Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

412 Abstracts 23352-23356

12th World AIDS Conference

23352 Looking beyond needle sharing: "Indirect" sharing and
viral infections among injecting drug users in Zurich,
Switzerland
Jen Wang1, J. Marcinko2, D. Meili2, U. Zellweger1, M. Olgiati3, B. Somaini1,
M. Flepp4. 1lnstiitute F Social + Preventive Medicine, Sumatrastr. 30, CH-8006,
Zurich; 2Zoll 1, Methadone Substitution Clinic, Zurich, ZH; 3Psychiatric
University Clinic, Zurich, ZH; University Hospital, Zurich, ZH, Switzerland
Objective: To obtain data on a broad palette of sharing behaviors associated
with injection drug use and constituting important risk for viral transmission.
Methods: The results presented here stem from two 1997 studies among
opiate-using clients in drug substitution clinics in Zurich, Switzerland. First, the
cross-sectional phase of the Zurich Prometheus Study where questions on drug
and sexual behaviors associated with risk for HIV and hepatitis infection were developed and collected. Second, an in-depth study on drug injection behaviors was
carried out at one of the participating methadone clinics among current injectors.
Results: The results presented from the Zurich Prometheus Study are based
on provisional analyses of data on the 369 ever injectors (77%) collected through
December 17, 1997. The male-female ratio was 3:2, and the mean age was 31
years (SD = 6). All have consumed heroin (mean = 9 years), and 83% cocaine
(mean = 7 years). Self-reported prevalence was 21% for HIV and 70% for HBV
and HCV. As for risk behaviors over their entire drug-using career, 57% reported needle sharing, 51% front- and/or back-loading, and 76% reported sharing
spoons, filters, or water - the latter two constituting "indirect" sharing. Among the
active IDUs (58% within the past 6 months), 13% have shared needles in the
past 6 months, comparable to rates in the early 1990s. However, if one also takes
"indirect" sharing into account, the prevalence of injection-related risk behavior
jumps to 44%. HIV-positive injectors were just as likely as their HIV-negative
counterparts to lend their needles or to load drug solute into another syringe. The
in-depth study provides further insight on the common practice of" L6ffelteilete" -
two or more IDUs simultaneously withdrawing solution from the same spoon with
their syringes - found among 80% of IDUs in this study population. The water
that goes into the spoon is measured (50% share water from the same container)
with their own syringes which 83% use more than once. Only half of the repeat
users clean their syringes after each use, the vast majority rinising ineffectually
with cold or hot tap water. Questions on the perceived risk of one's own behaviors
reveal that while IDUs appreciate the danger of infection associated with classical
needle sharing, forms of "indirect" sharing are considered completely harmless
by the majority.
Conclusion: In the 1990s, Zurich has been the setting of many unique interventions targeting IDUs- e.g., Needle Park, syringe automats, shooting rooms, heroin
substitution programs, and syringe filter distribution. Despite this wide spectrum of
interventions, viral transmission continues to occur as best evidenced by 60-70%
prevalence rates of hepatitis found even among the newest and youngest IDUs.
The results here show that "indirect" sharing is common and takes place at various stages leading to actual injection, and harm reduction will need to go beyond
needles and syringes and 1) educate about the risks associated with "indirect"
sharing and 2) target such behaviors with concrete interventions.
23353 Shedding of HIV-1 DNA subtypes C and A in genital
ulcers in patients attending STD Clinics in Pune, India
Deepak Gadkari1, T.C. Quinn2, R.R. Gangakhedkar1, S.M. Mehendale1,
A.R. Risbud1, M.E. Shepherd2, R.C. Bollinger2. 1National AIDS Research
Institute, Pune, India; 2John Hopkins University, Baltimore, USA
Background: Our previous studies have indicated high prevalence (20%) and
incidence (10%/yr) of HIV in patients attending two STD clinics Pune. The present
study was undertaken to 1) support clinical diagnosis of genital ulcer diseases
(GUDs) with multiplex PCR, m-PCR 2) detect HIV DNA shedding in ulcerative
lesions 3) define the risk factors associated with HIV DNA shedding and 4) to
identify HIV-1 subtype in shed DNA.
Methods: After obtaining the informed consent, HIV antibody positive and
antibody negative but p24 antigen positive individuals attending two STD clinics
were identified by respective EIAs and WB assay. Genital ulcer swab specimens
were collected and m-PCR (Roche) was performed for simultaneous detection
of chancroid, syphilis and herpes was performed. AMPLICOR assay was used
to detect HIV DNA in swab specimens. HIV-1 subtypes were determined by
heteroduplex mobility assay (HMA).
Results: Of the 302 consecutive patients attending the clinics with GUDs,
67 were HIV-1 seropositive and 4 were HIV antibody negative but p24 antigen
positive. DNA-mPCR confirmed the clinical diagnosis of 51/71 individuals (16
chancroid, 11 syphilis and 31 herpes) with seven individuals being dually infected.
HIV-1 DNA was detected in 24 (34%) GUD samples as revealed by amplification
of the gag gene portion. All four p24 antigenemic subjects were shedding HIV
DNA. Multivariate analyses indicated that chancroid, duration & depth of ulcer
and concurrent urethritis/cervicitis were independently associated with HIV-DNA
shedding. Entire env gene could be amplified from two p24 antigenemic subjects.
Analysis of the C2-V5 region of the env gene by HMA revealed that one individual
was infected with subtype C while the other with subtype A.

Conclusions: HIV infected individuals with GUDs appear to be efficient transmitters of HIV. 100% of the individual in 'window period' were found to shed HIV
DNA in ulcerative lesion. Detection of subtype A in shed DNA was unique since
HIV-1 C is the most prevalent subtype in Western India. This poses an opportunity
for generation of A and C recombinant viruses. Early and complete treatment of
GUDs and safe sex practices warrant immediate attention to control secondary
transmission.

I23354 A family study among relatives of HIV-1 persistently
seronegative prostitutes in Nairobi, Kenya
Joshua Kimani1, E. Njagi1, A. Myatha1, B. Ball2, R. Kaul1, J. Kreiss3,
F. Plummer3. Nairobi University Narobi, Kenya; 2University of Manitoba,
Canada; 3University of Washington Seattle, USA
Introduction: In a cohort of HIV-1 resistant prostitutes from Pumwani, Nairobi,
30% have been noted to be first degree relatives, suggesting that inherited genetic
factors may influence HIV-1 susceptibility. We are therefore conducting a family
study to determine whether there is a hereditary basis for HIV-1 resistance.
Method: The study is designed using the approach of King et al. For every one
index HIV-1 resistant prostitute, four index HIV-1 susceptible prostitutes, matched
for tribe and prostitution duration, are recruited. These women then ask their
first degree relatives to participate. A standard questionnaire is administered to
all participating relatives. Blood is then drawn for ABO grouping, RPR, HIV-1
serology, serologic HLA typing and genetic analysis.
Results: We have recruited 48 HIV-1 resistant and 89 HIV-1 susceptible prostitutes from the Pumwani cohort, and a total of 508 relatives. Among the relatives,
102 women are also prostitutes enrolled in the Pumwani cohort. Relatives of
resistant and susceptible prostitutes are similar in terms of demographic, social
and sexual parameters. However, relatives of resistant prostitutes are significantly
less likely to be infected by HIV-1. This applies both to high-risk relatives who are
enrolled in the prostitute cohort (OR = 0.03, 95% CI; 0.006-0.1); and to lower-risk
relatives from outside the cohort (OR = 0.46, 95% CI; 0.26-0.8).
Conclusion: Relatives of HIV-1 resistant prostitutes are less likely to be infected by HIV-1 than relatives of HIV-1 susceptible prostitutes, despite similar
risk-taking behaviours. This suggests that resistance to HIV-1 may be mediated
by inherited factor(s).
23355 Acceptability of male circumcision as a strategy to
reduce HIV and STD infections in Uganda
Robert C. Bailey, S. Neema. University of Illinois at Chicago SPH 2121 W Taylor
St Chicago, IL, USA; Makerere Inst for Social Research, Kampala, Uganda
Objectives: To assess the beliefs, knowledge and attitudes of Ugandans concerning male circumcision (MC) and the acceptability of MC as an intervention to
reduce HIV and STD infection.
Design: Cross-sectional. Focused discussions with separate groups of men
and women.
Methods: Adults (ages 18-67) were recruited through Local Councilors in
each of three districts in western Uganda where MC is not normally practiced.
Discussion groups (13 with women; 14 with men), lasting 50-95 minutes with
6-13 consenting participants were conducted in both rural and periurban settings
in the local language by a native moderator. Education ranged from none to university level. Discussions were recorded, then translated and transcribed within
24 hours, and covered the following topics: Reasons not to circumcise, reasons
to circumcise; preferred ages for MC; preferred circumcisors; genital hygiene;
sexual pleasure; and sexual attraction.
Results: The main barriers to MC were cultural identity, association of MC
with Islam, and fear of pain or infection. Exposure to risk of HIV infection, loss
of penile sensitivity, and tendency for circumcised men to be womanizers were
also concerns. The main facilitators, seen as related, were a strong desire for
increased penile cleanliness and reduced risk of STD and HIV infection. Most
men and women believed that circumcised men have increased sexual pleasure
and confer greater pleasure to their partners. Medical doctors were preferred
circumcisors. There was no consensus on the best age to perform MC.
Conclusion: We found widespread belief that circumcised men are less susceptible to STDs and HIV infection, a strong ethic among men and women favoring
better male hygiene, and trust in health professionals to perform MC safely and
with minimal pain. These findings suggest that many Ugandans in areas where
MC is not now widely practiced would be more accepting of MC than formerly
recognized if MC were discussed in the context of health in addition to culture
and religion.
S23356 Risk factors for HIV-1 infection in a rural population of
15,000 adults in southwest Uganda
Lucy Carpenter1, J.F. Kengeya-Kayondo2, P.M. Kintu1, M. Quigley3, A. Kamali1,
J.A.G. Whitworth1. 1 PO Box 49 Entebbe MRC Programme on AIDS, UVRI,
Uganda; 2 Tropical Diseases Research, WHO Geneva, Switzerland; 3London
School of Hygienea Tropical Medicine London, UK
Background: To estimate the effects of risk factors for HIV-1 infection in an adult
rural population in SW Uganda.
Methods: Data on risk factors for HIV infection were collected for a baseline
survey of a large community intervention trial to assess the efficacy of a behaviour
change programme with or without improved management of sexually transmitted diseases (STD) for reducing HIV transmission. Sera were obtained from all
consenting adults (13+ years) and tested for HIV infection. Age-adjusted odds

ratios (OR) were used to estimate relative risks associated with presence of a
risk factor versus absence of that factor on HIV prevalence with 95% confidence
intervals (95%CI).
Results: 14,818 adults were surveyed and bled. Overall HIV prevalence was
9.7% (95%CI 9.2-10.2) and was significantly raised (P < 0.001) among the widowed and divorced and among non-muslims. Prevalence was over 25% among
men and women reporting a genital ulcer in the last 12 months (OR = 2.7, 95%CI