Evaluation of Twenty-One Cases of Abdominal Tuberculosis: A Single-Center Experience

Abstract

Background:

Abdominal tuberculosis (TB) remains an issue as it masquerades as many malignant or benign abdominal conditions.

Objective

: To analyze the clinical and laboratory features of abdominal TB retrospectively and discuss its management.

Methods:

The data of patients with a histopathologic diagnosis of abdominal TB seen from January 1, 2008, to February 1, 2019 were collected in The First Affiliated Hospital of Nanjing Medical University. Nodal, visceral, peritoneal, and mixed TB cases were included while excluding other forms of extra-pulmonary TB (EPTB).

Results:

A total of 21 patients presented having a median age of 49 years (interquartile range 33–57 years) with 12 females and 9 males. Ten presented with abdominal pain, whereas four had abdominal pain and distention. Weight loss was present in five and type 2 diabetes mellitus (DM) in three. Every patient received contrast-enhanced computed tomography (CE-CT) with positive results in all the cases. Seven patients received endoscopic ultrasound-guided fine-needle aspiration cytology examination (EUS-FNAC) and five had results positive for TB. Pre-operative diagnosis of abdominal TB was possible in seven; however, the majority (n = 14) underwent exploratory laparotomy, and all obtained a definitive diagnosis of TB. No deaths occurred.

Conclusions:

Both CE-CT and EUS-FNAC can aid in the timely diagnosis. Laparotomy is an invasive but efficient tool for the final diagnosis of abdominal TB.

Previously known as “consumption,” the disease now known as tuberculosis (TB) is still prevalent, remaining in the top 10 leading causes of death worldwide, with an estimated 1.3 million deaths in 2017. Globally, it is the primary cause of death from a single infectious agent [1].

The incidence of TB infection has increased in recent years with greater globalization. Abdominal TB constitutes 12% of extra-pulmonary TB (EPTB) cases and 1%–3% of total TB cases [2 –4]. Despite being the sixth most common form of EPTB [5], the diagnosis of abdominal TB can be elusive, as it mirrors clinical features observed in many malignant and benign abdominal conditions. In addition, misconceptions loom about abdominal TB: It is thought to be a rare condition, although several reported series exist in the literature, and it is considered to occur mainly in under-developed or developing countries [6]. However, the increasing trends of migration from high TB-incidence countries to low-incidence nations has caused an increase in TB notification among 20 of 30 European countries [7]. Transmission from foreign-born to native-born persons of low-incidence countries remains uncommon; however, this situation might change in the future [7]. Tuberculosis therefore is an issue for both low- and high-incidence countries. In China, the burden of tuberculosis in 2018 was estimated to be around 866,000 [8]. Therefore, it can be hypothesized that the burden of abdominal TB is not negligible, so it is crucial to consider it in cases of suspected abdominal malignancies.

Abdominal TB includes lymphatic, peritoneal, visceral gastrointestinal, colorectal, hepatobiliary, pancreatic, and mixed types [3]. It has been known to imitate various generally benign conditions including acute cholecystitis, Crohn's disease, IgG4-related sclerosing mesenteritis, as well as various malignancies such as pancreatic cancer, peritoneal carcinomatosis, and colorectal cancer [9 –11]. There is no gold standard to diagnose abdominal TB, and no internationally accepted diagnostic algorithm exists. In our experience, EUS-FNAC along with CE-CT helped in the diagnosis in a relatively non-invasive manner. However, as a last resort, in cases where the diagnosis was still not clear, we found that surgery remains the only effective tool to confirm abdominal TB.

Patients and Methods

This retrospective study was approved by the Ethics Committee of Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University. The data were collected from patients with a discharge diagnosis of abdominal tuberculosis from January 1, 2008, to February 1, 2019. All types of abdominal TB (nodal, visceral, peritoneal, and mixed) were included while excluding other forms of EPTB. The clinical manifestations, characteristics, laboratory results, imaging features, and pathological results were analyzed retrospectively. Patients received a confirmed diagnosis of TB after satisfying the following criteria: (1) Histopathologic examination revealing caseating granuloma with multi-nucleated giant cells; (2) appropriate radiologic findings along with clinical manifestations and a positive interferon gamma release assay; and (3) satisfactory response to anti-TB treatment despite negative pathological and bacterial tests with a clinical picture suggestive of TB. The disease in 16 patients was diagnosed with criteria 1 and 2, while the disease in five patients was diagnosed with criterion 2 alone.

