Tubercular Abdominal Cocoon: Systematic Review of an Uncommon Form of Tuberculosis

Abstract

Background:

Abdominal cocoon formation is an uncommon manifestation of abdominal tuberculosis that is characterized by the formation of a fibrous membrane-like sac around the small intestinal loops. Appropriate treatment and outcomes are uncertain

Objectives:

To review the clinical presentation, treatment, and outcomes for tubercular abdominal cocoon (TAC).

Patients and Methods:

We included studies published in the English language and listed in EMBASE or PubMed. All case series or reports that reported patients with TAC were considered for inclusion. Details regarding demographic, clinical presentation, and treatment received were tabulated.

Results:

The clinical features included predominantly abdominal pain, abdominal distension and features of intestinal obstruction, loss of appetite, and weight loss. The diagnosis was usually established at surgery, however, computed tomography was a useful tool. In most patients the treatment reported was surgical, however, conservative therapy with anti-tubercular therapy (ATT) did succeed in a subset of patients. The published literature includes only case series and reports. The outcomes, especially long-term outcomes, have not been reported in most studies.

Conclusion:

Abdominal cocoon is an uncommon form of abdominal tuberculosis and conservative management with ATT may suffice in some patients whereas non-responsive cases require surgery.

Tuberculosis is a problem for the residents of the tropical world. The increase in human immunodeficiency virus (HIV) infection and the increasing use of immunosuppressant drugs may have resulted in an increase in number of cases from the Western world [1]. Abdominal tuberculosis is a form of extra-pulmonary tuberculosis that may be characterized by the involvement of luminal structures such as the intestine, peritoneum, lymph nodes, or visceral organs (such as pancreas and liver) [2,3]. Peritoneal tuberculosis can take many forms depending on the amount of fluid accumulation and the fibrotic reaction: it may occur as a predominantly wet form associated with free ascites or may be a result of a fixed form in which many peritoneal adhesions and fibrosis may accompany the disease [2,4]. Abdominal cocoon is a distinctive pattern that is observed uncommonly with peritoneal tuberculosis. Tubercular abdominal cocoon (TAC) is an uncommon pattern of peritoneal tuberculosis characterized by formation of a membranous structure that surrounds the small intestinal loops. Although the entity may be asymptomatic occasionally, abdominal pain and intestinal obstruction are the frequent clinical presentations [5].

Abdominal cocoon is also termed as sclerosing encapsulating peritonitis or encapsulating peritoneal sclerosis (when in chronic ambulatory peritoneal dialysis patients). This clinical condition has been described as occurring in either a primary form (idiopathic) or secondary form (related to chronic ambulatory peritoneal dialysis, malignancy, β-blockers) [5,6]. Tuberculosis is an uncommon cause of this condition and the published literature is limited to a few case series and case reports. The objective of this review is to summarize the available literature about TAC and suggest the diagnostic criteria and management approach for this entity [7 –26]

Patients and Methods

Literature search

We searched PubMed with the search strategy tuberculosis OR tubercular AND abdominal cocoon OR sclerosing encapsulating peritonitis OR encapsulating peritoneal sclerosis). We also searched EMBASE with the strategy abdominal AND cocoon OR sclerosing encapsulating peritonitis OR (encapsulating AND peritoneal AND sclerosis) AND tuberculosis. Of the 47 and 37 respective results retrieved from PubMed and EMBASE, 25 were duplicates (Fig. 1). In all 59 results were considered for inclusion and eventually four case series and 16 case reports were included in the final analysis (Tables 1 and 2) [7 –26]. Among these was one case report in which Mycobacterium bovis was implicated as the causative agent [26].

FIG. 1.

Selection of articles for inclusion in the present review.

Table 1.

