Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Mo.C.320 - Mo.C.324

Monday, July 8, 1996

participated in case-controlled interviews that examined risk-taking behaviors, sexual abuse
history and gynecological infection history
Results: The HIV seroprevalence rate at the prison during the study period ranged from 13
to 20% (voluntary testing) in the study period, and our best estimate of the seroconversion
rate among previously HIV seronegative women returning to prison was 6 to I 6 percent.
Latina WI were 1.5 times more likely (95% Cl 1.0 to 2.3) to be identified as HIV-seropositive than white WI; African American WI were less likely than white women (POR 0.4, 95%
Cl 0.2 to 0.8) to be identified as HIV-seropositive. HIV seropositivity was 2.0 times more
prevalent (95 % C 1.0 to 3.9) among women reporting exchanging sex for drugs or
money; 14.8 times more prevalent (95% CI 6.5 to 33.7) among women who reported
injection drug use; and less likely (POR 0.5, 95% CI 0.2 to 0.9) among women who reported having had unprotected sex at the time of their last encounter, than among women who
did not report these histories. HIV seropositivity was 2.8 times more prevalent (95% ClI.0
to 7.4) among women who reported a history of childhood sexual abuse, and these
women were 4.6 times more likely (95% CI 1.4 to 15.2) to have participated in three HIV
risks (sex work, injection drug use, and unprotected sexual contact), than women who did
not have a history of sexual abuse. HIV+ Latina WI and African American women were not
more likely to experience a decrease in their CD4 slope, than the reference group (HIV+
white WI). Sexually transmitted disease history (STD) was prevalent in this population: 62%
of HIV+ women and 68 % of HIV-women (n=88) reported having had an STD within
three months prior to the interview.
Conclusions: HIV risk reduction is a critically important intervention for women who are
incarcerated, and must take into consideration the long term effects of prior sexual abuse.
Differential risks for HIV infection and indicators of HIV progression by race and ethnicity
deserve further investigation.The opportunity to provide under-served and high risk
women access to effective HIV education, prophylaxis, and treatment in the prison setting
should not be missed.
Anne S. De Groot,TB/HIV Research Laboratory, International Health Institute, Brown
University School of Medicine, Box GB 473, Providence RI 029 I 2.
Telephone [phone] Fax 863- I 243 email Anne_DeGroot @ postoffice.brown.edu
Mo.C.320
THE ROLE OF SERUM MICRONUTRIENT LEVELS IN HIV-I DISEASE PROGRESSION
Tang Alice M*, Graham NMH*, Semba RD*, Chandra RK**, Saah AJ*. *Johns Hopkins University Baltimore MD, USA; **Janeway Child Health Centre, St. John's, Newfoundland, Canada.
Objective: To examine the associations between serum levels of vitamins A, B6, B12' E,
and folate with risk of progression to two key outcomes in the natural history of HIV- I
infection - first AIDS diagnosis and mortality
Methods: The study population was drawn from a well-established cohort of gay/bisexual
men in the Baltimore site of the Multicenter AIDS Cohort Study who have been followed
semiannually since 1984. Eligible subjects were HIV- I seropositive upon study entry and had
serum available in the serum repository from their 1984 baseline study visit Serum
micronutrient levels were assessed on 312 eligible subjects. Kaplan-Meier analyses and Cox
proportional hazards models were used to estimate the relative hazard of progression to
each outcome for low versus adequate, and quartiles of serum micronutrient levels.
Results: In Kaplan-Meier analyses, median AIDS-free time was estimated to be four years
shorter in subjects with low serum B 12 levels (< 120 pmol/L) compared to those with adequate B I2 levels (>120 pmol/L). Low serum B I2 levels were independently and significantly
associated with an increased risk of progression to AIDS, after adjusting for HIV- I related
symptoms, CD4+ cell count, age, serum albumin, serum folate, use of antiretroviral therapy
before AIDS, and frequency of alcohol consumption (relative hazard (RH)- 1.87, 95% confidence interval (CI) I. 14-3.08). Analyses of quartiles yielded an apparent threshold effect
for both vitamins A and E. Subjects in the highest quartile of serum vitamin E levels (>10 13
eg/dL) had a 33% decrease in risk of progression to AIDS compared to those in the lower
three quartiles combined (RH=0.67, 95% C1=0.45-1.00). Subjects in the highest quartile of
serum vitamin A level ( 90.2 pg/dL) showed a similar decrease in risk of progression to
AIDS, however this result was not statistically significant (RH=0.73, 95% CI0.49-1.07).
Similar trends were observed with mortality as the outcome.The associations were
strongest in subjects with low serum albumin levels (<35 g/L).