In this study, SPSS version 25 was used for data analysis. Continuous variables and categorical data were presented as mean ± standard deviation (SD) and percentage, respectively.

Results

Over the study period, an average of two cases of abdominal TB per year were observed in our institution. The majority were nodal or pancreatic TB (62% and 29%, respectively). Patients presented at a median age of 49 years (inter-quartile range 33–57 years old) with 12 females and 9 males (Table 1). The duration of symptoms ranged from 1 day to 2 years. The most frequent complaints were abdominal pain (47.6%; n = 10), abdominal pain and distention (19.1%; n = 4), abdominal distention (14.3%; n = 3), abdominal mass (9.5%; n = 2), and no symptoms (incidental finding) (9.5%). Among the two patients who were discovered incidentally, one was a case of dermatomyositis in which CT indicated a tuberculous abscess of the head of the pancreas, whereas the other was a patient with an elevated Ca-19-9 concentration in which CT showed a hypodense lesion on the head of the pancreas. Weight loss was a feature in five patients (23.8%). Also, 14.3% (n = 3) of our patients had concomitant ascites. Fever was present in 9.5% (n = 2). Only one patient was reported to be immunocompromised, a consequence of chronic corticosteroid therapy. Acute inflammatory markers (erythrocyte sedimentation rate [ESR]/C-reactive protein [CRP]) were high in six out of nine (67%) of the patients. In particular, ESR was high in four (average 35 mm/h), and CRP was elevated in two. Only 38% of patients (n = 7) were anemic (serum hemoglobin concentration <120 g/L), and 38% had hypoalbuminemia. Two (9%) had leukocytosis. The prevalence of type 2 diabetes mellitus (DM) was 14.2 %. Two had newly diagnosed type 2 DM and one had been found to have DM within one year of the abdominal TB diagnosis.

Table 1.

Characteristics of Patients with Abdominal Tuberculosis

	Total (n = 21)	Nodal (n = 13)	Pancreatic (n = 6)	Intestinal (n = 1)	Nodal and peritoneal (n = 1)
Median age	49 (IQR 33– 57)	49	49.5 (IQR 31–55)	70	25
Female (%)	12 (57)	8 (61.5)	3 (50)	0	1 (100)
Symptoms (%)	Abdominal pain n = 10 (48)Abdominal pain + distention n = 4 (19)Abdominal distention (n = 3) (14)	Abdominal pain + distention n = 9 (69)	Abdominal discomfort n = 1 (17)Abdominal pain + distention n = 3 (50)	Intermittent abdominal pain	Abdominal mass
No. (%) with weight loss	5 (24)	3 (23)	2(33)	0	0
Vomiting (%)	4 (19)	n = 3 (23)	0	1 (100%)	0
Abdominal mass (%)	2 (19.5)	1 (8)	0	0	1 (100)
Ascites (%)	3 (14)	1 (8)	2 (17)	0	0
Incidentally diagnosed (%)	2 (10)	0	2 (33)	0	0
Night sweats (%)	1	1 (8)	0	0	0
History of TB	1	1	0	0	0
Distribution		Abdominal lymph nodes	Pancreatic head (100)
CT appearance (%)		Peri-pancreatic (n = 1; (8)Celiac (n = 2; (15)Portal (n = 2; (15)Para-aortic (n = 1; (8)Mixed lymphadenopathy (n = 7; (54)	Hypoechoic lesion 83%; mixed solid-cystic lesion 17%	Splenomegaly stage 3, abdominal lymphadenopathy	Retroperitoneum cystic mass; nodular thickening of ascending mesocolon
Chest radiography suspect findings (%)	6 (28.6)	2 (15)	3 (50)	1 (100)	0
Type 2 diabetes mellitus (%)	3 (14)	2 (15) (1 with <1-year history)	1 (17) (newly diagnosed)	0	0
LFT disturbance (%)	6	4 (31)	1 (17)	0	1 (100)
High NSE (>16.3 ng/mL) (n = 9)(%)	5/9 (56)	2/5 (40)	3/4 (75)	NA	NA
Leukocytosis (%)	2 (9.5)	1 (8)	0	1 (100)	0
Low albumin at time of diagnosis	8 (38)	7 (54)	1	0	0
Anemia at time of diagnosis (%)	8 (38)	5 (38)	1 (17)	1 (100)	1 (100)
Immunodeficiency (%)	1 (5)	0	1 (17)	0	0

IQR = interquartile range; LFT = liver function tests; NA = not applicable; NSE = neuron-specific enolase.

