Review Article

ABDOMINAL TUBERCULOSIS : CURRENT STATUS

D.K. Bhargava
Senior Consultant, Department of Gastroenterology & Hepatology, Indraprastha Apollo Hospitals,
Sarita Vihar, New Delhi 110 076, India.
e-mail: maya@ndf.vsnl.net.in

Abdominal tuberculosis continues to be reported from developing coutries and re-surgence in western
countries due to HIV infection and immigrant population. Isolates in India are only mycobacterium
Tuberculosis. It can involve any part of GI tract however predominantly ileocaecal region.
Clinical diagnosis is correct only in 50% cases. Complications includes obstruction, perforation,
malabsorption, fistulae and lower GI bleeding. Definitive diagnosis is by demonstrating characteristic
granuloma or microbiologic proof. In absence of tissue diagnosis other procedures are helpful like strongly
positive PPD, positive findings in Radiological and imaging techniques and positive ELISA test. Endoscopic
procedures has been shown to be useful for the diagnosis (gross appearance and tissue diagnosis).
Peritoneal tuberculosis occurs in majority of cases in ascitic, form and uncommonly as fibroadhesive. Ascitic
flud is exudative and cells are predominantly lymphocytic. Adenosine deaminase level of more than 36 U/L is
quite specific for the diagnosis. Laparoscopy is helpful in doubtful cases.
Management with conventional antitubercular drugs are recommended atleast for 6 months. Surgical
procedures are mostly performed for associated complications.
Key words: Abdominal tuberculosis, Gastrointestinal and peritoneal, ELISA for tuberculosis (IgG, IgM and
IgA), Colonoscopy, Adenosine demainase (ADA), Laparoscopy.

TUBERCULOSIS involvement of gastrointestinal tract,
Peritoneum, glands and other organs in abdomen continues
to be reported from developing countries due to increasing
population, poor socioeconomic status and acquired
immuno-deficiency. In western countries resurgence is due
to increasing number of immigrants, in persons infected
with HIV virus and other conditions associated with
immuno-deficiency.
It’s a disease of young adults (21-40 years), females
suffer more than male and family history is available in 2.2%
cases. Recent studies shows that associated pulmonary
lesions are present in 20 to 30% cases and in 70% cases this
is possibly due to recent infection.
PATHOPHYSIOLOGY
Mycobacterial infection spread either by hematogenous
route or swallowing of infected sputum, contaminated milk
or food and from adjacent organs. Subsequently bacilli in
G.I. tract transverse from mucosa to submucosa. Later on
inflammatory changes including granuloma formation,
lymphangitis, endarteritis, mucosal ulceration, caseating
necrosis, fibrosis and stricture occurs.
In gastrointestinal tract, pathological lesions are
predominantly ulcerohypertrophic followed by ulcerative
and fibrotic. Studies on cell mediated immunity and on
virulence of organism showed that possibly
ulcerohypertrophic lesions are due to good degree of
immune response and low virulence of the organism
whereas ulcerative lesions are due to poor immune
response and highly virulent organisms. Isolates in India
are only mycobacterium Tuberculosis. In a study reported
from Sandiago USA mycobacterium bovis was responsible
for tuberculosis in children. M. Avium is also isolated in
patients infected with HIV virus.
AREAS OF INVOLVEMENT
Gastrointestinal tract is involved in 57% cases as
followed by peritoneal in 38% and mesenteric lymphnodes
in about 6% patient. Within GI tract in 70% patients disease
affects ileum and ileocaecal region ( Relative physiological
stasis, high rate of absorption and abundance of lymphoid
tissue). Isolated lesions frequently occur in other parts of
the intestines, including colon and jejunum. Stomach and
duodenal tuberculosis each constitutes around 1% cases
and mimics peptic ulcer disease. Esophageal tuberculosis
is still rare and can present as dysphagia , odynophagia
and ulcer.
SYMPTOMS OF GI TUBERCULOSIS
Despite a high index of suspicion: TB is difficult to

287 Apollo Medicine, Vol. 4, No. 4, December 2007

Review Article
diagnose (a) symptoms are vague (b) Signs are non-specific
(c) mimicks other diseases like crohn’s and cancer.
Clinically diagnosis is correct only in 50% cases. Colicky
mid abdominal and or RLQ pain suggesting partial small
bowel obstruction is the presenting complaint in 90-100%
patients, weight loss in 63% fever 61%, anorexia and
vomiting in 50% patients. Change in bowel habits occurs in
20% patients. Other complaints includes borborgmi,
amenorrhoea and rectal bleeding.