Case Series on Tubercular Abdominal Cocoon

Reference	Site study type	No. of cases	Gender (male)	Age (median, range)	Mantoux	Clinical presentation	Basis of diagnosis cocoon	Basis of diagnosis (TB)	Confirmed TB	Probable TB	Cocoon (partial or complete)	Microbiologic positivity (site)	Ascites present	Mortality	Treatment (medical alone/surgery)
Sharma et al. 2017	India Retrospective study	15^a	12	28	12	IO: 11 Pain: 4 Fever:10 Lump: 9 LOA: 12 LOW: 13	CT: 13 MRI: 1 Surgery: 2	Microbiologicl: 3 Granuloma: 7 ADA: 8	3	12	Partial: 3 Complete: 12	Peritoneal: 1 Sputum: 1 Spleen: 1	10	1	Medical in 12
Singh et al. 2013^b	India Retrospective Study	18	10	NA	NA	IO: 16 Perforation:2gut Lump: 2	Surgery: 17 Barium: 1	Microbiologic: NA Histology: 9	NA	NA	NA	NA	NA	1	Surgical in 17
Wani et al. 2010	India Prospective Study	11	3	18.3 (16 mon-58 years)	NA	IO: 11	Surgery: 11 CT: 1	Histology: Caseating granuloma 11	11	0	Partial: 10 Complete: 1	NA	NA	0	Surgical
Kaushik et al. 2006	India Retrospective study	6	4	21.8 (9–45)	NA	IO: 6 Lump: 1	Surgery: 6	Histology Caseating granuloma: 5 Pulmonary: 1	5	1	Partial: 4 Complete: 2	None	NA	1	Surgical

Fourteen had completed follow-up.

Describes 18 cases of tubercular and non-tubercular cocoon; seven patients had received empirical ATT of whom only two had definite TB on histology after surgery.

IO = intestinal obstruction; TB = tuberculosis; CT = computed tomography; MRI = magnetic resonance imaging; NA = not available; ATT = anti-tubercular therapy.

Table 2.

Case Reports on Tubercular Abdominal Cocoon

Reference	Site	Gender (male)	Age	Mantoux	Clinical presentation	OOI	Basis of diagnosis cocoon	Basis of diagnosis (TB)	Confirmed or probable TB	Cocoon (partial or complete)	Ascites present	Outcome (death/survival)	Treatment (medical alone/surgery)
Lalloo et al. 2002	South Africa	Female	36	NA	Pain Abdominal distension	Pulmonary	Surgery CT	Histology AFB+	C	Partial	Yes	Survival	Surgery
Makam et al. 2008	India	Female	21	NA	IO Lump	None	Surgery	PCR+	C	Partial	Yes	NA	Surgery ) Laparoscopic)
Jain et al. 2008	India	Male	13	Positive	IO Loss of weight Lump	Mesenteric LN Supra- clavicular Ln	CT	Caseating Granuloma in supraclavicular lymph node Mantoux +	C	Partial	No	Survival	Conservative
Rastogi et al. 2008	India	Male	30	NA	Pain	Pulmonary	CT Surgery	Histology	Uncertain	Partial	Yes	Survival	Surgery
Mohanty et al. 2009	India	Male	15	NA	Pain Lump	None	CT	Granuloma	P	Partial	No	Survival	Surgery
Bani-Hani et al. 2009	UK	Male	45	NA	Perforation	None	CT	AFB+ in lymph nodes	C	NA	NA	Survival	Surgery
Gadodia et al. 2011	India	Male	22	NA	Pain Distension	NA	CT Surgery	Granuloma in omentum Histology	P	Partial	Yes	Survival	Surgery
Singh et al. 2012	India	Male	15	NA	IO Lump	Pulmonary	Surgery	Lymph node: AFB+, caseating granulomas	C	Partial		Survival	Surgery
Sarmast et al. 2012	India	Male	30	NA	IOLump	NA	Surgery	Caseating granuloma in mesenteric lymph node	C	NA	Yes	Survival	Surgery
Tseng et al. 2012^a	China	Male	76	NA	PainCloudy effluent in PD	NA	CTSurgery	AFB + from peritoneal fluidCulture from membrane	C	Partial	Yes	Survival	Surgery
Gomes et al. 2012	India	Male	37	NA	Pain, distension, fever, weight loss	None	CT	Adenosine deaminase, PCR+ in ascites	C	NA	Yes	Survival	Conservative
Cisse et al. 2013^a	Senegal	Male	70	NA	PainDistensionLOA, LOW	None	Surgery	Culture in fluid	C	NA	Yes	Survival	Surgery
Puppala et al. 2014	India	Male	55	NA	PainDistensionFever	Pleural effusion	CT	Granuloma in peritoneal nodules	P	Partial	Yes	Survival	Conservative^b
Ibrarullah et al. 2016	India	Male	65	NA	DistensionLump	None	CT	Histology (?)	P	NA	Yes	NA	Surgery
Sousa et al. 2016	Portugal	Female	37	NA	DistensionLOWIO	Pulmonary	Surgery	HistologyCaseating granulomaAFB+	C	NA		Survival	Surgery
Anantha et al. 2012^c	Canada	Female	12	NA	IOWeight loss	None	Surgery	Culture (Mycobacterium bovis)Histology: caseating granulomas	C	NA	No	Survival	Surgery

