Clinical, Colonoscopic, and Histological Profile of Colonic Tuberculosis in a Tertiary Hospital

• Patients and Methods
• Results
• Discussion
• References

Background and Study Aims: Colonic tuberculosis is not uncommon in developing countries. As emigration to the West increases, it is worthwhile to recall the clinical, colonoscopic, and histopathological features of this condition.
Patients and Methods: The clinical, colonoscopic and histopathological findings were evaluated in 43 patients with colonic tuberculosis.
Results: Abdominal pain, weight loss, diarrhea, fever, and a lump in the abdomen were the commonest symptoms. Extraintestinal tuberculosis was present in 11 patients (26 %). Colonoscopy revealed ulcers in 30 patients (70 %), nodules in 24 (56 %), a deformed cecum and ileocecal valve in 17 (40 %), strictures in 10 (23 %), polypoid lesions in six (14 %), and fibrous bands forming mucosal bridges in three (7 %). The cecum and ascending colon were the commonest sites involved. Segmental tuberculosis was seen in six of the 32 patients (19 %) in whom full-length colonoscopy could be performed. Two or more sites were involved in 19 patients (44 %). Histopathology revealed well-formed granulomas in 23 patients (54 %). Fourteen of the above patients (61 %) had caseation and 11 (48 %) had confluence of the granulomas. Acid-fast bacilli were present in the biopsies from two patients (5 %). Ill-formed granulomas were seen in seven patients (16 %) and chronic inflammatory changes in 13 (30 %). Despite the various histopathological findings, all of the patients responded to antitubercular treatment and continued to remain asymptomatic during the follow-up period.
Conclusions: Colonoscopy with biopsy is a useful method for diagnosing colonic tuberculosis. Even in the absence of the classic histopathological features, a therapeutic trial may be indicated in a given clinical and colonoscopic setting. Follow-up is essential.

With the reemergence of tuberculosis in the West due to the epidemic of acquired immune deficiency syndrome (AIDS) and increased immigration from developing countries, physicians in the Western world should have a heightened awareness of colonic tuberculosis as a potential source of abdominal symptomatology. The clinical, colonoscopic, and histological features of the condition have been well described previously [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]; the purpose of this article is to increase awareness of this problem by reviewing the data for 43 patients treated in our hospital.

Patients and Methods

Seventy-four patients underwent colonoscopy for a suspected diagnosis of colonic tuberculosis over a period of 11 years (1988 - 1999). Of these, only 43 for whom regular follow-up data during the therapy were included. Others were excluded for the following reasons: normal colonoscopy (n = 15), no or inadequate follow-up (n = 12), malignancy (n = 2) and probable Crohn’s disease (n = 2). Chest radiographs were taken in all of the patients, and the erythrocyte sedimentation rate was assessed in 37. Colonoscopy was carried out with a fiberoptic colonoscope (Olympus CF-LB3W; Olympus Optical Co., Tokyo, Japan) after mannitol preparation in the initial years, and with a video colonoscope (Olympus CF-130L) after preparation with polyethylene glycol with balanced electrolytes in the later years. Four to six biopsies from each abnormal site were taken and examined histopathologically (hematoxylin-eosin staining) and with staining for acid-fast bacteria (Ziehl-Neelsen stain). Culture of biopsies for Mycobacterium tuberculosis was not carried out.

The diagnosis of colonic tuberculosis was based on colonoscopy with histological evidence, or evidence of tuberculosis elsewhere, coupled with a good response to antitubercular therapy. Well-formed granulomas were defined as well-defined clusters of epithelioid cells with or without Langhans giant cells and caseation. Ill-formed granulomas were defined as small, poorly organized collections of epithelioid cells without the other features of granulomas.

Results

The mean age of the patients was 35.3±12.3 years. There were 20 men and 23 women. The clinical data are shown in Table [1]. Eleven patients had extracolonic tuberculosis (TB), including pulmonary findings (n = 7), lymph nodes (n = 5), peritoneal (n = 2), pleural (n = 1), and spinal (n = 1). Colonoscopy as far as the cecum was possible in 32 patients. In the remaining 11, the area proximal to the stricture could not be visualized in nine patients, and full-length colonoscopy could not be achieved for technical reasons in two. The cecum and ascending colon were the commonest sites involved (n = 25), followed by the transverse colon (n = 7), descending colon (n = 4), sigmoid colon (n = 4) and rectum (n = 1) The entire colon was involved in one patient. Two or more sites were involved in 19 patients (44 %). The terminal ileum was involved in nine of the 21 patients in whom it could be intubated. Colonic tuberculosis without involvement of the ileocecal area (segmental tuberculosis) was seen in six of the 32 patients (19 %) in whom full-length colonoscopy was possible. The colonoscopic findings are described in Table [2] and illustrated in Figure [1]- [10]. The ulcers were transverse, measuring 0.5 - 3.0 cm in size, and the base was covered by an exudate. The nodules were 0.3 - 1.0 cm in size, with some appearing in clusters and others in isolation.