The CA-19-9 concentration was measured in 13 patients (all those with pancreatic TB and seven with nodal TB) and was elevated consistently for nine months (highest value 81.U/mL) in one patient who had peri-pancreatic lymphadenopathy, a hypoechoic lesion of the pancreas, and a final diagnosis of pancreatic TB.

The disease in seven patients was diagnosed by clinical, microbiologic, and radiologic means (33.3%), whereas 14 diagnoses (66.7%) were confirmed by laparotomy. All the patients who underwent surgery had intra-operative specimens sent for rapid frozen section review. Explorative and therapeutic laparotomy was warranted in patients in whom a definitive diagnosis could not be established on the basis of clinical features, laboratory results, or EUS-FNAC findings. In our study, 11 patients underwent EUS, seven received EUS-FNAC, of whom five (71%) had positive results. Among the patients who underwent EUS-FNAC, there were three with pancreatic TB and four of nodal TB. All the patients with a final diagnosis of pancreatic TB who underwent 4EUS-FNAC had cytology results indicating tuberculosis.

Computed tomography was the imaging modality of choice, with positive findings in all cases. Positive chest radiographs and CT were observed in six patients, and among them, three had pancreatic TB. Four patients (14%) had suspect findings on chest radiography alone. In the patients with pancreatic TB, CE-CT showed lesions occupying the head of the pancreas with four a hypodense lesion in five (83.3%) and mixed solid-cystic mass in one (16.6%) (Table 2). Two of the seven in the pancreatic TB group had concomitant intra-hepatic bile duct dilation and no case of dilated main pancreatic duct.

Table 2.

Characteristics of Six Patients with Pancreatic Tuberculosis^a

Feature	Percent of cases
Distribution	Pancreatic head (100)
Computed tomography appearance of pancreas	Hypoechoic lesion (83); mixed solid-cystic lesion (17)
Presence of lymphadenopathy on CT (n = 5)	Peri-pancreatic (20); celiac (20); mesenteric (20); mixed lymphadenopathy (40)
Signs of exocrine/endocrine pancreatic deficiency	Nil
Tumor markers	High Ca-19-9 (17); high NSE (75); normal Ca125 (100 for 2 patients)

One case of isolated pancreatitis and five with lymphadenopathy.

Mixed lymphadenopathy was the most common CT finding among patients with nodal TB (53.8%; n = 7) (Table 3). Enlarged lymph nodes appeared as irregular hypodense lesions with homogeneous or heterogeneous mild enhancement after contrast administration. In the visceral TB group, the pancreas was the most commonly involved organ (28.6%; n = 6). Pancreatic TB was always accompanied by lymphadenopathy except in one case, where the pancreas alone was involved (Table 3).

Table 3.

Location of Tuberculosis in 21 Patients

Isolated nodal TB	13
Peri-pancreatic	1
Para-aortic	1
Portal	2
Celiac	2
Mixed	7
Visceral TB	7
Pancreas and nodal	5
Pancreas alone	1
Intestine and nodal	1
Peritoneal and nodal	1
Pancreatic tuberculosis with lymphadenopathy	5
Peri-pancreatic	1
Celiac and portal	1
Celiac	1
Mesenteric	1
Portal and retroperitoneal	1

Acid-fast bacilli smear and culture were positive in 20%. Samples sent for evaluation were six intra-operative; three sputum; and one each of blood, ascitic fluid, and EUS-FNAC.

Positive T-SPOT.TB test results were elicited in nine of 11 patients (81.1%). QuantiFERON TB Gold In-Tube test was positive in five of seven patients (71.4%). One patient received PCR testing that yielded a positive result. Histopathologic examination through specimens obtained by FNAC or surgery confirmed or diagnosed TB in 16 of 17 patients (94.1%). Histologic analysis of specimens obtained intra-operatively or with FNAC indicated granulomatous inflammation in 76.5% (n = 13) and the presence of caseous necrosis in 23.5% (n = 4) of patients.

