SUMMARY
SETTING
Data from the mortality database, Rio de Janeiro City (RJC) Health Department, Rio de Janeiro, Brazil.
OBJECTIVES
To determine the role played by tuberculosis (TB) in Brazil’s human immunodeficiency virus (HIV) positive population, we investigated the frequency of TB as the primary cause of death among HIV-positive subjects in RJC.
DESIGN
Information about acquired immune-deficiency syndrome (AIDS) deaths from 1996 to 2005 in individuals aged >12 years was obtained from the Mortality Information System (SIM), and the cause of death was classified according to the International Classification of Diseases (ICD-10), through primary causes coded in Chapter I—B20 to B24 (HIV disease).
RESULTS
There were 8601 AIDS-related deaths in RJC between 1996 and 2005. TB was the primary cause of death in 9.0% of all AIDS-related deaths, while Pneumocystis carinii pneumonia (PCP) accounted for 4.7%. TB cases erroneously classified under other infectious diseases may have contributed to an underestimation of the number of TB deaths among HIV-positive patients.
CONCLUSION
Our study showed that TB is the leading cause of AIDS-related deaths and is responsible for twice as many deaths as PCP, in a scenario of free access to antiretrovirals. The potential benefits of TB preventive treatment and of the availability of highly active antiretroviral treatment could not be established by this analysis.
Keywords: tuberculosis, HIV, co-infection, mortality, antiretroviral
The tuberculosis (TB) epidemic is fueled worldwide by human immunodeficiency virus (HIV) infection, and HIV poses the greatest threat to TB control.1 The World Health Organization (WHO) estimated that there were 8.9 million new cases of TB in 2005. Of these, 741 000 were among HIV-infected adults. TB may have caused 1.7 million deaths in 2004 and 248 000 in TB-HIV co-infected individuals.2 The Joint United Nations Program on HIV/AIDS (acquired immune-deficiency syndrome) estimated that 2.9 million AIDS deaths would occur in 2006, 65 000 of which in Latin America.3
Brazil has an estimated 600 000 HIV-infected individuals aged between 15 and 49 years.4 According to the Brazilian Ministry of Health, 85 000 new TB cases occur annually, causing 6000 deaths.5
AIDS-related mortality decreased in Rio de Janeiro City (RJC) in the 2 years following the introduction (in 1997) of highly-active antiretroviral treatment (HAART), which is widely available free of charge. Since 1999, AIDS-related mortality has been stable.6 More than 26 000 AIDS cases have been reported since the beginning of the epidemic,7 and approximately 8000 TB cases are reported annually.8 Among TB patients seeking care in primary health units in RJC, one study showed that approximately 10% were infected with HIV.9
To determine the role played by TB among Brazil’s HIV-positive population, we investigated the frequency of TB as the primary cause of death among HIV-positive subjects in RJC. We also looked at the use of HAART by co-infected patients, aiming to use that as a proxy of access to care and to assess relationships between HAART and survival.
METHODS
Information about AIDS deaths in individuals aged >12 years was obtained from the Mortality Information System (SIM); causes of death were classified according to the International Classification of Diseases (ICD-10),10 through primary causes coded in Chapter I—B20 to B24 (HIV disease), from 1996 to 2005.
The mortality database is standardized; the information entered is collected from the declaration of death recorded by a physician after the outcome. Coding of the primary cause of death followed the ICD-10 algorithms.
Information on HAART was obtained from the RJC antiretroviral (ARV) drug database, beginning with each subject’s starting date of HAART. Information on specific drug usage, changes in regimens, immune and virological responses and adherence to therapy were not available. In Brazil, ARVs are only available from the public sector.
Linkage of the databases was performed using the software Reclink.11
Statistical analyses were performed using Epi Info 6.04d (Centers for Disease Control and Prevention, Atlanta, GA, USA) and SPSS 9.0 (SPSS Inc, Chicago, IL, USA). Comparison of proportions was done using the χ2 test (α = 0.05). Data were analyzed by year of death, and when homogeneous, pooled analyses were applied.
The study was approved by the Ethics Research Committee of the Rio de Janeiro City Health Department.
RESULTS
Between 1996 and 2005, there were 8601 AIDS-related deaths in RJC. TB was the primary cause of death in 9.0% of all AIDS-related deaths, while Pneumocystis carinii pneumonia (PCP) accounted for 4.7% (Table 1). The percentage of TB-HIV deaths decreased slowly from 1996 (11.1%), the last year without HAART in Brazil, to 6.8% in 2005. ‘Other’ infectious diseases accounted for 54.3% (all cases under the B20 coding, with the exception of B20.0, which is the code for TB-HIV co-infection). Table 1 also shows the proportional distribution of the three-digit ICD-10 coding for AIDS-related deaths. The contribution of infectious diseases to AIDS deaths in RJC has remained high over the years.
