Indian Pediatrics 2002; 39:458-462
Clinical Spectrum of Tuberculosis in BCG Vaccinated Children
S. Sushama Bai
With eight million new cases of pulmonary tuberculosis being identified in the world per year and 2.89 million deaths, tuberculosis (TB) is still a major public health problem(1). After the implementation of expanded and universal immunization programmes in India, there is substantial improvement in BCG vaccination coverage reaching up to 90% in urban areas(2). In spite of this improved vaccination coverage and timely revised treatment protocols the disease is still rampant, and drug resistant strains have emerged, the situation being aggravated by HIV infection. The protective efficacy of BCG vaccination is under debate(3,4). This study was undertaken to evaluate the clinical spectrum of TB and its outcome in properly BCG vaccinated children.
Subjects and Methods
This prospective study was conducted in the Department of Child Health of a Medical College from January 1998 to December 1999. Patients in the age group from 2 months to 12 years attending the Outpatient Department or admitted in the wards with symptoms suggestive of tuberculous disease were subjected for detailed evaluation. Asympto-matic children having close contact with adult TB patients were also investigated and included if found to have tuberculous disease. All had BCG vaccination within the first three days of life and the presence of BCG scar was verified. The clinical criteria for investigation of tuberculous disease were recurrent or prolonged fever, recurrent respiratory infec-tions, recurrent wheezing, poor weight gain, lymphadenopathy, organomegaly, meningitis, convulsions, serous effusions and arthritis. The diagnostic criteria for tuberculous disease were positive Mantoux reaction along with chest skiagram finding. In Mantous test negative clinically suspected patients, demonstration of acid fast bacilli in gastric aspirate, lymph node biopsy revealing TB pathology or CT scan appearance character-istic of tuberculoma brain were considered as diagnostic criteria. Asymptomatic Mantoux positive children with no evidence of disease, babies less than 2 months of age, those who had BCG vaccination after 3 days of life, children with BCG adenitis, those without BCG scar and those on empirical anti-tubercular drugs were excluded from the study. A case control study was not done in BCG unvaccinated group because of the non existence of such patients in the area.
Patients were assessed by detailed history, thorough physical examination and diagnostic investigations. In the history details regarding familial and extra familial contact with TB was enquired apart from the details of the ill- ness. Socioeconomic status was assessed by modified method of Kuppuswami(5). Nutritional assessment was done according to IAP classification(6). Mantoux test was done for all patients by the research person and induration exceeding 10 mm 48-72 hours after test was considered as positive reaction(7-11). X-ray chest, hemogram and urine routine examination were done for all patients. Radiologic findings were analyzed with the help of a competent radiologist in the institution. In relevant cases gastric aspirate for AFB smear examination for three consecutive days, lymph node biopsy, cerebrospinal, pericardial, pleural and peritoneal fluid studies including tubercular antibody and adenosine deaminase test were done. Other diagnostic possibilities were excluded by appropriate investigations. Confirmed cases were treated according to IAP Consensus on Treatment of Childhood TB(12). Patients were followed up at monthly intervals for minimum period of 9 months. During each visit they were clinically assessed for nutritional status, clinical improvement and evidence of any drug toxicity. For all children, SGPT was done at the end of one month after starting therapy and chest X-ray were repeated at the end of 3, 6 and 9 months of therapy. All the family members were screened with chest X-ray to exclude intrafamilial contact. Siblings of index cases were screened with Mantoux test and chest x-ray to exclude asymptomatic tuberculous disease.
Two hundred and fifty children who fulfilled the diagnostic criteria for tuberculous disease were initially included in the study but final analysis was done in only 95 patients who could be completely followed up. Male female ratio was 1.5 : 1. Maximum number of cases were in the 1 to 5 years age group (49.5%); 27.4% were in 5 to 10 years age group, 17.9% were under 1 year and 5.3% were in the age group 10 to 12 years. Slightly more than half (55.8%) belonged to low, 38.9% to middle and 5.3% to high socio-economic groups. Statistically significant association was observed between severe forms of TB and low socio-economic status. Half the cases revealed positive history of contact with adult case of tuberculosis. Screening the family members disclosed two new cases, altogether con-tributing to 52%. Screening the siblings of index cases identified three new cases of pulmonary tuberculosis.
The predominant symptoms were fever (65.3%) and cough (62.2%). One third had weight loss or poor weight gain and 24.2 % had wheezing. Significant cervical lymph-adenopathy was observed in 11.6% cases while 32% had initial presentation as seizures. Four patients were investigated because of positive contact history and had primary pulmonary complex. Hyper-sensitivity phenomenon was observed in 4 patients in the form of Poncets arthritis (2 cases), erythema nodosum (1 case) and phlyctenular conjunctivitis (1 case).
