Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

442 Abstracts ThPeC7414-ThPeC7417

XIV International AIDS Conference

HIV epidemics in international settings requires more effective preventive interventions.
Methods: Since Nov 1999, we enrolled women 18-35 years from Family Planning
and STD clinics and community settings in Thailand (TH), Uganda (UG) and Zimbabwe (ZIM). HIV non-infected women, provided informed consent, underwent a
pelvic examination and STI screening, completed a baseline interview, and returned for follow-up visits at 3-month intervals up to 24 months with repeat HIV
and STI testing. Data from 7705 screened and 4786 women enrolled through 15
August 2001 are presented.
Results: HIV prevalence at screening was 2.0% in TH, 18.6% in UG and 40.7% in
ZIM. HIV incidence varied among 4133 women followed for 3574.5 women years
(WY): 0.2/100 WY in TH, 1.2/100 WY in UG, and 3.3/100 WY in ZIM. Syphilis
prevalence at screening was 2.4% UG, 2.6% ZIM and 1.3% TH and correlated
with HIV prevalence (p<0.05). Herpes simplex virus-2 prevalence was common
in all countries. Prevalence of infection due to C. trachomatis (Ct) was: 5.4% in
TH, 3.0% in UG, and 3.3% in ZIM. Gonorrhea (GC) prevalence varied from 0.6%
(TH) to 2.1% (ZIM), and trichomoniasis ranged from 0.2% (TH) to 3.8% (ZIM).
Bacterial vaginosis was very common: 3.6% in TH, 9.3% in UG and 19.2% in ZIM
but was not related to age and yearly number of sex-partners at baseline. STI incidence rates varied for Ct (7.9/100 WY in TH, 2.8/100 WY in UG and 3.5/100 WY
in ZIM) and GC (3.8/100 WY in UG, 3.5/100 WY in ZIM and 1.8/100 WY in TH)
while GC and BV infection carried an increased risk for HIV acquisition (p<0.05).
Conclusion: HIV prevalence and incidence rates vary considerably in countries
with mature HIV epidemics. Control of STI may help reduce HIV in the general
population in some countries.
Presenting author: Robert Salata, CWRU Med School E202, 10900 Euclid Ave,
Celveland, Ohio, 44106-4984, United States, Tel.: [phone], Fax: [phone], E-mail: [email protected]
ThPeC7414 Biological mechanisms of reduced HIV acquisition
in circumcised males: susceptibility of human
foreskin and cervical tissue to HIV-1 infection
R.C. Bailey1, B. Patterson2, J. Siegel3, A. Landay4. 1University of Illinois at
Chicago, School of Public Health, University of Illinois at Chicago, 1603 W
Taylor St., Chicago, IL 60612, United States; 2Northwestern University, Chicago,
United States; 3Rush-Presbyterian St Lukes Medical Center, Chicago, United
States; 4Rush-Presbyterian-St. Lukes Medical Center, Chicago, United States
Background Over 35 epidemiological studies have found a significant protective
effect of male circumcision (MC) against HIV acquisition. We studied mechanisms
of HIV infection in foreskin tissue compared to cervical tissue by quantifying HIV
target cells, chemokine receptor expression, and target cell infectivity with HIV-1.
Methods Foreskins from pediatric and adult subjects with and without histories
of STI infection were evaluated. CD4 T-cells, macrophages, and Langerhans cells
(LC) were quantified using immunohistochemistry and image analysis. Cells expressing HIV-1 co-receptors CCR5 and CXCR4 were quantified using 2-color immunofluorescence and image analysis. Foreskin biopsies were infected ex-vivo
in organotypic culture with HIV-1. Infected cell phenotype was determined by simultaneous immunophenotyping/ultrasensitive in situ hybridization. The quantity
of HIV-1 DNA copies in foreskin and cervical mucosal tissue was compared.
Results Foreskin mucosa contained immune cells with higher mean proportions
of CD4 T cells (22%), macrophages (2%) and LCs (12%) in adults than in children (5%, 0.3%, 6%, respectively) or than in adult cervical mucosa (6%, 1%, 2%).
The highest prortions of CD4 T cells and LCs occurred in subjects with history of
infection. Foreskin mucosal cells expressed predominantly the CCR5 HIV-1 receptor. The keratinized outer surface of adult foreskin showed no uptake of HIV;
whereas mucosa from the inner surface had several-fold greater susceptibility to
infection with HIVbal than did cervical mucosa.
Conclusions Relative to cervical mucosa and external non-mucosal foreskin,
foreskin mucosa have high proportions of HIV-1 target cells, especially T-cells
and LCs, which express primarily CCR5, and are significantly more susceptible
to ex-vivo HIV-1 infection. Circumcision likely reduces risk of HIV-1 acquisition in
men by decreasing HIV-1 target cells and their receptors.
Presenting author: Robert Bailey, School of Public Health, University of Illinois
at Chicago, 1603 W. Taylor St., Chicago, IL 60612, United States, Tel.: +1 312
355 0440, Fax: [phone], E-mail: [email protected]
ThPeC7415 Alcohol consumption before sex and the risk of
HIV acquisition
l.B. Zablotska', R.H. Gray', M.J. Wawer2, D. Serwadda3, N.K. Sewankambo4,
T. Lutalo5. 'Johns Hopkins School of Public Health, 2720 St. Paul Str, Apt 3FF,
Baltimore, MD 21218, United States; 2Columbia University School of Public
Health, New York, United States; 3 Makerere University Institute of Public Health,
Kampala, Uganda; 4Makerere University School of Medicine, Kampala, Uganda;
5Uganda Virus Research Institute, Entebbe, Uganda

