Fistula

Accepted Manuscript

Open Abdomen with Concomitant Entero-Atmospheric Fistula: Attempt to Rationalizethe Approach to a Surgical Nightmare and Proposal of a Clinical Algorithm

Salomone Di Saverio , MD, Antonio Tarasconi , MD., Kenji Inaba , MD, FACS,Pradeep Navsaria , MBChB, FCS(SA), FACS, Federico Coccolini , MD, David CostaNavarro , MD, Matteo Mandrioli , Panteleimon Vassiliu , MD, FACS, Elio Jovine , MD,Fausto Catena , MD, Gregorio Tugnoli , MD

PII: S1072-7515(14)01845-6

DOI: 10.1016/j.jamcollsurg.2014.11.020

Reference: ACS 7650

To appear in: Journal of the American College of Surgeons

Received Date: 5 September 2014

Revised Date: 8 November 2014

Accepted Date: 12 November 2014

Please cite this article as: Di Saverio S, Tarasconi A, Inaba K, Navsaria P, Coccolini F, NavarroDC, Mandrioli M, Vassiliu P, Jovine E, Catena F, Tugnoli G, Open Abdomen with ConcomitantEntero-Atmospheric Fistula: Attempt to Rationalize the Approach to a Surgical Nightmare andProposal of a Clinical Algorithm, Journal of the American College of Surgeons (2014), doi: 10.1016/j.jamcollsurg.2014.11.020.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTOpen Abdomen with Concomitant Entero-Atmospheric Fistula: Attempt to Rationalize the

Approach to a Surgical Nightmare and Proposal of a Clinical Algorithm

Salomone Di Saverio, MD (1), Antonio Tarasconi, MD. (2), Kenji Inaba, MD, FACS (3), Pradeep

Navsaria, MBChB, FCS(SA), FACS (4), Federico Coccolini, MD (5), David Costa Navarro, MD (6),

Matteo Mandrioli (1), Panteleimon Vassiliu, MD, FACS (7), Elio Jovine, MD (1), Fausto Catena,

MD (2), Gregorio Tugnoli, MD (1),

(1) Maggiore Hospital – Bologna Local Health District, and Emergency Surgery and Trauma

Surgery Unit, Maggiore Hospital Trauma Center, Bologna, Italy.

(2) Department of Emergency and Trauma Surgery - Maggiore Hospital, Parma, ITALY

(3) Department of Surgery. University of Southern California, Los Angeles, USA

(4) Trauma Center, Groote Schuur Hospital & University of Cape Town, South Africa

(5) General and Emergency and Trauma Surgery, Hospital Papa Giovanni XXIII, Bergamo, Italy

(6) Department of Surgery, Alicante University General Hospital, Alicante, Spain

(7) Attikon University Hospital, Department of Surgery IV, Surgery and Traumatology, Athens,

Greece

Disclosure Information: Nothing to disclose.

Technique partially presented at the video session, 98th Clinical Congress of the American College of Surgeons, Chicago, IL, October 2012. Management Algorithm presented at the International Trauma & Emergency Surgery Week, Alicante, Spain, June 2014.

Correspondence address:

Salomone Di Saverio, MD

Maggiore Hospital, L.go Nigrisoli 2, Bologna Italy

Fax: +390516478412

email: [email protected]

Brief title: Entero-Atmospheric Fistula Management

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTAbbreviations and Acronyms AAST = American Association for the Surgery of Trauma EAF = entero-atmospheric fistula ECF = entero-cutaneous fistula IAS = intra-abdominal sepsis NPWT = negative pressure wound therapy POD = post-operative day

The development of an entero-atmospheric fistula (EAF) in the midst of an open abdomen

represents a surgical nightmare, with an extremely challenging critical care problem set including

the full spectrum of surgical, metabolic, nutritional and nursing issues. Spontaneous closure is rare

as there is no fistula tract and there is a lack of well vascularized overlying tissue. For these reasons,

the acute fistula management should aim to completely divert fistula output, protect the surrounding

viscera and allow for the granulation of exposed bowel. This is difficult to achieve, because of the

extreme frailty of the surrounding tissues, which are chronically injured by enteric fluids, and the

multiple systemic derangements of the patient, driven by severe dehydration, electrolyte and

acid/base disturbances, a hypercatabolic status and ongoing sepsis. Despite the advancements in

critical care and surgical management, EAF-related mortality is still upwards of 40%[1].

