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

XIV International AIDS Conference

Abstracts ThPeC7409-ThPeC7412 441

ThPeC7409 HIV transmission in a cohort of serodiscordant
heterosexual couples
J. Castilla', B. Marincovich2, J. del Romero3, S. Garcia 3, M.J. Belza1,
C. Rodriguez3, L. Olmos4. 1 Centro Nacional de Epidemiologia (ISCIII) y
Secretarfa del Plan Nacional sobre Sida, Madrid; 2Beca MUTIS, Agencia
Espahola de Cooperacidn Internacional, Madrid; 3Centro Sanitario Sandoval,
Servicio Madrileto de Salud, Madrid; 4 Universidad Complutense, Madrid, Spain
Background: To estimate risk of HIV transmission in a cohort of serodiscordant
heterosexual couples, and to identify variables associated with higher risk.
Methods: We studied steady heterosexual couples in which one of the members
was HIV infected (index case, IC) while the other was not, after excluding other
risk exposures. Couples were contacted during the 1987-1999 period in a clinic
of Madrid. Both members of each couple were followed by the same practitioner
and had a check-up every six months. Incidence rates were calculated according
to the number of person-years (PY) of exposure and according to the estimated
number of penetrations without condom. Multivariate analysis were carried out
using a Poisson regression.
Results: A total of 292 couples fulfilled criteria, 50 with woman IC and 242 with
man IC. They totalled 788 PY and 57,850 penetrations (anal or vaginal), of which
8,214 took place without condom. 4 cases of HIV transmission were confirmed,
which represented a rate of 5.1 per 1000 PY (CI 95%, 1.4-12.9), 6.9 per 100,000
coitus and 49 per 100,000 unprotected intercourses (CI, 13-125). In the multivariate analysis of incidence rates per PY, were independently associated with transmission: aids diagnosis for IC (rate ratio=33), diagnosis of STD for IC (RR=18),
genital infection for contact (RR=33), and intercourse with ejaculation and without
condom (RR=54). In the multivariate analysis of rates calculated with the number
of penetrations without condom as denominator, transmission was associated to
diagnosis of genital infection for contact (RR=14), and AIDS diagnosis for IC was
close to significance (RR=14, p=0.06).
Conclusions: HIV transmission was rare, due to frequent use of condom. The
risk per sexual act is similar to that described in literature. Intercourse without
condom, genital tract infections and advanced clinical state of IC were the main
determinants of transmission.
Presenting author: Jests Castilla, Centro Nacional de Epidemiologfa, Sinesio
Delgado, 6., 28029, Madrid, Spain, Tel.: [phone], Fax: [phone],
E-mail: jcastill @isciii.es
ThPeC7410 1 Sexual risk behavior of HIV+ persons aware versus
unaware of their infection status: refining the focus
of HIV prevention efforts
G. Marks, N. Crepaz. Centers for Disease Control and Prevention, Atlanta, GA,
United States
Background: New cases of sexually transmitted HIV infection may stem from
two groups: HIV+ persons unaware they are infected, and HIV+ persons aware of
their infection status. A better understanding of the contribution of these groups
to new infections will inform the focus of HIV prevention efforts.
Methods: A computer-assisted search identified 11 studies/databases from 1987
through 2000 (10 US, 1 Rwanda) with relevant data that were meta-analyzed.
The analysis included cohort designs in which behavior was assessed in the 12
months before and 12 or more months after learning HIV+ status (4 samples)
and between-group designs in which HIV+/aware persons were compared with
HIV+/unaware persons (7 samples), focusing on the prevalence of unprotected
vaginal/anal intercourse or new sexually transmitted infections (STIs).
Results: The 11 samples included 11,417 participants. Among HIV+/aware persons, the prevalence of unprotected vaginal/anal sex ranged from 13-33% and
the incidence of new STIs ranged from 6-18% across samples. The combined
findings demonstrated an average of 58% lower prevalence in sexual risk behavior/STIs among HIV+/aware persons compared to their prior behavior or
HIV+/unaware persons (RR 0.42, 95% Cl 0.35-0.50). In univariate analyses, the
reduction was larger in samples from 1993-2000 (RR 0.38, 95% Cl 0.30-0.47)
than 1987-1992 (RR 0.59, 95% Cl 0.43-0.76), and larger in between-group (RR
0.39, 95% CI 0.31-0.48) than cohort designs (RR 0.58, 95% Cl 0.40-0.77). There
were no differences by gender.
Conclusions: There is substantial reduction in sexual risk behavior after persons
become aware they are HIV+. Nevertheless, the unprotected sexual intercourse
among HIV+ persons aware of their status may contribute to new infections, although transmission risk may be moderated by viral load levels and partner selection factors. The findings argue for increased opportunities for HIV testing, linkage
to medical care, and on-going prevention services for HIV+ persons.
Presenting author: gary marks, 1149 hancock dr. n.e., atlanta, georgia, 30306,
United States, Tel.: [phone], Fax: [phone], E-mail: gmarks
@cdc.gov

