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

100 Abstracts WePeC6065-WePeC6067

XIV International AIDS Conference

Table 1
Variables n
32 1
11 110 68 1.93 (0.89-4.20)
HIV-RNA: <30000 32 27 1
30000-100000 24 20 1.06 (0.32-3.91)
>100000 62 53 0.98 (0.37-2.63)
presented in Table 2.
Table 2
Variables n % Univariate Multivariate
HR (95% CI) HR (95% CI)
S (TIS): SO 97 60 1 1
S1 66 40 2.20 (1.16-4.17) 2.18 (1.15-4.14)
Visceral involvement: No 95 61 1 1
Yes 61 39 2.52 (1.33-4.78) 2.48 (1.31-4.71)
Previous AIDS: No 118 72 1
Yes 45 28 2.57 (1.37-4.82)
Conclusions: In the era of HAART, the ACTG staging classification (TIS) does
not seem to provide useful prognostic information in pts with AIDS-related KS.
Supported by ISS and AIRC grants.
Presenting author: Guglielmo Nasti, Divisione di Oncologia Medica a, Centro di
Riferimento Oncologico, Via Pedemontana Occidentale 12, 33081 - Aviano (pn),
Italy, Tel.: [phone], Fax: [phone], E-mail: [email protected]
WePeC6065 CD4 count and viral load-strong, independent
predictors of time to death in clade E HIV-infected
Thais
C. Costello1, A. Duerr2, C.M. Heilig2, S.C. Shiboski3, S. Sennun4,
S. Tovanabutra4, K.E. Nelson5, V. Suriyanon4. 1TRW/CDC, Atlanta, GA, United
States; 2CDC, Atlanta, United States; 3University of California at San Francisco,
San Francisco, United States; 4 Chiang Mai University, Chiang Mai, Thailand;
5Johns Hopkins University Baltimore, United States
Background: Rates of disease progression and correlates of survival among
HIV-clade E infected persons have not been well defined.
Methods: HIV-positive (HIV+) male blood donors at participating Chiang Mai
blood banks from 1988-97 were screened for a couples study on heterosexual
HIV transmission. Most female partners had no other known risk factors for infection. At enrollment (1992-98), all seropositive subjects were antiretroviral naive
and most were asymptomatic. For seropositive subjects without a documented
HIV-negative test, seroconversion was estimated to be the midpoint between the
onset of the heterosexual epidemic (January 1st, 1989) (or if later, the date of
sexual debut) and the first documented HIV+ test. Baseline CD4 count and viral
load were available on 93%, and vital status was available on 96%, of the 866
HIV+ subjects.
Results: By the end of 1999, 286 (49%) men and 69 (25%) of the HIV+ wives
had died. The median time to death in years was 6.9 (95% CI=6.6-7.3) for men
and 8.3 (8.1-unbounded) for women. For men with CD4 counts < 200 cells/p l,
the median survival time in years was 5.4 (5.0-5.6) compared to 9.1 (8.8-9.4) if
CD4 count > 500; the respective survival times among women were 4.1 (3.7-6.3)
and 8.3 (8.1-8.5) years. Men with baseline viral loads < 10,000 copies/ml lived
a median of 8.8 years (7.8-unbounded) compared to 6.2 (5.7-6.6) if viral load >
100,000; respective survival times for women were 8.5 (8.1-8.5) and 5.3 (5.0-6.4)
years. A multivariate survival analysis on study men utilizing monthly male blood
bank prevalence and date of first positive blood donation to impute seroconversion date probabilities found that for each 100 cells/[l increase in CD4 count, the
risk of death decreases by 24% (16.1-30.4) and for each log increase in viral load,
the risk of death increases 40% (3.7-88.4).
Conclusions: CD4+ and viral load are independent, strong predictors of time to
death from clade E HIV, similar to reports on clade B.
Presenting author: Caroline Costello, CDC Mailstop K34, 4770 Buford Highway, Atlanta, Georgia 30341-3724, United States, Tel.: [phone], Fax:
[phone], E-mail: [email protected]
WePeC6066 Mortality during the first 24 months after delivery
in a cohort of HIV-1 seropositive women in Dar es
Salaam, Tanzania
C. Kilewo1, K. Karlsson2, A. Swai1, A. Massawe1, E. Lyumuya1, F Mhalu1,
G. Biberfeld2. 1Muhimbili University College for Health Sciences, Tanzania;
2Swedish Institute for Infectious Disease Control, Karolinska Institute,
Stockholm, Sweden

