A retrospective evaluation of a score system adopted by the Ministry of Health, Brazil in the diagnosis of pulmonary tuberculosis in childhood: a case control study

Avaliação retrospectiva de sistema de pontuação pelo Ministério da Saúde do Brasil, no diagnóstico de tuberculose pulmonar na criança: estudo controle de casos

BRIEF COMMUNICATION

A retrospective evaluation of a score system adopted by the Ministry of Health, Brazil in the diagnosis of pulmonary tuberculosis in childhood: a case control study

Avaliação retrospectiva de sistema de pontuação pelo Ministério da Saúde do Brasil, no diagnóstico de tuberculose pulmonar na criança: estudo controle de casos

Clemax Couto Sant'Anna^I; Cláudia T. S. Orfaliais^II; Maria de Fátima B. Pombo March^I

^IDepartamento de Pediatria da Faculdade de Medicina da Universidade Federal do Rio de Janeiro. Instituto de Puericultura e Pediatria Martagão Gesteira da UFRJ. Rio de Janeiro, RJ. Brasil

^IIUniversidade Federal Fluminense. Programa de Mestrado em Pediatria.

^{Correspondence}

SUMMARY

Based on a retrospective case-control study we evaluated the score system adopted by the Ministry of Health of Brazil (Ministério da Saúde - MS), to diagnose pulmonary tuberculosis (PTB) in childhood. This system is independent of bacteriological or histopathological data to define a very likely (> or = 40 points), possible (30-35 points) or unlikely (< or = 25 points) diagnosis of tuberculosis. Records of hospitalized non-infected HIV children at the Instituto de Puericultura e Pediatria Martagão Gesteira of Federal University of Rio de Janeiro (IPPMG-UFRJ), were reviewed. Patients were adjusted for age and divided in two different groups: 45 subjects in the case group (culture-positive) [mean of age = 10.64 mo; SD 9.66]; and 96 in the control group (culture-negative and clinic criteria that dismissed the disease) [mean of age = 11.79 mo.; SD 11.31].

Among the variables analyzed, the radiological status had the greater impact into the diagnosis (OR = 25.39), followed by exposure to adult with tuberculosis (OR = 10.67), tuberculin skin test >10mm (OR = 8.23). The best cut-off point to the diagnosis of PTB was 30 points, where the score system was more accurate, with sensitivity of 88.9% and specificity of 86.5%.

Keywords: Pulmonary tuberculosis; Child; Case-control study.

RESUMO

Avaliou-se o sistema de pontuação adotado pelo Ministério da Saúde do Brasil para o diagnóstico de tuberculose pulmonar (TP) na infância através de estudo caso-controle retrospectivo. Tal sistema independe de dados bacteriológicos ou histopatológicos e define o diagnóstico de tuberculose como muito provável (> ou = 40 pontos); possível (30 a 35 pontos) ou pouco provável (< ou + 25 pontos). Foram revisados os prontuários de crianças não infectadas pelo HIV internadas no Instituto de Puericultura e Pediatria Martagão Gesteira da Universidade Federal do Rio de Janeiro; os pacientes foram ajustados por idade e divididos em dois grupos: 45 casos (cultura positiva) [média = 10,64 meses; DP 9,66]; e 96 controles (cultura negativa e sem critérios clínicos para TP) [média = 11,79 meses; DP 11,31]. Dentre as variáveis analisadas, o quadro radiológico teve grande importância para o diagnóstico (OR = 25,39), seguido pelo contato com adulto com tuberculose (OR = 10,67), teste tuberculínico > 10 mm (OR = 8,23). O melhor ponto de corte para o diagnóstico de TP foi 30 pontos, no qual houve maior acurácia do sistema, com sensibilidade de 99,9% e especificidade de 86,5%.