Surgical intervention was advised in suspected TB cases but where malignancy could not be excluded (79%; 11 of 14 patients) and when a primary diagnosis of carcinoma was suspected (29%; n = 4). Surgery involved abdominal lymphadenectomy (85%; n = 12), resection of intestine (7%; n = 1), drainage of tuberculous abscess of the head of pancreas (7%), and splenectomy in the patient with intestinal TB.

Intra-operatively for nodal TB, enlarged lymph nodes (celiac, portal peri-pancreatic lymph nodes) that sometimes coalesced to form a mass or singularly enlarged lymph nodes were observed. On dissection of these lymph nodes or masses of lymph nodes, we could observe caseous necrosis as shown by an outflow of white-milky substance. A granuloma mass 6 cm in diameter was observed in one patient with outflow of a milky substance. In one patient with tuberculous abscess of pancreatic TB, the cystic pancreatic head had purulent discharge that was sent for analysis. In the cases of pancreatic TB, the mass or enlarged lymph nodes were seen encroaching on the duodenum or at the neck of pancreas; however, the pancreas was smooth. In our patient with peritoneal and nodal TB, we could observe ascites with multiple enlarged lymph nodes.

No deaths occurred in our series. One patient (intestinal TB) had a post-operative complication and underwent a second operation seven hours after the primary surgery (intestinal resection and splenectomy) because of abdominal bleeding likely attributable to a poor coagulation profile consistent with a history of chronic lymphocytic leukemia. No other post-operative complications were reported.

Discussion

Establishing a definitive diagnosis of abdominal TB remains a challenging task for even the most experienced clinicians because of the lack of pathognomonic clinical manifestations. In accord with other series, the most frequently reported complaints were abdominal pain (47.6%; n = 10) and abdominal pain with distention (19.1%; n = 4) [9,12]. A slight female predominance (57% versus 43%) was noted, as in other series. This also is observed in countries where TB is endemic, whereas male predominance is noted in many European countries [13]. The incidence of fever in our series was much lower than in previous series (9.5% versus 25%) [8]. In concordance with previous results, acute inflammatory markers (ESR/CRP) that were measured in nine patients were high in 67% [9].

In our series, 14.2% of patients had type 2 DM, among whom two thirds were patients with newly diagnosed type 2 DM and one patient found to have DM within one year of abdominal TB diagnosis. These results are consistent with those of a prospective study from China [14]. Our series demonstrated lower rates of newly diagnosed DM and reduced DM prevalence than in series from Iran, Malaysia, Nigeria, Saudi Arabia, and Indonesia [15 –19]. A prospective cohort study in Taiwan reported that the risk of TB among the general population was increased by type 2 diabetes, especially in patients who had a secondary complication of diabetes [20].

The EUS-FNA procedure is safe and efficient compared with percutaneous FNA, which has a risk of causing peritoneal dissemination in cases of malignancy [21]. Various series have reported high diagnostic accuracy of EUS-FNA in cases of intra-abdominal lymphadenopathy [19,22]. In our study, the diagnostic accuracy of EUS-FNAC was comparable to the results obtained in a South Korean study that had 80.5% diagnostic accuracy [19]. A superior accuracy of EUS-FNA could have been achieved in our series if pathology, culture, PCR, and acid-fast bacilli tests were included. Based on EUS alone, no distinction exists between the appearance of pancreatic head malignancy and pancreatic TB [23].

In our series, all patients underwent CE-CT (Table 3). Most of the patients in our abdominal lymphadenopathy group had clusters of enlarged lymph nodes that appeared as irregular hypodense lesions with homogeneous or heterogeneous mild enhancement after contrast administration. Tuberculous lymph nodes typically show peripheral enhancement after contrast administration. In pancreatic TB, the most common sites of involvement in descending order are the pancreatic body, head, and tail [24] yet all our patients had TB involving the head of the pancreas (see Table 2). It was noted that two of seven patients had concomitant intra-hepatic bile duct dilation; however, none of the patients with pancreatic TB had a dilated main pancreatic duct. The features of pancreatic TB and pancreatic malignancies are indistinguishable. Several studies have reported a higher rate of positive findings on chest imaging in patients with pancreatic TB [25 –27]. Chest imaging studies were abnormal in 50% of our six patients with pancreatic TB. Thus, in a suspected case of abdominal tuberculosis, the presence of an abnormal chest radiograph may hint at pancreatic involvement [9].

