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

XIV International AIDS Conference

Abstracts ThOrC1396-ThOrC1445 435

ThOrC 1396 The emergence of midazolam injection and its
association with needle sharing among injecting
drug users (IDUs) in Bangkok, Thailand
S. Vanichseni 1, F. van Griensven2, R Suntharasamai3, U. Sangkum4,
J. Tappero 2, W. Heyward 5, K. Choopanya1. 1Bangkok Vaccine Evaluation
Group, Bangkok, Thailand 2Thai MOPH - US CDC Collaboration, Thai MOPH -
US CDC Collaboration, DMS 6 Building, Ministry of Public Health, Nnthaburi
11000, Thailand; 3Mahidol University, Bangkok, Thailand; 4Bangkok Metropolitan
Administration, Bangkok, Thailand; I VaxGen, Inc., Brisbane, CA, United States
Background: Monitoring risk behavior in HIV vaccine efficacy trials is necessary
to identify changing patterns and to formulate prevention messages. In response
to reports of injection of midazolam (a sedative), we evaluated its use and association with needle sharing among participants in the AIDSVAX~B/E vaccine
efficacy trial in Bangkok, Thailand.
Methods: From March 1999 - August 2000, 2,545 IDUs were enrolled. Demographic and HIV risk behavior data were collected at baseline, and every 6 months
thereafter. Generalized estimation logistic regression analysis was used to evaluate trends and assess independent risk factors.
Results: 93.4% of participants were male, the median age was 26 years, 95.0%
had >primary education. At baseline, 2,388 (93.8%) reported injection drug use.
At 12 months follow-up, overall injection drug use decreased to 72.1% (p <0.001)
and needle sharing from 33.0% to 16.3% (p <0.001). Among those injecting at
baseline, 98.5%, 15.8% and 12.4% reported injecting heroin, methamphetamine
and midazolam, respectively Among those injecting at 12 month follow-up, these
were 96.6% (-1.9%, p <0.6), 17.4% (+10%, p <0.1) and 19.1% (+54%, p <0.001).
In multivariate analysis, younger age (< 40 yrs), higher education (>grade 9),
daily injection and not being in methadone treatment at baseline were independently associated with midazolam injection (all p <0.0001). Injection of midazolam, methamphetamine, younger age, daily injection and not being in methadone
treatment at baseline were independently associated with needle sharing (all p <
0.0001).
Conclusion: Overall, injection drug use risk decreased significantly over the first
12 months of the trial; however a marked increase in midazolam injection was observed., which was independently associated with needle sharing. Surveillance
of drug use practices is important to target those at highest behavioral risk and
formulate prevention messages.
Presenting author: Frits van Griensven, Thai MOPH - US CDC Collaboration, DMS 6 Building, Ministry of Public Health, Nnthaburi 11000, Thailand, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
ThOrC1443 Continued benefit from highly active antiretroviral
therapy: trends in AIDS-related opportunistic
illnesses in a public health care system, 1990-2001
D.L. Cohn, P.S. Breese, W.J. Burman, D.L. Lopez, B.L. Barth, C.A. Rietmeijer,
A.J. Davidson. Denver Public Health, denver public health, 605 bannock street,
denver co 80204, United States
Background: The use of highly active antiretroviral therapy (HAART) has had
a dramatic effect on HIV-related morbidity and mortality, but there are concerns
that the benefits of HAART may be waning because of drug resistance, side effects, and nonadherence. We evaluated the incidence of selected opportunistic
illnesses (Ols) and mortality in all patients in a public health care system in the
pre- and post-HAART era.
Methods: Since 1990, Denver Health, an integrated public health care system
and largest HIV care provider in Colorado, has recorded all HIV-related Ols and
mortality using CDC's HIV/AIDS Reporting System (active surveillance) and Adult
Spectrum of Disease Study (6-month chart review).
Results: Of 4118 HIV-infected patients, incidence-density of 01-specific first
episodes and mortality (per 100 person-years) from 1990 to 2001 was:
PCP MAC CMV CE WAS KS NHL Mortality
1990-91 6.0 5.0 2.8 3.8 3.3 4.3 0.7 12.4
1992-93 5.7 4.6 3.2 3.2 o.a 2.7 0.5 13.0
1994-95 4.6 4.0 3.5 3.3 0.3 2.4 0.6 15.6
1996-97 2.6 1.5 1.4 1.3 0.1 1.5 0.4 8.9
1998-99 1.6 0.5 0.2 1.2 0.4 0.6 0.1 4.2
2000-01 1.7 0.4 0.5 0.8 0.1 0.4 0.2 4.7
PCP = Pneumocystis carinii pneumonia, MAC = Mycobacterium avium complex, CMV = Cytomegalovirus, CE = Candida esophagitia, WAS = Wasting syndrome, KS = Kaposi's sarcoma,
NHL = Lymphoma
Conclusions: From 1990 to 2001, there was a decrease of 79-97% for selected
Ols, and 73% in mortality. These decreases were greatest between 1996 and
1997 (data not shown) and were sustained through 2001. We found no evidence
of waning clinical benefit after the implementation of HAART.