Fifty eight per cent of cases had normal nutritional status and 42% had protein energy malnutrition (PEM) of which 37% has Grade I or Grade II PEM and 5% had Grade III PEM. There was no child with Grade IV PEM. Statistically significant relation was observed between severe forms of tuberculosis and PEM.
Eighty nine (93.3%) patients had positive tuberculin test. Out of the tuberculin negative 6 cases, 4 had lymph node tuberculosis proved by biopsy, one had tuberculoma brain proved by CSF studies and CT scan and one had miliary tuberculosis proved by chest skiagram and AFB positive gastric aspirate. This three months old baby with miliary tuberculosis had her mother detected to have open tuberculosis on screening. There was no statistical relation between the extent of Mantoux positivity and age, nutritional status or disease severity.
Abnormal chest X-ray was seen in 93 cases. The commonest radiological abnormality was parenchymal lesion (66.2%), hilar adenopathy (13.2%) and parenchymal plus nodal lesion (8.4%). The two patients without any skiagram changes had gland TB proved by biopsy.
Clinical Types of TB
Most common type of tuberculosis was primary pulmonary complex occurring alone or in combination with other lesions (72.6%). Of this, 3.1% had progressive primary disease. Seventeen per cent had disseminated tuberculosis which included multiple organ involvement (4 cases), miliary TB (1 case), TB meningitis (1 case), tuberculoma (2 cases), peritoneal TB (1 case), TB pericarditis (2 cases) and lymph node associated with primary complex (5 cases). Table I shows the clinical types of tuberculosis encountered.
Ninety seven per cent cases showed clinical improvement at the end of 1 month and 100% at the end of 2 months treatment. Radiologically 22% showed complete clear-ance and 73% showed moderate clearance at the end of therapy. Five per cent did not reveal radiological clearance in spite of clinical recovery. None of the children revealed deterioration of illness or radiological abnormality. There was no case of drug toxicity, drug resistance or mortality.
Udani in his study on 2000 BCG vaccinated children with tuberculosis has observed that 91% had intra-thoracic lesions with majority having mediastinal lymph node tuberculosis(1). The incidence of neuro tuberculosis and other types of disseminated tuberculosis were 45% each in his study with modified clinical pictures of neuro tuberculosis like serous tuberculous meningitis and localized tuberculous disease in meninges and brain(13). Isolated splenomegaly, hepato-megaly, single organ involvement, bone and joint tuberculosis were also high(14). Mathur et al. in a comparative study between BCG vaccinated and non vaccinated groups of patients could not find any significant difference in clinical pattern or mortality rate(4). In this study, 24.4% had thoracic TB, 68.6% had TB meningitis and mortality rate was 27% in BCG vaccinated group. The clinical pattern in relation with nutritional status and socio economic conditions were not analyzed in these studies.Table I-Types of Tuberculosis
In the present study, 72.6% had intra-thoracic lesions alone, 16.8% had dis-seminated TB, 6.32% had isolated lymph node TB and 4.2% had hypersensitivity phenomena. Mediastinal lymph node TB, modified neuro TB and other forms of TB described by Udani(1) were not observed. A control study to compare the percentage of different forms of TB in BCG non vaccinated group was not possible due to the non existence of such a group in the area.
All cases of disseminated TB in our study were seen in children with PEM and in those belonging to low socio economic group. Various studies reveal the association between low socio-economic status and severe forms of TB(15-16). The impact of undernutrition on childhood tuberculosis progressing to severe and disseminated forms was not evaluated in these series whereas in the present study, 58% of patients had normal nutrition. It is stated that BCG vaccine has protective value against dissemination of tuberculosis because T cells in vaccinated children are highly sensitized preventing hematogenous spread(1). In under-nourished children, cell mediated immunity is greatly impaired and hence the vaccine fails in preventing dissemination of tuberculosis. The ICMR BCG trials in Chingleput also report that BCG offers no protection against primary tubercular infection or its progression to severe forms(17). Presently, BCG vaccination is advised to be continued in infants and children to reduce the risk of primary tubercular infection disseminating to severe forms(18).
Our study reveals that protective benefit of BCG vaccine against the dissemination of tuberculosis in children is possible only if they have normal nutrition and improved socio-economic conditions.
Contributors: SSB drafted the paper and co-ordinated the study (design and interpretation) and will act as the guarantor. RLD collected data and participated in drafting.
Competing interests: None stated.