Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

12th World AIDS Conference

Abstracts 22125-22130 287

22125 New perspective in the treatment of multi drug resistant
tuberculosis (MDR-TB) in HIV positive patients
undergoing antiretroviral therapy
Jaime Waisman1, M.B. Bouzas2, D. Curcio', A. Karnoubi1, I. Zapiola2,
E. Santucho', R. Negroni1. ' Hospital FJ Muhiz, Uspallata 2272, 1282 Buenos
Aires; 2Unidad de Virologia Hospital "FJ. Muniz", Buenos Aires, Aregentina
Introduction: HIV infection has been one of the principal factors that has contributed to the increase number of cases of MDR-TB.
Objective: To evaluate the evolution of HIV positive patients with MDR-TB
under treatment with zidovudine (AZT), lamivudine (3TC), indinavir (INV) and
second line agents (prothionamide, cycloserine, clofazimine, ofloxacin).
Material and Methods: During 1997, 17 hospitalized patients positive for HIV-1
with microbiogical diagnosis of MDR-TB were treated with AZT+3TC+INV and
with second line agents for MDR-TB. CD4 counts and HIV viral load in plasma
was evaluated at the baseline and three months after the initiation of the therapy.
HIV RNA levels were perfomed by the nucleic acid-based sequence (NASBA QT,
Organon Teknika). CD4 counts were measured by flow cytometry.
Results: The median age was 29 years (range 24-44), 16/17 (94%) were
male. At the admission all presented fever, cough, expectoration, constitutional
symptoms with roentgenographic alterations. At baseline the median CD4 counts
was 45 cel/ul and the median for HIV viral load was 5.2 log. After 3 months of
treatment the median for CD4 was 224 cel/ul and for HIV viral load 3.17 log.
methods
Conclusion: These preliminary results indicate, that in our series of patients
the addition of antiretoviral therapy with AZT+3TC+INV to the second line tuberculostatic agents in the treatment of MDR-TB resulted in a clinical and virological
improvement in all cases. We consider that a more extensive study will be
necessary in order to reafirm these findings..
S447* / 22126 A randomized trial of rifampin/pyrazinamide for 2
months vs. INH for 12 months in HIV+ tuberculin+
adults (CPCRA004/ACTG177/CDC/PAHO)
Richard E. Chaisson1, F. Gordin2, J. Matts3, L. Garcia4, R. Hafner5, R. Obrien6.
SJohns Hopkins University School Of Medicine, 600N. Wolfe St. Carnegie 292
Baltimore, Maryland 21287 - 6220; 2VA Medical Center, Washington, DC3Univ
Minnesota Minneapolis MN; 5NIAID Bethesda MD; 6CDC Atlanta GA, USA;
41INDRE Mexico City, Mexico
Objective: To compare the efficacy of 2 months of RIF/PZA to 12 months of INH
for preventing tuberculosis (TB) in tuberculin+, HIV-infected patients (pts).
Design: Randomized clinical trial.
Methods: HIV+ pts. with PPD-tuberculin responses >5 mm were randomized
to daily INH 300 mg x 12 months or daily RIF 600 mg/PZA 20 mg/kg x 2 months
and followed for TB and death.
Results: 1583 pts. were enrolled and followed a mean of 36 months. The mean
age was 37 years and the median enrollment CD4 count was 436. Treatment was
completed by 80% of RIF/PZA pts. and 68% of INH pts. (p <.001). Outcomes
are listed in the Table.

