Primary Pancreatic Tuberculous Abscess : A Report of Two Cases

Tuberculosis is an endemic disease in the developing countries and can affect all parts of the gastrointestinal tract. Primary involvement of the pancreas though rare, has been documented in immunocompromised patients. We report two cases of primary tuberculosis infection of the pancreas who presented as chronic abdominal pain, were diagnosed on tissue culture and polymerase chain reaction and were managed successfully by conservative treatment.

Introduction

Tuberculosis has been known to involve all parts of the gastrointestinal tract, from the oesophagus to the anal canal.¹ Approximately 12% of the patients show gastrointestinal involvement with tuberculosis.² Atypical forms of gastrointestinal tuberculosis have been commonly reported in immunocompromised patients.³ Even then, isolated pancreatic tuberculosis is extremely rare.⁴ Due to this clinical rarity, the importance of establishment of correct diagnosis and treatment and its prognostic implications is vital.⁴ We report two cases of primary pancreatic tuberculosis which were successfully treated by chemotherapy.

Case Reports

Case 1

A 28 year-old female patient presented with dull aching abdominal pain in the upper abdomen since 15 days. She had no history of nausea, vomiting, jaundice or any similar episodes in the past. There were no aggravating or relieving factors. She also gave history of anorexia, weight loss. She was non-alcoholic and had no other medical illness in the past. On examination, she was averagely nourished with pulse of 86/min and blood pressure of 120/80 mm of Hg. On palpation of the abdomen, there was no tenderness, guarding or lump palpable. Plain X-rays of the chest and abdomen were normal.


Fig. 1 : CT scan of the first patient showing pancreatic fluid collection along with peripancreatic lymph nodes.

Abdominal ultrasonography (USG) revealed a hypoechoic lesion in the pancreas along with enlargement of the peri-pancreatic lymph nodes. Investigations revealed a raised erythocyte sedimentation rate of 40 at the end of one hour with a slightly elevated serum amylase of 280 Somogyi units and were sero-negative for HIV. Patient was treated with antacids, pancreatic enzyme supplements, Abdominal computed tomography (CT Scan) of the abdomen was done which revealed a hypodensity in the region of the head of pancreas with a normal body and tail with peri-pancreatic lymphadenopathy (Fig. 1). USG- guide biopsy of the peri-pancreatic lymph nodes did not yield any tissue, while aspiration yielded caseous material in which mycobacterium was isolated. Polymerase chain reaction (PCR) for tuberculosis from the aspirated tissue was positive. The patient was started on four drug anti-tubercular chemotherapy (Rifampicin, Isoniazid, Pyrazinamide and Ethambutol) and was kept under observation. Patient showed improvement in symptoms after two months. CT scan of the abdomen repeated after 6 months revealed complete resolution of the lesion. Patient received the anti-tubercular treatment for 9 months and is totally a symptomatic on follow-up after one year.

Case 2

A 35 year-old male patient presented with chronic epigastric pain and nausea since 15 days. Patient had no history of vomiting, haematemesis, melaena or jaundice. He was non-alcoholic and had no previous episodes of a similar nature. There were no apparent aggravating or relieving factors. On examination, his vital parameters were normal with a soft abdomen and no palpable lump. Plain X-rays of the chest and abdomen were normal. USG of the abdomen revealed peripancreatic fluid collection. CT scan confirmed the findings (Fig. 2). Patient was started on antibiotics, analgesics and antacids. He developed fever with leucocytosis of 15,000/cmm. Acid fast bacilli were isolated from a USG guided aspirate and polymerase chain reaction was positive for mycobacterium tuberculosis. The patient was seronegative for HIV. Patient was started on four drug anti-tubercular chemotherapy and responded to conservative line of treatment. A repeat CT scan of the abdomen done after 3 months of anti-tubercular treatment showed a considerable decrease in the size of the collection. Patient was asymptomatic with complete resolution of the collection after 7 months of chemotherapy. Patient completed 9 months of anti-tubercular therapy and is asymptomatic on follow up after 6 months.


	Fig. 1 : CT scan of the first patient showing pancreatic fluid collection along with peripancreatic lymph nodes.

Discussion

Though tuberculosis endemic in the developing countries, involvement of the pancreas is rare.⁵ Primary involvement of the pancreas in bacterial and parasitic infections is rare and usually is secondary to pancreatic necrosis. However, recently cases are being reported with increased frequency of pancreatic tuberculosis attributed to an evolutionary change in the biology of mycobacterium, drug resistance and new populations with immunologic deficits.⁴ Pancreatic tuberculosis has been reported as 1) secondary infection as in generalised miliary tuberculosis with mycobacterium tuberculosis as the agent 2) spread from coeliac and other retroperitoneal lymph nodes with mycobacterium bovis as the main agent and 3) primary localized pancreatic tuberculosis with the origin most probably being the intestines.⁴ The exact way the mycobacterium reaches is not known. The accepted mode of transmission is of lymph-haematogenous dissemination with primary focus in the intestines.⁴ However, few authors support the fact that it may be toxic-allergic reaction of the pancreas in response to tuberculosis elsewhere in the body.⁶ Patients with pancreatic tuberculosis commonly present as pancreatic mass or masses, multiple peripancreatic lymphadenopathy, acute or chronic pancreatitis, stenosis of Wirsung duct or obstructive jaundice. Rare presentations include gastrointestinal bleeding due to splenic vein thrombosis, encasement of coeliac trunk, focal hepatic lesion, lytic bone lesions. However these patients present only with constitutional symptoms as anorexia, malaise, low-grade fever, acute or chronic abdominal pain, melaena and ascites.⁷ The commonest presentation in immunocompromised patients is tubercular abscess. This presentation is so common that Jabber and Gleckman have termed it as ‘AIDS - defining disorder’ for pancreatic tuberculosis after review of 20 patients.⁸ Investigations as such as chest X-rays and sputum examination may hardly contribute to the diagnosis.⁵