ThPeC7411 HIV transmission between gay men through oral
sex: Seroconversion Study, Sydney
J. Richters1, J. Ellard1, O. Hendry2, S. Kippax1. 1National Centre in HIV Social
Research, 2National Centre in HIV Epidemiology and Clinical Research,
University of New South Wales, Sydney, Australia
Background: US researchers reported in 2000 that 8/122 men recently diagnosed with HIV had probably acquired it through oral sex. This prompted concern

in Australia about whether the advice that oral sex was 'low risk' should be reviewed.
Methods: We examined interview data from a study of Sydney men who had
recently acquired HIV (n = 97, 1993-2001) to identify possible cases of oral transmission, according to the men's accounts of the events they believe led to their
seroconversion.
Results: For none of these men was oral sex unambiguously the primary risk for
HIV transmission, though some initially offered it as the route they thought most
likely. There were four men who reported no anal intercourse in the relevant period and whose infection must be attributed to blood or semen contact through
oral sex or nonsexual means. There were a further 18 men who reported protected anal sex only, for whom oral transmission was a (remote) possibility For
the other 75 men, unprotected anal intercourse with a partner who was or might
have been HIV-positive, or in one case needle-sharing, was a clearly higher risk.
Many men who had no unprotected anal intercourse felt their sexual practice was
safe and that their seroconversion was a surprise and perhaps unfair or even unbelievable. None used condoms for oral sex. Most assumed that protected anal
intercourse and oral sex without ejaculation were 'no-risk' rather than 'low-risk'.
Because the practice of unprotected oral sex is so common among these men, we
should expect to find cases of HIV acquired through receptive oral sex with ejaculation. Only one man, who had open wounds in his mouth, probably belonged to
this category
Conclusions: HIV transmission through oral sex appears to be very rare. Dentists should give clear advice after invasive procedures.
Presenting author: Juliet Richters, National Centre in HIV Social Research, University of New South Wales, Sydney NSW 2052, Australia, Tel.: +(61 2) 9385
6407, Fax: +(61 2) 9385 6455, E-mail: [email protected]
ThPeC7412I Risk factors for recent HIV infection among MSM:
The Importance of Delayed Application of
Condoms. Results of the Polaris HIV
Seroconversion Study
L. Calzavaral, A. Burchell1, R.S. Remis1, C. Major2, P. Corey3, T. Myers1,
P. Millson1, E. Wallace2, and the Polaris Study Team1. 1HIV Social, Behavioural
and Epidemiological Studies Unit, Faculty of Medicine, University of Toronto,
Toronto, ON, Canada; 2Ontario Ministry of Health and Long Term Care, Toronto,
Canada; 3Department of Public Health Sciences, Faculty of Medicine, University
of Toronto, Toronto, Canada
Background: To identify predictors of recent HIV infection among MSM.
Methods: Participants were recruited through Ontario's HIV diagnostic laboratory, physicians and community organizations. Recent seroconverters (cases)
were determined using HIV test results. Cases were gender-, risk-, and regionmatched to HIV-negative controls using a 1:2 case:control design. Cases were
asked about behaviour during the period of infection (median 6 months, range 3 -27 months) and for controls during an equivalent time period. A subsample of 183
MSM (62 cases and 121 controls) enrolled as of June 2001 was analysed. This
analysis focussed on sexual behaviours with casual, client, sex worker or HIVpositive/status unknown regular partners (i.e. partners other than HIV-negative
regular partners). Multiple logistic regression identified independent risk factors
for seroconversion.
Results: The average age was 34 years (range 18-64). Rates of unprotected receptive oral, insertive anal, and receptive anal sex and delayed application of condoms for receptive anal sex were significantly higher among cases (97%, 41%,
53%, 32%) than controls (73%, 19%, 14%, 2%). Independent correlates of HIV
infection were receptive anal sex without condoms (OR=3.74, p=0.005) and delayed application of condoms for receptive anal sex (OR=4.87, p=0.03). There
was a marginally significant correlation with breakage or slippage of condoms
during receptive anal sex (OR=2.96, p=0.09). No significant independent associations were observed with the number of sex partners or insertive anal sex.
Conclusions: Delayed application of condoms for receptive anal sex may result
in contact with pre-ejaculate. This behaviour, which to date has received little attention, may be as much a risk factor for HIV infection as unprotected receptive
anal sex. The finding has implications for future research and prevention messages.
Presenting author: Liviana Calzavara, 12 Queen's Park Cres. W., 3rd Fir, McMurrich Bldg., University of Toronto, Toronto, Ontario M5S 1A8, Canada, Tel.: [phone], Fax: [phone], E-mail: liviana.calzavara@ utoronto.ca
ThPeC7413 Prevalence and incidence of HIV and sexually
transmitted infections (STI) among young women
in Thailand, Uganda and Zimbabwe participating in
the hormonal contraception and risk of HIV
acquisition cohort
R. Salata1, P. Cornelisse2, T. Chipato3, F Mmiro4, C. Morrison5, N. Padian6,

S. Rugpao7, D. Celentano8. Case Western Reserve University Celveland, OH,
United States; 2FHCRC-SCHARP, Seattle, United States; 3University of
Zimbabwe, Harare, Zimbabwe; 4Makerere University, Kampala, Uganda; 5Family
Health International, Research Triangle Park, United States; 6UCSF, San
Francisco, United States; 7Chiang Mai University, Chiang Mai, Thailand; 8Johns
Hopkins University, Baltimore, United States
Background: The burden of STI among young women in societies with mature