Clinical Profile and Outcome of Childhood Tuberculosis at Dhulikhel Hospital

The corrected PDF for this article was loaded on 07/02/2011. Background: Children contribute a significant proportion of the tuberculosis (TB) burden in Nepal and suffer severe TB related morbidity and mortality, particularly in endemic areas. Diagnosis and management of pediatric TB especially Extra pulmonary TB (EPTB) is challenging. Objectives: The present study was designed to study clinical, laboratory characteristics and outcome of childhood tuberculosis. Methods: A prospective analysis of 60 cases of TB children within three years period was conducted at Pediatric Department, Dhulikhel Hospital and followed up till child completely recovered. Results: In the present study 60 children had clinical and lab evidence of tuberculosis and received anti tubercular therapy. Extra pulmonary tuberculosis was common (78.3%) than pulmonary tuberculosis (21.6%). BCG scar was absent in eight (13.33%) and absent BCG vaccination was significantly associated with disseminated TB (p<0.05). The most frequently seen symptoms were fever (65 %), cough (46.67 %) and abdominal distension (36.67%). Hepatomegaly (45%), ascites (33.33%), lymphadenopathy (23.33 %) and splenomegaly (11.67 %), and are common signs. Malnutrition seen in (33.3%) with more malnourished children with disseminated TB than in other diagnosis (p<0.001). Mantoux test was positive in (48.3%). Isolation of AFB was possible only in (8.33%). Among 60 cases (60%) were recovered completely with the primary regimen, (5%) recovered with extending the duration of primary regimen to three more months. (25%) failed to follow up, while (6.67%) died. Conclusion: EPTB is common than pulmonary TB. Among 60 cases 36 (60%) recovered completely with the primary regimen, three cases (5%) recovered with extending the duration with primary regimen. Fifteen cases (25%) failed to follow up, while four cases (6.67%) died. Key words: Children; Tuberculosis; Diagnosis; Outcome DOI: 10.3126/jnps.v31i1.4160 J Nep Paedtr Soc 2010;31(1):11-16


Introduction
T uberculosis has reemerged as a major public health challenge in the world.Children are mainly infected by adult pulmonary TB source and childhood TB therefore refl ects the intensity of the ongoing transmission of Mycobacterium tuberculosis within a community.
It is estimated that one third of the world's population is infected with Mycobacterium tuberculosis, and that each year, about 9 million people develop TB, of whom about 2 million die.Of the 9 million annual TB cases, about 1 million (11%) occur in children (under 15 years of age).Of these childhood cases, 75% occur annually in 22 high burden countries that together account for 80% of the world's estimated incident cases 1 .Despite this huge disease burden, children's access to anti tuberculosis treatment in most endemic areas remains poor, as tuberculosis control programs focus predominantly on the treatment of sputum smearpositive adults 2 .
Children can present with TB at any age, but the most common age is between 1 and 4 years.Case notifi cations of childhood TB depend on the intensity of the epidemic, the age structure of the population, the available diagnostic tools, and the extent of routine contact tracing.