Significant heterogeneity in the surgical techniques described for EAF management remains

and there is no single approach that has proven to be ideal in every circumstance: the aim of this

study is to develop a flowchart designed to help guide the surgeons who are caring for this

devastating complication, to choose the best approach on a case-by-case basis.

EAF definition, etiology and management

Entero-atmospheric fistula is defined as the occurrence of an enteric fistula in the middle of an open

abdomen. Specific characteristics defining EAF are: the absence of a fistula tract, the lack of well-

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTvascularized surrounding tissue and location within an open abdomen, resulting in spillage of

enteric content directly into the open peritoneal cavity.

Little is known about the risk factors associated with EAF development. A recent

multicenter prospective observational study by Bradley, et al.[2] has attempted to determine

independent predictors for the development of entero-cutaneous fistula (ECF), entero-atmospheric

fistula and abdominal sepsis (IAS) in patients undergoing damage control laparotomy after trauma.

Using the AAST (American Association for the Surgery of Trauma) Open Abdomen registry they

identified patients who developed ECF, EAF or IAS during open abdomen management following

damage control laparotomy and they compared the groups of patients with and without

complications. Interestingly, their data showed that large bowel resection, large volume

resuscitation and an increasing number of re-explorations were statistically significant predictors for

the development of a fistula within an open abdomen after trauma.

Another recent review of the available techniques for open abdomen management found a

higher incidence of EAF occurring in septic OA compared to non-septic OA (12.1% versus 3.7%

respectively). Furthermore, this review did not show any evidence of a relationship between use of

NPWT (Negative Pressure Wound Therapy) and fistula formation[3].For many years the application

of negative pressure has been considered the primary cause of fistula formation, but further studies

have demonstrated that fistula formation rate can be as low as 5%[4,5], even with a hand-made

dressing, and that in a good proportion of cases the fistula can spontaneously heal during NPWT[6].

In a recent relatively large series of EAF in patients treated with an OA for the management of

abdominal sepsis, the incidence of EAF was 54.5% (18/33)[7]. In this series the rate of spontaneous

closure with NPWT was 22% (4/18). The average hospital stay was 88.89 days, the median number

of VAC (Vacuum Assisted Closure) applications was 22.5, and the median duration of VAC

applications 43.6 days. Mortality was as high as 44.4% (8/18).

Many of the principles applied to ECF management can be also considered in the

management of the EAF. The latter however are often more difficult as the unprotected bowel is

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTfriable and located within a “frozen abdomen”, precluding any possible surgical manoeuvres or

resection, with the added burden of peritonitis/sepsis due to the bowel content pooling directly

within the exposed peritoneal cavity.

Although the optimal option would be to perform a bowel resection or at least a proximal

bowel diversion, this option is invariably not feasible for multiple reasons including mesenteric

retraction and edema, the “frozen abdomen” and loss of domain due to retraction of the abdominal

wall.

As bowel resection or proximal diversion is not feasible, EAF management should aim to[8-10]:

• decrease fistula effluent: the use of total parenteral nutrition, somatostatin and its

analogues[11-14] as well as a proton pump inhibitor[15] can be effective in decreasing the

bowel effluent;

• correct any fluid and electrolyte imbalances, acid/base derangements and reversal of the

hypercatabolic state (collection, quantification and characterization of bowel effluent are of

key importance);

• interrupt or prevent ongoing sepsis;

• divert fistula output, thus allowing granulation of the exposed bowel;

• if fistula closure is impossible, allow fistula to become chronic and controlled.

The goal should be to control and limit the acute phase of care. Fistula takedown and abdominal

wall reconstruction should be delayed optimally by at least 8 – 12 months, in order to allow

loosening of the visceral adhesions, and should be performed only when the patient is well

nourished and satisfactory physiological homeostasis has been achieved. Different methods for

definitive fistula takedown and bowel reconstruction are currently described in the literature[16-20].

