The annual incidence of tuberculosis (TB) is nearly 8 million, with 2 million deaths worldwide (1). In Tunisia, a developing country, TB in children represents 10% to 15% of the total TB cases. In the paediatric age group, the prevalence is 1 to 6/1000 (2). Abdominal TB is an uncommon presentation of TB, especially in children without any other debilitating disease such as cirrhosis, diabetes, and chronic renal failure on continuous ambulatory peritoneal dialysis (3). The most common forms of abdominal TB in children are adhesive peritonitis and nodal disease. It is estimated that peritoneal TB occurs in 0.1% to 3.5% of all patients with pulmonary TB and represents 4% to 10% of all extrapulmonary TB (4,5). Diagnosis of TB among children poses technical and operational issues, more so in the field of abdominal TB, in which the protean clinical manifestations continue to challenge physicians in diagnosis and therapy. We aimed to review the clinical features of abdominal TB in children who were followed up in our hospital.
PATIENTS AND METHODS
Clinical charts, including clinical and laboratory findings and treatment characteristics, of 13 paediatric patients diagnosed as having abdominal TB in the Departments of Pediatrics B and Pediatric Emergency at the Children's Hospital of Tunis, Tunisia, during 1995 to 2008 were evaluated retrospectively. The presentation symptoms, history of TB exposure, biochemical and microbiological tests, and clinical and histological features of the patients were recorded. The tuberculin skin test was evaluated 72 hours after intradermal injection of 5 tuberculin units of purified protein derivative. It was considered to be positive when the induration was ≥10 mm. The diagnosis of abdominal TB was based bacteriologically on identification of Mycobacterium tuberculosis through Ziehl-Neelsen acid-fast stain and/or culture in Lowenstein-Jensen, on histopathological demonstration of epithelioid granuloma with central casaeous necrosis in biopsy specimens, and/or by clinical, biological, and radiological features compatible with TB, in association with a good response to antituberculous medication and exclusion of other diseases.
Thirteen patients were diagnosed as having abdominal TB, and all of the patients were vaccinated with Bacillus Calmette-Guérin in Tunisia. They were 3 boys and 10 girls. All patients were older than 7 years and the mean age was 9.8 years (range 7–14 years). At presentation, abdominal pain was a common complaint in 8 patients, mimicking surgical abdomen in 1 patient. Abdominal distension was found in 8 patients and 2 patients had abdominal mass. In addition, asthenia and anorexia were found in 7 and prolonged fever in 4 patients (Table 1). The onset of symptoms was 4 days to 1 year before admission time (mean 4.2 months). The history of exposure to TB was found in 1 patient; the presumable index cases were the cousin and the uncle. Tuberculin skin test was positive in 8 patients and was phlyctenular in 1 patient. Chest x-ray was performed in all patients; it was normal in 11, showed pleural effusion in 1 patient, and pulmonary condensation with pleural effusion in another patient. No patient had positive gastric fluid culture and/or sputum for M tuberculosis. Nine patients had peritoneal TB and 7 had abdominal lymph node involvement (Figs. 1 and 2). Three patients had intestinal involvement attested to by abdominal imaging or biopsy; spleen involvement was seen in 2, and hepatic involvement in 1 patient (Fig. 3). One patient had multifocal TB (central nervous system [CNS], pleural, liver, and spleen involvement) and another had pleural effusion and parenchyma lung condensation associated with abdominal TB. Abdominal ultrasonography and/or abdominal computed tomography scan showed ascites in 9 patients and mesenteric lymph nodes suggestive of TB in 7 patients, spleen nodules in 2, and liver nodules in 1 patient. Ascitic fluid analysis was performed in 8 patients. All of the ascitic fluids were exudative. Direct examination of ascitic fluids revealed a predominance of lymphocytes, and positive cultures for M tuberculosis were present in 2 patients. No tumour cells were found on cytospin preparations of ascitic fluid. Barium study was performed in 4 patients; it was normal in 3 patients and showed deformed and narrowed caecum with dilated ileum in 1 (Fig. 4). Ileocolonoscopy was performed in 2 patients and caecal biopsy revealed caseating granuloma in 1. Laparotomy was performed in 3 patients: The first patient presented with acute abdominal pain, the second with abdominal mass, and the third with abdominal pain, fever, and abdominal lymph node. It revealed a thickened peritoneum and adhesions in 2 patients, and histological studies of lymph nodes or epiploon were compatible with the diagnosis of TB. A computed tomography scan-guided lymph node biopsy was performed in 1 patient, and the histopathological study revealed granuloma with caseum. Liver puncture biopsy was performed in 1 patient; it revealed M tuberculosis on direct examination and epithelioid granuloma on histopathology.
The remaining 5 patients were diagnosed by ascitic fluid diagnostic features (exudative fluid, predominance of lymphocytes, and no tumour cells), positive tuberculous skin test and/or tuberculous exposure, and radiological imaging, suggestive of TB and a positive response to antituberculous treatment.