PATTERNS OF PRESENTATION
Different patterns can be derived which are helpful in
suspecting diagnosis. These included non-specific vague
symptoms in 30%, ascites 37% abdominal masses in about
33% of patients. Other presentation includes symptoms
suggestive of obstruction, malabsorption, perforations and
fistulae (Enteroenteric, coloenteric, coloduodenal and
phyloduodenal fistulas may occur). External fistulas are
rare but may follow after surgery. Perforation can occur in 8
to 15% and lower GI bleeding in 4% patients.
DIAGNOSIS
Definitive diagnosis of tuberculosis is by demonstrating
characteristics granuloma or microbiological proof on
histological slides, culture or PCR techniques. Other
features are gross appearance of tissue with caseating
granuloma or caseation in lymph nodes, or non caseating
granulomas with tuberculosis elsewhere responding
to chemotherapy. For these criteria tissue has to be
obtained by endoscopy or surgery. In absence of tissue
diagnosis, certain procedures are supportive of clinical
diagnosis.
(a) Laboratory investigations: can be nonspecific or
normal. The most common abnormal findings are
elevated ESR in 90% and mild anemia in 80% patients.
Montoux test (by using PPD 5 Tu): Induration more
than 20 mm in 67% cases, induration between 10-20
mm in 30% and 10-15 mm in 3% was reported in our
studies. Indurations more than 15 mm is suggestive of
recent infection. Strongly positive PPD test should be
considered in favour of TB.
(b) Radiological Investigations: Pulmonary involvement
has been noted in 20-30% patients include active
infiltration or older lesions. Plain X-ray abdomen
reveals positive findings in 18-30% patients
(obstruction / calcification). Barium contrast studies
are traditional methods of evaluation. Findings include
stricture with dilatation, Deformed and pulled up
cecum, ulcers and masses. These features are non-
specific as similar findings are noted in other diseases.
Findings are present in 60% patients. However, still
useful in assessment of lesions and localizing the site.
Apollo Medicine, Vol. 4, No. 4, December 2007 288
Enteroclysis further improves that diagnostic accuracy
to 75%.
(c) Imaging Techniques: Ultrasound or CT scan often
reveals findings such as thickening and nodularity of
mesentry, enlarged lymphnodes, omental masses, high
density ascites, thickened peritoneum or intestinal wall,
interlacing septation, and dilated small bowel loops.
These features are non-specific.
(d) Serological methods: Improvement in serological
methods for diagnosing tuberculosis represent recent
advance. This has occurred with the advent of specific
antigens like antigen 5, KD 38 and A 60. False negative
results may be dur to variable antigenic load and false
positive results due to presence of latent, inactive or sub
clinical infection and cross reactivity among antigens
of mycobacterial, fungi and other bacteria. Soluble
antigen fluorescent antibody test (SAFA) and Enzyme
linked immunosorvent assay (ELISA) has been
evaluated. Both test differentiated TB from other
diseases though false positive and false negative results
were obtained. ELISA technique is simpler and
showed a sensitivity of 80-84% and specificity or 88 to
95% and diagnostic accuracy in 84% cases. Three types
of antibody response e.g., Igm, IgG and IgA can be
measured. IgM detectable for upto 10 days of infection,
IgG indicated established infection and IgA presence
indicates frequent contact with infectious foci. Golden
rule; serological data must be integrated with clinical,
radiological, imaging and endoscopic findings before
arriving at any conclusion.
ENDOSCOPIC PROCEDURES
Endoscopic procedures now can evaluate gastro-
intestinal tract. These include UGI Endoscopy for
oesophagus, stomach and duodenum. Push endoscope
(Enterscopy) evaluates jejunum. Colonoscopy with
intubation of terminal ileum evaluates 70 to 80 % lesions.
Capsule endoscopy for small bowel lesions is yet to
be evaluated. In recent years Colonoscopy with
colonoscopically obtained biopsy specimens has been
shown to be useful for the diagnosis of colonic and ileocecal
tuberculosis.
1. Lesions in primary esophageal tuberculosis includes
ulcer with whitish base and nodules in the vicinity of
ulcer margin. In gastric tuberculosis lesions are mostly
obstructive like deformed Pylorus, ulcers and nodules.