In our study, the majority of patients received a domestic alternative of the interferon-gamma release assay (approved in China). The T-SPOT.TB test was positive in 81.1% of 11 patients, whereas the QuantiFERON TB Gold In-Tube test was positive in 71.4% of seven patients. In adults, the sensitivities of T. SPOT.TB and QuantiFERON Gold IT are reported to be similar in the two studies [28,29].

The intra-operative or FNAC histopathologic findings (granulomatous inflammation in 76.5% and caseous necrosis in 23.5%) were consistent with those of other series [19,30]. Central necrosis is the only sign that is relatively specific to TB, as non-caseating granuloma may be present in other conditions such as sarcoidosis, Crohn's disease, and fungal infection [31].

The majority of our patients received a “definitive diagnosis” of abdominal TB after histopathologic confirmation of resected specimens. Pre-operative diagnosis of abdominal TB was possible in some cases; however, the majority of our patients (67%) underwent a diagnostic or therapeutic laparotomy. In the majority of patients in whom lymph nodes were involved, abdominal lymphadenectomy was performed for histopathologic analysis and for therapeutic purposes where enlarged lymph nodes encroached on or oppressed surrounding structures. Enlarged lymph nodes that coalesced to form a mass or granuloma were resected as well. Two patients with nodal TB received an omental patch repair because of an internal duodenal fistula created by removal of the cluster of nodes. The lower pole of the mass was tightly adherent to the posterior wall of the duodenal bulb. Resection of intestine along with splenectomy was performed in one patient with intestinal tuberculosis because of a history of chronic lymphocytic leukemia and splenomegaly.

Conclusion

Both EUS-FNAC and CE-CT are effective modalities that can aid in the timely diagnosis of abdominal TB. Laparotomy still has a role in the management and treatment of abdominal TB.

Footnotes

Authors' Contributions

TGW and SH contributed to data collection and analysis and drafting of the manuscript. JW, KZ, JW, WG, QL, KJ, and YM contributed to the conception and design of the study and to interpretation and analysis of the data. YM, KJ, JW, WG, and JW performed the operations. JW critically revised the manuscript for important intellectual content. All authors read and approved the final version to be submitted.

Funding Information

Funding for this study was received from the Development Program of Jiangsu Province Hospital (No. 511).

Author Disclosure Statement

All procedures performed in studies involving human participants were in accordance with the standards of the Ethics Committee of Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University and with the 1964 Helsinki Declaration and its later amendments. This retrospective study was approved by the Ethics Committee of Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University. The Registration number is IRB-GLI-AF03.

The authors declare that they have no competing interests.

References

World Health Organization. Global tuberculosis report. Geneva: World Health Organization, 2019.

Farer

, Lowell

, Meador

. Extrapulmonary tuberculosis in the United States. Am J Epidmiol, 1979; 109:5–15.

Sheer

, Coyle

. Gastrointestinal tuberculosis. Curr Gastroenterol Rep, 2003; 4:273–278.

Pescarini

, Rodrigues

, Gomes

, et al. Migration to middle-income countries and tuberculosis: Global policies for global economies. Global Health, 2017; 1:15.

Donoghue

, Holton

. Intestinal tuberculosis. Curr Opin Infect Dis, 2009; 5:490–496.

Abu-Zidan

, Sheek-Hussein

. Diagnosis of abdominal tuberculosis: Lessons learned over 30 years: Pectoral assay [published correction appears in World J Emerg Surg 2019;14:42]. World J Emerg Surg, 2019; 14:33.

Lönnroth

, Mor

, Erkens

, et al. Tuberculosis in migrants in low-incidence countries: Epidemiology and intervention entry points. Int J Tuberc Lung Dis, 2017; 6:624–637.

World Health Organization. www.who.int/tb/country/data/profiles/en/

Hasosah

, Satti

, Yousef

, et al. IgG4-related sclerosing mesenteritis in a 7-year-old Saudi girl. Saudi J Gastroenterol, 2014; 6:385–388.

10.

Sharma

, Rana

, Kumar

, Bhasin

. Pancreatic tuberculosis. J Gastroenterol Hepatol, 2016; 2:310–318.

11.

Shreshtha

, Ghuliani

. Abdominal tuberculosis: A retrospective analysis of 45 cases. Indian J Tuberc, 2016; 63:219–224.

12.