Techniques for fistula effluent diversion

A wide spectrum of treatment options for effluent diversion has been described in the literature and

is often used at the discretion of the attending surgeon, as a universally accepted management

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTalgorithm does not exist. The following techniques can be individualized to patients depending on

their particular characteristics.

1. Negative Pressure Wound Therapy (NPWT)– The use of negative pressure in the open

abdomen with hand-made dressings is well described and some authors extended this

technique to cases of EAF. This method is the easiest to apply and represents the foundation

for many of the subsequent techniques.

2. Biological dressing / Fibrin Glue or Cyanoacrylates / Primary Suture[21,22]–The application

of biological dressings (ie, Human Acellular Dermal Matrix or Cadaveric Split-thickness

Skin Graft) has been described with good results, particularly in case of a small size fistula

or deserosalization. Fibrin glue can improve the outcome of biological dressings, helping to

keep them in place and effectively sealing the fistula. Cyanoacrylates may be beneficial for

small EAFs, especially as an adjunct to primary suturing.[23]

3. Floating Stoma[24] –This is made with a plastic silo placed over the open abdomen, where

openings are created to best fit the fistula opening; the edge of these openings are then

sutured to the margins of the opening in the enteric loops with a continuous polypropylene

suture. A stoma bag is then placed over this “floating stoma” in order to collect the enteric

effluent.

4. Fistula VAC[25] –This method is easy and quick to apply, consisting of a single sheet of

Xeroform dressing placed to protect the wound bed, with a hole cut for the enteric opening.

The VAC sponge is accurately tailored to fit the wound and to match the fistula opening,

and is placed over the Xeroform layer. An adherent polyurethane drape is eventually placed

over the sponge and a hole is cut in this drape directly over the fistula, to allow placement of

the appliance of an ostomy bag for collecting the fistula effluent. A number of refinements

to this technique are described in literature, all attempting to obtain a better sealing of fistula

area from the rest of the wound[26-28].

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT5. Tube VAC[29] –The EAF is intubated with Malecot catheters and the surface of the open

abdomen is covered with petroleum-impregnated gauge. The wound is then covered with a

polyurethane sponge and the Malecot catheters are pulled through it. The entire device is

then sealed with an occlusive dressing and a continuous negative pressure is applied.

6. Baby Bottle Nipple VAC[30-38] –A baby bottle rubber nipple is placed over the fistula opening

and a Pezzer tube, Malecot or Foley catheter is placed through a small hole cut into the tip

of the nipple. A round layer of colostomy paste can be placed under the nipple in order to

ensure a better seal. A petroleum jelly-impregnated gauge or clear Telfa sheet is then placed

over the bowel and the entire wound is covered by a commercial VAC dressing. A small

hole is shaped into the VAC sponge, to hold the nipple in place.

7. VAC Chimney[31]– A chimney is created with a white sponge dressing, with the base of this

construct covering the adjacent healthy tissue for an area of 2-3 cm, to avoid leakage of

bowel effluent. A plastic tube, shorter than the chimney, is then placed into the chimney

itself, to avoid its collapse. The open abdomen is then covered with a conventional VAC

dressing, placing the connector to the pump directly over the chimney.

8. Fistula Patch[32]– A soft, flexible gel lamellar is shaped into a round form. Sutures are then

placed through the center of this circle and fixed to a drain tube placed across the fistula,

outside of the bowel. The silica gel lamellar is then folded and pushed into the fistula. Once

inside the bowel it will unwrap, acting like a patch inside the bowel. The fistula patch would

then remain in situ until the definitive fistula takedown operation. Some authors have

recently described a similar technique using a patch of hydrophilic polyvinyl alcohol foam

to cover small fistulas, combined with NPWT used on the remaining abdominal surface4.

