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

12th World AIDS Conference

Abstracts 60356-60360 1065

to work in brothel complexes, and to have more customers per week than other
sex workers. The prevalence rates of syphilis, gonorrhea, chlamydia, and trichomoniasis were higher among sex workers who received the routine penicillin
treatment than among those who had not received antibiotic treatment in the last
28 days. However, after adjustment for age, education, fee per sex act, number
of customers, and condom use in the previous 7 days, only trichomoniasis was
still significantly different (adjusted odds ratio of 3.2).
Conclusion: High risk women were more likely to participate in the routine
penicillin prophylaxis program. The lack of a demonstrable individual-level protection from this prophylaxis treatment program in this cross-sectional study appears
due to differential uptake of penicillin prophylaxis by women at higher presumptive
risk for STD. Randomized clinical trials and mathematical modeling, together with
observational data such as presented here, all can contribute to optimal understanding of a complex intervention like mass chemoprophylaxis for STD among
female sex workers.
S60356 | Adherence to HAART in clinical trials vs. clinical care
setting
Katrin Bergeron1, J. Gormley1, H. Sousa1, K. Tashima2, T.P. Flanigan2,
N.A. Merriman. ' The Miriam Hospital, 164 Summit Ave, Providence RI 02906;
2Miriam Hospital Brown University, Providence RI, USA
Issue: Do clinical trials enhance adherence to highly active antiretroviral therapy
(HAART) in a comprehensive care setting?
Methods: A retrospective chart review was conducted to compare adherence
among individuals enrolled in open-label drug trials to a randomly selected group
of individuals receiving prescription antiretrovirals in the same clinical care setting.
The two groups were similar with respect to age, sex, race, amount of time on
therapy, and viral load prior to initiating HAART. Plasma viral load (PVL) was used
as the primary measure of adherence.
Results: Participation in clinical trials enhanced adherence to HAART when
compared to a similar group receiving prescription antiretrovirals in the same clinical setting. Fifty-one individuals in clinical trials at our site completed >3 months
(median 6 months) of open label therapy and were included in this analysis. 84%
(43/51) of the clinical trial group had a most recent PVL of <.400 copies/mL. The
comparison group of patients received prescription antiretrovirals within the same
clinical setting. 59% (15/36) of those taking HAART by prescriptions had a most
recent PVL of -400 copies/mL. The prevalence of prior antiretroviral therapy
was higher in the group of individuals receiving prescriptions. Therefore, a subset
analysis was done on the individuals receiving prescriptions who were naive to
treatment prior to initiating HAART. Among these 60% (9/15) had a most recent
PVL of -400 copies/mL.
Conclusions: In a clinical trials setting, 84% of patients on HAART had PVL
-400. The adherence techniques employed in clinical trials need to be explored
in clinical settings to improve adherence and impact on viral load suppression.
60357 Prevention of Pneumocystis carinii pneumonia:
The incarcerated
Abe Macher1, E. Goosby2, D. Von Zinkernagel2. 1560 Fishers Lane Room
7A-55, Rockville MD 20857; 2U.S. Public Health Service, Washington DC, USA
Issue: As increasing numbers of HIV-infected substance abusers enter correctional facilities, their morbidity will be directly related to the implementation of standard-of-care prophylactic regimens for the prevention of opportunistic infections.
Project: We reviewed the medical records of four HIV-infected inmates with a
history of injection drug use who were admitted from prison to a city hospital for
diagnosis and treatment of Pneumocystis carinii pneumonia (PCP).
Results: Although all four inmates had CD4+ T-lymphocyte counts below 200,
none had been receiving prophylaxis for the prevention of PCP during their
incarcerations.

progression-free survival and changes in CD4 cell counts and serum viral RNA.
Methods: Data were used from a clinical endpoint study of IDV 800 mg q8h vs.
ZDV 200 mg q8h vs. IDV+ZDV in 996 treatment-naive HIV-1 infected patients with
50-250 CD4 cells/mm3. Clinical efficacy was assessed by evaluating time from
initiation of therapy to the first clinical event (AIDS-defining condition or death).
CD4 and vRNA were measured in all patients. Lamivudine (3TC) was added to
the regimens of most patients in the IDV+ZDV and ZDV groups during the study
(median 40 wks).
Results: 716 patients (72%) were male and 280 (28%) were female. 84 (11.7%)
and 23 (8.2%) of the male and female patients, respectively, experienced a clinical event. For both genders, the IDV-containing regimens resulted in marked
decreases in vRNA and increases in CD4 cell counts relative to the ZDV arm.