Also on peritoneal dialysis.

Received steroids also.

Related to Mycobacterium bovis

TB = tuberculosis; OOI = other organ involvement; C = confirmed TB; P = probable TB; CT = computed tomography; MRI = magnetic resonance imaging; AFB = acid fast bacilli; IO = intestinal obstruction./

Definitions

We have considered tuberculosis as a definitive possibility in cases in which microbiological evidence in the form of acid-fast bacilli positivity on smear, culture or positivity of polymerase chain reaction using specific primers from tissue in which the presence is not otherwise expected, e.g., peritoneal biopsy or ascites or histopathologic evidence of caseating granulomas in absence of other diseases.

A case has been labeled as having presumptive (or probable) tuberculosis when the patients have all of the following:

1. Clinical and radiologic profile is consistent with tuberculosis

2. Biochemical (elevated adenosine deaminase in fluids), histologic (granulamtous inflammation), or corroboratory evidence (positive Mantoux skin test, changes on chest radiograph consistent with active or old pulmonary tuberculosis)

3. Exclusion of other differential diagnosis.

These definitions are based on the terminology used in Indian Extra-Pulmonary Tuberculosis Guidelines (INDEX-TB Guidelines) [27]. The diagnosis of abdominal cocoon was established by the demonstration of a membrane around a part or whole of small intestine as visualized on computed tomography or magnetic resonance imaging or detected on surgery. Cocoon was labeled a partial cocoon if the membrane surrounded only the small intestine whereas if the presence of large intestine or other organs inside the cocoon was detected the cocoon was labeled as complete cocoon [5]. The diagnostic strategy is depicted in Figure 2.

FIG. 2.

Suggested diagnostic criteria for tubercular abdominal cocoon (TAC). PCR = polymerase chain reaction; CT = computed tomography; MRI = magnetic resonance imaging; ADA = adenosine deaminase.

Data synthesis and analysis

We tabulated all demographic information (age, gender, place of residence), clinical presentation (pain, intestinal obstruction, lump), diagnostic methods for cocoon (imaging or surgery and tuberculosis (microbiologic, biochemical analysis of fluid, or histologic features) and the management strategy (conservative with anti-tubercular therapy [ATT] alone or surgical therapy) for each of the cases. Wherever the information was available for interpretation, we also recorded whether the cocoon was partial (including only the small bowel) or complete (small bowel and large intestinal or other organ involvement). Risk of bias was not assessed formally because the reports were all case series or case reports.

Results

Clinical and demographic profile

We included four case series and 16 case reports in the present study reporting 66 cases. However, one of the case series by Singh and Gupta [8] included both tubercular and non-TAC and did not provide separate details for the patients with TAC and was therefore excluded from summative analysis. The summative analysis, therefore, is based on 48 patients with TAC. Occasional reports also suggest multiple underlying etiologic factors that could have caused cocoon formation (e.g., tuberculosis and chronic ambulatory peritoneal dialysis) [22].