Histopathology revealed well-formed granulomas in 23 cases (54 %) (Figure [11]), ill-formed granulomas in seven (16 %) (Figure [12]), and chronic inflammatory changes in 13 patients (30 %). Among the patients who had well-formed granulomas on endoscopic biopsies (n = 23), 14 (61 %) had caseation and 11 (48 %) had confluence of the granulomas. Acid-fast bacilli were seen in two patients (5 %) (Figure [13]).

Antitubercular therapy with rifampicin (10 mg/kg) p. o., isoniazid (5 mg/kg) p. o., pyrazinamide (30 mg/kg) p. o., and ethambutol (15 mg/kg) p. o. or streptomycin (0.75 - 1.0 g) i. m. was given for 2 months, followed by rifampicin (10 mg/kg) p. o. and isoniazid (5 mg/kg) p. o. for an additional 7 months. Two patients underwent a diagnostic laparotomy, as the colonoscopic biopsies showed only chronic inflammation, with no evidence of tuberculosis elsewhere, and the treating surgeon was reluctant to administer empirical antitubercular therapy. The surgical specimens showed well-formed granulomas with Langhans giant cells and caseation in both patients.

Follow-up of these 43 patients after the start of antitubercular therapy included a hospital visit every month until the end of the therapy. Two patients developed intestinal obstruction within 3 weeks after starting antitubercular therapy; laparotomy revealed an abscess in the right ileocolic region, with enlarged lymph nodes and an ileal stricture, respectively. Both patients were continued on antitubercular therapy. All of the 43 patients were asymptomatic at the end of 9 months of antitubercular therapy. Follow-up after completion of the 9 months of antitubercular therapy included a questionnaire sent by post to the patient’s address, inquiring as to whether they had had any recurrence of symptoms or had developed tuberculosis in other sites. The questionnaires were sent 1 - 11 years after completion of antitubercular therapy. Forty of the 43 patients replied, and all had been asymptomatic since the completion of therapy.

Discussion

The global prevalence of Mycobacterium tuberculosis infection is estimated to be 32 % (1.86 billion people) [16]. Each year, about 7.96 million people develop the disease and 1.87 million people die of the disease [16]. The resurgence of tuberculosis appears to be in part responsible for the increased involvement of extrapulmonary sites. Awareness of the myriad forms that tuberculosis may take will help expedite early diagnosis and treatment.

The pathophysiology of tubercular enteritis is attributed to hematogenous spread from active pulmonary or miliary TB, contiguous spread from adjacent organs, swallowing of infected sputum in active pulmonary TB, or ingestion of contaminated milk [17]. Hence, pulmonary involvement is not invariable, and can range from 23 % to 75 % [3] [18]. After entering the gastrointestinal tract, the tubercle bacillus traverses the mucosa, lodges in the submucosa, and induces inflammatory changes and eventually granuloma formation. Lymphangitis, endarteritis, and fibrosis ensue, leading to mucosal ulceration, caseating necrosis, and narrowing of the intestinal lumen. The predilection for the ileocecal region is due to the abundance of lymphoid tissue and relative physiological stasis in this site [17].

Tuberculosis is a great mimicker; occasionally, malignancy and ulcerative colitis need to be differentiated from it, but usually it has to be differentiated from Crohn’s disease and vice versa. Although the present study focused only on colonic tuberculosis, it may be useful to review the features described in the literature that differentiate tuberculosis from Crohn’s disease.