Table 1.
Total number of deaths and proportional distribution of primary cause of death in AIDS cases, Rio de Janeiro, Brazil, 1996–2005
| Year | Deaths n | TB (B20.0) % | PCP (B20.6) % | Other infectious diseases (B20, except for B20.0 and B20.6) % | Malignant neoplasms (B21) % | Specified diseases (B22) e.g., HIV, dementia, wasting, etc. | Other conditions (B23) % | Unspecified diseases (B24) % |
|---|---|---|---|---|---|---|---|---|
| 1996 | 1388 | 11.1 | 5.8 | 37.5 | 3.8 | 5.1 | 5.5 | 31.1 |
| 1997 | 1078 | 8.7 | 4.4 | 44.3 | 3.3 | 6.3 | 1.2 | 31.7 |
| 1998 | 900 | 8.1 | 5.3 | 47.6 | 2.9 | 9.6 | 0.3 | 26.2 |
| 1999 | 743 | 8.0 | 5.2 | 54.5 | 3.7 | 11.4 | 0.8 | 16.5 |
| 2000 | 782 | 9.3 | 3.7 | 57.9 | 4.1 | 9.7 | 0.6 | 14.6 |
| 2001 | 712 | 9.4 | 3.3 | 55.3 | 3.1 | 12.8 | 1.7 | 14.5 |
| 2002 | 776 | 9.4 | 4.4 | 50.6 | 3.0 | 13.3 | 1.4 | 17.9 |
| 2003 | 778 | 9.0 | 4.1 | 54.3 | 3.2 | 10.3 | 1.2 | 18.0 |
| 2004 | 766 | 8.0 | 3.6 | 52.8 | 3.7 | 12.7 | 2.1 | 17.2 |
| 2005 | 678 | 6.8 | 5.9 | 55.6 | 3.1 | 11.2 | 2.8 | 14.5 |
Source: Mortality Information System, Rio de Janeiro, Brazil, 1996–2005.
AIDS = acquired immune-deficiency syndrome; TB = tuberculosis; PCP = Pneumocystis carinii pneumonia; HIV = human immunodeficiency virus.
Patients who died from TB were younger (38.7% vs. 32.4% aged <35 years, P < 0.001), less educated (60.8% vs. 52.7%, P < 0.0001) and more likely to be non-white (63.7% vs. 46.1%, P < 0.0001) than patients who died from other causes (Table 2).
Table 2.
Demographics of AIDS-related deaths, comparing tuberculosis and other primary causes of death, Rio de Janeiro, Brazil, 1996–2005
| Characteristics | TB % | Other causes % | P value |
|---|---|---|---|
| Male | 71.7 | 70.5 | 0.49 |
| Age <35 years | 38.7 | 32.4 | <0.001 |
| Schooling <8 years | 60.8 | 52.7 | <0.001 |
| Race non-white | 63.7 | 46.1 | <0.001 |
Source: Mortality Information System, Rio de Janeiro, Brazil, 1996–2005.
AIDS = acquired immune-deficiency syndrome; TB = tuberculosis.
The distribution of patients on HAART is shown in Table 3. By the year 2000, around 50% of those who died had been on HAART at some time in their lives, and this percentage is sustained until 2005. Looking at all years pooled together (1997–2005), 27.3% of TB-related deaths were among patients on HAART (median 18 months) vs. 20.1% of those who died from PCP (median 20 months). In total, 33.8% of the cases had been treated with HAART (median time 22 months).
Table 3.
Distribution of patients on HAART by year of death related to TB or PCP, Rio de Janeiro, Brazil, 1996–2005
| Year | Total n (%) | TB (B20.0) n (%) | PCP (B20.6) n (%) |
|---|---|---|---|
| 1996 | 0 | 0 | 0 |
| 1997 | 108 (10.0) | 3 (3.2) | 3 (6.3) |
| 1998 | 155 (28.3) | 20 (27.4) | 5 (10.4) |
| 1999 | 255 (42.5) | 17 (28.8) | 14 (35.9) |
| 2000 | 316 (44.9) | 31 (42.5) | 5 (17.2) |
| 2001 | 351 (54.6) | 33 (49.3) | 8 (33.3) |
| 2002 | 389 (50.5) | 23 (31.5) | 12 (35.3) |
| 2003 | 392 (50.3) | 33 (47.1) | 12 (37.5) |
| 2004 | 391 (49.1) | 23 (37.7) | 12 (42.9) |
| 2005 | 331 (48.8) | 27 (58.7) | 10 (25.0) |
Source: Rio de Janeiro antiretroviral drug database.