IDV+ZDV vs. ZDV Overall
Males
Females
IDV vs. ZDV Overall
Males
Females

Percent Reduction in Hazards of Clinical Event
(95% Cl)
70% (50%, 82%)
77% (58%, 88%)
38% ( 58%, 76%)
61% (38%, 76%)
61% (45%, 76%)
67% ( 2%, 89%)

Conclusions: IDV provided clinical and surrogate marker benefit in both males
and females.
60359 Nurses education in occupational HIV transmission
Anna Bamford. Infection & Immunity Unit, Ealing Hospital, Uxbridge Road
Southall, Middlesex, UBI 3HW, England, England
Issue: Health Care Workers have a low but measurable risk of HIV infection after
accidental exposure to infected blood or body fluids. Although compliance with
infection control recommendations in handling sharps is the mainstay of prevention, additional prevention strategies now include post-exposure prophylaxis with
anti-retroviral therapy.
Project: The aim of my project was to review the latest research, data and
update nurses as to what guidelines are in place and the choices of post-exposure
prophylaxis that is recommended. I also collected information from questionnaires
sent to nurses working on an HIV unit, to monitor their knowledge of the above
issues.
Results: Nurses had a good understanding of univeral precautions and risk
fctors, but there was a lack of knowledge regarding transmission rates and
post-exposure prophylaxis.
Education is a fundamental necessity to empower a nurse with the relevant
knowledge to make an informed decision regarding risk assessments in patient
care.
60360[ Predictors of progression of the human
immunodeficiency virus (HIV) to first opportunistic
infection
Denis Keohane12, D.R. Warren2, M.E. Charlson2. 1 1320 York Avenue-Apt 31-D
New York, New York 10021; 2Cornell University Medical College New York NY,
USA
Purpose: Most predictive models for progression of HIV were developed in the
1980's and are currently not clinically relevant because of changes both in the
treatment of HIV and in the disease itself. Our objective is to identify clinical
variables related to occurrence of first opportunistic infection in order to create a
new model for disease progression.
Methods: Our study involved a retrospective longitudinal cohort of 139 HIV+
patients seen at the Cornell Internal Medicine Associates or the Center for Special
Studies (dedicated HIV unit) during the calendar year 1990. 50 patients were
excluded from the study because they either received less than one year of
follow-up care after initial presentation or had no chart documentation of HIV
disease. Clinical, laboratory and psychosocial data was recorded at baseline,
yearly intervals, first opportunistic infection (01) and at death for the 89 subjects
included in the study. Zerotime was established as the date of the first known
positive HIV serology. For this analysis, our primary outcome is time to first 01.
Results: Our study population is 60% male, mean age 36.4 years, 56% single,
45% African-american and 36% Latino. 74% had solo coverage with medicaid
and 87% primarily spoke English. Risk factors for HIV were intravenous drug use
(35%), heterosexual (23%) and homosexual (14%). Time to first HIV diagnosis
was, on average, 1.48 years.
28% of the subjects developed an 01 with a mean time of 1.53 years. Using proportional hazards models in SAS, the following variables were identified
individually as significant predictors of progression of HIV disease to first opportunistic infection: inpatient status within one year of zerotime (for any medical
reason), zerotime CD4% < 10%, zerotime history of syphilis, zerotime absolute
lymphocyte count <750, zerotime hemoglobin -10 and no treatment with AZT
(all p < 0.05). Controlling for the total number of medications (including antiretrovirals and prophylaxis), multivariate analysis identified baseline CD4% < 10%
and heterosexual risk as significant predictors of progression of HIV disease to
first 01.
This analysis suggests that, while controlling for the baseline HIV medications,
heterosexual risk and low zerotime CD4% were strong predictors of progression
of HIV disease to first opportunistic infection.

Patient CD4 count Duration of cough Respiratory rate Temperature
1 6 Month 40 102.3
2 121 Unknown 32 105.1
3 34 Months 40 102.0
4 89 Month 24 102.9

Pulse
126
122
114
106

Lessons Learned: Care and treatment available to HIV-infected inmates in
correctional facilities should be equivalent to that available to HIV-infected persons in the community. HIV-infected inmates should have access to clinicians
experienced with HIV-care.

160358j

Subgroup analysis of clinical and surrogate marker
efficacy by gender in a study of indinavir (IDV) alone and
with zidovudine (ZDV) in treatment-naive HIV-1 patients
with 50-250 CD4 cells/mm3

Randi Leavitt', D.S. Lewi2, D. Uip3, R. Pedro4, A. Kalichman5, J. Suleiman6.
Merxk Protocol 28 Study Group; ' Merck Research Laboratories BL3-4, 10 Sentry
Parkway, Blue Bell, PA 19422; 2Federal Univ of Sao Paulo, Sao Paulo, SP; 3Univ
of Sao Paulo Med School, Campinas, Campinas, SP; 4 Campinas State Univ
Med School, Sao Paulo, SP; 5Centro de Referencia E Trenamento AIDS, Sao
Paulo, SP; 6Emilio Ribas Infectious Disease Inst, Sao Paulo, SP, Brazil
Objective: To investigate the consistency across gender of treatment effects on