[Skip to Content]
[Skip to Content Landing]
July 22, 1992

Efficacy of Nonoxynol 9 Contraceptive Sponge Use in Preventing Heterosexual Acquisition of HIV in Nairobi Prostitutes

Author Affiliations

From the Departments of Epidemiology (Dr Kreiss), Medicine (Dr Holmes), and Biostatistics (Ms Roberts and Dr Fleming), University of Washington, Seattle; the Departments of Community Health (Ms Ngugi) and Medical Microbiology (Drs Ndinya-Achola and Anzala and Mss Ruminjo and Sajabi), University of Nairobi, Kenya; the Department of Medical Microbiology, University of Manitoba, Winnipeg (Drs Holton and Plummer); and the Center for Microbiology Research, Kenya Medical Research Institute, Nairobi (Dr Waiyaki and Ms Kimata).

JAMA. 1992;268(4):477-482. doi:10.1001/jama.1992.03490040053025

Objective.  —To determine the efficacy of the nonoxynol 9 contraceptive sponge in preventing sexual acquisition of the human immunodeficiency virus (HIV).

Design.  —Prospective, randomized placebo-controlled trial.

Setting.  —Research clinic for prostitutes in Nairobi, Kenya.

Patients and Interventions.  —One hundred thirty-eight HIV-seronegative women were enrolled, of whom 74 were assigned to nonoxynol 9 sponge use and 64 to placebo use. These two groups did not significantly differ with respect to demographic characteristics, sexual practices, or prevalence of genital infections at enrollment, except for a lower number of sex partners per week and a higher initial prevalence of genital ulcers among women assigned to nonoxynol 9 sponge use. Among the 116 women who returned for follow-up, the mean durations of follow-up were 14 and 17 months for the two groups, respectively.

Main Outcome Measure.  —HIV seroconversion.

Results.  —Nonoxynol 9 sponge use was associated with an increased frequency of genital ulcers (relative risk [RR], 3.3; P<.0001) and vulvitis (RR, 3.3; P<.0001) and a reduced risk of gonococcal cervicitis (RR, 0.4; P<.0001). Twenty-seven (45%) of 60 women in the nonoxynol 9 sponge group and 20 (36%) of 56 women in the placebo group developed HIV antibodies. The hazard ratio for the association between nonoxynol 9 sponge use and HIV seroconversion was 1.7 (95% confidence interval [CI], 0.9 to 3.0). Using multivariate analysis to control for the presence of genital ulcers at enrollment, the adjusted hazard ratio for the association between nonoxynol 9 sponge use and seroconversion was 1.6 (95% CI, 0.8 to 2.8).

Conclusions.  —Genital ulcers and vulvitis occurred with increased frequency in nonoxynol 9 sponge users. We were unable to demonstrate that nonoxynol 9 sponge use was effective in reducing the risk of HIV infection among highly exposed women.(JAMA. 1992;268:477-482)