HAART = highly-active antiretroviral treatment; TB = tuberculosis; PCP = Pneumocystis carinii pneumonia.
DISCUSSION
Our results show the magnitude of the impact of TB on survival in people with HIV. Even in a city with a low HIV burden such as RJC (<1% among pregnant women), the high burden of TB can reduce survival among people living with HIV despite free access to care and ARV drugs. One explanation is the occurrence of TB prior to HIV diagnosis, with TB initiating an HIV/AIDS diagnosis.
Although we have worked with secondary databases to assess the burden of TB in AIDS-related mortality, our data support other findings of the TB-HIV profile in RJC. For example, the percentage of HIV-positive cases among those with active TB was stable, at around 8%, throughout the study.12
The pattern of death among AIDS cases in RJC, when using only the three-digit ICD-10 coding B20 to B24, remains primarily composed of infectious diseases. A similar pattern is found in the developing world, where limited access to antiretroviral treatment (ART) prevents the reduction in mortality seen in the industrialized nations,13 where viral hepatitis, non-AIDS-related malignancies and cardiovascular events are now more common causes of death.14–16 Thus, infectious diseases are still a large component of AIDS-related mortality in RJC. More attention should be given to strategies that lead to earlier diagnosis of HIV in the general population, to wider prescription of those preventive treatments available and to the appropriate timing of starting HAART, all with the aim of reducing infectious disease-related mortality.17
Although it is well known that HAART reduces the incidence of opportunistic diseases, TB can occur at high CD4 cell counts, in contrast to PCP, which occurs mainly in patients with CD4 cell counts <200 cells/mm3. This may explain the lower proportion of PCP in our cases compared to TB, although we were unable to find a difference in the duration of treatment with HAART between the two conditions.
One limitation of our approach of secondary databases is that, by looking only at ICD-10 coding on the death certificate, it is not possible to address diagnostic misclassification or simultaneous diagnosis of TB with other illnesses. Therefore, TB cases who were erroneously classified under other infectious diseases could contribute to an underestimation of the number of TB-related deaths among HIV-positive patients. On the other hand, the availability of such a database allows us to quickly assess the burden of a disease on mortality, which, from the point of view of public health planning, is of great importance. It also helps formulate policies to implement prophylactic treatment, as can be done in the case of TB.
Furthermore, we did not have access to CD4 cell counts or viral load results, which could be used as covariates in our analysis. Nor were we able to analyze the use of TB preventive treatment by such subjects.
We previously reported that in almost 30% of the AIDS cases that were entered in the National Diseases Surveillance System in 2002, the diagnosis of AIDS came from death certificates.6 We were also able to show that among TB-HIV cases occurring in 2003 in RJC, 50% were on HAART before the diagnosis of TB.18 These previous findings showed that a proportion of our TB-HIV cases did not have the benefit of an opportune HIV diagnosis, thus eliminating any potential protective effect of HAART. On the other hand, another study in RJC has shown the protective effects of HAART and isoniazid preventive treatment on TB occurrence,19 suggesting that, besides receiving HAART, when HIV-infected patients are tested and treated for latent TB infection, a tremendous reduction in TB incidence may occur.
CONCLUSIONS
Analysis of the primary cause of death among AIDS cases using a secondary database showed that TB is the leading cause of AIDS-related deaths and is responsible for twice as many deaths as PCP. Infectious diseases still pose a great burden among HIV-positive patients in RJC, in contrast with other regions of the world where patients also have access to HAART. The TB burden may be underestimated due to potential misclassification among causes of death labeled as ‘other’. The potential benefits of TB preventive treatment and HAART availability could not be established by this analysis.