Antituberculous treatment was given to all of the patients. It consisted of isoniazid, rifampicin, ethambutol, and pyrazinamide for 2 months and isoniazid and rifampicin for a period of 9 to 12 months without any complications. No patient showed a relapse of disease during the follow-up period (6–15 months).
Abdominal TB is, for the most part, caused by M bovis, and the major route of transmission is the ingestion of infected milk or milk products. Abdominal TB may involve the gastrointestinal tract, peritoneum, lymph nodes, or solid viscera; however, peritoneum and abdominal lymph nodes are the most common sites. In the paediatric literature, abdominal TB has been described infrequently in 2 of 1000 and 5 of 1700 patients in industrialised countries (6–8); this is attributable to the eradication of M bovis by slaughter of infected cattle and pasteurisation of milk. However, in developing countries, this form of TB is still present. In our study, the most common sites were peritoneum and lymph nodes. One patient had multifocal disease with liver, spleen, CNS, and pleural involvement, and another patient had intestinal, lymph node, spleen, lung, and pleural involvement; these presentations are rare. The mean age of presentation in our study (9.8 years) was comparable to the reported common age group (6–11 years) (9,10).
Abdominal TB spreads through close contact and a positive family history is likely in most cases; however, in our study, positive family history was found only in 1 patient. We can speculate that the major cases of abdominal TB are probably due to M bovis, which has not been eradicated in Tunisia, and children are contaminated by drinking nonpasteurized milk.
The clinical spectrum of abdominal TB is wide and nonspecific. The initial symptoms of abdominal TB are nonspecific, such as abdominal distension caused by ascites, pain, fever, and weight loss (11–14). The clinical presentation of our patients was similar, and abdominal pain and distension were the most common presenting complaint in the present study (9 patients). Fever was seen in 4 and weight loss with asthenia in 7 patients.
TB is regarded as a disease with insidious onset and chronic presentation, most patients having symptoms for a few weeks to months, sometimes years; Lambrianides et al (15) even stated that TB is rarely an emergency. However, in our study, 1 patient presented with acute abdomen. Abdominal masses mimicking Burkitt lymphoma were reported in a paediatric group; this presentation makes the diagnosis of TB more difficult. In our series, 2 patients presented with abdominal mass mimicking lymphoma and the defined diagnosis was obtained by histopathological studies. Because of the nonspecific symptoms and physical findings, diagnosis is often delayed. Delay may range from 1 month to 6 years (16–18); mean delay in the present study was 3 months (range 4 days–1 year).
Abdominal ultrasonography is a noninvasive and easily available method of detecting abdominal fluid and lymphadenopathy. It can be used for the diagnosis of peritoneal TB as a first-step investigation method. The most specific sonographic findings of abdominal TB are ascites with fine septations and lymphadenopathy with hypoechogenic centres indicating caseating necrosis (19).
Barium studies are sensitive to ileocaecal and colonic lesions (20). Double-contrast barium enema in ileocaecal TB shows a shortened ascending colon, deformed (irregular, shortened, and narrowed) caecum, deformed and incompetent ileocaecal valve, dilated ileum, and a distorted ileocaecal junction with increased (obtuse) ileocaecal angle (21). Barium studies were performed in 4 patients and revealed ileocaecal changes in 1.
Analysis of ascitic fluid in peritoneal TB often shows exudative features with lymphocytic predominance and serum ascites albumin gradient of <1.1 g/dL (22). The ascites samples acquired from 8 patients had exudative features. High levels of adenosine deaminase in the ascitic fluid were shown to be compatible with the diagnosis of peritoneal TB with high sensitivity (100%) and specificity (97%); however, the analysis of adenosine deaminase activity is expensive and is not available in Tunisia (22–24).
In our study, abdominal TB could be confirmed bacteriologically only in 3 patients. Other studies (25–27) have also faced similar difficulties in the microbiological confirmation of the disease; most of them relied on histopathological diagnosis.
In children with relevant history, laparoscopy has been found to be a rewarding investigation tool (28) with a high success rate in histopathological diagnosis on the tissues retrieved for biopsy. It has brought down the rate of unnecessary laparotomies in children, thanks to practice and experience, its role may also be extended to other therapeutic purposes (stricturoplasty and adhesiolysis). In our study, histopathological diagnosis by laparotomy was performed in 3 patients.
Classical 4-drug chemotherapy was used in all children in the present study and a favourable outcome was observed in 12. Neurological sequelae were observed in only 1 patient who had multifocal TB with CNS involvement.
Abdominal TB should be strongly considered in children in the 5- to 10-year-old age group presenting with vague abdominal pain, weight loss, low-grade fever, and abdominal distension. In areas with a low prevalence of abdominal TB, tissue or microbiological diagnosis is highly justified before the start of a course of therapy. However, in places where the disease is common and confirmatory investigations are inadequately available, the treatment may be initiated, based on strong clinical diagnosis and supportive investigations. In such situations, it is the response to therapy that indirectly proves the diagnosis.
