmission. Presented at the International AIDS Society conference in Sydney, Australia, July 24, 2007; Cates W. Effectiveness of Cellulose Sulfate Gel For Prevention of HIV: Results of the Phase III Trial in Nigeria. Presented at the International AIDS Society conference in Sydney, Australia, July 24, 2007), and two contraceptive studies including this one. Safety studies in women and men have varied in terms of control group (inactive sexual lubricant such as K-Y Jelly [Johnson&Johnson, New Brunswick, NJ] or active spermicidal gel such as Conceptrol [Johnson&Johnson, Maniti, Puerto Rico]), frequency of use (once, twice, or four times daily), duration of use (from 6 days to 6 months), sexual activity (abstinent or active), and HIV status of participants (infected or uninfected). End points generally have included signs and symptoms of genital irritation, colposcopic findings, changes in vaginal microflora, and in some cases, vaginal inflammatory markers and responses to acceptability questionnaires. The end points studied before the HIV prevention trials indicated that cellulose sulfate gel was as safe and acceptable as marketed spermicides and sexual lubricants as well as the “universal placebo.” The latter is a placebo designed specifically for use in microbicide studies, being similar in appearance and viscosity to K-Y Jelly and showing negligible anti-HIV activity in vitro and in a nonhuman primate model of vaginal transmission of simian immunodeficiency virus, and inducing no signs of significant vaginal irritation or protection against HSV-2 in animal models.18
Two HIV prevention trials done among women at high risk for HIV were stopped prematurely due to the potential of harm observed in a scheduled interim analysis of one of them. In that trial, there was no difference in HIV seroconversions among cellulose sulfate gel users compared with users of the universal placebo in the final intent-to-treat analysis, but there were significantly more seroconversions among cellulose sulfate gel users in the preliminary per protocol analysis (censoring women at their first documented product interruption); in the other trial, there was no difference (van Damme, 2007; Cates, 2007). The reason for this finding is presently unknown. Data are undergoing additional analyses, and a new phase I safety study has been started to look for novel safety signals among users of cellulose sulfate, including cellular markers of inflammation in vaginal biopsies and cervicovaginal lavage cell pellets, and antiviral activity of cervicovaginal lavage fluid after cellulose sulfate gel exposure.
Two contraceptive effectiveness studies were carried out before the HIV studies were stopped. The first, a single-cycle, placebo-controlled so-called “True Efficacy” study, was stopped prematurely after enrollment of 67 out of the intended 150 couples, due to recruitment difficulties. Two pregnancies were conceived in the study, both in the placebo arm. The second study is the subject of this manuscript.
Cellulose sulfate gel appears to be effective, with only short-term adverse effects when used as a vaginal contraceptive gel among couples at low risk for STDs. The efficacy results of this noncomparative study are best viewed in light of the most recently completed study on the only vaginal contraceptive gel marketed in the United States, namely nonoxynol-9.19 That study was conducted by Family Health International in collaboration with the U.S. Food and Drug Administration and the National Institutes of Health using modern study methodology in a U.S. population. The study had a total of five arms, three of which involved doses of 52.5-, 100-, and 150-mg nonoxynol-9 gel. Each nonoxynol-9 gel arm enrolled between 295 and 300 women. For reasons already described, we felt it was inadvisable to include an nonoxynol-9 arm in our study. However, although not a direct comparison, the results of the cellulose sulfate gel and nonoxynol-9 studies can be evaluated in terms of enrollment, participant retention, and efficacy.
In our study, only one participant was enrolled for every 10 who contacted the study site expressing interest. Although loss to follow-up was much lower than expected in our study (1%), more women than expected discontinued early without becoming pregnant (49%). Most of these women discontinued for reasons unrelated to the study gel rather than because of gel intolerance or unacceptability; these discontinuations, therefore, do not reflect poorly on the study gel.