Differentiating colonic tuberculosis from Crohn’s disease can be problematic. Endoscopically, the ulcers in both conditions can be of variable size, shape, and depth; however, in tuberculosis, they are generally transversely oriented ulcers with the margins sharply defined, with surrounding erythema. The mucosa surrounding the ulcers may have features of inflammation such as erythema, nodularity, or edema [5]. In Crohn’s disease, linear ulcers parallel to long axis of the bowel are characteristic [15], and they are generally surrounded by normal-appearing mucosa. Deep, tortuous ulcers are more common in Crohn’s disease than in tuberculosis [5]. Aphthoid erosions or ulcerations, the hallmark of Crohn’s disease, rarely occur in tuberculosis, and are usually but not invariably associated with the classic transversely oriented ulcers [19]. The ulcers were transversely oriented in the patients in the present study. Sessile, firm polyps adjacent to the ulcers and multiple small diverticula scattered between the ulcers characteristically occur in tuberculosis, with no similar findings in Crohn’s disease [15]. Cobblestoning of the mucosa, although strongly suggestive of Crohn’s, has occasionally been reported in tubercular colitis [5]. Both diseases can involve the ileocecal valve in the form of nodularity, deformity, or ulceration, but a patulous ileocecal valve with surrounding heaped-up folds, or one that is destroyed by disease, is more likely to be due to tuberculosis than Crohn’s disease [5]. The strictures due to tuberculosis are generally, but not invariably, short (less than 3 cm) [5]. The mucosal bridges reported with inflammatory bowel disease can also occur in colonic tuberculosis [3] [12].

Granulomas occur in both tuberculosis and Crohn’s disease, with caseation characteristically occurring in tuberculosis. Caseation may be seen only in the lymph nodes without concomitant caseation in the colonic biopsy; additionally, it may be totally absent in those who have received antitubercular therapy in the past [17]. In noncaseating tuberculosis, the granulomas are usually large, multiple, and confluent, whereas in Crohn’s disease, they are noncaseating, infrequent, small, and with microgranulomas [20]. Focally enhanced colitis and fissures typically occur in Crohn’s disease [20]. A comparative study was not carried out here, as the number of patients with Crohn’s disease treated in the hospital was small.

When transversely oriented ulcers occur with caseating and confluent granulomas, with or without acid-fast bacilli, a diagnosis of tuberculosis seems certain. In its absence, or when histopathology reveals only chronic nonspecific inflammation, loose collection of epithelioid cells, or noncaseating granulomas, the diagnostic dilemma increases, and one should rely on multiple factors: corroborative evidence of tuberculosis elsewhere if present, including biopsies of accessible lymphadenopathy, orientation of ulcers on colonoscopy, size and confluence of the granulomas, history of contact with tuberculosis, ethnicity of patient, a Mantoux test, culture and polymerase chain reaction (PCR) of the biopsy specimen, radiologic assessment with computed tomography (CT) or ultrasonography, and a therapeutic response to antitubercular therapy. The Mantoux test is positive in 70 - 86 % of patients, but has limited usefulness in immunosuppressed patients [17].

Culture of colonic biopsy for Mycobacterium tuberculosis, the gold standard for diagnosing tuberculosis, has limitations: false-positive results may occur in patients with active pulmonary tuberculosis due to swallowed sputum, there is a variable yield (0 - 69 %) [3] [21], and the test takes 4 - 6 weeks with the traditional method of culturing. Cultures were not done in the present study, since in addition to being costly, the reports from our laboratory were not reliable. Rapid methods of culture such as those using radiolabeled CO₂ were not available in our institution. PCR, although highly specific is not markedly sensitive (26.5 - 75.0 %), especially for intestinal tuberculosis, but allows exclusion of Crohn’s disease with a specificity of 100 % [22] [23]; this was not available during most of the study period.

Ascites, diffuse omental and mesenteric infiltration, or nodules and lymphadenopathy with peripheral enhancement on CT scan or a hypoechoic center on ultrasonography, favor a diagnosis of tuberculosis [24]. Serological markers such as anti-cord factor antibodies, which are promising [25], and anti-Saccharomyces cerevisiae antibodies, which were not found to be useful in a preliminary report [26], require validation in further studies on differentiating tuberculosis from Crohn’s disease.

Occasionally, a therapeutic trial with antitubercular drugs may be needed in case of nondiagnostic histopathological features; Crohn’s disease is unlikely to respond to these drugs [27]. On receiving therapy with antitubercular drugs and during the follow-up period (1 - 11 years after completing antitubercular therapy), the patients in the present study were asymptomatic, a feature unlikely in Crohn’s disease.

Unfortunately, many of the above tests are not widely available, and differentiation between tuberculosis and Crohn’s disease still relies on histology, culture, a therapeutic response to antitubercular medications, and in recent years also PCR.

In conclusion, the possibility of colonic tuberculosis, especially in immigrants, should be entertained and in a given clinical and colonoscopic setting, a therapeutic trial with antitubercular drugs may be indicated even in the absence of typical histopathological features on the biopsy specimens. Follow-up is essential; tuberculous patients will respond to therapy and continue to remain asymptomatic.