Presenting author: david cohn, denver public health, 605 bannock street, denver co 80204, United States, Tel.: [phone], Fax: [phone], E-mail:
[email protected]

ThOrC1 444 Factors associated with survival and progression
to AIDS before and after the introduction of HAART
P. Pezzotti1, M. Dorrucci1, A. Sinicco2, G. Angarano3, M. Zazzara 1,
G. Mazzarello4, M. Cusini5, G. Rezza1. lstituto Superiore diSanit&, Istituto
Superiore di Sanitb, Reparto AIDS e MST Viale Regina Elena, 299, 00161
Roma, Italy; 2Universita di Torino, Torino, Italy; 3Universita di Bari, Bari, Italy;
4Universita di Genova, Genova, Italy; 5Policlinico di Milano, Milano, Italy
Background: We evaluated the risk of progressing to death and AIDS by calendar period and by age at seroconversion, exposure category, and gender, before
and after the introduction of HAART
Methods: We conducted a prospective cohort study (Italian Seroconversion
Study) of HIV-infected persons with an estimated date of seroconversion, considered as time-zero of analysis. A Cox model allowing for staggered entries was
used to estimate the RH of death and AIDS in the periods 1980-1996, 1997 -1998 (reference), and 1999-2000. To evaluate simultaneously the role of age at
seroconversion, gender, and exposure category before and after the introduction
of HAART, we performed multivariate analyses for the periods before and after
January 1997.
Results: We followed 1,847 persons for a median of 8.6 years. During this period
601 developed AIDS and 469 died. The median age at seroconversion was 27.0
years; 53.0% were injecting drug users (IDU), 26.5% men who have sex with men
(MSM), and 20.5% heterosexual contacts, 30.6% of whom were women. Since
the RHs of death and AIDS were consistently very similar, only the data for AIDS
are shown here. The RH of AIDS was 2.33 (95% CI: 1.79-3.04) in 1980-1996,
and 0.30 (0.19-0.47) in 1999-2000, compared to 1997-1998. Before HAART, the
RH of AIDS was 1.43 (1.28-1.60) per 10-year increment in age at seroconversion,
1.07 (0.86-1.33) for women vs. men, 0.93 (0.74-1.16) for MSM vs. IDUs, and 1.03
(0.80-1.33) for heterosexual contacts vs. IDUs. After the introduction of HAART,
the RH of AIDS was 1.44 (1.08-1.93) per 10-year increment in age at seroconversion, 0.82 (0.51-1.32) for women, 0.27 (0.14-0.51)] for MSM, and 0.45 (0.24-0.49)
for heterosexual contacts.
Conclusions: The beneficial effects of HAART seem to be increasing with time.
However, since the introduction of HAART, the risk of death and AIDS has apparently increased for IDUs, possibly as a result of poor adherence to therapy.
Presenting author: Patrizio Pezzotti, Istituto Superiore di Sanitb, Reparto AIDS
e MST, Viale Regina Elena, 299, 00161 Roma, Italy, Tel.: [phone], Fax:
[phone], E-mail: [email protected]
ThOrC1 445 Does HAART fully explain the improved survival of
injection drug users in recent years?
D. Vlahov1, N. Galai2, J.C. Bareta3, S. Cohn4, S. Galea1, T. Sterling 5,
J.B. Margolick3. 'New York Academy of Medicine, Johns Hopkins University,
Bloomberg School of Public Health, 627 N. Washington st, Baltimore, MD
21205, United States; 2Ben-Gurion University of the Negev, Beer-Sheva, Israel;
3Johns Hopkins Bloomberg School of Public Health, Baltimore, United States;
4Research Triangle Institute, Rockville, United States; 5Johns Hopkins School of
Medicine, Baltimore, United States
Background: Since the introduction of highly active antiretroviral therapy
(HAART), survival of HIV infected persons has increased substantially. At the
same time, more aggressive treatment and prophylaxis of opportunistic infections
(01) also occurred. We evaluated the survival difference attributable to HAART
among injection drug users (IDUs), adjusting for CD4 cell count and other known
prognostic factors.
Methods: 601 HIV+ IDUs were followed from 1988-2000 with semi-annual interviews and laboratory tests. Time to HIV-related death was estimated from the first
visit at which CD4 cell count was <200 cells/ul. Kaplan-Meier survival curves preHAART (1988-96) and post-HAART (1996-2000) were compared, accounting for
actual HAART use. Time-dependent Cox regression models were used to model
time to death, adjusting for these and other prognostic factors.
Results: Survival estimates for the pre-HAART period, post-HAART with no
individual HAART treatment, and post-HAART with actual treatment were respectively: 0.94, 0.93, 1.0 at one year and 0.57, 0.85, 0.96 at 3 years. A
significant residual survival benefit in the post- HAART period remained (adjusted hazard ratio (HR) 0.52, p=0.002) after adjustment for individual use of
HAART (HR=0.32, p=0.007), PCP-prophylaxis (HR=0.68, p=0.02), AIDS diagnosis(HR=10.3, p<0.001), CD4 cell counts (HR=0.99,p<0.001) and hemoglobin
levels (HR=0.78, p<0.001). Reported treatment for Ols other than PCP, frequent
injection, age and gender and viral load) did not have a significant effect on survival.
Conclusions: For IDUs with low CD4 cell counts, the rate of HIV-related death
decreased significantly in the post-HAART era. However, improved survival was
only partially explained by reported HAART and PCP prophylaxis. The additional
unexplained benefit could reflect the selective mortality of 'heavy' drug users of
non HIV-related causes and the parallel observed reduction in drug injection frequency, but warrants further study.

Presenting author: Joseph Bareta, Johns Hopkins University, Bloomberg School
of Public Health, 627 N. Washington st, Baltimore, MD 21205, United States, Tel.:
+[phone], Fax: +[phone], E-mail: [email protected]