Conclusions: These data show that subjects with low serum B12 levels had an approximate
two-fold increase in risk of progression to AIDS, while subjects with increased serum vitamin E levels had a greater than 30% decrease in risk of progression to AIDS.The inclusion
of vitamins A,. and B12 in clinical trials of nutrient supplementation should be considered
in HIV infected persons.
A.M.Tang, 624 N. Broadway, Suite 394, Baltimore, MD, 21205, USA Telephone: [phone]
Fax: [phone] email: [email protected]
Mo.C.32I
FRACTION OF PATIENTS FREE OF CLINICAL AIDS AFTER CD4 CELL COUNT
DROPPING LESS THAN 200X106/I IN JAPANESE HEMOPHILIACS INFECTED
WITH HIV-I
Shinobu Tatsunami, Mimaya*, j, Meguro,T, Nakamura, 5, Kuwabara, RYago, NYamada, K. St.
Marianna University School of Medicine, Kawasaki, JAPAN and *Children's Hospital of
Shizuoka Prefecture, Shizuoka, JAPAN
Objective: To evaluate the distribution of time from the first date of the CD4 cell count
dropping less than 200X 106I to the onset of clinical AIDS by 1987 Centers for Disease
Control and Prevention case definition in Japanese hemophilia patients infected with HlV- I.
Methods: We extracted the variables from the data base constructed by Natural History
Committee affiliated to The Research Committee on Prevention of Developing Illnesses and
Therapy for HIV Infected Patients that has been sponsored by the Japanese Ministry of
Health and Welfare.The first date when the recorded count of CD4 cell was less than
200X I06/j was used as the initial point of analysis ti.The cumulative incidence of clinical
AIDS by I1987 definition was analyzed by proportional hazard model.The incorporated
covariates were age and CDC stage of each patient at time i as well as CD8 cell count,
levels of IgA, GOT and GPT
Results: A total of 21I0 patients was analyzed.The level of IgA at ti was the only one variable that showed statistical significance (p<c0.001I) in the present covariates.The time at
which the 6raction of patients fr-ee of clinical AIDS reached 50% by the Kaplan-Meier

method was 44.3 years.The estimated mean period T from ti to cinical AIDS onset
was 5.7~4.4 years
Conclusions: The resent inalysis reveals that there is a broad distribution in the time from
ti to clinical AIDS rset in, Japanese hemophi iacs, and it evaluates the distribution quantitatively Higher level of IyA i ssociated with shorter period from t to clinical AIDS onset.
Especially noticeable is the value of estirnmated mean period that is longer than 5 years. The
present results should be taken into consideration in various desion making on the therapy and care of HIV I infected people as well as on the medical plannin g for them.
Shinobu Tatsunami, Radioisotope Res. Inst St.M arlanna Un. Scho ol of Med. 2-16I Sugao,
Miyamae-ku, Kawasaki, Japan 216Tel [phone] E t 361 Fax.[phone]
Mo.C.322 C
O
PROGRESSION OF HIV INFECTION AND STABILITY OF IMMUNOLOGIC MARKERS ~
PRIOR TO AND AROUND SEROCONVERSION
Munoz, Alvaro*, Margolick JB, Chu C, Don nenbe rg AD, G orgi JV, Phair JP *Johns Hopkins
School of Public Health, Baltimore, Maryland, USA.
Objectives: To determine the prognostic information for -IIV disease progress on of high
within-individual variances of immunologic markers prior to HIV-I seroconversion (SC) and
changes of markers around seroconversion.
Methods: The study population was 370 part cipants of the Multicenter AIDS Cohort Study
with known (+- 6 months) dates of SC and wi h T-cell subsets measured at the ast negative
and first positiye visits. Up to July 1995, clinical AIDS has occurred in 153(4 I%), and 132
(36%) have died. Stability of CD4+, CD8 and CD3+ lymphocyte counts and percentages
among lymphocytes was quantified by the withc individual variance and coefficent of variation of measurements prior to the last negati vscvsit. Mult pie linear regresson was used to
relate stability with changes between last negatve and first positve sits rn the og-transformed CD4 cell count and the CD4 percent. Parametric survival regression was used to
relate stability and changes ofT-cell subsets around SC with AIDS-fee and survival times.
Results: The medians of the pre-SC within-ind ividual standard deviations of CD4, CD8 and
CD3 cell counts were I 82, I 27 and 289, respectively; and 4.2, 2.3 and 4.7 for the corresponding percentages among lymphocytes. Median 6 -morth CD4 declines around SC were
226 cells (or 26%) in the count and 7.0 in the percent. Neither counts nor percentages
pre- SC stability were signficantly associated with decline of CD4 between the last negative
and first positive visits. Individuals with large (40%; N-= 93), moderate ( i0%-40%; N= 139)
and small (< 10%; N= 138) declines in CD4 cell count around SC had median (95% C.I.)
