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

MOPE0394
Contrasting patterns of HIV and TB co-morbidity in
Europe
I. Devaux', J. Alix', A. Nardone', D. Falzon2. 'Institut de Veille Sanitaire,
EuroHIV, St Maurice, France, 'Institut de Veille Sanitaire, EuroTB, St Maurice,
France
Background: Tuberculosis (TB) is the commonest AIDS indicative disease in
Europe. Monitoring HIV prevalence in TB patients is useful in assessing the
spread of HIV in the population and its impact on the burden of TB.
Methods: HIV/AIDS and TB surveillance data in the WHO European Region are
collated annually by two projects: EuroHIV and EuroTB. We present data for
1998-2004 from countries in the European Union (EU, which includes the Baltic
States) and the former Soviet Union (FSU). Indicators used are (i) AIDS cases
initially defined by TB (AITB) (31/37 countries), (ii) the proportion of all TB
notifications being AITB, and (iii) the prevalence of HIV seropositivity among
TB notifications (21/37 countries).
Results: In 2004, of 12,264 AIDS cases reported in 31 European countries,
3,813 had AITB (24% in EU, 54% in FSU). The absolute number of AITB has
decreased in most EU countries since the late 1990s, but there was eightfold
increase in Ukraine between 1998 (172 cases) and 2004 (1,397 cases). In
the EU in 2004, 3% of TB notifications were diagnosed AITB (country range:
0-11.6%), whilst in the FSU, it represented 0.5% (range: 0-3.6%), the large
majority of which deriving from Ukraine. HIV seropositivity among TB cases
has remained stable at low levels in most of the 17 EU countries with data
(mean for latest year: 4 %), with highest levels reported from Portugal (15.7%
in 2004). A steady increase in prevalence has been noted in Estonia (0.1 % to
4.4% between 1999 and 2004) and Latvia (0.5% to 2.5%).
Conclusions: The recent increase in HIV/TB co-morbidity in Ukraine and Baltic
States necessitates reinforcement of control measures for both diseases in
the ex-Soviet countries. We recommend all countries should not only monitor
HIV and TB co-morbidity using suitable surveillance systems and report these
results.
MOPE0395
Tuberculosis (TB) incidence by HAART and isoniazid
prophylactic therapy (IPT) in HIV-infected patients
in Rio de Janeiro, Brazil
J.E. Golub', V. Saraceni2, S. Cavalcante3, A.G. Pacheco3, B. King',
R.D. Moore', R.E. Chaisson', L.H. Moulton', B. Durovni'. 1Johns Hopkins
Center for Tuberculosis Research, Baltimore, United States, 2Rio de Janeiro
Secretariat of Health, Rio de Janeiro, Brazil, 3Rio de Janeiro Secretariat of
Health, National School of Public Health - FIOCRUZ, Rio de Janeiro, Brazil
Background: Brazil provides combination antiretroviral therapy free of charge
to HIV-infected patients who meet clinical criteria. IPT is recommended for
patients co-infected with Mycobacterium tuberculosis, but uptake is very low.
The THRio study is a cluster randomized trial on the impact of implementing IPT
in 29 HIV clinics in Rio de Janeiro, Brazil in an effort to reduce TB incidence. We
report preliminary analysis from baseline data for 6,391 patients.
Methods: We calculated TB incidence rates among patients seen between
September2003-September2005. Person time prior to TB diagnosis was
calculated and Poisson regression compared incidence rates (IR) between
patient groups.
Results: There were 336 cases of TB diagnosed in the time period under
study; 12 excluded because exact HIV or TB diagnosis dates were unknown.
Of 324 evaluable TB patients, 66% were male, and the incidence was greater
for men compared to women (IRR=1.29,p=0.04). Rates by age did not
differ. Only 787(12.8%) received IPT, of whom 76% completed 6 months of
treatment. TB incidence in IPT patients was 0.7/100 PY vs. 3.0/100PY in nonIPT patients (IRR=0.27,p<0.001); 0.6/100PY in IPT completers vs 2.0/100PY
in non-completers (IRR=0.29,p=0.07). Of 1673 patients who never received
HAART, 203 TB cases occurred (IR=6.6/100PY) compared to 121 among 4495
patients who received HAART (IR=1.4/100PY;IRR=0.22,p<0.001); HAART>1
year was significantly protective (IRR=0.13,p<0.001). Baseline CD4 count
was not associated with TB risk. Compared to patients receiving neither IPT
nor HAART (IR= 7.4/100PY); HAART/no IPT patients had a decreased risk
(IR=1.5/100PY;IRR=0.20,p<0.001); IPT/no HAART patients a decreased risk
(1.0/100PY;IRR=0.13,p<0.001); and patients receiving both had the lowest
(0.6/100PY; IRR=0.08, p<0.001).
Conclusions: While HAART reduces TB rates in this population, use of IPT
is very effective at further reducing TB incidence in co-infected patients. In
populations where HAART is widely implemented, concurrent IPT can have a
significant impact in reducing TB incidence.
MOPE0396
Early diagnosis of pulmonary tuberculosis in HIV
positive children by scoring system during regular
outpatient follow-up