These lesions mimicks cancerous lesions. Lesions of
duodenal tuberculosis are like duodenal ulcer. Lesions
in small bowel tuberculosis also revealed ulcers, nodule
and stricture. Biopsies obtained for histology and
cultures of the tissue are rewarding.

2. Colonoscopy: Lesions of ileo- cecal and colonic
tuberculosis manifest in the from of:
(a) Deformed ileo-cecal valve: Edematous labial folds
along with nodules. No change due to ileal
movements.
(b) Mucosal nodules measuring 2 to 6 mm in size with
pinkish surface scattered and at places densely
packed. Multiple small ulcers are located between
the nodules. These represent tubercle formation
along with cellular infilitration and edema in
submucosa and mucosa. As a result of deprivation
of blood supply probably through end arteries the
overlying mucosa, further swells and mucosal
breakdown leads to formation of ulcers.
(c) Ulcers are single or multiple. They are located on
abnormal mucosa. Surrounding mucosa is
inflamed (edema, nodules). Ulcers are located
either on transverse axis or circumferential.
Margins of ulcers are sharply defined and
erythematous and base is covered with whitis to
yellowish slough.
(d) Heaped up folds and fibrous bands, mimicking
polyps are also seen.
(e) Strictures at times also covered with nodules and
ulcers.
Colonoscopic Biopsies: Establishes histopathological
diagnosis (granulomas) in about 40 % patients. It has also
been shown that cultures alone were positive in 46% cases.
Combined culture and histology in same patients can yield
diagnosis in 60% cases. Thus the tissue should be utilized
for both the methods. AFB on histology slides were rarely
seen. Six to eight biopsies should be obtained preferably
form same spot. This may result in obtaining submucosal
tissue though granulomas were detected in mucosa and
submucosa. In our study more granulomas were detected
form ulcer margins and mycobacteria from mucosal
nodules. Cultures were positive even in those patients
where histology was not positive for TB. All our isolates
were Mycobacterium tuberculosis.
Differential Diagnosis: Tuberculosis has to be
differentiated from neoplasm, amoebic lesions and Cronic
disease. Colonic neoplasm includes carcinoma and
lymphoma. These can be differentiated on endoscopic
appearance and biopsies (positive in more than 90 % cases).
Amoebic ulcers are either pinhead size or large and have
rolled up margins and in between, the ulcers the mucosa is
normal. E. histolytica trophozoites are present in biopsy
material. Amoeboma and rarely stricture are seen. Crohns
289
Review Article
disease, which use to be rarely seen in India now frequently
encountered. IC valve is deformed in TB and usually spared
in crohns. In TB ulcers are single/multiple on an abnormal
mucosa with sharply on normal mucosa, deep linear and
located on longitudinal axis. Nodules are present in TB and
cobblestonig and Pseudopolyps are the features of Crohns
disease. On biopsy specimens; caseating granulomas are
the features of TB. Non caseating granulomas can be
differentiated; in TB they are confluent large with
peripheral cells and hyalinization. Fibrosis usually
accompanies. In Crohns disease granulomas are discrete
and small. Fibrosis and hyalinization if rare. Sub mucosal
widening and fissure are common in Crohn’s disease.
TUBERCULOUS PERITONITS
It is a frequent cause of ascites in India and other
underdeveloped countries. Peritoneum involvement is the
second most common site (37.6%) after GI tract.
Simultaneous involvement of intestines can occur as 28 %
patients can present with symptoms and signs of
obstruction. Majority of patients (97%) presents in ascetic
form and about 3% as fibroadhesive (intestines, measentry,
omentum and peritoneum glued together).
The illness often occurs quite insidiously. Most patients
will have symptoms for several weeks to months at the time
of presentation Abdominal swelling is the most common
symptoms (82%), fever in 74 % followed by anorexia
weight loss and abdominal pain.
DIAGNOSIS
(i) Ascitic fluid is exudative (protein 2.5 to 3 g/dL) in
nature. The serum –ascites albumin difference is <1.1
g/dL. Ascitic fluid white blood count is 150 to 4000
mm3 and consists predominantly of lymphocytes.
Findings on abdominal CT scan and ultrasound are
nonspecific and includes high density ascites,
adenopathy, omental and mesenteric thickening.
Examination of acid-fast stained smear of ascitic fluid
will identify the organism in less than 3% cases. Culture
of fluid will be positive in less than 20% cases and that
too takes 4-8 weeks.
