Soumya Swaminathan, Aarti Raghavan, Manjula Datta*,
C.N. Paramasivan and K.C. Saravanan†
From the Tuberculosis Research Center (Indian
Council of Medical Research), Chennai, *The Tamil Nadu Dr.
M.G.R. Medical University, Chennai and †Government Stanley
Medical College and
Hospital, Department of Radiology, Chennai, India.
Correspondence to: Dr. Soumya Swaminathan, Division
of HIV/AIDS, Tuberculosis Research
Center, Mayor V. R. Ramanathan Road, Chetpet, Chennai 600 031, India.
E-mail: [email protected]
Manuscript received: May 11, 2004, Initial review
completed: August 4, 2004;
Revision accepted: September 21, 2004.
Abstract:
This report is based on observations during
the conduct of a larger study to develop diagnostic criteria for
childhood tuberculosis (TB). Of 20l children confirmed to have
pulmonary or lymph node TB, 84 had normal chest radiographs.
Computerized tomography (CT) of the chest was performed in nine
of them, seven of whom had normal chest radiographs while two
had visible calcification. Eight of the nine children had
definitive lesions detected by computerized tomography of the
chest. While five children had primarily hilar lymph node
enlargement, three had pulmonary parenchymal lesions. The use of
more sensitive diagnostic tests like computed tomography helps
to detect tuberculosis lesions not otherwise visualized on chest
radiographs. This report highlights the difficulty in excluding
active tuberculosis in children. More studies are required on
the role of CT scans in the diagnosis of tuberculosis in
children.
Key words: Computerized tomography, Tuberculosis.
Tuberculosis (TB) in the pediatric
population produces short-term morbidity and mortality and also
serves as a reservoir for adult forms of the disease. The
natural history of tuberculosis in any age group comprises of
exposure, infection and disease. In children, the
differentiation between infection and disease is not as
clear-cut as in adults. The situation is compounded by the
absence of signs and symptoms as well as the lack of specific
radiographic features at the stage of primary infection. The
tuberculin reaction is the only known method of detecting TB
infection but is known to be falsely negative in a variety of
situations including malnutrition and in the presence of severe
viral infections. Radio-graphic changes are often found later in
the disease spectrum and are influenced by the host inflammatory
reaction. Detection of acid-fast bacilli is rare and
mycobacterial cultures take time(1). There is thus a need for
better diagnostic methods to detect tuberculosis infection and
disease in children. This report describes the computerized
tomography (CT scan) findings in nine children who showed
strongly reactive tuberculin reactions, had Mycobacterium
tuberculosis in gastric lavage cultures but had chest
radiographs that were normal. They were observed during the
course of a larger study on diagnostic strategies for childhood
tuberculosis in Chennai, South India.
Subjects and Methods
Children between the ages of six months and
twelve years attending the outpatient department of the following
hospitals were included in this study: (a) Institute of
Social Pediatrics, Government Stanley Medical College, Chennai; (b)
Department of Pediatrics, Sri Ramchandra Medical College and
Research Institute, Porur, Chennai and (c) the Kanchi
Kamakoti Child’s Trust Hospital, Chennai.
Children presenting with cough or fever for
more than two weeks, more than six episodes of respiratory
infection in three months, failure to thrive or unexplained loss
of weight were referred for the study. A strong clinical suspicion
in the absence of these symptoms was also considered as sufficient
grounds for investigation. Detailed history including all symptoms
and a history of contact with TB was recorded. Clinical
examination was conducted and findings recorded.
Chest radiographs and tuberculin tests were
performed. The pediatrician and radiologist reviewed radiographs
independently–both were blinded as to the clinical status of the
patient. In case of disagreement between the two, a third opinion
was obtained from another pediatrician. Tuberculin test was done
using 1TU PPD RT 23 with Tween 80 and induration was recorded
after 48 to 72 hours. The diagnosis of tuberculosis was based on
sputum or gastric lavage smear/culture (on two consecutive days
early in the morning) or lymph node biopsies with
histopathological examination and culture obtained from those who
had significantly enlarged superficial nodes. CT scans could be
performed only in nine children due to lack of availability. All
scans were performed at the Government Stanley Medical College and
Hospital, Chennai, after culture results had become available. All
children with TB diagnosed were treated with a full course of anti
tuberculous therapy. The study was approved by the Institute’s
Ethics committee and informed consent was obtained from the
parents.
Results
2352 children were referred for the study;
tuberculosis was confirmed in 201 either bacteriologically or
histopathologically. The chest radiographs of 84 of these children
were read as normal by the pediatrician and radiologist. We were
able to perform CT scans in nine of these children and details of
their investigations are shown in Table I. While seven of
them had normal chest radiographs (confirmed by reading the films
again), two had evidence of calcification. All were HIV-negative.
Eight had Mycobacterium tuber-culosis isolated from gastric lavage
cultures while one child had a lymph node biopsy suggestive of
tuberculosis. The CT scan was abnormal in eight of the nine
patients. While five children had primarily hilar lymph node
enlargement, the other three had pulmonary parenchymal lesions.
Table I
Findings in Nine Children with TB Disease and Normal Radiographs.