S. Bhattacharyya. Medical College, Calcutta, India, Department of Pediatrics,
Kolkata, India
Background: Tuberculosis (TB) is one of the commonest opportunistic infection
among HIV+ children. Rate of TB is 30 times higher and often progressive.
Establishing diagnosis of TB is difficult because of falsely negative tuberculin
test(TST), difficult bacterial isolation and pulmonary manifestations may mimic
HIV related conditions.An OPD based scoring system has been developed for

early suspicion, evaluation and treatment of TB in HIV+ children.
Methods: Scoring was applied on 100 HIV+ (confirmed by ELISA and WB)
children (2-12yrs)(M=65,F=35) during monthly follow-up for 1.5years at
Pediatric HIV OPD.None were on ART. Total 10 scoring points were divided
as i) 2 for TST(1TU with Tween-80) >_ 10mm induration, ii) 1 for TST 6 -9 mm, iii) 3 for TB contact with adult family members, iv) 1 for persisting
cough, fever > 3 weeks, v) 3 for abnormal chest x-ray (enlarged nodes,lober
infiltration,nodules). Points>5 possibility of pulmonary TB, points _4 rule out
other HIV related causes As sputum is less likely to be produced or to contain
visible acid fast bacteria, hence was excluded.
Results: With this scoring 45 new cases were diagnosed and started therapy,
they responded significantly, clinically and radiologically with antitubercular
drugs during follow-up. Five cases were missed inspite of scoring, only to
diagnose by further investigations. In 4 cases TB was suspected by scoring
but subsequent follow-up excluded TB. In rest 46 cases TB was excluded by
scoring even with related symptomatology, they continued to be free from all
forms of TB during follow-up. None of these 97 children(3 lost in follow-up)
developed any progressive/dessiminated TB during follow-up. 95% confidence
interval(CI) of sensitivity of TB scoring (0.822 to 0.978) and 95 % of CIs of
specificity (0.845 to 0.995).
Conclusions: Early diagnosis is important to prevent dissemination.Pulmonary
TB is the commonest TB. TB scoring effectively helps in early suspicion,
diagnosis and treatment in HIV+ children in resource constraint OPD/field
setting.
MOPE0397
Expanding TB/HIV joint activities: from local
success to a national action plan
S. Akksilp', O. Karnkawinpong2, W. Wattana-amornkiat2, D. Wiriyakitjar3,
Y.-R. Lo4, P. Nunns. 1Office of Disease Prevention and Control #7, Department
of Disease Control, Ubon Ratchatani, Thailand, 2Office of Disease Prevention
and Control #7, Department of Disease Control, Ubon ratchatani, Thailand,
3Department of Disease Control, Nonthaburi, Thailand, 4World Health
Organization, New Delhi, India, 'World Health Organization, Stop TB
Department, Geneva, Switzerland
Issues: WHO advocates for TB and HIV programs to work together, but few
successful models of such collaboration have been reported from countries with
dual epidemics. Thailand is 17th on WHO's list of 22 high TB burden countries
and has > 600,000 persons living with HIV.
Description: Thailand's Office of Disease Prevention and Control Region 7
began offering HIV counseling and testing, CD4 testing, and HIV care, including
antiretroviral treatment (ART), to TB patients. The project began in February
2003 in one large province (Ubon-ratchathani, population 1.7 million) and, in
February 2005, expanded to six additional provinces in the same region (total
population 6.7 million). To evaluate the performance of TB/HIV joint activities,
we compared an 8-month cohort of 1,539 newly registered TB cases from 2003
to an 8-month cohort of 5,849 from 2005. The number of TB patients with HIV
test results, CD4 measurements, and ART prescriptions increased from 501,
101, and 39 in 2003 to 3417, 384, and 206 in 2005. The proportion of eligible
patients receiving these interventions increased from 33% to 58% (p < 0.01),
50% to 57% (p = 0.35) and 19% to 31% (p = 0.02), respectively for each
intervention. In 2003, 13% of tested patients were HIV-infected compared with
11.5% in 2005.
Lessons learned: We report the successful implementation of TB/HIV joint
activities covering a large area in Thailand. The success of this activity led
Thailand's Ministry of Public Health to adopt a national TB/HIV action plan that
calls for 100% of TB patients to receive HIV counseling, > 75% of HIV-infected
persons to receive TB screening, and all TB patients with HIV to have access
to ART.
Recommendations: TB/HIV joint activities implemented at the local level
can be used to develop national policies that promote greater collaboration
between TB and HIV programs.
MOPE0398
High rate of negative tuberculin skin tests in the
group of HIV-positive patients who underwent full
BCG-vaccination schedule
J.D. Kowalska, E. Firlag-Burkacka, A. Horban. AIDS Diagnosis and Therapy
Center, Out-Patient Clinic, Hospital for Infectious Diseases, Warsaw, Poland
Background: Since 1976 vaccination program in Poland includes 2-5 BCG
doses. Each 18-year-old citizen is expected to be TST reactive (5-12mm
diameter) in the result of immunization schedule. Although TST is the
standard of TB screening for HIV(+) population, its usefulness in the group of
immunocompromised patients previously BCG vaccinated is of unclear benefit.
Therefore we decided to evaluate response to PPD and the rate of active TB
(ATB) in TST(+)group in Warsaw Cohort.
Methods: Information on TST results was obtained from Warsaw Out-Patient
Clinic's database, prospectively designed to collect it. Chi-squared and Kruskal
Wallis tests were used for univariable analysis as appropriate. In multivariable
logistic regression CD4 count and infection mode were used as covariates. CI
of 95% was accepted.
Results: There were 535 patients screened with TST. Only first TST result for
each patient was included into the analysis. 36 patients (6,7%) were found
to be TST(+), among them 8 were diagnosed and treated for ATB during the
follow-up (22,2%). Mean age was 33,26 (range 16,0-68,3) years. CD4 count
was significantly higher for TST(+), median 389 cells/mm3 than for TST(-)

Monday
14 August
Lxh<b.O~

XVI INTERNATIONAL AIDS CONFERENCE * 13-18 AUGUST 2006 * TORONTO CANADA * ABSTRACT BOOK VOLUME 1