Ultrasonography, endoscopic retrograde pancreaticography (ERCP), computed tomography may be suggestive of a mass or abscess but are unable to rule out malignancy or make a specific diagnosis. Therapeutic ERCP may be necessary if the patient presents with obstructive jaundice. Percutaneous biopsy, combined with Ziehl-Neelsen staining for acid-fast bacilli, culture of the biopsy material are the diagnostic methods.⁸ Preoperative biopsy is worthwhile, even though the sensitivity is low as it may avoid an unnecessary surgery. Acid-fast bacilli have been isolated in 38% of all the cases in all types of the biopsy material however; false negative cultures results are as high as 30%.⁷ Recently introduced polymerize chain reaction (PCR) may be useful in the confirmation of the diagnosis of early lesions and in patients with negative microscopy and culture results.⁸ However false negative results have also been reported in patients with positive culture reports.⁷ Pancreatic tuberculosis responds very well to anti-tubercular chemotherapy. A course of 6-9 months in drug susceptible cases and a 12-month course in isoniazid and/or rifampicin resistant cases have been recommended.⁹ Jenney et al have reported that about 35 out of 36 patients responded to anti-tubercular chemotherapy.¹⁰ Laparotomy may serve as the diagnostic test, however, it is debatable whether laparotomy should just be diagnostic or resection of pancreatic undiagnosed lesion should be done which may very well turn to be benign and curable with chemotherapy. Reports of pancreaticoduodenectomy (Whipple’s procedure) and distal pancreatectomy¹⁰ for pancreatic tuberculosis have been reported, but it also has to be remembered that even in expert hands these procedures carry a small but definite probability of mortality as well as high morbidity.⁷ Hence, it is recommended that an aggressive attempt be made to diagnose pancreatic lesions and pancreatic tuberculosis forms the important differential. Both our patients had no detectable primary source of tuberculosis, were not immunocompromised and presented only with chronic epigastric pain. USG and CT scan suggested a pathological lesion, which could be diagnosed only after aspiration followed by culture and polymerase reaction. None of our patients required intervention.

Conclusion

Tuberculosis infection of the pancreas should be kept as an important differential for any of the undiagnosed lesion of the pancreas and a thorough attempt should be made to attain a correct diagnosis before any surgical intervention is contemplated.

Acknowledgement

We are thankful to our Dean, Dr. Kshirsagar for allowing us to publish hospital data.

References

1. Kapoor VK, Sharma LK. Abdominal tuberculosis. Br J Surg 1988; 75 : 2-3.

2. Haddad FS, Ghossain A, Sawaya E, Nelson AR. Abdominal tuberculosis. Dis of Colon Rectum 1987; 30 : 724-35.

3. Dunderam G, McDonald RJ, Maniatis T, Oleske J, Kapila R, Reichman LB. Tuberculosis as a manifestation of the acquired immunity syndrome (AIDS). JAMA 1986; 256 : 362-6.

4. Elizondo ME, Arratibel JA, Compton CC, Warshaw AL. Tuberculosis of pancreas. Surg 2001; 129 : 114-6.

5. Vershney S Johnson (II). Tuberculosis of pancreas. Postgrad Med J 1995; 71 : 564-6.

6. Stock KP, Reiman JF, Stadler W, Rosch W. Tuberculosis of pancreas. Endoscopy 1978; 129 : 432-6.

7. Ozden I, Emre A, Demir K, Balci C, Poyanli A, Ilhan R. Solitary pancreat. Surg 2001; 8 : 279-83.

8. Jaber B, Gleckman R. Tuberculous pancreatc abscess as initial AIDS-defining disorder in a patient with the human immunodeficiency virus: case report and review. Clin Infect Dis 1995; 20 : 890-4.

9. Levine R, Tenner S, Steinberg W, Ginsberg A, Borum M, Huntington D. Tuberculous abscess of pancreas: case report and review of literature. Dig Dis Sci 1992; 37 : 1141-4.

10. Jenney AWJ, Pickles RW, Hellard ME, Spelman DW, Fuller AJ, Spicer WJ. Tuberculosis pancreatic abscess in an HIV-antibody-negative patient: case report and review. Scand J Infect Dis 1998; 30 : 99-104.