Tuberculosis and the risk of opportunistic infections and cancers in HIV-infected patients starting ART in Southern Africa

Authors

  • Lukas Fenner,

    Corresponding author
    • Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland
    Search for more papers by this author
  • Stewart E. Reid,

    1. Centre for Infectious Disease Research in Zambia (CIDRZ), Lusaka, Zambia
    2. Department of Medicine, University of North Carolina, Chapel Hill, NC, USA
    Search for more papers by this author
  • Matthew P. Fox,

    1. Health Economics and Epidemiology Research Office, Department of Medicine, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
    2. Center for Global Health and Development, Boston University, Boston, MA, USA
    Search for more papers by this author
  • Daniela Garone,

    1. Khayelitsha ART Programme, Médecins Sans Frontières, Cape Town, South Africa
    Search for more papers by this author
  • Maureen Wellington,

    1. Newlands Clinic, Harare, Zimbabwe
    Search for more papers by this author
  • Hans Prozesky,

    1. Division of Infectious Diseases, Department of Medicine, University of Stellenbosch and Tygerberg Hospital, Cape Town, South Africa
    Search for more papers by this author
  • Marcel Zwahlen,

    1. Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland
    Search for more papers by this author
  • Michael Schomaker,

    1. Centre for Infectious Disease Epidemiology and Research, School of Public Health and Family Medicine, University of Cape Town, Cape Town, South Africa
    Search for more papers by this author
  • Gilles Wandeler,

    1. Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland
    2. Department of Infectious Diseases, Bern University Hospital, Bern, Switzerland
    Search for more papers by this author
  • Nzali Kancheya,

    1. Centre for Infectious Disease Research in Zambia (CIDRZ), Lusaka, Zambia
    Search for more papers by this author
  • Andrew Boulle,

    1. Centre for Infectious Disease Epidemiology and Research, School of Public Health and Family Medicine, University of Cape Town, Cape Town, South Africa
    Search for more papers by this author
  • Robin Wood,

    1. The Desmond Tutu HIV Centre, Institute for Infectious Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa
    Search for more papers by this author
  • German Henostroza,

    1. Centre for Infectious Disease Research in Zambia (CIDRZ), Lusaka, Zambia
    2. Department of Medicine, University of Alabama, Birmingham, AL, USA
    Search for more papers by this author
  • Matthias Egger,

    1. Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland
    2. Centre for Infectious Disease Epidemiology and Research, School of Public Health and Family Medicine, University of Cape Town, Cape Town, South Africa
    Search for more papers by this author
  • IeDEA Southern Africa


Corresponding Author Lukas Fenner, Institute of Social and Preventive Medicine (ISPM), University of Bern, Finkenhubelweg 11, CH-3012 Bern, Switzerland. Tel.: +41 31 631 3867; Fax: +41 31 631 3520; E-mail: lfenner@ispm.unibe.ch

Abstract

Objectives

To investigate the incidence of selected opportunistic infections (OIs) and cancers and the role of a history of tuberculosis (TB) as a risk factor for developing these conditions in HIV-infected patients starting antiretroviral treatment (ART) in Southern Africa.

Methods

Five ART programmes from Zimbabwe, Zambia and South Africa participated. Outcomes were extrapulmonary cryptococcal disease (CM), pneumonia due to Pneumocystis jirovecii (PCP), Kaposi's sarcoma and Non-Hodgkin lymphoma. A history of TB was defined as a TB diagnosis before or at the start of ART. We used Cox models adjusted for age, sex, CD4 cell count at ART start and treatment site, presenting results as adjusted hazard ratios (aHR) with 95% confidence intervals (CI).

Results

We analysed data from 175 212 patients enrolled between 2000 and 2010 and identified 702 patients with incident CM (including 205 with a TB history) and 487 with incident PCP (including 179 with a TB history). The incidence per 100 person-years over the first year of ART was 0.48 (95% CI 0.44–0.52) for CM, 0.35 (95% CI 0.32–0.38) for PCP, 0.31 (95% CI 0.29–0.35) for Kaposi's sarcoma and 0.02 (95% CI 0.01–0.03) for Non-Hodgkin lymphoma. A history of TB was associated with cryptococcal disease (aHR 1.28, 95% CI 1.05–1.55) and Pneumocystis jirovecii pneumonia (aHR 1.61, 95% CI 1.27–2.04), but not with Non-Hodgkin lymphoma (aHR 1.09, 95% CI 0.45–2.65) or Kaposi's sarcoma (aHR 1.02, 95% CI 0.81–1.27).