9. Fistula Plug[33] – Thisisa plug designed to seal the EAF from inside consisting ofa circular

disk of 1mm thick silicone with a diameter of 2-5 cm. A Vicryl suture is passed through the

center of the silicon circle and then tied to a rubber band, with a bridge of foam-covered

aluminium. The silicone plug is then rolled and inserted into the fistula hole, the plug is

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPThung on the bridge by means of the suspension suture and rubber band. The bridge is fixed

to the abdominal wall using a self-adhesive plaster. Eventually, when the fistula is closed,

the tread will be cut off and the silicon plug will be discharged at the time of defecation.

10. Pedicled flap– Different types of flaps have been described in the literature[18,34-36]. Although

feasible and possibly effective, they are highly technically demanding and require a well-

stabilized patient, without ongoing sepsis or physiological derangements. For these reasons,

flaps remain a good option for definitive management, after the acute phase has resolved.

11. Suspension / Stoma Conversion[36,37] – In selected cases, when the fistula is located near to

the edge of the abdominal wound, the fistula edges can be gently mobilized and tacked to

the lateral aspect of the dermis. This allows the fistula to maturate into a definitive ostomy,

while negative pressure is applied to the granulating open abdomen, which is now protected

from bowel effluent.

The pros and cons, estimated cost and technical difficulty of every approach are summarized in

Table 1. Costs have been estimated, because the prices of every device and surgical item vary

widely between different countries and hospitals. A NPWT device rental price of 997,50 € was

utilized for the estimates.

Baby Bottle Nipple Diversion – Description of the technique

We have developed a refinement[38] of the technique previously described by Layton[30] et al.,

improving what we have found to be the practical downsides of this interesting management option

for an EAF. This has been named the baby bottle nipple diversion. First, the fistula is exteriorized

as much as possible outside the abdominal cavity, thus creating a flat surface that will allow optimal

nipple placement. At this time, irrigation and exploration of the abdominal cavity should be

performed, to reduce peritoneal contamination and decrease ongoing sepsis.

After these preliminary steps, a baby bottle nipple of soft silicone is placed over the fistula

(Video 1). The tip of the nipple is cut and a small size Pezzer tube is pulled through the hole of the

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTrubber nipple and secured in place by non-absorbable stitch tied around the tip of the nipple (Video

1). In order to achieve a better seal, an underlying layer of colostomy paste (Figure 1) can be placed

around the fistula and under the nipple, which is then surrounded by a jelly gauze, covering the

bowel and excluding the area of the fistula. We have modified Layton’s technique to further

optimize the external diversion of the fistula and minimize nipple dislocation by inserting a small

(10 Fr) Foley catheter (Figure 2) into the fistula opening and inflating the balloon gently with 2 to 3

cc of water. The Foley catheter is then passed through and secured to the silicone nipple neck.

Finally a NPWT dressing system is placed over the exposed viscera to promote surface

granulation (Video 1: Renasys™, Smith & Nephew Corporate). The plastic drape is placed over the

open abdomen and the nipple–tube complex is passed through a hole shaped within the drape and

the sponge. The plastic drape is then used to cover the sponge and shaped around the Pezzer tube.

The negative pressure is used to compress the dressing, hold the nipple in place and sealoff the

fistula (see Figure 3)

The baby bottle nipple diversion has the advantage of being effective even in cases of large, high

output fistulae and, unlike other techniques, can also be effective in case of multiple fistulae.

Clinical validation, patient 1

In October 2010 we applied this novel approach to the case of a 46-yearold male admitted for

complete adhesive small bowel obstruction and strangulation of a parastomal hernia, who

developed on postoperative day 7 two ileal perforations within a completely frozen abdomen.

These were initially treated with fibrin glue and an open abdomen with Bogotá Bag. After 6 days of

NPWT therapy, a leak from one of the previously repaired perforations appeared. A further attempt

at direct repair with Vicryl stitches and fibrin glue was not successful and the baby bottle nipple

diversion was utilized.

After three months of treatment with NPWT and baby bottle nipple diversion, the fistula

output was diverted, effectively functioning as an ileostomy. The abdominal surface was almost

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTcompletely covered by skin and subcutaneous tissue granulating from the edges of the abdominal

wall (Video 2 – CT scan); the patient was therefore discharged to a rehabilitation facility. After

three year follow-up, (Figure 4a, Figure 4b, Figure 4c) the patient has returned to a normal active

life with good quality of life. The previous EAF is functioning well as a natural ileostomy. The

open abdomen has evolved into a large herniation of intra-abdominal organs and only a thin skin

layer is covering the viscera (Figure 5 – three years follow-up CT scan).

