This prospective observational study was conducted on patients attending surgery outpatient department and emergency care of our tertiary care institute from a period of Nov-2020 to Nov-2021 having a clinical and radiological diagnosis of suspected abdominal tuberculosis. Informed and written consent was obtained. Ethical clearance for the study was obtained from the Institutional ethics committee. Patients who were medically unfit for laparoscopic surgery was excluded from the study. All the patients who presented with a chronic abdominal pain of unknown aetiology with symptoms suggesting of tuberculosis like abdominal distention, non-localised vague abdominal pain, nausea, and vomiting, abdominal lump were included in the study.

Patients after satisfying the inclusion criteria were subjected to diagnostic laparoscopy, biopsy of specimen from peritoneum, omentum, tissues, mesenteric lymph nodes for histo-pathological examination, AFB staining and cartridge based nucleic acid amplification test (CBNAAT) of peritoneal aspirates for confirmation of abdominal tuberculosis. As there is no tissue culture facility in our institute, specimen obtained are not subjected for culture. Pre-operative body weight and haemoglobin (Hb) % were recorded in all patients especially to compare with post-operative response to treatment.

All the patients underwent laparoscopy and were started on standard anti-tuberculosis treatment. Positive diagnosis of abdominal tuberculosis was considered if there was: 1) typical tuberculosis granulomata containing Langhan’s giant cells with caseation, 2) non-caseating necrosis with the demonstration of AFB in the biopsied tissues, 3) positive CBNAAT for Mycobacterium tuberculosis gene from the pelvic aspirate or biopsied tissues. These criteria can reliably equate to establish diagnosis. The data were evaluated to see the diagnostic yield of laparoscopy in the form of macroscopic appearance, histological or microbiological results and the response to medications in the form of sensitivity and specificity.

After administration of General anaesthesia, abdominal cavity was accessed by creating the pneumoperitoneum. Both the open (Hasson’s) and closed (Veress needle) method was used to create the pneumoperitoneum according to the cases. If adhesions were suspected as like the previous history of surgery, Veress needle was inserted in the Palmers point to create the pneumoperitoneum to avoid the inadvertent bowel injury. We prefer 10 mm camera port in the infra or supra umbilical region, but camera port may vary according to the suspected abdominal pathology. 2 or 3 working ports were made according to the therapeutic intervention as per laparoscopy. [Figure 1]

The whole abdominal cavity was carefully surveyed in a routine sequence. [Figure 2] Intra-operative findings were noted and interventions were planned according to that. Some interventions like adhesiolysis, mesenteric lymphadenectomy, biopsy, ascitic fluid analysis were done if required. In the follow up for 3 months, patients were evaluated for the clinical improvement, degree of pain relief, recurrence and complication.

Figure 2

All participants were evaluated for tuberculosis treatment response by assessment at week 4 and week 8 following laparoscopy. Therapeutic response was considered if there were at least two or more of the following criteria present: 1) weight gain ≥5%, 2) haemoglobin increase ≥1 gm. % and 3) at least half of the symptoms were much better or resolved, including assessment of adherence to tuberculosis treatment using the TB clinic treatment card were evaluated. All the patients had repeat abdominal ultrasonography at follow up to compare the findings with the preoperative reports.

A total of 59 patients (39 male & 20 female) who were suspected of having abdominal tuberculosis underwent diagnostic laparoscopy after obtaining informed & written consent. The highest incidence was seen in 20-29 year age group 24 (40.67%) and lowest in 50-59 year. Mean age of incidence in males was 27.45±10.19 years and 34.11±16.54 years in females with overall mean of 30.78 ±13.80 years. Males were more affected than females (male to female ratio of 1.95).

Only 7 out of 59 patients (11.86%) gave past history of pulmonary TB and 14(23.73%) patients gave positive history of presence of or treatment for pulmonary TB in other family member. None had prior history of abdominal TB.

It has been established that the incidence of all tuberculosis, including abdominal tuberculosis is rising and the diagnosis is still difficult.

The mortality of tuberculous peritonitis is 47- 49% if untreated, although it can be less than 5% with treatment, several studies report a mortality of up to 60% mainly because of delayed or missed diagnosis. ^{5, 6} To avoid any delay or missed diagnosis we wished to look at the utility of diagnostic laparoscopy in abdominal tuberculosis. Although the concept of diagnostic laparoscopy in tuberculosis is not new. ⁷ Historically peritoneal biopsy or peritoneoscopy. ⁸ Used to be done under local anaesthetics with a limited view of the abdominal cavity and higher complication rates, whereas at present laparoscopy is performed under general anaesthesia with better intra-operative views and less complication rates.