1. Dye C, Scheele S, Dolin P, et al
. Consensus statement. Global burden of tuberculosis
: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA 1999; 282:677–686.
2. Direction de Soins de Santé de Base, Ministère de la Santé publique, République Tunisienne. Guide Pratique Programme Lutte Antituberculeuse. Tunis: SAGEP; 1999.
3. Lazarus AA, Thilagar B. Abdominal tuberculosis
. Dis Mon 2007; 53:32–38.
4. Demir K, Okten A, Kaymakoglu S, et al
. Tuberculous peritonitis – reports of 26 cases, detailing diagnostic and therapeutic problems. Eur J Gastroenterol Hepatol 2001; 13:581–585.
5. Sochocky S. Tuberculous peritonitis. A review of 100 cases. Am Rev Respir Dis 1967; 95:398–401.
6. Snider DE, Reider HL, Combs D, et al
in children. Pediatr Infect Dis J 1988; 7:271–278.
7. Lee S, Bloch A, Onorato I. Changes in reported tuberculosis
cases in children less than 15 years old, U.S., 1988–1991. Paper presented at the 33rd Interscience Conference on Microbiological Agents and Chemotherapy. October 1993; New Orleans.
8. Sanai FM, Bzeizi KI. Systematic review: tuberculous peritonitis –presenting features, diagnostic strategies and treatment. Aliment Pharmacol Ther 2005; 22:685–700.
9. Lin YS, Huang YC, Chang LY, et al
. Clinical characteristics of tuberculosis
in children in the north of Taiwan. J Microbiol Immunol Infect 2005; 38:41–46.
10. Erkan T, Cam H, Ozkan HC, et al
. Clinical spectrum of acute abdominal pain in Turkish pediatric patients: a prospective study. Pediatr Int 2004; 46:325–329.
11. Cruz AT, Starke JR. Clinical manifestations of tuberculosis
in children. Paediatr Respir Rev 2007; 8:107–117.
12. Tanrikulu AC, Aldemir M, Gurkan F, et al
. Clinical review of tuberculous peritonitis in 39 patients in Diyarbakir, Turkey. J Gastroenterol Hepatol 2005; 20:906–909.
13. Gurkan F, Ozates M, Bosnak M, et al
. Tuberculous peritonitis in 11 children: clinical features and diagnostic approach. Pediatr Int 1999; 41:510–513.
14. Maltezou HC, Spyridis P, Kafetzis DA. Extra-pulmonary tuberculosis
in children. Arch Dis Child 2000; 83:342–346.
15. Lambrianides AL, Ackroyd N, Shorey BA. Abdominal tuberculosis
. Br J Surg 1980; 67:887–889.
16. Collado C, Stirnemann J, Ganne N, et al
. Gastrointestinal tuberculosis
: 17 cases collected in 4 hospitals in the northeastern suburb of Paris. Gastroenterol Clin Biol 2005; 29:419–424.
17. Das P, Shukla HS. Clinical diagnosis of abdominal tuberculosis
. Br J Surg 1976; 63:941–946.
18. Bernhard JS, Bhatia G, Knauer CM. Gastrointestinal tuberculosis
: an eighteen-patient experience and review. J Clin Gastroenterol 2000; 30:397–402.
19. Yilmaz T, Sever A, Gür S, et al
. CT findings of abdominal tuberculosis
in 12 patients. Comput Med Imaging Graph 2002; 26:321–325.
20. Dandapat MC, Mohan Rao V. Management of abdominal tuberculosis
. Indian J Tubercul 1985; 32:126–129.
21. Kapoor VK, Chattopadhyay TK, Sharma LK. Radiology of abdominal tuberculosis
. Australas Radiol 1988; 32:365–367.
22. Rasheed S, Zinicola R, Watson D, et al
. Intra-abdominal and gastrointestinal tuberculosis
. Colorectal Dis 2007; 9:773–783.
23. Riquelme A, Calvo M, Salech F, et al
. Value of adenosine deaminase (ADA) in ascitic fluid for the diagnosis of tuberculous peritonitis: a meta-analysis. J Clin Gastroenterol 2006; 40:705–710.
24. Hillebrand DJ, Runyon BA, Yasmineh WG, et al
. Ascitic fluid adenosine deaminase insensitivity in detecting tuberculous peritonitis in the United States. Hepatology 1996; 24:1408–1412.
25. Uygur-Bayramicli O, Dabak G, Dabak R. A clinical dilemma: abdominal tuberculosis
. World J Gastroenterol 2003; 9:1098–1101.
26. al-Quorain AA, Facharzt, Satti MB, et al. Abdominal tuberculosis
in Saudi Arabia: a clinicopathological study of 65 cases. Am J Gastroenterol
27. Pfaller MA. Application of new technology to the detection, identification, and antimicrobial susceptibility testing of mycobacteria. Am J Clin Pathol 1994; 101:329–337.
28. Rai S, Thomas WM. Diagnosis of abdominal tuberculosis
: the importance of laparoscopy. J R Soc Med 2003; 96:586–588.