Most reports seemed to emerge from India as do all of the case series (Fig. 3, Tables 1 and 2). Only occasional cases have been reported from the Western world. Even in such cases the disease was reported in individuals who were immigrants from Asian or African countries [16,25,26]. The disease can afflict patients with any age and either gender. However, a male preponderance was apparent (64.6% of 48 cases) in most of the reports except in a case series from Kashmir, India, in which females contributed to eight of the 11 cases [9]. The clinical manifestations include abdominal pain, abdominal distension, and features of intestinal obstruction, loss of appetite and weight loss. Of the 48 cases in which the clinical details were summated, 34 (70.8%) had episode(s) of intestinal obstruction, 19 (39.6%) has loss of appetite or weight, 16 (33%) had a palpable abdominal lump, 12 (25%) had abdominal pain without intestinal obstruction whereas 13 had history of fever. One patient was reported to have presented with features of perforation peritonitis [16]. The frequency of human immunodeficiency virus (HIV) infection was not reported and no case with concurrent HIV infection could be identified in our literature search.

FIG. 3.

Geographic locations from which cases of tubercular cocoon have been reported.

Diagnostic evaluation

The positivity of Mantoux skin test has been reported infrequently. However, in one case series 80% of patients had a positive Mantoux skin test [7]. The use of interferon gamma release assays (IGRA) has not been reported in the published literature.

For the diagnosis of abdominal cocoon the use of computed tomography and diagnosis at time of surgery was reported in various reports. In the four case series (Table 1) only one series reported pre-surgical diagnosis using computed tomography whereas in others pre-operative diagnosis was reported infrequently [7]. In series from Srinagar, India, only one of the 11 patients had a pre-operative diagnosis made on imaging [9]. In our recent series, all except one of the 15 patients had a diagnosis on imaging [7]. Although most cases suggest a radiologic and surgical correlation, occasional cases have been reported suggesting that computed tomography may miss the cocoon on some occasions [19]. Most studies have not reported whether the underlying cocoon was partial or complete. In our series, most patients (80%) had a complete abdominal cocoon [7]. In reports by Wani et al. [9] and Kaushik et al. [10] most patients seemed to have a partial cocoon on basis of description provided.

The diagnosis of tuberculosis was based on combination of criteria. In a series by Sharma and Gupta [7] only three patients had confirmed tuberculosis, whereas all of the patients in series by Wani et al. [9] and Kaushik et al. [10] had confirmed tuberculosis. This is primarily based on the presence of caseating granulomas although the microbiologic evidence was not available. Most of the case reports that have reported the use of polymerase chain reaction have not identified the primers used.

Treatment and outcomes

Most reports have described surgical therapy for the TAC. However, some reports suggest that a conservative approach with ATT may also improve symptoms. In a case report, Jain et al. [13] described a 13-year-old male patient who improved with ATT. Similarly a 37-year-old male reported by Gomes et al. [21] improved with ATT. Also, in a series from India a majority of patients who were started on ATT showed improvement [7]. However, the course is often punctuated by recurrent episodes of intestinal obstruction and occasionally conservative therapy may fail necessitating surgery [7,21,22]. In a series of six surgical patients of TAC, apart from one death, four patients developed post-operative enterocutaneous fistulae [10]. Other surgical series, however, report excellent outcomes [8,9]. Occasional reports also describe utility of a laparoscopic approach [12]. The recurrent episodes of obstruction may warrant a long hospital stay or re-admission [26]. Unrecognized cocoon may be complicated by development of intestinal perforation [16]. Use of steroids has been reported in three patients in one series with good outcome [7].

