SciELO - Scientific Electronic Library Online

vol.29 issue6Delay in the diagnosis and surgical treatment of lung cancerHistological features and survival in idiopathic pulmonary fibrosis author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Jornal de Pneumologia

Print version ISSN 0102-3586On-line version ISSN 1678-4642

J. Pneumologia vol.29 no.6 São Paulo Nov./Dec. 2003 



Profile and follow-up of patients with Mycobacterium sp. at the Hospital das Clínicas da Universidade Federal de Minas Gerais



Giselle Carvalho FroesI; Rosane Luiza CoutinhoI; Marcelo Nardy de ÁvilaI; Leandra Rocha CançadoI; Silvana Spíndola de MirandaII

IMedical Student – Scientific Initiation
IIMycobacteriosis Research Coordinator





BACKGROUND: Surveys of patients diagnosed with mycobacteriosis, taken at the Laboratory of Mycobacteriology of the Hospital das Clínicas da Universidade Federal de Minas Gerais, have shown that lack of information concerning diagnosis prevented 42% from seeking treatment.
OBJECTIVE: To evaluate the profile of mycobacteriosis patients who were attended to at the Hospital das Clínicas, actively searching for these patients and subsequently performing follow-up exams before reintroducing them into the healthcare system, in accordance with proposals made by the National Program for the Control of Tuberculosis.
METHODS: The files of patients testing positive for Mycobacterium sp. during 2002 were selected from the archives of the Laboratory of Mycobacteriology of the Hospital das Clínicas da Universidade Federal de Minas Gerais. An active search for these patients was carried out and follow-up exams were then performed.
RESULTS: Of the 66 patients selected, 62 (94%) were positive for Mycobacterium tuberculosis and 4 (6%) were positive for other mycobacteria. Another 4 (6%) had been transferred to other institutions. Of the remaining 58, 37 (63%) had been cured, 1 (2%) had refused treatment, 4 (6%) had nontuberculous mycobacteriosis, 11 (19%) had died and 9 (16%) were not found. Of the 11 deaths, 7 (64%) had tested positive for HIV.
CONCLUSION: Patients failed to receive treatment not due to lack of diagnosis but to inadequate structure, low awareness levels on the part of all parties involved and lack of tuberculosis control organization at the hospital level. Due to the high number of HIV-positive patients within this group, the number of patients cured was lower than that required by the National Health Ministry. Deaths were attributed to HIV infection and lack of knowledge about the disease. In order to identify and address the problems associated with clinical laboratory practice, laboratory professionals must work in concert with their clinical counterparts when carrying out operational research on tuberculosis.

Key words: Tuberculosis/diagnosis. Mycobacterium infections/complications.



Abbreviations used in this paper:
TB – Tuberculosis
AIDS – Acquired Immunodeficiency Syndrome
HIV – Human Immunodeficiency Virus
NTM – Nontuberculous mycobacteria
Mtb – Mycobacterium tuberculosis
HC/UFMG – Hospital das Clínicas da Universidade Federal de Minas Gerais
CTR/HIV/AIDS – Centro de Referência Orestes Diniz/HIV/AIDS



For people in the 15 to 44 age bracket worldwide, tuberculosis (TB) is currently the leading cause of death among women and the second leading cause among men. In developing countries, TB is responsible for more deaths than are all other infectious diseases combined, despite the fact that TB treatment is 97% effective.(1)

It is estimated that, in the United States, there are 40,000 to 50,000 TB-related deaths per year. This is a serious situation, especially in light of the fact that 20% to 30% of patients with severe forms of TB are not notified or treated. In 2001, 82,866 cases of TB were reported in Brazil, corresponding to a rate of 42.28 per 100,000 inhabitants. The rate of pulmonary cases was 36.15/100,000, of which 24.40/100,000 presented positive sputum cultures and 11.75/100,000 were cases in which there was no bacteriological confirmation. In addition, there were 5.88/100,000 cases of extrapulmonary TB.(2) Between 1998 and 2000, the ratio of TB in men to TB in women was 1.8:1.(3) In Brazil, the confirmed TB mortality rate for 2001 was 3.07/100,000 inhabitants, corresponding to 5,294 deaths. However, the estimated rate is far higher: 11/100,000.(1, 2) In spite of the fact that mortality rates in Brazil have decreased significantly since 1977, the situation is still worrisome. In the state of Minas Gerais, the incidence of TB in the year 2001 was 34.1/100,000 inhabitants.(4) In the city of Belo Horizonte (the capital of Minas Gerais), it was 61.71/100,000 inhabitants.