Age
(yrs)
|
Sex
|
Tuberculin
reaction (mm) |
Bacteriology
X-ray
|
Chest
|
CT scan
|
5 |
M |
24 |
Culture+ |
N |
Right paratracheal gland with
calcification. |
11
|
F
|
15
|
Lymph node
biopsy +
|
N
|
Righ tracheobronchial nodes with anterior
mediastinal mass.
|
3
|
F
|
15
|
Culture+
|
N
|
Normal
|
3
|
F
|
17
|
Culture +
|
N
|
Calcification in left hilar gland
|
3
|
F
|
14
|
Culture
|
Calcific
density
right lower
zone
|
Right retrocardiac patchy pneumonia
including calcific density
|
11
|
M
|
15
|
Culture +
|
N
|
Right hilar node enlargement
|
5
|
M
|
15
|
Culture +
|
N
|
Right mediolateral calcific node with rounded
opacity. Left lower lobe with varying density
(consolidation leading to liquification)
|
8
|
F
|
0
|
Culture +
|
N
|
Right basal lateral segment nodular
opacity at the level of diaphragm
|
6
|
M
|
15
|
Culture +
|
Calcifica-
tion in
hilar area
|
Bilaterial hilar adenitis producing obstructive
emphysema
|
Discussion
The role of CT scans in pediatric TB has been
explored in a few studies; none are from resource-restricted
settings. Delacourt, et al.(2) performed CT scans in 15
children with positive tuberculin reactions and found evidence of
enlarged intra-thoracic nodes in 60%. They suggest that CT scans
could play a role in the early detection of tuberculosis where
active disease has not been diagnosed due to absence of clinical
findings and normal chest X-rays. Similarly Gomez Pastrana,
et al.(3) detected intrathoracic lymphadenopathy in 14 of 22
(63%) children with tuberculous infection without disease. Kim,
et al.(4) while desrib- ing the CT findings of pulmonary
tuberculosis in children noted that in 20% of patients the
diagnosis was suggested only on CT scans; the characteristic CT
finding was mediastinal or hilar lymphadenopathy with rim
enhancement or calcification. Neu, et al. (5) also
described six children in whom CT scans revealed hilar/mediastinal
adenopathy with equivocal or negative chest radiographs.
During the course of our study we observed that
a fair proportion of children with gastric lavage culture positive
for Mycobacterim tuberculosis had normal chest radiographs. CT
scans done on a subset of these children picked up lesions in the
majority. Only two of the nine children reported here had any
abnormality detected on plain radiograph and both these were
calcifications. Pulmonary CT scans showed lymph node enlargement
in five, consolidation in four (with liquefaction in one) and an
anterior mediastinal mass in one patient. While there was no
classical or universal feature diagnostic of tuberculosis, the CT
scan was able to pick up significant abnormalities not detected by
plain radiographs.
Our findings raise two important issues.
Firstly, the role of CT scans in the diagnosis of pediatric TB.
Though it is a more sensitive test than plain radiograph of the
chest, it cannot be routinely recommended due to the cost and
inaccessibility in developing countries where tuberculosis is most
common. However, it can be used as an adjunct in high-risk groups
where current modalities are inconclusive and clinical suspicion
is high.
The other issue is regarding treatment. As more
sensitive tests are developed, it is becoming easier to detect
evidence of mycobacterial infection, where in the past it was not
possible. It is clear that there is a continuum between
tuberculous infection and disease, particularly in early
childhood. Six of the nine children in this study were under the
age of six years. The demonstration of unrecognized active disease
in many infected children raises the question of adequate
treatment for these children. Since it is impossible in many cases
to differentiate clearly between infection and disease with the
current diagnostic modalities, it may be advisable to treat
infection also (preventive therapy) with two drugs. A two-drug
regimen isoniazid and rifampicin) has been shown to be effective
in treating uncomplicated pulmonary tuberculosis and can
effectively be used for chemoprophylaxis also(6).
Contributors: SS and MD were responsible
for study design, monitoring and supervision. SS and AR were
responsible for data analysis and report writing. CNP was involved
in laboratory aspects and KCS helped with interpretation of
radiographs and scans.
Funding: None.
Competing Interests: None.
Key Messages |
• Children with active
tuberculosis may have normal chest radiographs.
• CT Scan is a sensitive diagnostic test to detect pulmonary
lesions.
|
|
1. Khan EA, Starke JR. Diagnosis of
tuberculosis in children: Increased need for better methods.
Emerg Infec Dis 1995; 1: 115-123.
2. Delacourt C, Mani TM, Bonnerot V, Blic
J, Sayeg N, Lallemand D, et al. Computed tomography
with normal chest radiograph in tuberculous infection. Arch
Dis Child 1993; 69: 430-432.
3. Gomez-Pastrana D, Caro Mateo P,
Torronteras Santiago R. Computed tomography and polymerase
chain reaction in tuberculosis infection in children. Arch
Bronchoneumol 1996; 32: 500-504.
4. Kim W S, Moon W K, Kim I O, Lee H J,
Im J G, Yeon K M, et al. Pulmonary tuberculosis in
children: evaluation with CT. Am J Roentgenol1997; 168:
1005-1009.
5. Neu N, Saiman L, San Gabriel P,
Whittier S, Knirsch C, Ruzal Shapiro C, et al.
Diagnosis of pediatric tuberculosis in the modern era.
Pediatr Infect Dis Journal 1999; 18: 122-126.
6. Ramachandran P, Kripasankar AS, Duraipandian M. Short
course chemotherapy for pulmonary tuberculosis in children.
Ind J Tub 1998; 45: 83-87.
.
|