Causes of death in HIV/TB coinfected patients: results from the HIV/TB collaborative study
Presented by Daria Podlekareva (Denmark).
A. Panteleev1, D. Podlekareva2, W. Bannister3, V. Riekstina4, A. Rakhmanova5, F. Post6, J.M. Miro7, H. Furrer8, M. Bruyand9, R.F. Miller10, E. Girardi11, M. Losso12, J. Toibaro12, J. Caylá13, N. Obel14, A. Skrahin15, N. Chentsova16, J.D. Lundgren2,14, A. Mocroft3, O. Kirk2, HIV/TB Study Group
1St. Petersburg TB Hospital #2, St. Petersburg, Russian Federation, 2University of Copenhagen, Copenhagen HIV Programme, Copenhagen, Denmark, 3University College London, Medical School Royal Free Campus, Department of Infection and Population Health, London, United Kingdom, 4State Agency of TB and Lung Diseases, Riga, Latvia, 5Botkin Hospital of Infectious Diseases, St. Petersburg, Russian Federation, 6King's College London School of Medicine, London, United Kingdom, 7Hospital Clinic - IDIBAPS, University of Barcelona, Barcelona, Spain, 8University Hospital of Bern, Bern, Switzerland, 9INSERM, U 897, 'Epidemiology and Biostatistics', Bordeaux, France, 10Mortimer Market Centre, London, United Kingdom, 11Instituto Nazionale Malattie Infettive L Spallanzani, Rome, Italy, 12Hospital JM Ramos Mejia, Buenos Aires, Argentina, 13Servicio de Epidemiología, Agencia de Salud Pública de Barcelona, CIBER Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain, 14Rigshospitalet, Copenhagen, Denmark, 15Research Institute of Pulmonology and Pulmonary Tuberculosis, Minsk, Belarus, 16Kiev City AIDS Centre, Kiev, Ukraine
Background: One-year mortality rate of HIV/TB-patients in Eastern Europe (EE) is 3-5 fold higher than in Western Europe and Argentina (WEA).
Objectives: To assess causes of death (COD) in HIV/TB-patients and to compare COD across regions as well as to investigate factors associated with TB-related death.
Methods: CoDe procedure (http://www.chip.dk/CoDe/tabid/55/Default.aspx) was used to ascertain underlying and immediate COD. Factors associated with TB-related death (underlying COD) were assessed using logistic regression.
Results: 201 of 1075 HIV-patients diagnosed with TB between 1/2004-12/2006 died within one year of TB diagnosis. Of these, 167 (83%) had CoDe forms available.
142 and 25 deaths occurred in EE and WEA respectively. In 124 (74%) cases, TB was the underlying COD, with TB-sepsis/multi-organ failure (31%), TB-meningitis (26%), and pulmonary TB (10%) the commonest immediate COD. In EE, TB was more common as underlying COD compared to WEA, whereas non-TB-AIDS, non-AIDS-infections and other/unknown causes were more common underlying COD in WEA (Table). Combination antiretroviral therapy (cART) at the time of death was less common in EE (11% vs. 32%, p=0.012).
In multivariable models, factors associated with TB-related death were EE-region [OR for EE vs WEA 6.11 (95% CI 1.29-29.04), p=0.023], and having disseminated TB [OR disseminated vs. other 3.06 (1.27-7.37), p=0.013].
Summary: Death in HIV/TB-patients in EE was more often due to advanced TB compared to WEA. Prompt TB diagnosis and timely introduction of optimal TB-treatment and cART may limit dissemination of TB and reduce the risk of death in HIV/TB-patients, particularly in EE.
Back to the session -
Back to the Programme-at-a-Glance