According to the World Health Organization, there were an estimated 637,000 cases of TB related to human immunodeficiency virus (HIV) in 1999. In Latin America, in Africa and in some developed countries, 30% to 60% of HIV-infected patients are co-infected with the TB bacillus.(1) Between January 1980 and June 2000, 190,532 cases of acquired immunodeficiency syndrome (AIDS) were reported in Brazil. It is estimated that 20% to 40% of those developed TB.(1) In the city of Belo Horizonte, 9% of the patients who are identified as having TB are also HIV positive.(5)

The Brazilian Health Ministry proposes to look for TB cases proactively, diagnose especially the bacilliferous cases, adequately treat the disease, prevent new cases and establish biosafety measures. The objective of this program is to achieve a diagnosis rate of 90% and a cure rate of 85% by 2005 (6).

At the Hospital das Clínicas of the Universidade Federal de Minas Gerais (HC/UFMG), we diagnose approximately 60 new TB cases each year using bacilloscopy or culture exams. These patients come from the walk-in clinic, emergency room, wards and outpatient clinics of the HCUFMG and from the Centro de Referência Orestes Diniz/HIV/AIDS (CTR/HIV/AIDS).

Surveys performed in the HCUFMG Mycobacteria Laboratory in previous years showed that about 42% of the patients who had mycobacteriosis had not initiated their treatments because they were not aware of their diagnosis. This demonstrates the level of disorganization within the TB-control program at the hospital level.(7)

We decided to initiate the reorganization of the sector of our hospital that receives patients with respiratory symptoms. We evaluated the profiles of patients with bacteriological diagnosis of mycobacteriosis during the year 2002, giving particular attention to follow up and reintroduction of these patients into the Health System, in accordance with the proposals put forth by the National Program for the Control of Tuberculosis.(8)



We selected patients with positive samples (bacilloscopy and/or culture) from the Mycobacteria Laboratory registry encompassing January to December 2002. We obtained 2,279 samples, corresponding to approximately 700 individuals. Diagnoses of both Mycobacterium tuberculosis (Mtb) and Nontuberculous mycobacteria (NTM) were included. Bacilloscopy was performed using Ziehl-Neelsen technique and fluorescence and the specimens were cultured on Lowenstein-Jensen slants. Identification tests were performed using biochemical methods and the sensitivity tests by the proportion method. Both were performed on all cultured samples and positive results were obtained.(9, 10)

Samples were positive in 66 patients (bacilloscopy or culture). As shown in Table 1, 4 samples (6%) tested positive for NTM. Culture testing was responsible for diagnosing 19.5% of the patients with Mtb.



As shown in Figure 1, using the laboratory registries, we began our search for the patients.

We evaluated the profiles of 66 patients with a bacteriological diagnosis of mycobacteriosis, taking into account gender, age, previous medical history, treatment history, symptoms suggestive of TB, radiological profile, HIV status (ELISA and Western Blot) and follow up (cure, treatment failure, treatment discontinuation, transfer to other institutions, death and patients lost to follow up). Patients who had not been tested for HIV and those who were waiting for the results of HIV tests were included.

Of the 66 patients, 27 (41%) came from the Hospital das Clínicas (21% from the walk in clinic/emergency room and 20% from the wards), 37 (56%) came from the outpatient clinics and treatment history was unknown in 2 (3%).

Patient files were revised 6 months after the beginning of treatment in order to conclude each case. After giving their written informed consent, all the patients found were invited to answer a questionnaire, which included questions about social, clinic and epidemiological issues (UFMG Ethics Committee, number 162/01).



Of the 62 patients with Mtb, 35 (56.5%) were male and 27 (43.5%) were female (1.3:1). Of the 4 patients with NTM, 3 were male and 1 was female.

We were able to find the files of 58 (88%) of the 66 patients. Of those, 54 had Mtb. Ages varied from 19 to 84 (median, 35), and most of the infected patients were between 25 and 57. Regarding treatment history, of the 54 Mtb patients, 32 (59%) were from the region surrounding Belo Horizonte (MG), 9 (17%) were from the city of Belo Horizonte, 12 (22%) were from other cities in the state of Minas Gerais, and one (2%) was from another state.

Of the 54 patients with Mtb, 53 (98%) had presented clinical signs and symptoms suggestive of TB. This information was missing from the file of one patient (2%).

One patient had renal TB. Of the 53 patients diagnosed with pulmonary TB, 46 (97%) had chest radiographs consistent with the diagnosis. The radiographs of 7 patients were not found.

Of the 54 patients with Mtb, 29 (54%) were interviewed and only 4 reported having had TB previously. Of the 54 Mtb patients, 25 (46%) were HIV-positive Of those, 17 (68%) were male and 8 (32%) were female. Of the same 54 patients, nine (17%) were confirmed HIV-negative and 20 (37%) had unknown serology (3 were waiting for results of HIV tests, and serology had not yet been requested for 17). Of the 11 Mtb patients who died, 7 were HIV-positive.

