7/31/2019 1-s2.0-S174391910700129X-main

1/6

Treatment of high output entero-cutaneous stulaeassociated with large abdominal wall defects: Single centerexperience

G. Dionigia ,*, R. Dionigia , F. Rovera a , L. Bonia , P. Padalino a , G. Minojab , S. Cuffari b ,G. Carraello ca Department of Surgical Sciences, University of Insubria, Viale Borri 57, 21100 Varese, Italyb Department of Anesthesiology and Critical Care Medicine, University of Insubria, Viale Borri 57, 21100 Varese, ItalycDepartment of Radiology, Vascular and Interventional Radiology, University of Insubria, Viale Borri, 57-21100, Varese, Italy

a r t i c l e i n f o

Article history:Published online 2 August 2007

Keywords:Entero-cutaneous stulaSurgeryMalnutrition

CancerSepsisVAC therapy

a b s t r a c t

Background and aim: Enteric stulas are dened by their sites of origin, communication andow. We evaluate the treatment of complex patients with entero-cutaneous stulae withlarge abdominal wall defects.Materials and methods: Retrospectivecase note reviewof 19 patients (15males,median age46years) treated at the Department of Surgical Sciences, University of Insubria, Varese, Italy.Thesewere distinguished by multiple/wide gastrointestinalstula orices, withtotal discon-tinuity of bowel. Fistulas were not covered by abdominal wall thus presenting with a giantabdominal wall defects. Surgery was planned once adequate nutritional status was present.Results: Allstulas resulted from previous surgery forIBD in 7 cases (37%), abdominaltrauma4 (21%),acute necrotic infected pancreatitis3 (16%),intra-abdominalmalignancy 3 (16%),anddiverticular disease 2 (10%). Themost commonsite of presentation was ileum (80%). Medianstula output was 800 ml/day (range 4001600 ml/day). Seltzers prognostic index identiedmalnutrition in 70% of patients at the time of presentation. The elapsed mean time fromonset of stula and elective time of surgical management were 184 days (range 202190days). The VAC system was used in the last 7 patients preoperatively and in 6 patients withpostoperative abdominal wound dehiscences that could not be closed immediately andwho were at high risk for healing complications. There were no complications from theVAC therapy. Surgerywas successful in 69%of cases. Mortality rate was 21%. Factors relatedto mortality were persistent malignancy, malnutrition and sepsis.Conclusions: After optimization of nutritional status surgery with en bloc resection of stulaoffers best results. In this series, cancer and sepsis were unfavourable factors for outcome.These stulas may be successfully managed with a multidisciplinary approach.

2007 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Gastrointestinal (GI) stula is an abnormal leaks of the bowelcontents to other organs (i.e. colovesical stula), other parts of

the intestine (entero-enteral) or the skin (entero-cutaneous). 1

The majority of stulas are consequences of a surgical proce-dure. Causes include disruption of the anastomotic sutureline, unintentional enterotomy or inadvertent bowel injury

* Corresponding author. Tel.: 390332278450; fax: 390332260260.E-mail address: gianlorenzo.dionigi@uninsubria.it (G. Dionigi).

www. t h e i j s . c om

1743-9191/$ see front matter 2007 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.ijsu.2007.07.006

i n t e r n a t i o n a l j o u r n a l o f s u r g e r y 6 ( 2 0 0 8 ) 5 1 5 6

mailto:gianlorenzo.dionigi@uninsubria.ithttp://-/?-http://-/?-mailto:gianlorenzo.dionigi@uninsubria.it

7/31/2019 1-s2.0-S174391910700129X-main

2/6

at the time of closure. Less frequently GI stulas are the re-sults of trauma. Inammatory processes, such as inamma-tory bowel disease (IBD), may also cause stulas. 2

By convention, GI stulas are usually dened by their sitesof origin, communication and ow 3; moreover, complex s-tula are associated with an intra-abdominal abscess cavity. 4

