Abstract Book Vol. 1 [International Conference on AIDS (16th: 2006: Toronto, Canada)]

Monday
14 August
O-al
ALstr ct
Sessioris

respectively. > 98% received NNRTI-based ART. 71.1% had been on ART < 12
months, while 22.2% and 6.8% were under therapy for 12-24 months and > 24
months, respectively. Overall, mortality on ART was 6.0% and lost to follow-up
(LFU) was 4.3%. Probability of survival on ART (not deceased or LFU) at 6, 12,
18, and 24 months was 0.91 (IQR: 0.90-0.93), 0.88 (0.86-0.90), 0.86 (0.84 -0.88), and 0.84 (0.80-0.87), respectively. Median CD4% and CD4% gain at 6
months were 17.5 (11.1-24.5) and 8.9 (5.4-13.6); at 12 months: 20.2 (14.2 -27.7) and 11.4 (4.4-16.9); at 18 months: 21.5 (12.2-28.4) and 13.8 (5.7 -20.0); and at 24 months: 19.8 (11.2-27.6) and 12.5 (5.2-19.1), respectively.
4.3% (79/1824) had at least one drug stopped due to side effects.
Conclusions: In RPCs, children < 13 years of age appear to do well on ART,
based on survival and immunologic criteria, with a low incidence of significant
adverse side effects. Continuing efforts at treatment scale-up should include
increased attention on ART for children.
MOABO205
Forecasting HIV treatment needs in children to
guide policy, planning and scale up: a multi-country
experience from India, Malawi, Cameroon, Rwanda
and Cote d'Ivoire
N. Ngashi', C. Luo1, D. Mulenga2, K. Little3, R. Gass'. 'United Nations
Children's Fund (UNICEF), Health, New York, United States, 2United Nations
Children's Fund (UNICEF), HIV/AIDS, New York, United States, 3Institute of
Child Health, Epidemiology, London, United Kingdom
Background: Globally, less than 5% of people receiving ART are children (WHO
2005). Factors of low access include lack of appropriate drug formulations,
diagnostic technologies and competencies to treat children. Furthermore,
countries lack estimates of the number of HIV-infected children eligible for
HIV treatment from a life cycle approach. UNICEF supported the estimation of
disease burden and treatment needs in five countries.
Methods: The London Institute of Child Health developed for UNICEF, a model
for estimating the number of children eligible for HIV treatment, based on WHO
clinical criteria, HIV prevalence in ANC settings, MTCT rates, infant and child
mortality rates, and access to PMTCT services, cotrimoxazole prophylaxis and
ART. The model was validated in Rwanda, Cameroon, Cote d'Ivoire, Malawi and
India, in collaboration with country epidemiologists, WHO and UNAIDS.
Results: In India, 2% (616) out of 27,300 children requiring ART receive it,
4% (1,149/28,000) children in Malawi, 5% (427/9,000) in Cameroon, 11%
(710/6,400) in Rwanda and 13% (2,309/18,400) in Cote d'Ivoire. With the
exception of Cote d'Ivoire and Malawi, cotrimoxazole prophylaxis is not included
in the package of scaled up care interventions for children in all countries. The
model demonstrated that cotrimoxazole prophylaxis could reduce child mortality
by 43% (36.8% vs. 21%). Where early infant diagnosis is available, further
mortality reductions could be achieved with a combination of cotrimoxazole and
ART: from 36.8% to 12.4% by year one and 75% to 26% by year 10.
Conclusions: National programs have started using this information to revise
national policies on pediatric care and to define realistic population-based
targets, scale up plans and strategies for achieving these targets. The estimation
of country-specific burden of disease and treatment needs is a powerful tool for
advocacy and scale up planning and should be roll out.
MOABO301