Discussion

Tubercular abdominal cocoon is an uncommonly described variant of abdominal tuberculosis. Because the literature regarding this entity is scarce, we attempted to collate the literature to seek a diagnostic and management strategy based on the literature. The difficulty in diagnosis of abdominal tuberculosis is well known and microbiologic evidence of tuberculosis is difficult to obtain in this entity. Therefore, a response to therapy (Logan modification of Paustian criteria) is often used as one of the criteria to establish the diagnosis [2,28]. However, the appropriate definition of “response to therapy” is uncertain and a recent article suggests that mucosal (ulcer) healing is the appropriate definition for patients with intestinal tuberculosis. Even so, the patients may continue to remain symptomatic because of formation of stricture [29]. The use of these criteria in TAC is even more difficult because of the lack of any objective criteria for response. To date, no evidence suggests that the cocoon or the membrane disappears with ATT [7]. Even with ATT, the patients may continue to be symptomatic and need surgery. Furthermore, if surgery is done after a period of ATT, the pathognomonic findings of tuberculosis could have disappeared as has been noted in one series in which patients who were on ATT prior to surgery did not have evidence of tuberculosis on histology [8]. We therefore propose that TAC may be described to be either a confirmed or a clinically diagnosed case on the basis of microbiologic and histologic findings (Fig. 2). Also, the diagnosis of cocoon has been established on basis of imaging (computed tomography or magnetic resonance imaging) or surgical findings. The use of ultrasound has been described for diagnosis of cocoon but the modality is operator dependent [30]. Also, the presence of pulmonary tuberculosis (past or current) may not be apparent in all the patients. In one series of 15 patients with abdominal cocoon related to tuberculosis, only four had evidence of pulmonary infection [7]. When present, the evaluation of pulmonary lesions with sputum examination for acid-fast bacilli may help in the diagnosis in a given case. However, in a majority of cases there may be no pulmonary lesions and the clinician may encounter diagnostic difficulty in such cases.

The treatment of TAC should include effective ATT and management of episodes of intestinal obstruction. The episodes of intestinal obstruction should be managed with nasogastric aspiration, monitoring of abdominal girth, and intravenous fluids. If the situation improves, one can suggest dietary modifications and hope that inflammatory component would be addressed by the ATT. In non-responsive cases, surgical intervention with resection of membrane and adhesiolysis may be needed. All attempts should be made to avoid surgery because the surgery for this condition is associated with significant morbidity. As of now, the utility of steroids is anecdotal and cannot be recommended. However, steroids have been used in encapsulating peritoneal sclerosis observed with peritoneal dialysis and may be considered along with ATT as an option.

To conclude, abdominal cocoon formation is an uncommon clinical manifestation of abdominal tuberculosis and usually presents with abdominal pain and intestinal obstruction. The diagnosis is on basis of a combination of features: diagnosis of cocoon with imaging or surgical findings and the diagnosis of tuberculosis using clinical, radiologic, histologic, biochemical, and microbiologic findings. Once the diagnosis is established management of intestinal obstruction and use of ATT may suffice in a subset of patients, however, recalcitrant cases may need surgical intervention.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Centers for Disease Control. Reported tuberculosis in the United States, 2015. Atlanta, GA: U.S. Department of Health and Human Services, CDC, November 2016. www.cdc.gov/tb/statistics/reports/2015 (Last accessed July 13, 2017).

Sharma

, Bhatia

. Abdominal tuberculosis. Indian J Med Res, 2004; 120:305–315.

Sharma

, Bhatia

, Malik

, et al. Visceral scalloping on abdominal computed tomography due to abdominal tuberculosis. Ther Adv Infect Dis, 2017; 4:3–9.

Sanai

, Bzeizi

. Systematic review: Tuberculous peritonitis—Presenting features, diagnostic strategies and treatment. Aliment Pharmacol Ther, 2005; 22:685–700.

Machado

. Sclerosing encapsulating peritonitis: Review. Sultan Qaboos Univ Med J, 2016; 16:e142–151.

Shah

, Kumar

, Singh

, et al. Cocoon carcinomatosa: An unusual cause of intestinal obstruction. Drug Discov Ther, 2017; 11:51–53.

Sharma

, Mandavdhare

, Rana

, et al. Role of conservative management in tubercular abdominal cocoon: A case series. Infection (in press).

Singh

, Gupta

. Abdominal cocoon: A case series. Int J Surg, 2013; 11:325–328.