Background: Alcohol consumption reduces social and sexual inhibitions and
concerns about HIV/STI prevention and safe sexual behavior. It facilitates" casual sex and increased number of sexual partners. We explored the association
between alcohol consumption in conjunction with sex and the risk of HIV-1 acquisition, in the population-based cohort in Rakai district, Uganda.
Methods: All consenting residents aged 15-59 enrolled into the trial were tested
for HIV-1. Sociodemographic and behavioral characteristics were ascertained by

home interviews. We examined the association between gender-specific prevalence and incidence of HIV-1 infection during 1996-1998, and alcohol consumption before sex, reported by both partners, either or neither partner. Multivariate
prevalence rate ratios (PRR) and incidence rate ratios (IRR) with 95% confidence
intervals (95% CI) for HIV prevalence and incidence were estimated using no
alcohol consumption by either of partners as a reference group.
Results: 7109 males and 8549 females were included in this analysis. The baseline prevalence of HIV-1 infection was 16.9%. Alcohol use with sex was reported
by 75%. HIV-1 prevalence in those consuming alcohol before sex was 18.1% compared with 9.2% in persons not using alcohol (adjusted PRR=1.58(1.39-1.81)).
HIV-1 incidence in individuals consuming alcohol with sex was 1.8 per 100 PY
compared with 1.0 per 100 PY in those not consuming alcohol (IRR=1.75 (1.21 -2.53) per 100 PY). The population attributable fraction (PAF%) of incident HIV-1
associated with alcohol use during sex was 36%.
Conclusion: Alcohol consumption in conjunction with sex is significantly associated with the risk of HIV-1 acquisition in this rural Ugandan population.
Presenting author: Irina Zablotska, 2720 St. Paul Str, Apt 3FF, Baltimore, MD
21218, United States, Tel.: [phone], E-mail: [email protected]
ThPeC7416 Baseline associations between vaginal practices
and vaginal infections in Zimbabwean, Ugandan
and Thai women
J.H.H. van de Wijgert', P.G.A. Cornelisse2, C. Morrison3, S. Chitsungo4,
M. Munjoma4, S. Rugpao5, J. Byamugisha6, A. Muchini4, S. Sriplienchan5,
N.S. Padian7. 'Population Council, The Population Council, One Dag
Hammarskjold Plaza, New York, NY 10017, United States; 2Fred Hutchinson
Cancer Research Center, Seattle, WA, United States; 3Family Health
International, Research Triangle Park, NC, United States; 4UZ-UCSF Womens
Health Program, Harare, Zimbabwe; 5Chiang Mai University, Chiang Mai,
Thailand; 6Makerere University, Kampala, Uganda; 7University of California, San
Francisco, CA, United States
Background: There is limited evidence that vaginal practices (i.e. cleansing and
drying/tightening) may be associated with vaginal infections, which in turn may be
associated with HIV acquisition.
Methods: As part of a prospective cohort study on hormonal contraception and
HIV acquisition in Zimbabwe, Uganda and Thailand, women were evaluated for
vaginal infections and interviewed about vaginal practices at baseline.
Results: Vaginal cleansing and drying/tightening were practiced by 60% of