Diagnosing tuberculosis in childhood is a challenge and it should take into consideration the value of clinical, radiological and epidemiological aspects. Various diagnostic approaches for diagnosis of childhood tuberculosis have been published. Most of them are based on clinical experience^3,5,6 but FOURIE et al.⁴ used a mathematic model. Based in well-established diagnosis criteria, a arbitrary score system, currently adopted by the Ministry of Health (Ministério da Saúde, MS), was created in Brazil to diagnose pulmonary tuberculosis (PTB) in childhood¹. This system is independent of bacteriological or histopathological data to define a very likely (> or = 40 points), possible (30-35 points) or unlikely (< or = 25 points) diagnosis of tuberculosis (Table 1). This score system was evaluated based on a retrospective case-control study. Records of hospitalized non-infected HIV children at the Instituto de Puericultura e Pediatria Martagão Gesteira of Federal University of Rio de Janeiro (IPPMG-UFRJ), that were pulmonary tuberculosis suspects and were submitted to a culture of acid fast bacillus (AFB) on their gastric smear between 1987 and 1994, were reviewed. Patients were adjusted for age and divided in two different groups: 45 subjects in the case group (culture-positive) [mean of age = 10.64 mo; SD 9.66]; and 96 in the control group (culture-negative and clinic criteria that dismissed the disease) [mean of age = 11.79; SD 11.31]. The score system was applied to the case and control subjects, in order to calculate the values of sensitivity and specificity and the odds ratio (OR) of each one of the variables of the system: clinical radiological status, exposure to adult with tuberculosis, tuberculin skin test and nutritional status. Patients younger than one-year-old (69.5%) and males (61.7%) were predominant. Among the variables analyzed, the radiological status had the greater impact into the diagnosis (OR = 25.39), followed by exposure to adult with tuberculosis (OR = 10.67), tuberculin skin test >10 mm (OR = 8.23). It was noticed that 40 (88.9%) cases and 23 (24.0%) controls had radiologic abnormalities suggesting tuberculosis, as well, 36 (80.0%) cases and 38 (39.6%) controls had prolonged fever. It was detected that 16/44 (36.4%) case subjects and 20/96 (20.8%) control subjects had not been vaccinated with BCG and that, among the 36 patients that were not vaccinated, 29 were younger than one-year-old. The absence of BCG vaccination indicated risk (OR = 2.17) of diagnosis of pulmonary tuberculosis. The tuberculin skin test was < 5 mm in 16/27 (59.52%) case subjects and 47/49 (95.9%) control subjects. Malnutrition was noticed in 28 (62.2%) cases and 61 (63.5%) controls. The best cut-off point to the diagnosis of PTB was 30 points, where the score system was more accurate, with sensitivity of 88.9% and specificity of 86.5% (Table 2). The score system proved to be important in this retrospective case control study in the diagnosis of PTB in hospitalized children. In the future, the implementation of new studies to evaluate this score in a prospective study including hospitalized as well out patients will be appropriate².

ACKNOWLEDGMENT

The authors thank to Dr. Mônica Tessinari R. Tura for collecting data and support.

Received: 27 November 2002

Accepted: 2 April 2003

• ¹
  BRASIL. MINISTÉRIO DA SAÚDE. FUNDAÇÃO NACIONAL DE SAÚDE - Controle da Tuberculose: uma proposta de integração ensino-serviço. 5ª ed. Rio de Janeiro, Centro de Referência Prof. Hélio Fraga, 2002.
• 2. CARREIRA, M.N. & SANT'ANNA, C.C. - Estudo comparativo de critérios para o diagnóstico de tuberculose em crianças atendidas em centro de saúde. J. Pneumol., 26: 219-226, 2000.
• 3. CHAULET, P. - Proposal of an improve score method for the diagnosis of pulmonary tuberculosis. Tuberc. Lung Dis., 73: 309, 1992.
• 4. FOURIE, P.B.; BECKER, P.J.; FESTENSTEIN, G.B. et al - Procedures for developing a simple scoring method based on unsophisticated criteria for screening children for tuberculosis. Int. J. Tuberc. Lung Dis., 2: 116-123, 1998.
• 5. STEGEN, G.; JONES, K. & KAPLAN, P. - Criteria for guidance in the diagnosis. Pediatrics, 43: 260-263, 1969.
• 6. TIDIJANI, O.; AMEDONNE, A. & TEN DAM, H.G. - The protective effect of BCB vaccination of the newborn against childhood tuberculosis in an African community. Tubercle, 67: 269-281, 1986.

Publication Dates

• Publication in this collection
  12 Dec 2006
• Date of issue
  Apr 2003

History

• Received
  27 Nov 2002
• Accepted
  02 Apr 2003