Similar difficulties in recruiting and retaining women were seen in the nonoxynol-9 study, in which total recruitment for all five arms together was stopped at 1,536 of the intended 1,800. In that study, only 15–18% of the women in the three gel arms discontinued early, but 20–25% were lost to follow-up. The combined discontinuation and loss-to-follow-up rate for that study of 35–42% is somewhat lower than ours (50%). However, having a low loss-to-follow-up rate and a relatively high discontinuation rate is preferable because it avoids the problem of having missing data about a large number of women, at least some of whom may have become pregnant while using the study product. These high rates of discontinuation and loss to follow-up probably reflect the short-term nature of vaginal products used for contraception; many women view them as temporary or even “weekend” methods, making it difficult to find study volunteers who will continue exclusive use for 6 months.
Cellulose sulfate gel performed comparably to nonoxynol-9 as a contraceptive. Figure 2 presents the pregnancy probabilities and 95% confidence intervals for the three gel arms of the nonoxynol-9 study and cellulose sulfate gel. The typical use pregnancy probability for cellulose sulfate gel is almost identical to that of the 150-mg gel in the nonoxynol-9 trial. Both cellulose sulfate gel and the 150-mg nonoxynol-9 gel had upper bounds of the 95% CI below 20%. The perfect-use pregnancy probability for cellulose sulfate gel is lower than all the nonoxynol-9 gel doses.
Cellulose sulfate gel had a reasonably good safety profile in this study, with no serious gel-related adverse events being reported. The incidence of abnormal cytology, candidiasis, UTI, and bacterial vaginosis was well within what may be expected in a sexually active population. Cellulose sulfate gel also appeared to be acceptable, with over three quarters of women and men reporting that they would buy it for contraception if it were available.
There is a need for a vaginal contraceptive to be available on the U.S. market in addition to or instead of nonoxynol-9. This study suggests that cellulose sulfate gel may be an effective vaginal contraceptive product. However, the results to date of the HIV prevention trials indicate that cellulose sulfate gel might not be appropriate for use by couples at risk of HIV acquisition.
1. Scordi-Bello IA, Mosoian A, He C, Chen Y, Cheng Y, Jarvis GA, et al. Candidate sulfonated and sulfated topical microbicides: comparison of anti-human immunodeficiency virus activities and mechanisms of action. Antimicrob Agents Chemother 2005;49:3607–15.
2. Anderson RA, Feathergill K, Diao XH, Chany C. 2nd, Rencher WF, Zaneveld LJ, et al. Contraception by Ushercell (cellulose sulfate) in formulation: duration of effect and dose effectiveness. Contraception 2004;70:415–22.
3. Cheshenko N, Keller MJ, MasCasullo V, Jarvis GA, Cheng H, John M, et al. Candidate topical microbicides bind herpes simplex virus glycoprotein B and prevent viral entry and cell-to-cell spread. Antimicrob Agents Chemother 2004;48:2025–36.
4. Neurath AR, Strick N, Li YY. Anti-HIV-1 activity of anionic polymers: a comparative study of candidate microbicides. BMC Infect Dis 2002;2:27. Epub 2002 Nov 21.
5. Simoes JA, Citron DM, Aroutcheva A, Anderson RA Jr, Chany CJ 2nd, Waller DP, et al. Two novel vaginal microbicides (polystyrene sulfonate and cellulose sulfate) inhibit Gardnerella vaginalis and anaerobes commonly associated with bacterial vaginosis. Antimicrob Agents Chemother 2002;46:2692–5.
6. Anderson RA, Feathergill KA, Diao XH, Cooper MD, Kirkpatrick R, Herold BC, et al. Preclinical evaluation of sodium cellulose sulfate (Ushercell) as a contraceptive antimicrobial agent. J Androl 2002;23:426–38.
7. Christensen ND, Reed CA, Culp TD, Hermonat PL, Howett MK, Anderson RA, et al. Papillomavirus microbicidal activities of high-molecular-weight cellulose sulfate, dextran sulfate, and polystyrene sulfonate. Antimicrob Agents Chemother 2001;45:3427–32.