AIDS- free times of 7. I (6.0, 8.4), 9.6 (8.2, I I.1) and 9.8 (813.4, I 13) years; and median survival times of 8.2 (7.0, 9.6), 10.0 (8.8, I.5) and 12.I (I0.4 14.0) years, respectively
Conclusions: Pre-SC variability of immunologic markers was not informative for progres
sion of HIV infection; but changes around SC were signficantly assocated with AIDS-free
and survival times. A dose-response relation was present with survival times: the larger the
decline in CD4 count around SC, the shorter the survival.
Alvaro Munoz, Ph.D., JHU, Room 797 Hampton House, 624 N. Broadway Baltimore, MD
21205, USA (410 955-2792:[phone]; munozolstaepi.sph.jhu.edu
Mo.C.323
HIV- I RNA LEVELS IN EARLY CHRONIC INFECTION ASSOCIATION WITH AIDS
AND LONG TERM NON-PROGRESSION
O'BrienThomas R*, BlattnerWA*, Kroner BL*U, Goedert JJ*. for- the Mutcenter
Hemophilia Cohort Study. *Viral Epidemiology Branch, Ntional Cancer Instit ute, Bethesda,
MD; **Research Triangle Institute, Rockville, MD
Objective: To determine the risk of AIDS and proportion of subjcts who were long-term
non-progressors (LTNP), by HIV-I RNA levels measured in early chron ic infection.
Methods: We measured HIV-I RNA by PCR (Roche M ortor assay) n irchived serum
specimens collected fom 1i88 persons enrolled in a ong-er m study of hemoph ia and
AIDS. Specimens were collected 12 to 36 months (meda-, 25.5 months) after the estimated date of HIV- I seroconversion (age at seroconverson, to 66 years). We used KaplanMeier methods to examine the risk of AIDS and proportioanal hazards models to determine
relative hazards. We defined a LTNP as a person wi th a Cr4+ lym ph ocyt e count >500
cells/mm3 and no HIVI associated condition ten years after seroconverson.
Results: The proportions of subjects who developed AIDS within ten years aer seroconversion were: 70 %for subjects with 100,000 copies/l (n- 0); 52% for subjects with
10,000-99,999 copies/ml (n55); 19% for subjects with 1,000-9,999 copies/m (94), and
0% for subjects with < I,000 copies/ml (n--29) (p<0.O0001), The age-adjusted relative hazard for AIDS for subjects wrth >10,000 copies/ml was 13.5 (95% confidence vnterval 3.9
46.38) compared to subjects with < 1,000 copies/ml. Five of nine subects,with <200
copies/ml were LTNP compared with10 of 179 subjects with hiher eves (p <0.00001).
Conclusions: HIV-I RNA levels during early chronic HIV-I infection are strong, age-independent predictors of clinical outcome; very ow levels de ine persons wth a high probability of becoming a LTNP Factors operating early in the cours of HIV- nfect on stron gly
influence long-term prognosis.
T.R. O'Brien Natioeal Cancer Institute. EPN 434, 6130 Execuative Blvd., Rckville. MD 20852.
[phone]. Fax. [phone]!7, r-ra r: OBRIENToT DNDCE.NC I.NIM.rOV
Mo.C.324
INCIDENCE AND PROGNOSTIC SIGNIFICANCE OF SYMPTOMATIC PRIMARY
HIV-I INFECTION IN HOMOSEXUAL MEN.
Veugelers, Paul J*, Strathdee SA0, Kaldor JM#, Page 5/hafer K A&, Srhechtec MT@ Ofoutinho
IRA*, Keet IPM*, van Griensven GJPC. *Muncipal Health Se'rvace, Amsterdam, Nether lands;
@ B.C. Centre for excellence in HIV/AI[DS, Var, uses: Carada; 8Unrversty of Ness South
Wales, Sydney, Australia &Unsversicy of ( a ifornisa. S ur Fr anoso US.
Objective: To investigate incidence and prognostic sgnrfrcau e of symptocctSt prmary HIVI infection in homosexual men from prospective cohoct s udiecv n Vancouver: Sydncey,
Amsterdam and Sac Francisco.
Methods: We compared incidence rates of self-reposrted diarrhoec, fever. n ght sweas,
cough and fatigue during the snterval of DIV-I ser orconeeson (SC) vs5/ races during biannual intervals prior So and abter SC rn men wits welI docur-csted and prospecively determined dates of SC, participating in theTricont nental Sesocreverter Study. We cornducted
Cox proportional hazard analyses to evaluate crc isk fo AdlDS and deat, ard modelled the 35