Clinical validation, patient 2

In June 2013 we had to face a similar surgical nightmare: a 73-yearold male was admitted for acute

bleeding from multiple angiodysplastic foci and underwent emergent right colonic resection.

On POD #7 the hospital course was complicated by evisceration and ileal non-anastomotic

perforation. This was treated with ileal resection and primary anastomosis. Later on, the patient

underwent two subsequent re-laparotomies for evisceration and small bowel perforations. This was

further complicated by a massive abdominal wall necrotizing fascitis, which caused a great loss of

tissue within the abdominal wall and required NPWT dressing.

After 32 days and 14 NPWT dressing changes, two ileal EAF appeared within a few centimeters

from each other. At this point the abdomen was frozen and completely inaccessible, due to heavy

matted adhesions. Based on the satisfactory results of our previous experience, the Baby bottle

nipple diversion technique was utilized, even in the presence of these multiple fistulas (Video 3).

Two separate baby bottle nipples were placed over the fistulae, allowing effective and separate

isolation of the two different EAFs (Figure 3). After four dressing changes, both fistulae were well

controlled and both their size and output had strongly decreased. The surrounding viscera were

covered with Renasys™ dressing, (Smith & Nephew Corporate) and treated with low pressure (-60

to -80 mmHg) NPWT. In the absence of significant enteric contamination, a clean granulation bed

was formed.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTOn POD #65, the patient died of acute pulmonary edema, secondary to myocardial infarction. At

this stage, the fistulae were well controlled and the abdomen was clean and rapidly granulating

(Figure 6).

Our proposal

The management of an EAF is extremely challenging and should be tailored to the specific

characteristics of each case. We believe that a systematic approach to the management of this

devastating problem is required. Therefore, after performing a comprehensive review of the

literature and analyzing the different options for EAF management, we have developed a flowchart,

merging all the techniques described to date and including our experience and techniques (Figure

7). This is one rational, practical method for approaching this surgical challenge.

The acute management of an EAF should aim to: 1) divert the fistula output, thus preventing the

spreading of intestinal fluid into the peritoneal cavity; 2) collect and quantify the fistula output, in

order to address and correct fluids and electrolyte imbalances; 3) protect and keep clean the

surrounding granulating viscera.

A thorough classification of EAF is essential for choosing the correct approach. According to the

literature, an EAF can be classified based on anatomic location (proximal/distal); fistula output

volume (low/moderate/high) and location in the open abdomen (superficial/deep) [10,16,39](Table 2).

It is well known that distal and low-output fistulas have a higher spontaneous closure rate than

those that are more proximal and with high-output. For this reason, in the case of de-serozalisation

and distal, small or low output fistulas, a first attempt at primary closure with sutures and sealants

(fibrin glue, cyanoacrylates)[21,23] should be made. If this first attempt fails, the fistula must be

approached in the same manner as a proximal, large or moderate/high output fistulae.

The latter fistulae type is very unlikely to close with a glue/suture approach. The presence of

ongoing peritonitis should be the factor that drives the choice of the management technique.

Biologic dressings and flaps or skin grafting should be avoided in these cases with ongoing

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTuncontrolled peritonitis, because of the high probability of failure (ie, graft rejection, flap necrosis

and lack of sepsis control).

If there is ongoing peritonitis, the primary aim of treatment should be the diversion of fistula

output, allowing improvement of the physiological condition of the patient.

In the most recent review on the management of EAF4, Tavusbay et al. proposed different

techniques according to the size/output of the fistula. Small fistulae are first covered with a patch of

hydrophilic polyvinyl alcohol foam and the entire abdominal wound is then covered with

polyurethane (PU) foam to seal the OA, preventing further spillage of the enteric contents. The

foam is covered with an adhesive drape and continuous negative pressure is applied. In case of large

fistulae with protruding mucosa, Tavusbay et al. suggest that the OA is covered with a perforated

polyethylene sheet, cutting a hole in the PU foam to match the fistula mouth, and the PU foam is

placed onto the polyethylene sheet.