Severe malnutrition and uid, electrolyte or metabolic dis-

turbances are the main effect of GI stulas depending on theirlocation along the intestine. 57 Furthermore, they may also bea source of skin problems. 8

Recent evidences support the fact that stula output isa signicant factor inuencing closure of GI stula, with theodds of spontaneous closure 3 times greater for lowoutput s-tulas (efuent < 200 ml/24 h) than for high output stulas(efuent > 200 ml/24 h).9,10 Despite advances in metabolicand nutritional care, mortality rates remain high. 11

This paper reviews our experience on the treatment of entero-cutaneous stula to evaluate current managementpractice and outcome. Because GI stulas display so muchvariability from patient to patient (high versus low output,site of origin, presence or absence of associated organ dys-function, and malnutrition), we evaluate the treatment of a specic group of patients with highoutputentero-cutaneousstulae associated with a large dehiscence of the abdominalwound.

2. Materials and methods

A retrospective descriptive study of 19 patients (15 males, 4females; median age 46 years, range 2179) with entero-cutaneousstulasassociated with a large dehisced abdominalwound treated at a single institution from 2000 to May 2007.This is the Department of Surgical Sciences, University of Insubria, in Varese (Italy) which is a university hospital andreferral centre.

At examination all patients presented with grossly visibleintestinal mucosa; in details, these were distinguished bymultiple and wide stula orices (more than 1 cm in diame-ter), and complete anastomotic disruption with total disconti-nuity of bowel ends (end stulae) ( Fig. 1). All these stulaspresented with an efuent > 200 ml/24 h and the orices of the stulas were not covered by abdominal wall, thus present-ing with a large abdominal wall defect.

Upper gastrointestinal tract (esophagus and stomach), bil-iary and pancreatic stulas were excluded from this study. 12

Outside referrals accounted for all patients. Patients were fol-lowed-up for a median of 22 months (range 1131 months)after denitive surgical treatment.

Patients were managed by a multidisciplinary team. Adaily record was kept of stula, urine, stomal and/or faecaloutput.

Sources of sepsis were identied and treated quickly, using appropriate radiological investigation and culture of all sitesof potential infection. All patients had more than one investi-gation; small bowel contrast follow-through studies weremost commonly used, followed by stulography and retro-grade studies of the distal bowel. Suspected collections wereinvestigated using computer tomography (CT) or ultrasonog-raphy (US). 13

We used the Seltzer instant prognostic index to identifynutritional status. 14 All patients were supported with articialalimentation as total parental nutrition (TPN) or total enteralnutrition (TEN). We consider adequate indexes of good nutri-tional statusserum albumin level ! 3.5 g/dl and blood lympho-cytic count cells ! 1500/mm 3.

We dene the following clinical events: T1 the time of rstlaparotomy; T2 onset/rst recognition of stula; and T3the elective time of denitive surgical treatment of entero-cutaneous stula.

Skin protection was achieved at an early stage using a vari-ety of barriers, adhesives and wound drainage bags (Bogotabag),15,16 allowing containment, drainage, continuous irriga-tion of the wound with saline, and measurement of efuentand to facilitate healing.

The vacuum assisted closure (VAC) system (Kinetic Con-cepts, Inc., San Antonio, TX, USA) was used preoperativelyand postoperatively in patients with an abdominal woundthat could not be closed immediately and who were at highrisk for healing complications. The VAC device was main-tained on a continuous mode with a negative pressure from 75 to 125 mmHg (Fig. 2). The dressing was changed every2 days. In most cases, protection of skin was directed by thesurgeon and ostomy nurses.

Initial treatment was conservative except for patients withan abscess who needed urgent drainage.Subcutaneous or intra-venous injection octreotide wasusedin 4 patientspreoperativelyin an attempt to decrease stula output. 17 In nonepatientsoctreotide was administered as prophylaxis postoperatively.