(ii) Adenosine deaminase activity ( ADA)
The major recent advance in the diagnosis of
tuberculous peritonitis is determination of ascitic fluid
adenosine deaminase activity. Adenosine deaminase is
an enzyme of purine catabolism, which catalizes
deamination of adenosine to inosine and ammonia.
Enzyme activity is ten times higher in lymphocytes than
RBC and 3-10 time more in T-lympocytes than
B-lymhocytes. In tuberculosis T-cell proliferation
occurs in response to cell mediated immunity. Our
Apollo Medicine, Vol. 4, No. 4, December 2007
Review Article
studies followed by number of other studies have
shown that at a cutoff of >36U/L the sensitivity and
specificity in tuberculous ascits to be about 100% and
97% respectively. ADA test differentiates TB from
other peritoneal diseases at this level. Meta-analysis of
four prospective studies concludes that using cut of
values from 36-40 IU/L showed high sensitivity
(100%) and specificity (97%). In cirrhosis of the liver
ADA levels may be lower. Thus, it is a useful and
inexpensive test and potentially supersede invasive
studies. Anti tubercular treatment can by started on the
basis of positive ADA.
(iii) Laparoscopy: Laparoscopy with directed biopsy is an
excellent method of diagnosing TB, in contrast to blind
percutaneous peritoneal biopsy. The laparoscopic
appearances include thickened peritoneum with
tubercles, thickened peritoneum alone and in some
fibroadhesive changes. Tubereles are multiple,
yellowish white in colour and of uniform size (2 to 5
mm) thickened peritoneum appears as hyperemic with
loss of usual shiny luster. In fibroadhesive parietal
peritoneum is thick, shows yellowish white nodules
and cheesy material. Multiple thick adhesions fixes the
viscera to abdominal wall.
The gross appearance itself is diagnostic. In our study
sensitivity of gross appearance was 94%, positive
tubercular histology in 80% and cultures of the tissue
yielded microbacterium tuberculosis in 40% cases.
Laparoscopy is a safe procedure in these patients. Thus
characteristic changes on laparoscopy can differentiate
them from other diseases and even in absence of
histology chemotherapy for TB can be instituted.
POLYMERASE CHAIN REACTION (PCR)
PCR in tissue provides rapid diagnosis usually within 48
to 72 hours. It can be helpful in differentiating TB from
Crohn’s disease and for it spositive result presence of
minimal number of bacilli will suffice. Sensitivity of this
test is 75-80% and specificity varies between 85 to 90%
patients. However, implementation of effective system for
monitoring specificity and sensitivity is required before
PCR can be used reliably. Further contamination of tissue
should be avoided.
HIV AND TUBERCULOSIS
In presence of HIV disease, tuberculosis is more florid.
The changes of tuberculosis in X-rays and CT scan are more
severe. There are more AFB in pathological specimen
which are Mycobacterium Tuberculosis and avium in
nature. Immunological response is poor and complications
of TB are more prevalent with higher mortality.
Apollo Medicine, Vol. 4, No. 4, December 2007 290
TREATMENT
(i) MEDICAL: Chemotherapeutic drugs are same as for
pulmonary TB. Current treatment consists of used of
INH + Rifampicine + Ethambutol + Pyrazinamide for 2
months followed by INH + Rifampian for 4 months. In
a randomized trial 6 months treatment was compared
with 12 months treatment. Both schedules were equally
efficacious. Role of corticosteriods is controversial.
Physicians are using them for reduction of adhesions
or fibrosis.
(ii) SURGICAL: For complications like obstruction,
perforation, fistula and massive haemorrhage.
ACKNOWLEDGEMENTS
Studies on Abdominal Tuberculosis were supported by
grants from Indian Council of Medical Research and All
India Institute of Medical Sciences and were conducted at
AIIMS, New Delhi.
BIBLIOGRAPHY
1. Bhansali SK. Abdominal Tuberculosis. Experiences with
300 cases. Am J Gastroenterol 1977, 67: 324-327.
2. Tandon HD, Prakash A. Pathology of Intestinal
Tuberculosis and its distinction from Crohns disease. Gut
1972, 13 ; 260-269.
3. Kapoor VK. Abdominal Tuberculosis: The Indian
Contribution. Indian J Gastroenterol 1998, 17: 141-147.
4. Bhargava DK, Chawla TC, Tandon BN, Shriniwas, Kapur
BML, Tandon HD. Cell mediated immune response in
intestinal tuberculosis. Indian J Medical Research 1984,
80 : 277-282.