Background: The objective of this study was to analyse mortality in a cohort of
HIV-1 seropositive women at the Dar es Salaam site of the UNAIDS multicentre
Petra trial (a mother-to-child HIV-1 transmission intervention study using antiretroviral therapy). Antiretroviral treatment was not available in this setting apart from
the short treatment given within the trial around delivery to prevent mother-to-child
transmission.
Methods: HIV-1 seropositive pregnant women enrolled into the Petra trial in Dar
es Salaam were followed at a research clinic within Muhimbili National Hospi
tal. The women were followed up at enrolment (36 weeks of gestation), delivery, weeks 1, 3, 6 and at months 3, 6, 9, 12, 15, 18, 21 and 24 postdelivery. Tlymphocyte subsets were determined by flow cytometry Plasma HIV-1 RNA was
quantified by Amplicor HIV-1 RNA Monitor v 1.5 (donated by Roche Diagnostic
Systems). Mortality was analysed using the Kaplan-Meier technique.
Results: Of 288 mothers enrolled into the Petra trial at the Dar es Salaam site, 21
mothers moved or withdrew their consent to participate before the week 1 visit.
Among the remaining 267 mothers 96% breastfed their children. The cumulative
probability of maternal death at 24 months after delivery was 7%. The CD4 cell
counts at enrolment were <200 cells/mm3 in 39 of 267 mothers (14.6%). Ten of
these 39 mothers (25.6%) died. Of the 16 mothers who died within 24 months
after delivery, 10 (62.5%) had CD4 cell counts <200/ mm3 at enrolment. 15 of 16
mothers who died were tested for viral load at enrolment and 11 (73.3%) of them
had > 100 000 copies/mL.
Conclusions: A considerable proportion (14.6%) of the pregnant women in this
study fulfilled the laboratory criteria for AIDS (<200 CD4 cells/ mm3) at enrolment. The mortality among these women was high. Enrolment viral loads were
high in women who died.
Presenting author: Gunnel Biberfeld, Swedish Institute for Infectious Disease
Control, Nobelsv.18, S-17182, Solna, Sweden, Tel.: [phone], Fax: +46 8
337460, E-mail: [email protected]
WePeC6067 Comparison of HIV RNA changes, CD4+ count
evolution and clinical progression under highly
active antiretroviral therapy (HAART) between
naive and pre-treated HIV-infected
C. Droz1, F Raffi2, RP Mercie3, J.B. Hubert4, L. Meyer4, C. Leport5, M. Egger6,
F. Dabis1, G. Chine, on behalf of the ANRS AC7 CASE Collaboration1.
1INSERM U330, Bordeaux, France; 2Hotel-Dieu Hosp, Nantes, France;
3Haut-Lev6que Hosp., Bordeaux, France; 4INSERM U292, Kremlin-Bic6tre,
France; 5University of X Bichat, Paris, France; 'University of Bristol, Bristol,
United Kingdom
Background: After starting HAART, differences in HIV RNA and CD4+ count
have been observed between patients initially naive of antiretroviral therapy and
pre-treated. Whether this early difference is translated in subsequent progression
to AIDS or death is unclear. We compared the evolution of HIV-1 RNA, CD4+ cell
count, rates of progression to AIDS and death after the start of HAART between
naive and pre-treated patients.
Methods: Data on 3,731 patients from 3 ANRS-sponsored cohorts in France
were combined as part of the CASE (Cohorte Aquitaine, Aproco, SEroco) Collaboration. Comparisons between naive and pre-treated patients used Cox proportional hazards regression models.
Results: Mean baseline CD4+ cell count did not differ between naive and pretreated patients (277 vs 278/mm3), whereas HIV-1 RNA did: 4.6 log10 copies/mL
vs 4.0 log10 copies/ml, respectively (p<0.0001). At 12 months, the rate of progression to AIDS was 4.2% among the naive group vs 6.3% among the pretreated patients (p=0.01) and the mortality rate was 3.9% vs 5.4% (p=0.06),
respectively. In multivariate analysis, after adjusting for gender, age, baseline
clinical status, haemoglobinemia, HIV RNA and CD4+ count, naive patients remained at a lower risk of progression to AIDS (Hazard Ratio [HR]=0.5, 95% Confidence Interval [CI]: 0.3-0.7, p=0.001) or to death (HR=0.7, CI 0.4-1.0, p=0.06).
Mean increase in CD4+ count at Month (M)6 was: +128 vs +75/mm3, respectively
(p<0.0001) and mean reduction of HIV RNA was -2.13 vs -0.96 Iog10910 copies/ml,
respectively (p<0.0001). After adjusting for CD4+ count and HIV RNA changes
at M6, rates of clinical progression and mortality did not differ between naive and
pre-treated patients.
Conclusion: Response to HAART defined by CD4+ count and HIV RNA variation
is better in patients initially naive to HAART than among others. The short-term
changes (M6) of CD4+ count and HIV RNA under HAART accounts in part for
this difference.
Presenting author: Cecile Droz, Inserm U330, 146, rue Ldo Saignat, case 11,
33076 Bordeaux Cedex, France, Tel.: [phone], Fax: [phone], Email: Cecile.Droz @ isped.u-bordeaux2.fr
WePeC6068 Survival analysis of HIV-infected neurologic
patients: role of HAART and of specific neurologic
disorders
A. Antinori1, P. Lorenzini1, A. Cingolani2, P. Cinque3, F Soldani1,
S. Bossolasco3, S. Grisetti1, M.G. Finazzi4, V. Tozzi1, M. Bongiovanni5,
F Morettie, B. Vigo7, B. Gigli, G. Mazzarello8, S. Foresti9, L. Cristianoo,
M. Mena"1, G.C. Fibbial2, S. Artiolil3, G. Fasulol4, A. d'Arminio Monforte5,
A. Ammassari2. 'NI/D "L. Spallanzani", Roma, Italy; 2Clinic Infectious Diseases,
Catholic University, Roma, Italy; 3HSR "San Raffaele Hospital", Milano, Italy;
" Department of Infectious Diseases, Bergamno, Italy; 5 Clinic Infectious Diseases,

Milano, Italy; 6Clinic Infectious Diseases, Brescia, Italy; 7,"Niguarda" Hospital,
Milano, Italy; 8 "San Martino" Hospital, Genova, Italy; 9Department of Infectious
Diseases, Monza, Italy; ~Department of Infectious Diseases, Taranto, Italy;
" Department of Infectious Diseases, Cuggiono (MI), Italy; 2Department of
Infectious Diseases, Mantova, Italy; 3Department of Infectious Diseases, La
Spezia, Italy; "'Department of Infectious Diseases, Bologna, Italy
Background: The exact impact of HAART on survival of patients with HIV