8. Van Damme L, Ramjee G, Alary M, Vuylsteke B, Chandeying V, Rees H, et al.; COL-1492 Study Group. Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: a randomised controlled trial [published erratum appears in: Lancet 2002;360:1892]. Lancet 2002;360:971–7.
9. Mauck C, Weiner DH, Ballagh S, Creinin M, Archer DF, Schwartz J, et al. Single and multiple exposure tolerance study of cellulose sulfate gel: a phase I safety and colposcopy study. Contraception 2001;64:383–91.
10. Schwartz JL, Mauck C, Lai JJ, Creinin MD, Brache V, Ballagh SA, et al. Fourteen-day safety and acceptability study of 6% cellulose sulfate gel: a randomized double-blind Phase I safety study. Contraception 2006;74:133–40. Epub 2006 May 2.
11. Malonza IM, Mirembe F, Nakabiito C, Odusoga LO, Osinupebi OA, Hazari K, et al. Expanded Phase I safety and acceptability study of 6% cellulose sulfate vaginal gel. AIDS 2005;19 (18):2157–63.
12. Doh AS, Ngoh N, Roddy R, Lai JJ, Linton K, Mauck C. Safety and acceptability of 6% cellulose sulfate vaginal gel applied four times per day for 14 days. Contraception 2007;76:245–9. Epub 2007 Jul 25.
13. El-Sadr W, Mayer K, Maslankowski L, Hoesley C, Justman J, Gai F, et al. Safety and acceptability of Cellulose sulfate as a vaginal microbicide in HIV-infected women. AIDS 2006;20:1109–16.
14. van der Straten A, Napierala S, Cheng H, Mauck C, Depineres T, Dhlakama P, et al. A randomized controlled safety trial of the diaphragm and cellulose sulfate microbicide gel in sexually active women in Zimbabwe. Contraception 2007;76:389–99.
15. Mauck C, Frezieres R, Walsh T, Robergeau K, Callahan M. Cellulose sulfate: tolerance and acceptability of penile application. Contraception 2001;64:377–81.
16. Jespers V, Buvé A, Van Damme L. Safety trial of vaginal microbicide cellulose sulfate gel in HIV positive men. Sex Transm Dis 2007;34:519–22.
17. Barnhart KT, Pretorius ES, Shaunik A, Timbers K, Nasution M, Mauck C. Vaginal distribution of two volumes of the novel microbicide gel cellulose sulfate (2.5 and 3.5 mL). Contraception 2005;72:65–70.
18. Schwartz JL, Ballagh SA, Kwok C, Mauck CK, Weiner DH, Rencher WF, et al. Fourteen-day safety and acceptability study of the universal placebo gel. Contraception 2007;75:136–41. Epub 2006 Oct 25.
19. Raymond EG, Chen PL, Luoto J; Spermicide Trial Group. Contraceptive effectiveness and safety of five nonoxynol-9 spermicides: a randomized trial. Obstet Gynecol 2004;103: 430–9.
Figure. No caption available.
This article has been cited 8 time(s).
ContraceptionContraception technology: past, present and futureContraception
ContraceptionEffectiveness of an optimized benzalkonium chloride gel as vaginal contraceptive: a randomized controlled trial among Chinese womenContraception
ContraceptionEstimating contraceptive efficacy: the case of spermicidesContraception
European Journal of Contraception and Reproductive Health CareState-of-the-art of non-hormonal methods of contraception: II. Chemical barrier contraceptivesEuropean Journal of Contraception and Reproductive Health Care
ContraceptionPublic-private partnerships advance contraceptive research and developmentContraception
Plos OneEffectiveness of Cellulose Sulfate Vaginal Gel for the Prevention of HIV Infection: Results of a Phase III Trial in NigeriaPlos One
Clinical Pharmacology & TherapeuticsMethods for Female Contraception: A Model for Innovation in Drug Delivery SystemsClinical Pharmacology & Therapeutics
New England Journal of Medicine
Lack of effectiveness of cellulose sulfate gel for the prevention of vaginal HIV transmission
New England Journal of Medicine, 359(5):
© 2008 The American College of Obstetricians and Gynecologists