In looking for the 8 (15%) patients who were unaware of having been diagnosed with TB, we found the following data: 1 was discharged after being cured, 1 refused to be treated, 3 were not found and 3 had died.

During follow-up treatment, there were no treatment failures or discontinuations. Of the 62 patients diagnosed with Mtb, 4 were sent to other institutions. Follow-up data for the other 58 patients is shown in figure 2. The 4 patients who were being retreated were HIV-negative, and all 4 were cured. If we exclude these patients, the total cure rate would be 61% (33/54). Table 2 shows the cure rates of patients whose files were found. In 4(6%) of the 66 cases, the mycobacteria were nontuberculous, and 1 of these patients, who was HIV-positive, died.






In this study, we used the bacteriological diagnosis of Mtb as a starting point (Table 1). However some other issues must be considered. In our laboratory testing, 80.5% of the TB diagnoses were made through bacilloscopy alone. Culture testing raised the diagnosis rate by 19.4%, attesting to its importance. One sample was insufficient to be cultured, but we considered the patient had TB, because a cure was effected using the treatment I regimen.(8) The implementation of new, more effective and faster techniques in TB diagnosis is of utmost importance.(11, 12)

In the HC/UFMG, we receive patients mainly from the metropolitan region of the city of Belo Horizonte. Many patients (21%) were from the walk in clinic or the emergency room, and we received a similar percentage (20%) from the hospital wards. Therefore, all of these patients first came to the hospital through the walk in clinic/emergency room. This reflects the deficiencies in the primary care services, increases morbidity and-mortality and exposes health professionals to the risk of being infected with Mtb since most hospitals have not adopted biosafety policies.(11)

Among the patients who had TB, we observed no significant difference in the ratio of males to females (1.3:1). This fact might be explained by the small number of cases or by the fact that more women are working outside the homes and are therefore more exposed to the risks, biological or otherwise, of contemporary society.

Tuberculosis affected mainly individuals between 25 and 57 years of age, which is a productive phase of life regarding work. This reflects the importance of the disease in terms of the Brazilian economy.(1, 11)

Of our TB patients, 46% were also HIV-positive, and the frequency of HIV positivity was greater among males, which is in agreement with data in the literature.(11-13) We observed that the rate of HIV positivity in TB patients in the hospital was far superior to that of patients seen at the primary care health clinics. In addition, most of the deaths occurred in HIV-positive patients. It should be noted, however, that the rate of HIV positivity among our patients was influenced by the fact that they were referred to us by the CTR/HIV/AIDS.

Of the patients with Mtb, 37% were not tested for HIV. That was mainly due to the fact that the test had not been requested. This may reflect a lack of information given to the health professionals and the absence of an adequate diagnostic routine in general hospitals.

In this study, we observed that the rate of cure was below the 85% target established by the Ministry of Health for cases in which the treatment I regimen is used.(8) This result might be justified by the great percentage of HIV-positive patients and by the fact that most of the patients had no knowledge of their test results. If we excluded patients who were being retreated, all of whom were cured, the cure rate would be lower (61%).

In the files reviewed, the signs and symptoms, as well as the chest radiographs, were consistent with TB – even in HIV-positive patients. This suggests that, through the use of the high-potency antiretroviral regimen introduced in Brazil at the end of 1996, the patients started to become immunocompetent, in contrast to what was observed in another university hospital in the state of Rio de Janeiro.(14)

One of the main problems in searching for patient files was the lack of information about exam requests, such as clinical chart numbers and the full names of patients and physicians. The search for files from the emergency room/walk in clinic was also problematic because charts are not created for patients who stay less than 24 hours, and the charts that are created are not filed systematically, making it impossible to find a particular one. The fact is that an evident lack of proper organization exists. However, this issue is being addressed.

The inadequate reporting on the charts has been a problem historically, and has many times prevented studies from being conducted. In order to minimize this problem, we proposed the creation of charts that would be easy to complete. Some patients refused to fill out the questionnaire. This is to be expected in any study. Despite that, all the patients received the same standard of care and attention from the investigators.

We looked exhaustively for the 15% of patients who were unaware of their diagnosis, but some were found too late and others had already died without treatment. A great deal of effort must be made to reorganize the health system in order to facilitate the location of patients whose bacilloscopy tests were negative in order to inform these patients of the need to return and be given the results of their culture exams.

One patient refused treatment despite the efforts that were made to convince him to be treated. Nevertheless, the healthcare team is still working on this case.

Five cases were diagnosed with NTM. Three were suspected of colonization or contamination (1 of them had Sjörgen’s syndrome, another had systemic lupus erythematosus and the other was HIV-positive). All of them evolved in a satisfactory way without treatment. The HIV patient had disseminated disease caused by NTM (isolated from a sterile site) and died within few weeks.