Methods
The prospective study was conducted in the pediatric department of Dhulikhel Hospital, Kathmandu University Hospital from January 2007 to December 2009.All the new cases or diagnosed as tuberculosis from outside and referred for reevaluation in the age group of 0-14 years admitted in the ward with symptoms suggestive of tuberculosis disease were evaluated and diagnosed cases were enrolled into the study.Children were assessed by detail history, thorough physical examination and familial and extra familial contact with TB was enquired, apart from the details of the illness.Mantoux test was done and indurations ≥ 10mm after test will be considered a positive reaction.Chest Xray, haemogram and ESR was done in all patients.In relevant cases, sputum for AFB (acid fast bacilli) or in children who cannot expectorate sputum gastric aspirate for AFB smear examination for 3 consecutive days, lymph node biopsy, cerebrospinal, pleural and peritoneal fl uid test were done.Confi rmed cases were treated according to WHO 1 guidance for national tuberculosis programme on the management of tuberculosis in children and IAP consensus on treatment of Childhood 4 whenever feasible.Patients were followed up at monthly intervals till patient is completely recovered.CXR was repeated at the end of treatment.Malnutrition was classifi ed according to Indian Academy of Pediatrics IAP classifi cation 3 .Complete recovery defi ned by weight gain, subsidence of the symptoms and signs, radiological improvement and improvement in the nutritional status at the end of the treatment.All these information and type of tuberculosis and follow up details were recorded in the proforma.Fisher exact test was used to test the signifi cant difference between the proportions.The data were analyzed using SPSS software (SPPS, Chicago, IL).The ethical consent for the study was taken from the Institutional Review Committee, Kathmandu University School of Medical Sciences.
Isolation of AFB was possible only in fi ve cases (8.3%), three from lymphnode biopsy and two from sputum sample.Among fi ve TB lymphadenitis cases FNAC revealed granulomatous lymphadenitis in four (80%) cases.
Among 60 cases thirty six (60%) were recovered completely with the primary regimen, 3(5%) who were diagnosed disseminated tuberculosis from outside on Category I was extended for 3 more months and was recovered.Two cases which were diagnosed as disseminated tuberculosis after ruling out malignancy by bone marrow had persistence of symptoms even after completion of ATT for total of 9 months hence HIV (human immunodefi ciency virus) test was also performed but came out to be negative hence were referred to National Tuberculosis Centre.
Fifteen cases (25%) failed to follow up, while four cases (6.67%) expired.Out of which three were diagnosed as disseminated TB and one was tubercular meningitis.Outcome of tuberculosis children is shown in Table 3.