References

• 1 Bhansali S K. The challenge of abdominal tuberculosis in 310 cases.  Indian J Surg. 1978;  40 65-77
  PubMedSearch in Google Scholar
• 2 Tabrisky J, Lindstrom R R, Peters R, Lachman R S. Tuberculous enteritis: review of a protean disease.  Am J Gastroenterol. 1975;  63 49-57
  PubMedSearch in Google Scholar
• 3 Singh V, Kumar P, Kamal J. et al . Clinicocolonoscopic profile of colonic tuberculosis.  Am J Gastroenterol. 1996;  91 565-568
  PubMedSearch in Google Scholar
• 4 Bhargava D K, Tandon H D, Chawla T C. et al . Diagnosis of ileocecal and colonic tuberculosis by colonoscopy.  Gastrointest Endosc. 1985;  31 68-70
  CrossrefPubMedSearch in Google Scholar
• 5 Shah S, Thomas V, Mathan M. et al . Colonoscopic study of 50 patients with colonic tuberculosis.  Gut. 1992;  33 347-351
  CrossrefPubMedSearch in Google Scholar
• 6 Aoki G, Nagasako K, Nakae Y. et al . The fibercolonoscopic diagnosis of intestinal tuberculosis.  Endoscopy. 1975;  7 113-121
  PubMedSearch in Google Scholar
• 7 Bhargava D K, Kushwaha A KS, Dasarathy S. et al . Endoscopic diagnosis of segmental colonic tuberculosis.  Gastrointest Endosc. 1992;  38 571-574
  PubMedSearch in Google Scholar
• 8 Kalvaria I, Kottler R E, Marks I N. The role of colonoscopy in the diagnosis of tuberculosis.  J Clin Gastroenterol. 1988;  10 516-523
  PubMedSearch in Google Scholar
• 9 Pettengell K E, Pirie D, Simjee A E. Colonoscopic features of early intestinal tuberculosis (report of 11 cases ).  S Afr Med J. 1991;  79 279-280
  PubMedSearch in Google Scholar
• 10 Ferentzi C V, Sieck J O, Ali M A. Colonoscopic diagnosis and medical treatment of ten patients with colonic tuberculosis.  Endoscopy. 1988;  20 62-65
  PubMedSearch in Google Scholar
• 11 Radhakrishnan S, Nakib B, Shaikh H, Menon N K. The value of colonoscopy in schistosomal, tuberculous and amebic colitis: two year experience.  Dis Colon Rectum. 1986;  29 891-895
  PubMedSearch in Google Scholar
• 12 Misra S P, Misra V, Dwivedi M, Gupta S C. Colonic tuberculosis: clinical features, endoscopic appearance and management.  J Gastroenterol Hepatol. 1999;  14 723-729
  CrossrefPubMedSearch in Google Scholar
• 13 Das P, Shukla H S. Clinical diagnosis of abdominal tuberculosis.  Br J Surg. 1976;  63 941-946
  PubMedSearch in Google Scholar
• 14 Singh V, Jain A K, Agrawal A K. et al . Clinicopathological profile of abdominal tuberculosis.  Br J Clin Pract. 1995;  49 22-24
  PubMedSearch in Google Scholar
• 15 Naga M I, Okasha H H, Ismail Z. et al . Endoscopic diagnosis of colonic tuberculosis.  Gastrointest Endosc. 2001;  53 789-793
  CrossrefPubMedSearch in Google Scholar
• 16 Dye C, Scheele S, Dolin P. et al . Global burden of tuberculosis: estimated incidence, prevalence and mortality by country.  JAMA. 1999;  282 677-686
  CrossrefPubMedSearch in Google Scholar
• 17 Horvath K D, Whelan R L. Intestinal tuberculosis: return of an old disease.  Am J Gastroenterol. 1998;  93 692-696
  PubMedSearch in Google Scholar
• 18 Kim K M, Lee A, Choi K Y. et al . Intestinal tuberculosis: clinicopathologic analysis and diagnosis by endoscopic biopsy.  Am J Gastroenterol. 1998;  93 606-609
  CrossrefPubMedSearch in Google Scholar
• 19 Tarumi K, Koga H, Lida M. et al . Colonic aphthoid erosions as the only manifestation of tuberculosis: case report.  Gastrointest Endosc. 2002;  55 743-745
  CrossrefPubMedSearch in Google Scholar
• 20 Pulimood A B, Ramakrishna B S, Kurian G. et al . Endoscopic mucosal biopsies are useful in distinguishing granulomatous colitis due to Crohn’s disease from tuberculosis.  Gut. 1999;  45 537-541
  CrossrefPubMedSearch in Google Scholar
• 21 Sakai Y. Colonoscopic diagnosis of the intestinal tuberculosis.  Mater Med Pol. 1979;  11 275-278
  PubMedSearch in Google Scholar
• 22 Gan H T, Chen Y Q, Ouyang Q. et al . Differentiation between intestinal tuberculosis and Crohn’s disease in endoscopic biopsy specimens by polymerase chain reaction.  Am J Gastroenterol. 2002;  97 1446-1451
  CrossrefPubMedSearch in Google Scholar
• 23 Anand B S, Schneider F E, El-Zaatari F A. et al . Diagnosis of intestinal tuberculosis by polymerase chain reaction on endoscopic biopsy specimens.  Am J Gastroenterol. 1994;  89 2248-2249
  PubMedSearch in Google Scholar
• 24 Ha H K, Kim J K. The gastrointestinal tract. In: Haaga JR, Lanzieri CF, Gilkeson RC (eds) CT and MR imaging of the whole body, vol. 2. St. Louis; Mosby 2003: 1154-1169
  Search in Google Scholar
• 25 Kashima K, Oka S, Tabata A. et al . Detection of anti-cord factor antibodies in intestinal tuberculosis for its differential diagnosis from Crohn’s disease and ulcerative colitis.  Dig Dis Sci. 1995;  40 2630-2634
  PubMedSearch in Google Scholar
• 26 Patel N, Amarapurkar D N, Kulshreshta P P. et al . Utility of serological markers in inflammatory bowel disease.  Indian J Gastroenterol. 2002;  21 A37
  PubMedSearch in Google Scholar
• 27 Thomas G AO, Swift G L, Green J T. et al . Controlled trial of antituberculous chemotherapy in Crohn’s disease: a five year follow up study.  Gut. 1998;  42 497-500
  PubMedSearch in Google Scholar

