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

436 Abstracts ThOrC1446-ThOrC1485

XIV International AIDS Conference

ThOrC 446 Marked improvment in survival following AIDS
dementia in the era of highly active antiretroviral
therapy
G.J. Dore, A. McDonald, Y. Li, J. Kaldor, B. Brew. National Centre in HIV
Epidemiology and Clinical Research, NCHECR, Level 2, 376 Victoria Street,
Darlinghurst NSW 2010, Australia
Background: The study aimed to determine the effect of introduction of highly active antiretroviral therapy (HAART) on survival following AIDS Dementia Complex
(ADC).
Methods: Australian AIDS notification data over the period 1993-2000 was examined. Analyses were based on all initial AIDS illnesses. In order to examine
the impact of HAART, two periods of AIDS diagnosis were chosen: pre-HAART
(1993-1995) and HAART (1996-2000). Median survival was based on KaplanMeler estimates, with examination of factors influencing survival in a Cox proportional hazards model.
Results: Over the period 1993-2000 in Australia, 5017 initial AIDS illnesses were
diagnosed among 4351 AIDS cases. The proportion of AIDS cases with ADC
increased from 5.2% in 1993-1995 to 6.8% in 1996-2000 (p=0.029). Median survival following AIDS increased from 19.6 months for AIDS cases diagnosed in
1993-1995 to 39.6 months for AIDS cases diagnosed in 1996-2000 (p<0.0005).
Median survival following ADC increased more than all other AIDS illnesses, from
11.1 month in 1993-1995 to 48.2 in 1996-2000 (p<0.0005). Most striking was the
seven-fold increase in survival among ADC cases with a CD4 count <100mm3
at diagnosis; 5.1 months in 1993-1995 to 38.5 months in 1996-2000 (p<0.0005).
In contrast to the marked improvement in ADC survival, median survival following other predominantly CNS AIDS illnesses (cryptococcosis plus toxoplasmosis)
increased significantly, but only from 13.1 months to 25.5 months (p=0.013).
Conclusion: Although there has been a proportional increase in ADC at AIDS diagnosis, survival following ADC has improved markedly in the era of HAART The
relatively smaller and greater effects of HAART on ADC incidence and survival,
respectively, when compared to other AIDS illnesses may increase the prevalence
of ADC in many settings.
Presenting author: Gregory Dore, NCHECR, Level 2, 376 Victoria Street, Darlinghurst NSW 2010, Australia, Tel.: [phone], Fax: [phone], Email: [email protected]
ThOrC1447 Survival of perinatally HIV-infected children in
Canada
S.M. King', J. Forbes2, J. Singer3, N. Lapointe4, L. Samson5, J. Embree6,
W. Vaudry7. 1IHospital for Sick Children and University of Toronto, Hospital for
Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada;
2University of British Colombia, Vancouver, Canada; 3Canadian HIV Trials
Network, Vancouver, Canada; 4 Universityde Montreal, Montreal, Canada;
5University of Ottawa, Ottawa, Canada; 6University of Manitoba, Winnipeg,
Canada; 7University of Alberta, Edmonton, Canada
Objectives: To determine the cumulative survival and factors affecting survival
among perinatally HIV-infected children in Canada.
Methods: The Canadian Perinatal HIV surveillance program, started in 1992, is
an active surveillance program in which pediatric centres and HIV clinics, where
HIV-exposed or HIV-infected infants receive care, provide a yearly update on their
perinatal cases.
This national surveillance program reports on approximately 95% of the diagnosed HIV-exposed infants. Information collected includes; maternal data - risk
factors for HIV acquisition, country of birth, ethnicity and ART in pregnancy; infant data - place and date of birth, infant's ART and HIV outcome. Cumulative
data to December 2001 is reported. Combination antiretroviral therapy became
available for children in Canada in 1996, so survival for birth cohorts pre- and
post-1996 are compared.
Results: The cumulative number of identified perinatally HIV-infected children in
Canada is 394. For infants who were symptomatic at diagnosis (A,B or C) their
five-year survival is to 64.4% compared to 89.8% for those who were either identified prior to delivery (E) or were asymptomatic at diagnosis (N), p<0.001. The 5
year survival rates for infants born before 1996 compared to those born in 1996
or later are 71.7% vs 91.7%, p = 0.002.
Conclusion: The 5yr survival rates pre-1996 are similar to those reported in the
United States (75%), Italy (75%) and France (65%) for the same period. One
study in the United States indicated that since 1996 there has been a decrease in
mortality to 0.7% per year, which would be similar to these findings. This improvement in survival is likely due to several factors, such as, identification of infants
before symptoms develop, prevention of PCP with chemoprophylaxis and use of
combinations of antiretroviral agents.
Presenting author: Susan King, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada, Tel.: [phone], Fax: +416 813
5032, E-mail: [email protected]