5. Bhargava DK, Shriniwas, Chawla TC, Kapur BML, Tandon
HD. Intestinal Tuberculosis: Bacteriological study of
tissue obtained by colonoscopy and during surgery.
J Tropical medicine & Hygiene 1985, 88: 249-252.
6. Gupta SP, Arora A, Bhargava DK. An unusual presentation
of esophageal tuberculosis.Tubercle 1992, 73: 174-176.
7. Kapoor VK, Chattopadhyay TK, Sharma LK. Radiology of
abdominal tuberculosis. Australas Radiol 1988, 32: 365-
367.
8. Kedar RP, Shah PP, Shivde RS. Sonographic findings in
gastrointestinal and peritoneal tuberculosis. Clin Radiol
1994, 49: 24-29.
9. Jain R, Sawhney S, Bhargava DK, et al. Diagnosis of
abdominal tuberculosis: Sono-graphic findings in patients
with early disease. Am J Roentgenol 1995, 165(6): 1391-
1395.
10. Gulati MS, Sarma D, Paul SB. CT appearances in
abdominal tuberculosis. A pictorial assay. Clin Imaging
1999, 23 : 51-59.
11. Bhargava DK, Shriniwas, Tandon BN, Kiran, Chawla TC,

Kapur BML : Serodiagnosis of intestinal tuberculosis by
soluble antigen flurescent antibody test (SAFA test).
Tropical med & Hygiene 1986, 89 : 61-65.
12. Bhargava DK, Dasarathy S, Shriniwas, Kushwaha AKS,
Dhupare H, Kapur BML. Evaluation of Enzyme linked
immunosorbent assay using Mycobacterial Saline
extracted antigen for the serodiagnosis of abdominal
tuberculosis. Am J Gastroenterology 1992, 87: 105-108.
13. Chawla TC,Sharma A, U Kiran, Bhargava DK , Tandon BN.
Serodiagnosis of Intestinal Tuberculosis by Enzyme
Immunoassay and soluble antigen fluorescent antibody
tests using a saline extract antigen. Tubercle 1986, 76:
55-60.
14. Gandhi BM, Bhargava DK, Irshad M, Chawla TC, Dibe A.
Enzyme linked protein A – An ELISA for detection or IgG
antibodies against M.Tuberculosis. Tubercle 1986, 67:
219 -224.
15. Bhargava DK, Tandon HD. Ileocaecal tuberculosis
diagnosed by colonoscopy and biopsy. Australian NZ surg
1981, 21: 225.
16. Bhargava DK, Tandon HD, Chawla TC, Shriniwas, Tandon
BN, Kapur BML. Endoscopic Diagnosis of ileocaecal and
colonic tuberculosis by colonoscopy. Gastrointestinal
Endoscopy 1985, 31: 68-70.
291
Review Article
17. Bhargava DK, Kushwaha AKS, Dasarathy S,Shriniwas,
Chopra P. Endoscopic diagnosis of segmental colonic
tuberculosis. Gastrointestinal Endoscopy 1992, 38: 571 -
574.
18. Bhargava DK, Shriniwas, Chopra P, Nijhawan S,
Dasarathy S, Kushwaha AKS, Peritoneal tuberculosis:
Laparoscopic patterns and its diagnostic accuracy.
American J Gastroenterology 1992, 87: 109-112.
19. Bhargava DK, Nijhawan S, Gupta M. Adenosine
deaminase and Tuberculous peritonitis. Lancet 1989, 1:
1261.
20. Bhargava DK, Gupta M, Nijhawan S, Dasarathy S,
Kushwaha AKS, Adenosine deaminase in peritoneal
tuberculosis: Diagnostic value in ascitic fluid and serum:
Tubercle 1990, 71: 1121-126.
21. Riquelime A, Calvo M, Salech F, et al. Value of adenosine
deaminase (ADA) in ascitic fluid for the diagnosis of
tuberculosis peritonitis: A meta-analysis. J Clin Gastro-
enterol 2006, 40: 705-710.
22. Balasubramanian R, Nagarajan M, Balambal R, et.al.
Randomised controlled clinical trial of short course
chemotherapy in abdominal tuberculosis: a five
year report. Int J Tuberc Lung Disease 1997, 1:
44-51.
Apollo Medicine, Vol. 4, No. 4, December 2007