J. F. Alvares, M. D., D. M., D. N. B.

c/o Adv C. F. Alvares · Near the Holy Spirit Church · Margao-Goa · India ·

Fax: +91-832-2711883

Email: filipealvares@yahoo.co.in

References

• 1 Bhansali S K. The challenge of abdominal tuberculosis in 310 cases.  Indian J Surg. 1978;  40 65-77
  PubMedSearch in Google Scholar
• 2 Tabrisky J, Lindstrom R R, Peters R, Lachman R S. Tuberculous enteritis: review of a protean disease.  Am J Gastroenterol. 1975;  63 49-57
  PubMedSearch in Google Scholar
• 3 Singh V, Kumar P, Kamal J. et al . Clinicocolonoscopic profile of colonic tuberculosis.  Am J Gastroenterol. 1996;  91 565-568
  PubMedSearch in Google Scholar
• 4 Bhargava D K, Tandon H D, Chawla T C. et al . Diagnosis of ileocecal and colonic tuberculosis by colonoscopy.  Gastrointest Endosc. 1985;  31 68-70
  CrossrefPubMedSearch in Google Scholar
• 5 Shah S, Thomas V, Mathan M. et al . Colonoscopic study of 50 patients with colonic tuberculosis.  Gut. 1992;  33 347-351
  CrossrefPubMedSearch in Google Scholar
• 6 Aoki G, Nagasako K, Nakae Y. et al . The fibercolonoscopic diagnosis of intestinal tuberculosis.  Endoscopy. 1975;  7 113-121
  PubMedSearch in Google Scholar
• 7 Bhargava D K, Kushwaha A KS, Dasarathy S. et al . Endoscopic diagnosis of segmental colonic tuberculosis.  Gastrointest Endosc. 1992;  38 571-574
  PubMedSearch in Google Scholar
• 8 Kalvaria I, Kottler R E, Marks I N. The role of colonoscopy in the diagnosis of tuberculosis.  J Clin Gastroenterol. 1988;  10 516-523
  PubMedSearch in Google Scholar
• 9 Pettengell K E, Pirie D, Simjee A E. Colonoscopic features of early intestinal tuberculosis (report of 11 cases ).  S Afr Med J. 1991;  79 279-280
  PubMedSearch in Google Scholar
• 10 Ferentzi C V, Sieck J O, Ali M A. Colonoscopic diagnosis and medical treatment of ten patients with colonic tuberculosis.  Endoscopy. 1988;  20 62-65
  PubMedSearch in Google Scholar
• 11 Radhakrishnan S, Nakib B, Shaikh H, Menon N K. The value of colonoscopy in schistosomal, tuberculous and amebic colitis: two year experience.  Dis Colon Rectum. 1986;  29 891-895
  PubMedSearch in Google Scholar
• 12 Misra S P, Misra V, Dwivedi M, Gupta S C. Colonic tuberculosis: clinical features, endoscopic appearance and management.  J Gastroenterol Hepatol. 1999;  14 723-729
  CrossrefPubMedSearch in Google Scholar
• 13 Das P, Shukla H S. Clinical diagnosis of abdominal tuberculosis.  Br J Surg. 1976;  63 941-946
  PubMedSearch in Google Scholar
• 14 Singh V, Jain A K, Agrawal A K. et al . Clinicopathological profile of abdominal tuberculosis.  Br J Clin Pract. 1995;  49 22-24
  PubMedSearch in Google Scholar
• 15 Naga M I, Okasha H H, Ismail Z. et al . Endoscopic diagnosis of colonic tuberculosis.  Gastrointest Endosc. 2001;  53 789-793