Calendar time Southwest Southeast Central North
Europe Europe Europe Europe

P-value

09-94-09-95
09-95-09-96
09-96-09-97

13.6
2.9
3.1

1.6
2.0
1.0

0.6
0.8
0.4

1.2 0.001
0.5 -.30.001
0.4 0.001

In multivariate models, and including all time periods, patients from SWE had
a relative risk of 12.7 (95% Cl: 9.7-16.6) of acquiring tb compared with patients
from NE. If the three time-periods shown in the table were analyzed separately,
the relative risk of SWE vs. NE was 14.8 (10.7-20.5), 7.9 (4.4-14.3), and 7.7
(3.6-16.4), respectively.
Other significant risk factors in multivariate models were CD4 count at baseline
(<100 vs. >200/mm3: RR = 3.2 (2.6-4.0)), injecting drug use (RR = 1.7 (1.3-2.1),
and white vs. other ethnicities (RR = 0.4 (0.3-0.6).
Conclusion: Incidence of tuberculosis in Europe has decreased markedly in
recent years. This decrease was most pronounced in Spain and Portugal, where
the incidence of tuberculosis is higher than in other regions of Europe. A possible
role of therapeutic interventions, which could explain these changes in the epidemiology of HIV-associated tb, is currently under investigation. Other significant
risk factors were level of immunodeficiency, injecting drug use, and ethnicity.
22128 1 Tuberculosis (TB) in HIV-infected patients with CD4 below
50/1l
Immaculada Grau Garriga1, D. Podzamczert, J. Ramon2, B. Sanuyt,
F. Alcaide3, G. Rufi1, F. Gudiol1. l'nfectious Diseases Department, Hospital de
Bellvitge, Barcelona; 2Preventive Medicine, Hospital Bellvitge, Barcelona;
3Microbiology, Barcelona, Spain
Objective: To describe the clinical findings and survival in HIV+ patients with
culture-positive TB and CD4 < 50/uL.
Methods: Of 305 HIV+ patients diagnosed of TB between 1985 and 1995, 71
(23%) had CD4 < 50/uL at the time of TB and were included in the analysis.
Follow-up period was considered from TB diagnosis up to last visit or death
before December 1996. Survival rates were calculated by Kaplan-Meier method
and factors associated to mortality by Cox model.
Results: In 51 (72%) patients, TB was the first AIDS-defining disease. Median
CD4 count was 18 (3-49) cells/uL. Although 36 (51%) patients presented with
fever without focal symptoms, 61/68 (86%) had pulmonary TB confirmed by
positive cultures in respiratory samples (Ziehl-Neelsen stain showed AFB in 59%
of tested sputa). 62% had extrapulmonary TB. No strains of multidrug-resistant
MT were found. Median follow-up was 16 (1-46) months. 53 (75%) patients died
during the study period, none of them from TB. Cumulative survival rates at 12,
24 and 36 months were 71%, 33% and 13%, respectively. Only the existence of
prior AIDS diagnosis was independently associated to mortality (RR 2.3, 95%CI
1.1-4.6).
Conclusions: Our data show the relevance of suspecting TB in severely
immunosuppressed HIV-infected patients with fever, as most of them are potentially contagious. Mortality depends on advanced HIV-infection and not on TB
presentation.
448*/22129 Adherence and compliance in tuberculosis
preventive therapy trial at the JCRC in Uganda
Mary Nasira Mulindwat, D. Tumukundel, M.W. Kiyimbal, K.A. Byamukama1,
M.S. Wanyana', K. Kataliwa2, C. Opit1. Joint Clinical Research Centre, PO Box
10005, Kampala; 2Cluru-Makerere University Research Collab., Kampala,
Uganda
Objective: Find reasons for defaulting
Method: 1324 adults (8% HIV negative status) and (92% HIV positive status
with no evidence of TB were enrolled in a placebo controlled randomised TB
preventive therapy study over a period of 5 years. Recognising adherence and
compliance were one of the requirements for validity of results in this clinical
trial. Home visits, counselling and provision of transport to all were some of the
measures instituted to facilitate compliance and adherence.
Results: 1324 people were enrolled at JCRC on the study, 65.94% completed therapy. 7.7% died during the study, 7.7% died after the therapy, 18.66%
defaulted.
Reasons for defaulting were: 1. Lack of transport, 2. III health or attending to
a sick family member, 3. Change of address without notifying the JCRC home visitor, 4. Stigmatisation, 5. Fear of losing their jobs, 6. Misplacement of appointment
cards.
Conclusion: Counselling at screening and a good working relationship between clients and health care givers gives them confidence and readiness to
comply with studies. Budgeting for transport could improve compliance.
1 22130 Severe inflammatory reaction (SIR) after starting HAART
in AIDS patients treated for disseminated tuberculosis:
Role of corticosteroids
Hansjakob Furrer, R. Malinverni. Med Poliklinik Inselspital, CH-3010 Bern,
Switzerland
Case Report: AIDS patients with drug susceptible tuberculosis (MTB) usually
respond quickly after starting standard MTB-therapy. New onset of symptoms

and signs such as fever, malaise and weight loss while under successful therapy

No. Confirmed TB Confirmed/Probable TB
RIF/PZA 791 19 (0.8/100 py) 28 (1.2/100 py)
INH 792 24 (1.0/100 py) 28 (1.2/100 py)
RR (95% CI) 0.78 (0.4-1.4) 0.99 (0.6-1.7)

Death
138 (5.7/100 py)
158 (6.6/100 py)
0.87 (0.7-1.1)

Susceptibility testing was done for 21 INH pts.; 2 were INH monoresistant and
2 INH/RIF resistant. For 18 tested RIF/PZA pts., 1 was INH resistant, 1 RIF
resistant, and 1 INH/RIF resistant. Adverse events were similar in both groups,
except 12 RIF/PZA pts. had opiate withdrawal vs. 0 INH pts. and 52 RIF/PZA pts.
had Grade III or higher uric acid elevations vs. 2 INH.
Conclusion: Both INH and RIF/PZA pts. had low rates of tuberculosis with
no significant difference in efficacy. The use of RIF/PZA resulted in better overall adherence but more opiate withdrawal syndromes. RIF/PZA is a promising
alternative to INH for TB prophylaxis.
SI22127 Changes in regional differences within Europe in
incidence of Mycobacterium tuberculosis infections
among HIV-positive patients: The EuroSIDA study
Ole Kirk. Coord. Cent. Eurosida, Dept. Inf. Dis. 144 Hvidoure Univ. Hospital,
2650 Hvidoure, Denmark
Objectives: To analyze changes in incidences of tb in recent years in regions of
Europe.
Methods: Analysis of data from EuroSIDA, a prospective, European multicentre
cohort initiated in May 1994 (total n = 4491). Patients were stratified by ethnicity,
origin, transmission category, and region of Europe (Southwest (SWE; n = 600),
Southeast (SEE; n = 1112), Central (CE; n = 1133), or North (NE; n = 1646)).
Incidences of tb were assessed as cases per 100 person-years of follow-up (PY)
or in a Cox model.
Results: 124 patients were diagnosed with tb; 90 had pulmonary (tbp) and 71
extrapulmonary (tbep) involvement. Incidence of tb varied substantially according
to region (7.8 cases per 100 PY (SWE) vs. <2 cases per 100 PY in the other
regions; p < 0.001). Comparable trends were observed for tbp and tbep analyzed
separately. Incidences - shown as cases per 100 PY - in one calendar year
intervals are tabulated beneath: