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

12th World AIDS Conference

Abstracts 23179-23183 377

syringes weighed heavily in local syringe ecology, but many syringes circulated
long enough to have been reused.
Conclusions: Teamed with information on numbers of injectors using and not
using the NSEP, the results of this simple approach can reveal how complete an
NSEP's coverage is and how much additional sterile equipment must be introduced; implemented serially, this information can be used to determine whether
the NSEP is effective at reducing the time each syringe spends in circulation; coupled with data on prevalence of HIV or other viruses, this method can demonstrate
impact of the NSEP as a preventive information.
23179 Methadone maintenance treatment and incident HIV
infection among persistently injecting drug users
Miranda Langendam, G.H.A. Van Brussel, R.A. Coutinho, E.J.C. Van Ameijden.
Nieuwe Archtergracht 100, 1018 WT Municipal Health Service, Amsterdam, The
Netherlands
Objective: In Amsterdam, the Netherlands, methadone treatment for drug users
is prescribed in low-, medium- and high-threshold methadone programs. Promoting to higher threshold programs is encouraged, but 'demotion' is also possible.
The present study evaluated harm reduction methadone maintenance treatment
(MMT) in Amsterdam in relation to incident HIV-infection among HIV-negative
injecting drug users (IDU) who persisted in injecting behavior and with a history
of methadone treatment.
Methods: Data of the Central Methadone Register in Amsterdam, which registers all methadone prescriptions in Amsterdam, were linked to the Amsterdam
AIDS cohort study among drug users. Poisson regression analysis was used to
model HIV-incidence.
Results: 258 IDU contributed 755 person years, 46 IDU seroconverted, the
overall HIV-incidence was 6/100 person years with a strongly declining trend
(1985-1996). Low frequency of program attendance (RR 2.7, 95% CI 1.0-6.9,
ref. 100% program attendance), change in frequency of program attendance
(higher frequency RR 2.7, 95% Cl 1.3-6.7, lower frequency RR 1.9, 95% Cl
0.8-4.2, ref. no change), decrease in methadone dosage (RR 2.9, 95% Cl
0.9-9.0, ref. no change) and consistently dispensing via general practitioner
(medium-threshold, RR 0.3 95% Cl 0.7-1.1, ref. low-threshold program) were
associated with HIV-incidence. Methadone dosage was not significantly associated with HIV-incidence, but some indications were found for a protective effect
of dosages >80 mg/day. In multivariate analysis decrease in methadone dosage
was no longer associated with HIV-incidence. The results did not change after
adjustment for potential confounders other than methadone treatment modalities.
Conclusion: The high HIV-incidence indicates that MMT is not sufficient to
stop the spread of HIV among IDU who persist in injecting. Further increase
of the methadone dosage and more strict program regimes to increase patient
retention may lower the HIV-incidence among IDU. Because HIV-incidence was
lower among higher threshold levels of dispensing, the policy to circulate between
the different thresholds appeared to be well implemented.
23180 1 Do A32 CCR-5 deletions protect intravenous drug users
(IDUs) against HIV infection?
Raffaele Malinverni1, D. Germann2, R. Haemmig3, H.J. Furrer1, L. Matter2,
A. Christen1, C. Dahinden4. 1Med. Poliklinik, Inselspital Bern, Bern; 2Institute
Medical Microbiology, Bern; 3University Psychiatry Services, Bern; 4Institute for
Immunology, Switzerland
Objectives: To assess the frequency of A 32 CCR-5 deletions in a cohort HIVnegative IDUs enrolled in the heroin maintenance prescription research program
(HMPRP) in Bern, and to compare it with a cohort of HIV-positive IDUs.
Design: Cross-sectional analysis of two cohorts
Methods: HIV-negative IDUs were recruited from HMPRP in Bern. Only IDUs
with long term, ongoing and heavy intravenous drug use who have failed other
treatment attempts are eligible for this program (irrespective of HIV-status). HIVseropositive control IDUs were recruited from our HIV outpatient clinic. All subjects
were tested for serum antibody against HIV (Genscreen), hepatitis C virus, HBsAg
and anti HBc (AxSYM). CCR-5 genotype was determined by PCR using primers
based on the published CCR-5 sequence.
Results: Preliminary analysis involved 99 HIV-negative and 53 HIV-positive
IDUs. Mean age was 34 and 35 years in the HIV+ and HIV-cohort respectively.
Mean duration of intravenous drug use before enrolment in this study was 12 and
13 years for the HIV- and HIV+ subjects respectively. Serum markers for hepatitis
B and hepatitis C were present in 96% and 97% of HIV- subjects respectively.
Homozygous A 32 deletion of CCR-5 gene was found in 3/99 HIV-IDUs and
0/53 HIV+ IDUs. Heterozygous A 32 deletion of CCR-5 gene was found in 22/99
HIV- and 7/53 HIV+ IDUs. Overall, A 32 deletions of the CCR-5 gene was more
frequent in HIV-seronegative than in seropositive IDUs (p = 0.08).
Conclusion: These preliminary data suggest that not only homozygous but
possibly also heterozygous A32 deletions of the CCR-5 gene provide protection
against transmission of HIV in heavy IDUs with risk behaviours. Polymorphisms
in other chemokine receptors may contribute to protection.