References
- 1.Corbett EL, Watt CJ, Walker N, et al. The growing burden of tuberculosis. Arch Intern Med. 2003;163:1009–1021. doi: 10.1001/archinte.163.9.1009. [DOI] [PubMed] [Google Scholar]
- 2.World Health Organization. Global tuberculosis control: surveillance, planning, financing. Geneva, Switzerland: WHO; 2006. WHO report 2006. WHO/HTM/TB/2006.362. [Google Scholar]
- 3.Joint United Nations Programme on HIV/AIDS. Report on the global HIV/AIDS epidemic, December 2006. Geneva, Switzerland: UNAIDS/WHO; 2006. [Google Scholar]
- 4.Szwarcwald CL, Souza PRB., Jr Estimativa da prevalência de HIV na população brasileira de 15 a 49 anos, 2004. Boletim Epidemiológico AIDST. 2006;1:11–15. [Brazilian] [Google Scholar]
- 5.Ministério da Saúde, Secretaria de Vigilância em Saúde. Programa de controle da tuberculose. Brasilia, Brazil: Ministério da Saúde; [Accessed November 2006]. http://portal.saude.gov.br/portal/svs/visualizar_texto.cfm?idtxt=21445. [Brazilian] [Google Scholar]
- 6.Saraceni V, Cruz MM, Lauria LM, Durovni B. Trends and characteristics of AIDS mortality in the Rio de Janeiro city after the introduction of highly active antiretroviral therapy. Braz J Infect Dis. 2005;9:209–215. doi: 10.1590/s1413-86702005000300003. [DOI] [PubMed] [Google Scholar]
- 7.Prefeitura da Cidade do Rio de Janeiro. Co-infecção tuberculose/HIV. Brasilia, Brazil: Secretaria Municipal de Saúde; 2007. [Accessed November 2006]. http://www.saude.rio.rj.gov.br/aids/ [Brazilian] [Google Scholar]
- 8.Secretaria Municipal de Saúde do Rio de Janeiro. Boletim informativo do programa de controle da tuberculose do município do Rio de Janeiro. Rio de Janeiro, Brazil: Municipal de Saúde; 2004. [Brazilian] [Google Scholar]
- 9.Toledo AS. Dissertação (Mestrado em Tisiologia e Pneumologia) Rio de Janeiro, Brazil: Universidade Federal do Rio de Janeiro; 1997. Estudos da associação entre tuberculose e a infecção pelo HIV em unidades ambulatoriais do município do Rio de Janeiro. [Brazilian] [Google Scholar]
- 10.World Health Organization. International classification of diseases—tenth revision. Geneva, Switzerland: WHO; 1990. [Accessed November 2006]. http://www3.who.int/icd/currentversion/fr-icd.htm. [Google Scholar]
- 11.Camargo KR, Jr, Coeli CM. Reclink: an application for database linkage implementing the probabilistic record linkage method. Cad Saude Publica. 2000;16:439–447. doi: 10.1590/s0102-311x2000000200014. [DOI] [PubMed] [Google Scholar]
- 12.Prefeitura da Cidade do Rio de Janeiro. Boletim informativo do programa de controle da tuberculose do município do Rio de Janeiro. Brasilia, Brazil: [Accessed November 2006]. http://www.saude.rio.rj.gov.br. [Google Scholar]
- 13.Groenewald P, Nannan N, Bourne D, Laubscher R, Bradshaw D. Identifying deaths from AIDS in South Africa. AIDS. 2005;19:193–201. doi: 10.1097/00002030-200501280-00012. [DOI] [PubMed] [Google Scholar]
- 14.Porter K, Babiker A, Bhaskaran K, et al. Determinants of survival following HIV-1 seroconversion after the introduction of HAART. Lancet. 2003;362:1267–1274. doi: 10.1016/s0140-6736(03)14570-9. [DOI] [PubMed] [Google Scholar]
- 15.Lewden C, Salmon D, Morlat P, et al. Mortality 2000 study group. Causes of death among human immunodeficiency virus (HIV) infected adults in the era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS. Int J Epidemiol. 2005;34:121–130. doi: 10.1093/ije/dyh307. [DOI] [PubMed] [Google Scholar]
- 16.Crum NF, Riffenburgh RH, Wegner S, et al. Triservice AIDS Clinical Consortium. Comparisons of causes of death and mortality rates among HIV-infected persons: analysis of the pre-, early, and late HAART (highly active antiretroviral therapy) eras. J Acquir Immune Defic Syndr. 2006;41:194–200. doi: 10.1097/01.qai.0000179459.31562.16. [DOI] [PubMed] [Google Scholar]
- 17.Smit C, Geskus R, Walker S, et al. Effective therapy has altered the spectrum of cause-specific mortality following HIV seroconversion. AIDS. 2006;20:741–749. doi: 10.1097/01.aids.0000216375.99560.a2. [DOI] [PubMed] [Google Scholar]
- 18.Saraceni V, King BS, Soares ECC, Cavalcante SC, Golub J, Durovni B. Tuberculosis, HIV and antiretroviral therapy in Rio de Janeiro City, Brazil. XVI International AIDS Conference; Toronto, Canada. Geneva, Switzerland: International AIDS Society; 2006. Abstract no. CDC0142. [Google Scholar]
- 19.Golub JE, Saraceni V, Cavalcante SC, et al. The impact of antiretroviral therapy and isoniazid preventive therapy on tuberculosis incidence in HIV-infected patients in Rio de Janeiro, Brazil. AIDS. 2007;21:1441–1448. doi: 10.1097/QAD.0b013e328216f441. [DOI] [PMC free article] [PubMed] [Google Scholar]