Conclusions

Our study suggests that there may be interactions between different OIs in HIV-infected patients.

Abstract

Objectifs

Etudier l'incidence de certaines d'infections opportunistes et de cancers et le rôle d'antécédents de tuberculose (TB) comme facteur de risque de développer ces affections chez les patients infectés par le VIH commençant un traitement antirétroviral (ART) en Afrique australe.

Méthodes

Cinq programmes ART au Zimbabwe, en Zambie et en Afrique du Sud ont participé à l’étude. Les résultats mesurés étaient la cryptococcose extra pulmonaire (CM), la pneumonie due à Pneumocystis jiroveci (PCP), le sarcome de Kaposi et le lymphome non hodgkinien. Un antécédent de TB a été défini comme un diagnostic de TB avant ou au début de l’ART. Nous avons utilisé des modèles de Cox ajustés pour l’âge, le sexe, le nombre de cellules CD4 au début de l’ART et le site de traitement, en présentant les résultats sous forme de rapports de risque ajustés (aHR) avec des intervalles de confiance (IC) à 95%.

Résultats

Nous avons analysé les données de 175.212 patients inscrits entre 2000 et 2010 et identifié 702 patients avec un incident de CM (dont 205 avec une histoire de TB) et 487 avec un incident de PCP (dont 179 avec une histoire de TB). L'incidence pour 100 personnes années au cours de la première année sous ART était de 0,48 (IC95%: 0,44–0.52) pour la CM; 0,35 (IC95%: 0,32–0,38) pour le PCP; 0,31 (IC95%: 0,29–0,35) pour le sarcome de Kaposi et 0,02 (IC95%: 0.01–0,03) pour le lymphome non hodgkinien. Un antécédent de TB a été associé à la cryptococcose (aHR: 1,28; IC95%: 1,05–1,55) et à la pneumonie à Pneumocystis jirovecii (aHR: 1,61; IC95%: 1,27–2,4), mais pas avec le lymphome non hodgkinien (aHR: 1,09; IC95% : 0,45–2,65) ou le sarcome de Kaposi (aHR: 1,02; IC95%: 0,81–1,27).

Conclusions

Notre étude suggère qu'il peut y avoir des interactions entre les différentes infections opportunistes chez les patients infectés par le VIH.

Abstract

Objetivos

Investigar la incidencia de ciertas infecciones oportunistas (IO) y cánceres y su papel en la historia de la tuberculosis (TB) como factor de riesgo para el desarrollo de estas condiciones en pacientes infectados con VIH y comenzando terapia antirretroviral (TAR) en el sur de África.

Métodos

Participaron cinco programas de TAR de Zimbabwe, Zambia y Sudáfrica. Los resultados eran la enfermedad criptocócica extrapulmonar (CM), la neumonía por Pneumocystis jirovecii (PCP), el sarcoma de Kaposi y el linfoma no-Hodgkin. La historia de TB se definió como un diagnóstico de TB antes o al comienzo del TAR. Utilizamos modelos de Cox ajustados por edad, sexo, conteo de CD4 al comienzo del TAR y lugar del tratamiento, presentando los resultados como una razón de riesgos ajustada (RRa) con intervalos de confianza (IC) del 95%.

Resultados

Hemos analizado datos de 175,212 pacientes incluidos entre el 2000–2010 e identificado 702 pacientes con incidencia de CM (incluyendo 205 con una historia de TB) y 487 con incidencia de PCP (incluyendo 179 con una historia de TB). La incidencia por 100 personas-años durante el primer año de TAR era del 0.48 (IC 95% 0.44–0.52) para CM, 0.35 (IC95% 0.32–0.38) para PCP, 0.31 (IC95% 0.29–0.35) para el sarcoma de Kaposi y 0.02 (IC95% 0.01–0.03) para linfoma no Hodgkin. Una historia de TB estaba asociada con enfermedad criptocócica (RRa 1.28, IC95% 1.05–1.55) y neumonía por Pneumocystis jirovecii (RRa 1.61, IC95% 1.27–2.04), pero no con el linfoma no Hodgkin (RRa 1.09, IC95% 0.45–2.65) o sarcoma de Kaposi (RRa 1.02, IC95% 0.81–1.27).

Conclusiones

Nuestro estudio sugiere que podría haber interacciones entre diferentes IOs en pacientes infectados con VIH.

Ancillary