Wani

, Ommid

, Waheed

, Asif

. Tuberculous abdominal cocoon: Original article. Ulus Travma Acil Cerrahi Derg, 2010; 16:508–510.

10.

Kaushik

, Punia

, Mohan

, Attri

. Tuberculous abdominal cocoon—A report of 6 cases and review of the literature. World J Emerg Surg, 2006; 1:18.

11.

Lalloo

, Krishna

, Maharajh

. Case report: Abdominal cocoon associated with tuberculous pelvic inflammatory disease. Br J Radiol, 2002; 75:174–176.

12.

Makam

, Chamany

, Ramesh

, et al. Laparoscopic management of abdominal cocoon. J Minim Access Surg, 2008; 4:15–17.

13.

Jain

, Nijhawan

. Tuberculous abdominal cocoon: A case report and review of the literature. Am J Gastroenterol, 2008; 103:1577–1578.

14.

Rastogi

. Abdominal cocoon secondary to tuberculosis. Saudi J Gastroenterol, 2008; 14:139–141.

15.

Mohanty

, Jain

, Agrawal

, et al. Abdominal cocoon: Clinical presentation, diagnosis, and management. J Gastrointest Surg, 2009; 13:1160–1162.

16.

Bani-Hani

, Al-Nowfal

, Gould

. High jejunal perforation complicating tuberculous abdominal cocoon: A rare presentation in immune-competent male patient. J Gastrointest Surg, 2009; 13:1373–1375.

17.

Gadodia

, Sharma

, Jeyaseelan

. Tuberculous abdominal cocoon. Am J Trop Med Hyg, 2011; 84:1–2.

18.

Singh

, Pandey

, Jindal

, Sandhu

. Tubercular abdominal cocoon: A rare cause of intestinal obstruction. J Surg Case Rep, 2012; 2012:10.

19.

Sarmast

, Showkat

, Sherwani

, et al. Abdominal tuberculosis with a cocoon. Iran Red Crescent Med J, 2012; 14:503–504.

20.

Tseng

, Tarng

. Cocoon-like fibroadhesive tuberculous peritonitis in a peritoneal dialysis patient. Chin J Physiol, 2012; 55:361–365.

21.

Gomes

, Mhetras

, Gundawar

. Tuberculous abdominal cocoon. Ann Gastroenterol, 2013; 26:279–280.

22.

Cisse

, Dia

, Seck

, et al. Encapsulating peritoneal sclerosis. Saudi J Kidney Dis Transpl, 2013; 24:1203–1206.

23.

Puppala

, Sripathi

, Kadavigere

, et al. Abdominal cocoon secondary to disseminated tuberculosis. BMJ Case Rep, 2014(Sep 19);2014.

24.

Ibrarullah

, Mishra

. Abdominal cocoon: “Cauliflower sign” on CT scan. Indian J Surg, 2016; 78:243–244.

25.

Sousa

, Batista

, Pacheco

, Nunes

. Abdominal tuberculosis: An old disease surprising young doctors. BMJ Case Rep, 2016(Nov 2);2016.

26.

Anantha

, Salvadori

, Hussein

, Merritt

. Abdominal cocoon syndrome caused by Mycobacterium bovis from consumption of unpasteurised cow's milk. Lancet Infect Dis, 2015; 15:1498.

27.

Working definitions of cases and outcomes. In: INDEX-TB Guidelines: Guidelines on Management of Extrapulmonary Tuberculosis of India. World Health Organization, 2016: 11–14.

28.

Logan

. Anorectal tuberculosis. Proc R Soc Med, 1969; 62:1227–1230

29.

Pratap Mouli

, Munot

, Ananthakrishnan

, et al. Endoscopic and clinical responses to anti-tubercular therapy can differentiate intestinal tuberculosis from Crohn's disease. Aliment Pharmacol Ther, 2017; 45:27–36.

30.

Rokade

, Ruparel

, Agrawal

. Abdominal cocoon. J Clin Ultrasound, 2007; 35:204–206.