The mean interval between diagnosis and the beginning of treatment (11.4 days) is too long. Much needs to be done in order for patients with confirmed TB to be treated as soon as possible, thereby breaking the bacillus transmission chain. However, the human and financial resources allotted to the control of the disease effectively do not always reach the pertinent sectors.

The development of a flow chart allowed a better evaluation of the profile and follow up of patients with Mycobaterium sp. from the HC/UFMG. Such a chart might also be useful for other institutions. We can add the search engines of the Information System of the State Health Department to this flow chart, which would allow us to recover cases not found but which were reported by other institutions.

As long as adequate policies for the control of TB at the hospital level are not implemented by the Health Ministry, mistakes will continue to be made. These policies should use multidisciplinary approaches and take diagnosis, treatment and biosafety issues into account. This study showed that it was not lack of diagnosis that prevented the patients from getting treated but the inadequate structure, with low levels of awareness of the part of all involved. We found no other study in the national medical literature describing what happens after laboratory-testing diagnosis. It is of utmost importance that the professionals who work at the laboratory level perform operational researches regarding TB, working together with the clinical practice professionals. This is the only way to find the solutions for the problems that exist between the laboratories and the clinical practice.



We wish to thank the staff of the Central Laboratory of the Hospital das Clínicas da UFMG, Mycobacteria sector, the other professionals and mainly the patients who contributed directly or indirectly to this study. Also, we wish to give special thanks to Professor Afrânio Lineu Kritski of the Universidade Federal do Rio de Janeiro for reviewing the text.



1. Ministério da Saúde. Fundação Nacional de Saúde. Centro de Referência Prof. Hélio Fraga/Sociedade Brasileira de Pneumologia e Tisiologia. Controle da tuberculose – Uma proposta de integração ensino-serviço. 5ª ed. Rio de Janeiro; 2002.         [ Links ]

2. Ministério da Saúde. Disponível em:         [ Links ]

3. Ministério da Saúde. Secretaria de Políticas de Saúde. Situação da tuberculose no Brasil. Brasília, 2002.         [ Links ]

4. DATASUS. Disponível em:         [ Links ]

5. Dados do Sistema de Informação da Secretaria Estadual de Saúde de Minas Gerais. Belo Horizonte, 2001.         [ Links ]

6. Ministério da Saúde. Fundação Nacional de Saúde. Tuberculose Guia de vigilância epidemiológica. Brasília, 2002.         [ Links ]

7. Anjos Moreira J, Froes GC, Spindola de Miranda S. Cases of tuberculosis diagnosed at Federal University Hospital, Belo Horizonte, MG, Brazil. In: 33rd World Conference on Lung Health of the International Union Against Tuberculosis and Lung Disease (IUATLD). Montreal, Canada, 6-10 Oct. 2002. Int J Tuberc Lung Dis 2002;6(10 Suppl 1): S1-210.         [ Links ]

8. Ministério da Saúde. Fundação Nacional de Saúde. Plano Nacional de Controle da Tuberculose Normas técnicas, estrutura e operacionalização. Brasília; 2000.         [ Links ]

9. Ministério da Saúde. Fundação Nacional de Saúde. Centro de Referência Prof. Hélio Fraga. Manual de bacteriologia da tuberculose. 2ª ed. Rio de Janeiro, 1994.         [ Links ]

10. Canetti G, Rist N, Grosset J. Mesure da la sensibilité du bacille de la tuberculeux aux antibacillares par la méthode des proportions. Rev Tuberc Pneumol 1963;27:217.         [ Links ]

11. Kritski AL, Conde MB, Souza GRM. Tuberculose: do ambulatório à enfermaria. 2ª ed. São Paulo: Atheneu, 2000.         [ Links ]

12. Kritski AL, Lapa e Silva JR, Conde MB. Tuberculosis and HIV: renewed challenge. Mem Inst Oswaldo Cruz 1998;93:417-21.         [ Links ]

13. Muzy de Souza GR, Gonçalves M, Carvalho ACC. Controle de infecção hospitalar por tuberculose. Pulmão RJ 1997;6:220-7.         [ Links ]

14. Mello FCQ, Lacerda APM, Gounder C, Salles CLG, Madeira FB, Gomes SA, Kritski A. The impact of institution of HAART in the co-infection tuberculosis and HIV in a reference hospital for AIDS in Rio de Janeiro, Brazil. Am J Respir Crit Care Med 2001;163:A496.         [ Links ]



Correspondence to
Silvana Spíndola de Miranda
Av. Alfredo Balena, 120. Bairro Santa Efigênia. Departamento de Clínica Médica/Pneumologia. 4º andar. Cep: 30.103-100
Tel: (31)32489599

Submitted: 16/06/2003. Accepted, after revision: 25/08/2003.



* Study performed at the Hospital das Clínicas da Universidade Federal de Minas Gerais

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License