23181 Recruitment of injection drug users (IDUs) in preparation
for HIV vaccine efficacy trials
Helen Navaline, D.S. Metzger, A. Davis-Vogel, G.E. Woody. PVAMC Bldg#3
Univ. & Woodland Aves. Philadelphia, 19104; University of Pennsylvania
Philadelphia PA, USA
Background: In 1995, the Vaccine Preparedness Study (VPS) enrolled 4892
HIV-seronegative individuals from 8 US cities. In Philadelphia, IDUs (N = 542)
were recruited from existing studies, methadone programs and needle exchange
sites. While these strategies had identified cohorts with incidence rates of >2%
per year in prior initiatives, the incidence rate of the 1995 cohort was approximately.5% per person year. New recruitment strategies were initiated in 1997.
This presentation will compare the risk characteristics of these two cohorts.
Methods: In 1997, recruitment strategies and eligibility criteria were adjusted
to target injectors considered to be at highest risk of HIV infection. Screening
occurred in drug detoxification facilities as well as via street outreach conducted
on foot and using a mobile assessment unit.
Results: A total of 507 individuals were screened in Philadelphia between
September, 1997 and January 1, 1998. Four hundred and twenty-one subjects
reported being HIV seronegative and injecting recently and exhibited physical
evidence of recent venipuncture. Twenty-nine percent were women, 42% AfricanAmerican, 17% Latino, and 40% white. Among the new cohort, 66% reported
injecting more than once a day, compared to 39% of the 1995 cohort. Forty-one
percent reported using a needle after another user, compared to 15% of the
earlier cohort. Approximately 20% of those screened reported getting most of
their needles from a needle exchange and 6% reported being in methadone
treatment, as compared to 48% and 45%, respectively, of the 1995 cohort.
Conclusion: The shifting patterns of HIV transmission among IDUs require
new recruitment strategies in order to identify those at greatest risk of HIV infection. Adjustments in recruitment strategies have resulted in the identification of
IDUs at higher risk of HIV infection.
123182 HIV-1 infection among drug users attending a needle
exchange program (NEP) and participating to a STD
screening program on urine samples in Quebec City
Michel Alary1, C. Poulin2, F. Bernier3, J. Ringuet2, C. Claessens4. ICHA -
Pavillion St - Sacrement 1050 Chemin Ste - Foy, Quebec, G1S 4L8; 2Centre De
Sante Publique De Quebec QC; 3Dept. of Microbiology, U. of Montreal Montreal
QC; 4Laboratorie De Sante Publique Du Quebec Montreal QC, Canada
Background: The objective of this study was to determine HIV prevalence and
associated risk factors among drug users attending a NEP in Quebec City in the
context of a STD screening program on urine samples.
Methods: Subjects attending NEP were asked to answer a questionnaire on
risk behaviors and to provide a urine specimen and an oral fluid sample (Orasure,
Epitope, OR, USA). The Calypte HIV-1 urine EIA kit was used for HIV testing on
urine specimens. Cambridge Biotech HIV-1 Western-Blot kit was used to confirm
all urine specimens found positive twice by EIA. Oral fluid samples were tested
for HIV using Vironostika HIV-1 (Organon Tecknica, NC, USA). Prevalences and
their 95% Confidence Intervals (CI) were calculated whereas risk factors were
assessed using logistic regression in a multivariate analysis.
Results: A total of 742 drug users (491 men and 251 women) participated in
the study. Among them, 67% of men and 52% of women had ever injected drugs
of whom 75% have injected in the previous 6 months (current injectors). One
subject (current injector) was excluded because HIV testing yielded discrepant
results on urine and oral fluid samples (respectively positive and indeterminate).
The overall prevalence of HIV was 5.4% (95% CI:3.9%-7.3%). No case of HIV
was found in non-injecting drug users but the prevalence was 5.5% (6/110) among
the ex-injectors and 9.8% (34/347) among the current injectors (p < 0.01). In a
logistic regression model, risk factors for HIV infection among current injectors
were (all p < 0.05): being a woman (OR:3.0); injecting since >5 years (OR:3.3);
needle sharing with a HIV+ person (OR:7.5); history of hepatitis (OR:2.6) and
having <1 heterosexual partner in the previous 6 months (OR:4.4). Needle sharing in general was borderline significant (OR:2.3, p = 0.06). Only one HIV positive
person had also a STD at the moment of the study and the overall prevalences
of chlamydia and gonorrhoea were respectively 3.4% and 0.4% using Roche
Amplicor TM CT/NG PCR on urine samples.
Conclusion: In this population of drug users, the prevalence of HIV infection
was moderate among ex- and current injectors and relatively stable when compared with previous studies. The non-injecting drug users seem still spared by
HIV infection. However, their closeness to infected injecting drug users may place
them at high risk for HIV.
582*/23183 HIV/AIDS risk in non-hospitalized injection drug
users in Japan
Timothy Sankary1, H. Ohashi2, K. Soda3. 1UCLA School of Public Health,
Epidemiology, Los Angeles, USA; 2Medical Care, Corrections, Justice, Tokyo;

3 Yokohama Univ School Medicine, Yokohama, Japan
Objectives: Since hospitalized methamphetamine users probably represent a
very small and unusual group of all IDUs, the purpose of this study, the first of its
kind in Japan, was to investigate the HIV risk behavior and reasons for needle
sharing among non-hospitalized IDUs in Japan.
Design: Cross-sectional study.