Discussion
The extent of tuberculosis problem in children is a refl ection of infectious pool of adult smear positive cases in the community.21(35%) were in the 5-9 years and 18(30%) were less than 5 years old.Study conducted in Bangalore by Vishwanatha KG 5 reported 48.2% were in 5-9 years age and 38.2% in < 5 years.This indicates the high prevalence of childhood tuberculosis in the younger group.The probable reasons for this may be due to low resistance of the host, increased prevalence of malnutrition and close contact with the infected adults.
There has been a rise in the prevalence of tuberculosis throughout the world with propensity of extra pulmonary involvement 6,7 .In our study extra pulmonary tuberculosis was found to be higher 47(78.3%)than pulmonary tuberculosis 13(21.7%),similar to recent studies in tertiary centers where extra pulmonary TB comprises 51-78% 8,9 .In our study among Extrapulmonary TB disseminated tuberculosis was the commonest followed by abdominal tuberculosis, contradicting to other studies where TB lymph node was commonest followed by TB pleurisy 10,11 .
There were no signifi cant association between age and the extra pulmonary tuberculosis however age <5 years was signifi cantly associated with pulmonary tuberculosis.In the study done in Philippines in culture positive tuberculosis patients the distribution of pulmonary tuberculosis was not age related, but there were signifi cantly more subjects in the younger age group <5 years old who had TB meningitis 12 .Eighteen (30%) were malnourished.Study has shown malnutrition adversely affects the host's defense mechanism by suppressing cell mediated response to the infection 13 .In malnourished children there is a higher risk of a localized lesion to become progressive or generalized disseminated tuberculosis, because of poor immune response to infection.Malnourished children had more extra pulmonary tuberculosis (p=0.05) and there were more malnourished children who had disseminated TB which was statistically signifi cant.Fifty two (86.67%) had received BCG, absent BCG scar was signifi cantly associated with disseminated TB (p<0.05).Study conducted by Vishwanatha KG 5 comparing 89 BCG vaccinated and 23 unvaccinated children concluded that all types of TB occur in BCG vaccinated children and 22.5% progress to severe disseminated forms.A meta-analysis of 14 prospective BCG trials and 12 case control studies concluded that the average protective effects of BCG vaccine are 50% 14 .Protection is higher for meningitis and disseminated disease than for pulmonary tuberculosis.This may explain the higher prevalence of disseminated TB in our study.
Among 60 cases only 20(32.3%) of the respondents, history of exposure to TB from immediate family member was elicited.Study in South Africa 15 in culture confi rmed childhood tuberculosis revealed only 49.5% had contact history.This may be due to social stigma attached to the disease of being not accepted by family member.
The diagnosis of the childhood TB presents a major challenge as it is complicated by the absence of practical "gold standard" 16,17 .Bacteriologic confi rmation, the accepted gold standard is of limited use in the children because of the paucibacillary nature of their disease and poor bacteriologic yield.Most children already infected may not show any sign and symptoms at all even with signifi cant abnormalities in the chest radiographs 18 .The diagnosis of tuberculosis in children from endemic areas depend mainly on clinical features and the subjective interpretation of the chest radiograph 19,20 .
The clinical presentation of TB in children takes many forms.The diagnosis in most case is still based on clinical evidence alone 21 .
Several studies showed cough, fever, loss of appetite, diarrhoea, vomiting, and weight loss are the most common symptoms in the children with TB, even in the culture positive cases 12,[22][23][24][25] .In our study also these symptoms were common.Our study revealed signs suggestive of generalized disease including lymphadenopathy, hepatomegaly and splenomegaly similar to study done by M. Burroughs et.al in culture positive tuberculosis cases 25 .Study done in Philippines 12 also revealed most common fi ndings as lymphadenopathy(55.2%),symptoms referable to gastrointestinal tract (26.9%), central nervous system(17.9%).
All the children were tested for Mantoux and only 48.3% showed positive results similar to other studies where it was reported to be 30-40% 8,9 .In endemic areas a positive Mantoux is not uncommon in randomly selected healthy children 26 which limit its diagnostic value.No signifi cant association was found between malnutrition and results of Mantoux.Study by Cherry Lyn 12 in culture positive TB showed most common fi nding in chest x-ray to be hazy densities 23.6% and cystic lucencies 21.8%, no signifi cant fi nding 21.8%, mediastinal lymphadenopathy 5.4%.In our study Chest X-ray revealed most common fi nding of hilar lymphadenopathy 16(23.3%),unilateral pleural effusion 14(23.3%),no signifi cant fi nding 13(21.7%),consolidation nine (15%).
Bacteriological or histological confi rmation has been reported to be 10-35% in studies at tertiary centers 8,9 .In the present study the diagnosis was often based on the indirect evidence of infection as AFB could be demonstrated in only fi ve (8.33) % of the cases.Two (3.33%) was in sputum AFB and AFB lymph node 3(5%).Study conducted by S.K Kabra 9 in AIMS (India) among tuberculosis patients without conducting culture, AFB could be isolated in 11% and was more commonly in lymph node tuberculosis which was comparable to our study.
However study conducted in Philippines 12 all culture positive tuberculosis patients AFB was isolated in 62.7% with common sources being gastric aspirate(44.8%)and sputum (38.8%) with older subjects > 5 years was signifi cantly associated with a positive AFB (p=0.024).This study highlights culture is mandatory in all patients in order to avoid over diagnosis of tuberculosis children especially in children though studies of childhood tuberculosis TB shows only 40-50% of cases are culture positive 12 .Poor yield of AFB may be due to paucibacillary nature of the illness and poor sample collection due to inability to produce appropriate sputum sample.Only 32-40% of gastric aspirates from children with pulmonary TBN are culture positive.Only 5-13% of gastric aspirate are AFB smear positive 27 .However in our study gastric aspirate could not demonstrate any AFB.Studies of childhood TB showed that only 40-50% of cases are culture proven 12 .Follow up among 45 cases showed 60% were completely recovered from the disease, three(5%) who were diagnosed disseminated tuberculosis from outside on Category I was extended for three more months and was recovered, two (3.33%) had persistence of symptoms even after completion of treatment.6.67% died and 25% lost to follow up.Follow up studies done even in culture positive cases 12 revealed 10.4 % cured of disease, 38.8% improved and 13.4% expired and more than one third were lost to follow up.
Limitation of our study were lack of data on the HIV status especially in disseminated TB as HIV test was not done on all cases due to economic factor.Contact survey for establishing a defi nite evidence of TB in the family members couldn't be done.

Conclusion
Extra pulmonary tuberculosis is common with higher incidence of disseminated tuberculosis in our set up.Bacteriological yield is very low.Outcome of tuberculosis revealed 60% were completely recovered from the disease, 5% recovered while extending duration of primary regimen, 3.33% had persistence of symptoms even after completion of treatment.6.67% died and 25% were lost to follow up.

Table 1 :
Spectrum of types of tuberculosis seen in children.

Table 2 :
Showing summary of clinical profi le of tuberculosis in children.

Table 3 :
Study outcomes of children with tuberculosis.