Surgery was planned once adequate nutritional status wasachieved. Surgical procedure comprised re-laparotomy, enbloc resection of the involved bowel and overlying skin(Fig. 3a, b), and anastomosis. Careful hand-sewn 2-layeredlateral-to-lateral anastomosis created to ensure good mucosalapposition and serosal closure. Finally we ensured that theentire intestine have been mobilized and free from injury.The bowel was temporary defunctioned at stula surgery by

Fig. 1 A case of multiple and wide entero-cutaneousstulas with total discontinuity of bowel ends (endstulae), associated with a large dehisced abdominal

wound.

i n t e r n a t i o n a l j o u r n a l o f s u r g e r y 6 ( 2 0 0 8 ) 5 1 5 652

7/31/2019 1-s2.0-S174391910700129X-main

3/6

fashioning a small bowel stoma proximal to the anastomosesin those patients presenting with complex GI stulas withmultiple intra-abdominal abscess and entero-enteral stulas,in patients requiring more than one bowel anastomoses.

Beforeclosing the abdomen, we considered placing a trans-gastric jejunostomy for early postoperative enteral feeding.We placed soft, closed-suction drains only for abscess cavitiesdiscovered intraoperatively, with removal following a CT scanor US that demonstrated obliteration of the space. During ab-dominal closure, omentum was placed between the anasto-mosis and the midline wound, if possible.

We have developed a staged approach for abdominal wallreconstruction: plastic closure with aps was preferable toprosthetic mesh in an effort to minimize the incidence of postoperative stula caused by foreign material. VAC therapycontinued until the integrity of the abdominal wall was re-established by surgical procedures or secondary healing withdelayed primary wound closure.

All procedures were performed by one senior surgeon (RD).

3. Results

Fistulae distribution is reported in Table 1 . Fifteen (79%) stu-las were complex: stula involved multiple bowel loops withan intra-abdominal chronic abscess cavity and internal stu-las. Allstulas (100%) resulted from previously abdominalsur-gery due to inadvertent enterotomies (unrecognized bowelinjuries) or anastomotic dehiscences; in details these hadIBD (Crohns) in 7 cases (37%), previous surgery for abdominaltrauma 4 (21%), acute necrotic infected pancreatitis 3 (16%),intra-abdominal malignancy (2 colorectal cancer, 1 gyneco-logic tumor) 3 (16%), and diverticular disease 2 (10%) ( Table2). These operations consisted also in lysis of adhesions in 8/19 of cases (42%). Mean number of previous laparotomieswas4 (range 27). Median stulaoutput was800 ml/day (range

4001600 ml/day).

Fig. 2 Skin treatment with VAC system in a case of entero-cutaneous stulas associated with a large dehiscedabdominal wound.

Fig. 3 (a, b) Laparotomy with en bloc resection of theinvolved bowel and overlying skin.

Table 1 Site of entero-cutaneous stula ( N [ 19)

Organ site Frequency (%)

Ileum 80Colon 15 Jejunum 5

Table 2 Postoperative surgical cause of entero-cutaneous stula ( N [ 19)Cause Frequency (%)

IBD 37Abdominal trauma 21Intra-abdominal malignancy 16Infected pancreatitis 16

Diverticular disease 10

i n t e r n a t i o n a l j o u r n a l o f s u r g e r y 6 ( 2 0 0 8 ) 5 1 5 6 53

7/31/2019 1-s2.0-S174391910700129X-main

4/6

The elapsed meantime fromT1 toT2 was 54 days(range 3485). The elapsed mean time from onset (T2) and elective time(T3) of surgical management were 184 days (range 202190days).

The majority of thepatients were malnourished at the timeof presentation (T2). Seltzers prognostic index identied mal-nutrition in majority of patients (70%) at the time of presenta-tion (T2). Serum albumin and lymphocytic count showedhigher levels at the end of treatment (T3) than at the beginning (T2). In particular, the mean serum albumin and blood lym-phocytic count cells were 3.32 g/dl, 1350 cells/mm 3 and3.45 g/dl, 1560 cells/mm 3 , respectively, at T2 and T3 ( Table 3 ).Between T2 and T3, 13 patients (69%) were supported witharticial alimentation (TPN 70% and TEN 30%).