ThOrC1448 Gender, disease progression and response to
HAART
J.B. Hubert1, C. Rouzioux2, F. Boufassal, J.F. Delfraissy3, L. Meyer', The
SEROCO Study Group4. 1lnserm Unitd 292, Service d'Epidemiologie, CHU de
Kremlin-Bic6tre, Inserm unitd 292 - H6pital de Bic6tre, Secteur bleu - Porte 26,
82 rue du gendral Leclerc, 94276 le Kremlin-Bic6tre cedex, France; 2Laboratoire
de Virologie, CHU Necker-Enfants Malades, Paris, France; 3Service de
Mddecine Interne, CHU de Kremlin-Bic6tre, Le Kremlin-Bic6tre, France; 4Inserm
Unite 292, Service d'Epidemiologie, CHU de Kremlin-Bic6tre, Le Kremlin-Bicdtre
Background: It has been recently suggested to revise downwards recommendations for viral load (VL) thresholds to initiate HAART in women. We investigated
whether men and women have a similar risk of progression at an equal level of
VL 1) during the natural course of the infection, 2) after initiation of HAART
Methods: Patients:
1) 424 seroconverters (SC) enrolled in the SEROCO cohort; median follow-up
before 1996, 75 months.
2) 531 individuals who started HAART during their follow-up in the SEROCO cohort; median follow-up after HAART initiation, 37 months. The effect of gender
was assessed on 1) time since infection to clinical AIDS, CD4 cell count <200/I1,
and death using Cox models, 2) evolution of CD4 and VL at 12 months after
HAART initiation; time since initiation of HAART to an AIDS event.
Results: 1) As compared to male SC, female SC (n=94; 22%) were younger
(28 vs 31 yrs), had higher CD4 cell count (mean, 688 vs 569 cells/Rl) and lower
VL (3.7 vs 4.1 log copies/ml) at enrolment. Progression to AIDS was lower in
women (RR=0.4, p<0.01). This benefit persisted after adjustment for age, baseline CD4 and VL (RR=0.5, p<0.04). Since differences in CD4 and VL across
sexes tended to diminish after 6 yrs of infection, we fitted models with updated
CD4 and VL: RR=0.6 (p<0.11), 0.8 (p=0.4) and 0.7 (p=0.3) for progression to
AIDS, CD4<200/l1 and death, respectively 2) At start of HAART, women (n=167,
31%) and men had similar CD4 (246 vs 247) and VL (4.1 vs 4.2). They experienced similar increase in CD4 (102 vs 86) and decrease in VL (0.9 vs 1.2) at
12 months. At least one AIDS event occurred in 50 patients (3.6 per 100 PY),
unrelated to sex (RR=1.1, p=0.8). Adjustment for age, CD4 and VL at HAART
initiation, and prior use of ART did not modify this result.
Conclusion: Our data do not support evidence that women experience more
rapid progression of HIV infection than men at an equal level of VL, during the
natural course of the disease and after HAART initiation.
Presenting author: Jean-Baptiste Hubert, Inserm unite 292 - H6pital de Bicitre,
Secteur bleu - Porte 26, 82 rue du genbral Leclerc, 94276 le Kremlin-Bicitre
cedex, France, Tel.: [phone], Fax: [phone], E-mail: hubert
@vjf.inserm.fr
ThOrC1485 Mobility is an important risk factor for HIV infection
in rural Tanzania
R. Isingqol, M. Urassa1, C. Kishamawel, E. Knoops2, H. Voeten2, G. Mwalukol,
T. Boerma3, B. Zaba4. TANESA, Centre for Population Studies, London School
of Hygiene & Tropical Medicine, 49-51 Bedford Square, London WC1B 3DP,
Tanzania; 2Erasmus University Rotterdam, The Netherlands; 3Measure
Evaluation, Chapel Hill, United States; 4 CPS, London, United Kingdom
Background: Mobility facilitates sexual mixing and is considered a key risk factor in the spread of HIV, but few studies assess incidence effects. Data from a
longitudinal community-based study in Kisesa ward, Mwanza region are used to
measure its direct contribution to HIV infection.
Methods: Socio-demographic characteristics, sexual partnership information and
HIV status collected for 11,340 persons in 1994/5, 1996/7 and 1999/0 were linked
to migration histories from 13 rounds of demographic surveillance between 1994
and 2000. Incidence estimates are based on 21,728 person-years of observation contributed by 5,647 individuals aged 15-49 tested at least twice. Hazard
analysis with time varying co-variates is used to assess the impact of mobility on
incidence.
Results: Overall 58% of men and 70% of women changed residence at least
once; for those with two HIV tests mobility rates were 50% and 61% respectively. Among the mobile 61% of men and 25% of women reported non-marital
sexual partnerships, compared to 50% and 12% respectively for non-movers.
270 sero-converters were identified, yielding an overall annual incidence of 1.2%
(confidence limits 1.1-1.4). Incidence was significantly higher in mobile individuals: 1.8% (1.6-2.1), than for non-movers: 0.6% (0.5-0.8). Restricting mobility to
movement occurring before sero-conversion, and allowing for time variant effects
of risk factors such as age, marital status, place of residence and male circumcision, suggests that mobility per se significantly increases the risk of HIV infection,
with hazard ratios of 5.1 (3.5-7.5) for males and 4.9 (3.5-7.5) for females independent of the excess risk associated with non-marital partnerships, for which male
and female hazard ratios were 2.3 (1.3-2.7) and 2.2 (1.5-3.2).
Conclusion: Mobility is associated with sexual partnership formation and substantially increased HIV infection rates. Incomplete reporting of non-marital part
nerships may complicate risk attribution.
Presenting author: basia zaba, Centre for Population Studies, London School of
Hygiene & Tropical Medicine, 49-51 Bedford Square, London WC1B 3DP, United
Kingdom, Tel.: +(44) 207 299 4699, Fax: +(44) 207 299 4637, E-mail: basia.zaba
@ lshtm.ac.uk