women in Zimbabwe (N=1438), 67% in Uganda (N=1819), and 13% in Thailand
(N=1578). The most common practice was vaginal cleansing with water, or water
and soap, using a finger. In Zimbabwe, Uganda, and Thailand, BV (Amsel criteria)
was diagnosed in 19%, 9%, and 4% of the women respectively; yeast (wet mount)
in 17%, 5% and 4%; trichomoniasis (wet mount) in 4%, 3% and 0.2%; and 17%,
45% and 36% of the women reported having had vaginal symptoms in the last
12 months. Vaginal practices were significantly associated with the presence of at
least one vaginal infection after controlling for use of hormonal contraception, sexual behavior, smoking, and alcohol use (OR=1.52; 95% CI=1.31-1.76; p<0.001).
However, there were differences between the study sites. The association between vaginal practices and vaginal infections was not statistically significant in
Zimbabwe or Uganda (OR=1.00 and 0.80 respectively), but was highly significant
in Thailand (OR=2.23; 95% CI=1.42-3.49; p=0.001).
Conclusions: Vaginal practices and vaginal infections were common, particularly
at the African sites. In Thailand, women who recently engaged in vaginal practices were more likely to have vaginal infections at baseline. In the prospective
phase of the study, we will evaluate the effect of both on HIV acquisition.
Presenting author: Janneke van de Wijgert, The Population Council, One Dag
Hammarskjold Plaza, New York, NY 10017, United States, Tel.: [phone],
Fax: [phone], E-mail: [email protected]
IThPeC7417 Men attending public STD clinics in Mumbai have
high rates of HIV, exposure to sex workers,
male-male sex, and herpes simplex 2 infection
C.R Lindan', H.R. Jerajani2, M.S. Mathur2, A. Gogate3, B. D'Souza4,
M.L. Ekstrand', S. Kumta2, M. Setia2. 'University of California, Center for AIDS
Prevention Studies, 74 New Montgomery Street, Suite 600, San Francisco, CA,
94105, United States; 2Sion LTMG Hospital and Medical School, Mumbal, India;
3Mumbai District AIDS Control Society, Mumbai, India; 'Center for AIDS
Research & Training (TISS), Mumbai India
Background Rates of HIV among commercial sex workers (CSW) in India are
high, but few interventions target their male clients. STD clinics provide access to
these high-risk men. We evaluated male STD patients as part of a randomized
controlled trial of a behavioral intervention to reduce HIV and STD incidence.
MethodsWe enrolled consecutive patients attending 2 municipal STD clinics;
men completed a survey, a clinical exam and HIV counseling and testing (C&T).

STDs were evaluated with urine PCR for gonorrhea (GC) and chlamydia (CT),
VDRL/TPHA (syphilis), Hepatitis BsAg, herpes simplex 2 IgG (HSV2) and PCR
for H. ducreyi (HD). We report on the first 386 of 2000 to be enrolled.
Results 49% of men were 19-25, 39% had less than 4 years of school, 54% had
visited CSW in the last 3 months, 47% were married, 24% had some MSM activity
(receptive or insertive anal sex) in the past and 46% and 12% of those who were
married visited CSW and MSM, respectively, in the last 3 months. 20% of men