Cho

, Choi

, Lee

, et al. Clinical features and outcomes of abdominal tuberculosis in southeastern Korea: 12 years of experience. BMC Infect Dis, 2018; 1:699.

13.

Abdallah

, Larbi

, Hamzaoui

, et al. Abdominal tuberculosis: A retrospective series of 90 cases. Rev Med Interne, 2011; 4:212–217.

14.

, Lin

, Mi

, et al. Screening of patients with tuberculosis for diabetes mellitus in China. Trop Med Int Health, 2012; 10:1294–1301.

15.

Ozma

, Rashedi

, Poor

, et al. Tuberculosis and diabetes mellitus in northwest of Iran [published online ahead of print, July 15, 2019]. Infect Disord Drug Targets.

16.

Sulaiman

SAS

, Khan

, Ahmad

, et al. Impact of diabetes mellitus on treatment outcomes of tuberculosis patients in tertiary care setup. Am J Med Sci, 2013; 4:321–325.

17.

Alebel

, Wondemagegn

, Tesema

, et al. Prevalence of diabetes mellitus among tuberculosis patients in sub-Saharan Africa: A systematic review and meta-analysis of observational studies. BMC Infect Dis, 2019; 1:254.

18.

Singla

, Khan

, Al-Sharif

, et al. Influence of diabetes on manifestations and treatment outcome of pulmonary TB patients. Int J Tuberc Lung Dis, 2006; 1:74–79.

19.

Alisjahbana

, Sahiratmadja

, Nelwan

, et al. The effect of type 2 diabetes mellitus on the presentation and treatment response of pulmonary tuberculosis. Clin Infect Dis, 2007; 4:428–435.

20.

Baker

, Lin

, Chang

, et al. The risk of tuberculosis disease among persons with diabetes mellitus: A prospective cohort study. Clin Infect Dis, 2012; 6:818–825.

21.

Kim

, Lee

, Kim

, et al. Peripancreatic tuberculous lymphadenopathy masquerading as pancreatic malignancy: A single-center experience. J Gastroenterol Hepatol, 2014; 2:409–416.

22.

Puri

, Mangla

, Eloubeidi

, et al. Diagnostic yield of EUS-guided FNA and cytology in suspected tubercular intra-abdominal lymphadenopathy. Gastrointest Endosc, 2012; 75:1005–1010.

23.

Rana

, Bhasin

, Srinivasan

, et al. Distinctive endoscopic ultrasound features of isolated pancreatic tuberculosis and requirements for biliary stenting. Clin Gastroenterol Hepatol, 2012; 3:323–325.

24.

Lee

, Van Tonder

, Tartaglia

, et al. CT appearances of abdominal tuberculosis. Clin Radiol, 2012; 67:596–604.

25.

SS Rana, V, V Sharma, R

Sharma

, et al. Involvement of mediastinal/intra-abdominal lymph-nodes, spleen, liver, and left adrenal in presumed isolated pancreatic tuberculosis: An endoscopic ultrasound study. J Digest Endos. 2016; 6:15–18.

26.

Rao

, Pandey

, Rana

MK et al

. Pancreatic and peripancreatic tuberculosis presenting as hypoechoic mass and malignancy diagnosed by ultrasound-guided fine-needle aspiration cytology. J Cytol, 2013; 2:130–135.

27.

Nagar

, Raut

, Morani

, et al. Pancreatic tuberculosis: A clinical and imaging review of 32 cases. J Comput Assist Tomogr, 2009; 1:136–141.

28.

Takasaki

, Manabe

, Morino

, et al. Sensitivity and specificity of QuantiFERON-TB Gold Plus compared with QuantiFERON-TB Gold In-Tube and T-SPOT.TB on active tuberculosis in Japan. J Infect Chemother, 2018; 3:188–192.

29.

, Xie

, Zhang

, et al. Prospective comparison of QFT-GIT and T-SPOT.TB assays for diagnosis of active tuberculosis. Sci Rep, 2018; 1:5882.

30.

Xia

, Poon

, Wang

, et al. Tuberculosis of pancreas and peripancreatic lymph nodes in immunocompetent patients: Experience from China. World J Gastroenterol, 2003; 6:1361–1364.

31.

Shah

, Pritt

, Alexander

. Histopathologic review of granulomatous inflammation. J Clin Tuberc Other Mycobact Dis, 2017; 7:1–12.