by decreasing differences in developmental functioning between HIV+ and HIVchildren during the first three years of life.
MOABO302
Neurocognitive impairment, symptomatic peripheral
neuropathy and depression are highly prevalent
in HIV-infected outpatients within the Asia Pacific
region: findings of the Asia Pacific NeuroAIDS
consortium (APNAC) study
E. Wright', B. Brew2, S. Kongsaengdao3, A. Arayawithchanont4,
K. Samintarapunyas, D. Imran', W. Lun', A. Kamarulzaman', P. Li', G. Tau",
S. Vonthanak", C. Sarim", S. Huffam", K. Kishore2, S.T. Ali'3, K. Robertson4,
L. Lals, M. Lim"s, D. Devadsons, P. Bain2, R. Dwyer'16, G. McCormack', C.
Scholten', C. Cherry", J. McArthur'', S. Wesselingh". 'The Alfred Hospital,
Melbourne, Australia, 2St Vincent's Hospital, Sydney, Australia, 3Rajavithi
Hospital, Bangkok, Thailand, 4Sappasithiprasong Hospital, Ubon Ratchathani,
Thailand, sLampang Hospital, Lampang, Thailand, 6Cipto Mangunkusomo
Hospital, Jakarta, Indonesia, 7Ditan Hospital, Beijing, China, "University of
Malaya Medical Centre, Kuala Lumpur, Malaysia, 9Queen Elizabeth Hospital,
Hong Kong, Hong Kong, "Port Moresby Hospital, Port Moresby, Papua New
Guinea, "National Centre HIV/AIDS, Dermatology and STD, Phnom Penh,
Cambodia, 2Fiji School Medicine, Suva, Fiji, "3Ministry of Health, Suva, Fiji,
14University of North Carolina at Chapel Hill, Chapel Hill, United States,
"Burnet Institute, Melbourne, Australia, 16Monash University, Melbourne,
Australia, "Johns Hopkins School Medicine, Baltimore, United States
Background: The prevalence of HIV-related neurocognitive impairment
(NCI) in the AP region is unknown. HIV dementia and symptomatic peripheral
neuropathy (PN) are reportedly uncommon. APNAC undertook a cross-sectional
study in 8 countries of the region to determine prevalence of NCI and PN. Study
hypothesis: NCI and PN are prevalent but under-diagnosed. Results from sites
in Thailand, Indonesia, China and Malaysia are presented.
Methods: HIV positive outpatients were screened for NCI, PN and depression.
The neurocognitive test battery comprised grooved pegboard, finger tapper,
timed gait and category fluency. NCI definition: normal: all tests > -1SD;
equivocal: <-1SD on 1 test; mild-moderate: <-ISD on 2 tests, or <-2SD on
1 test, up to -4SD total; severe: >-4SD. Results were analysed using US and
APNAC site control norms. The ACTG PN screening tool was used. Definite PN:
symptoms + vibration at great toes <10 seconds + absent ankle reflexes;
probable PN (pPN): symptoms + one of remaining 2 criteria. CES-D 20 was
used for depression screening. Student t-tests and tests for comparison of
proportions were used.
Results: 504 outpatients were enrolled and 220 evaluated: median age 33
years; male 73%; median CD4 cell count 177/mL; prior AIDS 61%; 148/220
patients (67%) were receiving HAART with 37% receiving didanosine, stavudine,
or both. Mild-moderate and severe NCI were found in 24% and 64% patients,
respectively. PN and pPN were found in 6.3% and 52% patients, respectively.
32% of patients reached the CES-D cutoff. Patients with severe NCI were older
and had lower nadir CD4 counts than those without NCI (p<0.05). Patients
with pPN were significantly likelier to have used dideoxynucleosides than those
without PN (p<0.05). <5% of patients received pain relief or antidepressants.
Conclusions: HIV-related NCI, symptomatic PN and depression are common
under-diagnosed conditions in HIV-infected outpatients at sites in Thailand,
Indonesia, China and Malaysia and require further study

Neurodevelopmental functioning in HIV-infected
children before and after the introduction of highly
active antiretroviral therapy (HAART) OAB303
Trends and risk factors for community-acquired