No spontaneous closure/healing of stulae was observed.In 4 patients, the effect of octreotide was monitored; in nonepatients, octreotide was of benet in output reduction orspontaneous resolution.

Six patients (31%) underwent radiologically CT-guideddrainage procedures for abscess cavities before denitive sur-gery as part of initial resuscitation.

Denitive surgery was successful in 13 (69%) cases. Thebowel was defunctioned at stula surgery in 6 patients(31.5%) by fashioning a small bowel stoma proximal to theanastomoses (end/loop ileostomy). There were no intraopera-tive complications. Three patients underwent radiologicallyCT-guided drainage procedures for abscess cavities after de-nitive surgery (16%). Postoperative catheter-related sepsisoccurred in 7 patients (36%). Metabolic complications werecommon, and included hypoalbuminaemia (78%), hypocal-caemia (73%), anemia (63%), and deranged liver function(63%).

The VAC system was used in the last 7 patients preopera-tively and in 6 patients with postoperative abdominal wounddehiscences that could not be closed immediately and whowere at high risk for healing complications. There were no

complications from the VAC therapy. Stable cutaneous cover-age was subsequently achieved in all patients by mesh graft-ing (13) (Fig. 4), or secondary intention healing (6). Nopatients had part of their VAC therapy as outpatients.

Six patients (31.5%) developed recurrence. Of these 6 cases3 had low output < 200 ml/24 h stulas all recovering conser-vatively with spontaneous closure. Three patients with high

output re-stulation required new surgery.No patient died during the course of treatment for their in-

testinal stulas (between T2 and T3). The overall in-hospitalmortality rate after denitive surgery was 21% (4/19). Two pa-tients developed a disseminated intravascular coagulopathysecondary to sepsis, and died from multiple organ dysfunc-tion. Two patients died from cardiorespiratory arrest. Threefourth of these patients were affected by stulas resultedfrom surgery for intra-abdominal malignancy (persistent/recurrence of cancer).

4. Discussion

This paper is a descriptive, retrospective report of a smallnumber but complex patients with high output entero-cutaneous stulas associated with a large dehisced abdominalwound. This condition is a challenging clinical problem.

GI stulas areassociated with prolonged hospital stay, highmorbidity and mortality. 11 In this study group in which stulaorices were not covered by the abdominal wall no spontane-ous closure/healing was observed. The length of the stulatract is important because greater the distance between thebowel and the skin, higher the incidence of spontaneous clo-sure. 18,19 A longer tract not only decreases the likelihood of skin epithelialization but also provides a greater resistance to

ow through the tract, promoting closure.18,19

Furthermore,the exposed bowel is at risk for furtherstula formation in un-protected loops because of desiccation, dressing changes andlacerations. Moreover, drainage from entero-cutaneous stu-lae is associated with severe inammatory skin reactionssuch as maceration and erythema. In ourexperience, success-ful and simple techniques of external control of the stula in-cluded laparostoma and the VAC system. 20,21 Thesedevices allow quantication and characterization of the

Table 3 Trend of nutritional index (according to Seltzersnutritional prognostic index) and clinical evolution. Scalefrom low to high represents nutritional index

T2: rst recognition of stula; and T3: time of denitive surgicaltreatment. Patients (PT) 11, 15, and 17 had recurrence of a high out-put stula (PT15 died). PTs 6, 12, and 16 died (all from previous sur-

gery for malignancy). Five sixth of these patients had medium tolow nutritional index in T3.