The male female ratio is highly variable from equal in the series reported by Ramesh et al to marked male predominance reported in Abdelaal A et al and to a marked female predominance in the study of Safarpor Faizollah et al. ^{9, 10, 11} In present study , disease incidence was found to be more in males 39(66.10%) cases, females were 20(33.90%) with male to female ratio of 1.95:1 which is comparable with study by Rai et al. ¹²

The most common symptom in present study was abdominal pain in 35 patients (59.32%) followed by loss of appetite and abdominal distention (51.22% each), fever (48.90%) and weight loss (45.66%) which is comparable with studies by Khan R et al, Islam J et al. ^{13, 14} Pallor and abdominal tenderness was common findings present in 25 (42.37%) and 24 (40.67%) patients respectively. Finding of abdominal tenderness in present study is comparable with study by Safarpor Faizollah et al. In the present study, association of old pulmonary TB with abdominal TB was found in 11.86% patients which comparable with study by Rai et al and more than in study by Ramesh J et al while family history of pulmonary TB or treatment for pulmonary TB in family member was found in 23.73% patients with abdominal TB which is comparable with study by Rai et al.

In the present study, 68.41% patients had normocytic normochromic anemia which is comparable to study by Wells AD et al ¹⁵ and Ramesh J et al while 28% patients showed raised TLC count which is comparable with Islam J et all .

The chest radiographs of only 10 patients (16.94%) were suggestive of tuberculosis with pulmonary infiltrates, though sputum cultures were negative in all 10 patients; a percentage one fourth of that reported in studies by Nafeh et al and Ramesh J et al. 8, 10 Ultrasonography of abdomen was suggestive of tuberculosis in all patients, but histology/CBNAAT results were positive in 20 patients. Though the sensitivity of CT scan was high (96%), the specificity was very low (46.15%) in our series. Caseating lymph nodes with hypodense centres and peripheral rim enhancement along with calcification are highly suggestive of tuberculosis. ¹⁶ Histology for tuberculosis was positive in 70.73% of patients with high ESR which is comparable with study by Islam J et all. 10 patients (33.33%) had positive CBNAAT from peritoneal aspirate which is comparable with study by Jin et al.

Mesenteric lymphadenopathy was the common finding in our series and 27 (71.05%) of our patients had lymph nodes biopsied which is comparable with study by Islam J et all. Tubercles were present only in 45.76% of our patients, whereas Nafeh et al had 58% and Al-Mulhim et al had 91% of their patients with tubercles. 8, 17 Adhesions were present in 12 patients (31.57%) in our series which was less than in the series by Nafeh et al (42%) and Al- Mulhim et al (52%) and more than in study by Islam J et all (22%).

The most tuberculosis positive histology specimens are tubercles (82.35%) and lymph nodes (74.07%). The most sensitive specimen is omentum (84%) and the most specific specimen is omentum (100%). The least sensitive specimen is peritoneal aspirate (40%). Thirteen patients had diagnosis other than tuberculosis (22.03%) and these 13 patients would have been started on anti-tuberculosis therapy based on clinical and radiological findings.

Our main concern is these 13 patients (22.03%) who are tuberculosis negative and had no other alternative diagnosis. 12 of them had non-specific chronic inflammation and one had no pathology. Al-Mulhim et al had 19% non-specific chronic inflammation, Nafeh et al had 3% of their patients with non-specific chronic inflammation and 7% with unsatisfactory biopsy and Islam J et al had 18% non- specific chronic inflammation.

Three patients (5.08%) had peri operative complications e.g. bleeding and difficult access and had to be converted to laparotomy. These laparotomies are for bleeding and adhesions both (7.89%). Mimica et al reported 16% conversion rate to laparotomy due to adhesions while Islam J et al reported conversion rate (11%) to laparotomy due to bleeding (33%), adhesions (22%), bowel perforation (22%) and technical difficulties (22%). ¹⁸

This study has shown the feasibility of performing laparoscopy in the majority of patients with suspected abdominal tuberculosis. It has a high yield to establish the diagnosis of abdominal tuberculosis (65.78%) by sampling macroscopically pathological tissues.

We are thankful to all the patients, OT staffs, anaesthesiologists and OT attendants for their immense help during the conduct of this study

None Declared

None