J. Lindsey', K. Malee2, P. Brouwers3, M. Hughes'. 'Center for Biostatistics in
AIDS Research, Biostatistics, Harvard School of Public Health, Boston, United
States, 2Children's Memorial Hospital/Northwestern University Feinberg School
of Medicine, Infectious Diseases/Child and Adolescent Psychiatry, Chicago,
United States, 3Division of AIDS & Health and Behavior Research NIMH, NIH,
Center for Mental Health Research on AIDS, Rockville, United States
Background: Recent investigations suggest a reduced prevalence of
encephalopathy in HIV-infected (HIV+) children. Howeverthe factors responsible
for this decline remain unclear. Early use of highly active antiretroviral therapy
(HAART) in children has improved survival and immunologic status. The
purpose of this investigation was to determine the impact of protease inhibitor
(PI)-containing HAART regimens on neurodevelopmental functioning during the
first three years of life.
Methods: This study examined profiles of neurodevelopmental functioning, as
measured by the Bayley Scales of Infant Development, in children participating
in a cohort of long-term outcomes (PACTG 219C). Random effects models were
used to compare mental and motor functioning during the first three years of
life by HIV infection status and before and after initiating HAART with a PI.
Results: In the pre-HAART era (before 6/1997), mean mental (85) and motor
(77) scores in HIV+ (n=54) infants less than one year of age were significantly
lower than among HIV- infants (105 and 107, n=221) and remained lower up to
two years, with similar negative trajectories. After HAART became available, the
mean mental (85) and motor (83) functioning of HIV+ infants (n=91) before
one year of age were still significantly lower than that of HIV- infants (92 and
90, n=838). However, against a not unexpected background of declining scores
among the HIV- infants (-6.2 points/yr for mental scores and -1.4 points/yr for
motor scores), there was evidence of limited improvement in the HIV+ infants
relative to their un-infected peers (declines of only -3.2 points/yr for mental
scores (p=0.01) and increases of +1.2 points/yr for motor scores (p=0.03)).
Conclusions: Suppression of HIV-1 RNA levels and subsequent benefits in
survival and immunologic status brought about by HAART have been followed

pneumonia among HIV-lnfected and HIv-unlnTected
intravenous drug users
K. Stein', J. Astemborski2, S. Mehta2, D. Vlahov3, N. Galai2, G. Kirk2. 'Johns
Hopkins University, Department of Medicine, Baltimore, United States, 2Johns
Hopkins University, Department of Epidemiology, Baltimore, United States,
3New York Academy of Medicine, New York, United States
Background: Both HIV infection and injection drug use increase risk for
community-acquired pneumonia (CAP). We examined CAP diagnoses among
HIV-infected and uninfected injection drug users (IDUs) in the ALIVE study.
Methods: From1988 to 2004, bacterial CAP diagnoses were confirmed through
record abstraction to have radiographic infiltrates and at least 1 finding of
productive cough, fever, or leukocytosis; 38% had positive sputum or blood
cultures. Cox time-dependent regression models were used to estimate risk for
initial CAP diagnosis.
Results: Among 2,833 IDUs (1,150 HIV-infected and 1,683 HIV-uninfected),
346 developed CAP; 273 were HIV-infected and 73 HIV-uninfected. CAP
incidence rates declined from the pre-HAART (1988-1996) to early (1997-2000)
and current (2001-04) HAART periods (2.2/100p-yrs, 1.6/100p-yrs, 0.8/100pyrs, respectively); similar trends were observed stratified by HIV status or CD4
levels. In multivariate analyses, time-period was not significantly associated
with CAP but smokers (adjRH 1.9; 95% CI, 1.3-2.8) and females (1.3; 1.0 -1.7) had increased risk. Recent injection drug use (1.5; 1.2-1.9) and related
complications including prior sepsis/bacteremia (4.6; 3.1-6.8) or endocarditis
(2.7; 1.8-4.1) were strongly associated with CAP, while inhaled drugs had
limited effect. HIV infection was strongly associated with CAP (8.5; 6.5-11.1)
with increasing risk observed with progressively lower CD4 counts. Further,
HIV-infected with CD4 >500 still had significantly increased risk compared to
HIV-uninfected (4.1; 2.7-6.1). CAP predictors among pre-HAART HIV-infected
participants were similar to HIV-uninfected (female, smoking, heavy drug use,
endocarditis/bacteremia) except for the strong CD4 effect seen in all time
XVI INTERNATIONAL AIDS CONFERENCE * 13-18 AUGUST 2006 * TORONTO CANADA * ABSTRACT BOOK VOLUME 1