Fig. 4 Stable cutaneous coverage by skin grafting.

i n t e r n a t i o n a l j o u r n a l o f s u r g e r y 6 ( 2 0 0 8 ) 5 1 5 654

7/31/2019 1-s2.0-S174391910700129X-main

5/6

7/31/2019 1-s2.0-S174391910700129X-main

6/6

12. Kazanjian KK, Hines OJ, Eibl G, Reber HA. Management of pancreatic stulas after pancreaticoduodenectomy: results in437 consecutive patients. Arch Surg 2005;140 (9):84954.

13. Gutierrez A, Lee H, Sands BE. Outcome of surgical versuspercutaneous drainage of abdominal and pelvic abscesses inCrohns disease. Am J Gastroenterol 2006;101 (10):22839.

14. Seltzer MH, Bastidas JA, Cooper DM, Engler P, Slocum B,Fletcher HS. Instant nutritional assessment. JPEN J ParenterEnteral Nutr 1979;3 (3):1579.

15. Alci R, Ashkenazi I, Kessel B, Zut N, Sternberg A. Temporarybowel diversion using the Bogota bag (Hadera stoma):technical details. J Am Coll Surg2004;199 (2):3446.

16. Cro C, George KJ, Donnelly J, Irwin ST, Gardiner KR. Vacuumassisted closure system in the management of enterocutaneous stulae. Postgrad Med J 2002;78 (920):3645.

17. Hesse U, Ysebaert D, de Hemptinne B. Role of somatostatin-14and its analogues in the management of gastrointestinalstulae: clinical data. Gut 2001;49 (Suppl. 4):iv1121.

18. Foster 3rd CE, Lefor AT. General management of gastrointestinal stulas. Recognition, stabilization, andcorrection of uid and electrolyte imbalances. Surg Clin NorthAm 1996;76 (5):101933.

19. Rolandelli R, Roslyn JJ. Surgical management and treatmentof sepsis associated with gastrointestinal stulas. Surg ClinNorth Am 1996;76 (5):111122.

20. Dearlove JL. Skin care management of gastrointestinalstulas. Surg Clin North Am 1996;76 (5):1095109.

21. Heller L, Levin SL, Butler CE. Management of abdominalwound dehiscence using vacuum assisted closure in

patients with compromised healing. Am J Surg 2006;191 (2):16572.

22. Chamberlain RS, Kaufman HL, Danforth DN. Enterocutaneousstula in cancer patients: etiology, management, outcome,and impact on further treatment. Am Surg 1998;64 (12):120411.

23. Cinat ME, Wilson SE, Din AM. Determinants for successfulpercutaneous image-guided drainage of intra-abdominalabscess. Arch Surg 2002;137 (7):8459.

24. Men S, Akhan O, Koroglu M. Percutaneous drainage of abdominal abscess. Eur J Radiol2002;43 (3):20418.

25. Alivizatos V, Felekis D, Zorbalas A. Evaluation of theeffectiveness of octreotide in the conservative treatment of postoperative enterocutaneous stulas.Hepatogastroenterology 2002;49 (46):10102.

26. Alvarez C, McFadden DW, Reber HA. Complicatedenterocutaneous stulas: failure of octreotide to improvehealing. World J Surg 2000;24 (5):5337.

27. Hollington P, Mawdsley J, Lim W, Gabe SM, Forbes A,Windsor AJ. An 11-year experience of enterocutaneousstula. Br J Surg 2004;91 (12):164651.

28. Poritz LS, Gagliano GA, McLeod RS, MacRae H, Cohen Z.Surgical management of entero and colocutaneous stulae inCrohns disease: 17 years experience. Int J Colorectal Dis2004;19 (5):4815.

29. Lynch AC, Delaney CP, Senagore AJ, Connor JT, Remzi FH,Fazio VW. Clinical outcome and factors predictive of recurrence after enterocutaneous stula surgery. Ann Surg2004;240 (5):82531.

i n t e r n a t i o n a l j o u r n a l o f s u r g e r y 6 ( 2 0 0 8 ) 5 1 5 656