  CrossrefPubMedSearch in Google Scholar
• 16 Dye C, Scheele S, Dolin P. et al . Global burden of tuberculosis: estimated incidence, prevalence and mortality by country.  JAMA. 1999;  282 677-686
  CrossrefPubMedSearch in Google Scholar
• 17 Horvath K D, Whelan R L. Intestinal tuberculosis: return of an old disease.  Am J Gastroenterol. 1998;  93 692-696
  PubMedSearch in Google Scholar
• 18 Kim K M, Lee A, Choi K Y. et al . Intestinal tuberculosis: clinicopathologic analysis and diagnosis by endoscopic biopsy.  Am J Gastroenterol. 1998;  93 606-609
  CrossrefPubMedSearch in Google Scholar
• 19 Tarumi K, Koga H, Lida M. et al . Colonic aphthoid erosions as the only manifestation of tuberculosis: case report.  Gastrointest Endosc. 2002;  55 743-745
  CrossrefPubMedSearch in Google Scholar
• 20 Pulimood A B, Ramakrishna B S, Kurian G. et al . Endoscopic mucosal biopsies are useful in distinguishing granulomatous colitis due to Crohn’s disease from tuberculosis.  Gut. 1999;  45 537-541
  CrossrefPubMedSearch in Google Scholar
• 21 Sakai Y. Colonoscopic diagnosis of the intestinal tuberculosis.  Mater Med Pol. 1979;  11 275-278
  PubMedSearch in Google Scholar
• 22 Gan H T, Chen Y Q, Ouyang Q. et al . Differentiation between intestinal tuberculosis and Crohn’s disease in endoscopic biopsy specimens by polymerase chain reaction.  Am J Gastroenterol. 2002;  97 1446-1451
  CrossrefPubMedSearch in Google Scholar
• 23 Anand B S, Schneider F E, El-Zaatari F A. et al . Diagnosis of intestinal tuberculosis by polymerase chain reaction on endoscopic biopsy specimens.  Am J Gastroenterol. 1994;  89 2248-2249
  PubMedSearch in Google Scholar
• 24 Ha H K, Kim J K. The gastrointestinal tract. In: Haaga JR, Lanzieri CF, Gilkeson RC (eds) CT and MR imaging of the whole body, vol. 2. St. Louis; Mosby 2003: 1154-1169
  Search in Google Scholar
• 25 Kashima K, Oka S, Tabata A. et al . Detection of anti-cord factor antibodies in intestinal tuberculosis for its differential diagnosis from Crohn’s disease and ulcerative colitis.  Dig Dis Sci. 1995;  40 2630-2634
  PubMedSearch in Google Scholar
• 26 Patel N, Amarapurkar D N, Kulshreshta P P. et al . Utility of serological markers in inflammatory bowel disease.  Indian J Gastroenterol. 2002;  21 A37
  PubMedSearch in Google Scholar
• 27 Thomas G AO, Swift G L, Green J T. et al . Controlled trial of antituberculous chemotherapy in Crohn’s disease: a five year follow up study.  Gut. 1998;  42 497-500
  PubMedSearch in Google Scholar

J. F. Alvares, M. D., D. M., D. N. B.

c/o Adv C. F. Alvares · Near the Holy Spirit Church · Margao-Goa · India ·

Fax: +91-832-2711883

Email: filipealvares@yahoo.co.in