Patients with multiple fistulae provide an even more challenging situation. Most of the

techniques described in the literature are complex and time consuming and often require a wide free

surface to properly position the diverting device. In contrast, the Baby Bottle Nipple Diversion is

quick and easy to apply, low in cost and requires very little space, thus allowing its application in

cases of multiple fistulae, whether they are far apart or in close proximity to each other.

In our experience, the Baby Bottle Nipple Diversion approach offers the best chance of

diverting the fistula output where there are multiple fistulae, either in close proximity or when

located far from each other. This allows for effective control of ongoing sepsis and promotes clean

granulation formation over the exposed bowel. It also allows exact quantification of the fistula

output and assessment of its characteristics. Table 3 provides the appropriate take home messages.

CONCLUSIONS

To the best of our knowledge, an algorithm for EAF management has never been proposed or

validated. The ever expanding variety of management techniques, as well as the increasing number

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTof review papers[16,40]] attempting to summarize them, mirror the great confusion that accompanies

the finding of an EAF and reflects the need for a rationale approach to EAF management.

The proposed flowchart in Figure 7, is derived from our surgical experience in managing the

open abdomen and EAF and from an extensive review of the available literature. Given the rarity of

this condition and the low likelihood of a controlled trial, our conclusions are not “evidence based”.

Rather, they are based on small case series and expert opinion. Because of the peculiarity of this

field, and the absolute lack of management guidelines, the opinion of expert surgeons is a major

resource in EAF treatment.

This flowchart may be an invaluable tool for those surgeons facing this surgical challenge.

Furthermore, the standardization of the approach to an EAF can lead to a better understanding of

the available literature. Speaking the same language all over the world is the first step in creating a

rational approach to EAF management and, eventually, the promulgation of internationally accepted

guidelines.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTREFERENCES

1. Becker HP, Willms A, Schwab R. Small bowel fistulas and the open abdomen. Scand J Surg

2007;96:263-71.

2. Bradley MJ, Dubose JJ, Scalea TM, et al. Independent predictors of enteric fistula and

abdominal sepsis after damage control laparotomy: results from the prospective AAST Open

Abdomen registry. JAMA surgery 2013;148:947-54.

3. Bruhin A, Ferreira F, Chariker M, et al. Systematic review and evidence based

recommendations for the use of Negative Pressure Wound Therapy in the open abdomen.

Int J Surg 2014;12:1105-14.

4. Navsaria P, Nicol A, Hudson D, et al. Negative pressure wound therapy management of the

"open abdomen" following trauma: a prospective study and systematic review. World J

Emerg Surg 2013;8:4.

5. Navsaria PH, Bunting M, Omoshoro-Jones J, et al. Temporary closure of open abdominal

wounds by the modified sandwich-vacuum pack technique. Br J Surg 2003;90:718-22.

6. Kirshtein B, Mizrahi S. Vacuum-assisted management of enteroatmospheric fistula within

the open abdomen. Am Surg 2014;80:209-10.

7. Tavusbay C, Genc H, Cin N, et al. Use of a vacuum-assisted closure system for the

management of enteroatmospheric fistulae. Surg Today 2014.

8. Schecter WP. Management of Enterocutaneous Fistulas. Surg Clin North Am 2011;91:481-

91.

9. Davis KG, Johnson EK. Controversies in the Care of the Enterocutaneous Fistula. Surg Clin

North Am 2013;93:231-50.

10. Evenson A, Fischer J. Current Management of Enterocutaneous Fistula. Journal of

gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract

2006;10:455-64.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT11. Coughlin S, Roth L, Lurati G, Faulhaber M. Somatostatin analogues for the treatment of

enterocutaneous fistulas: a systematic review and meta-analysis. World J Surg

2012;36:1016-29.

12. Stevens P, Foulkes RE, Hartford-Beynon JS, Delicata RJ. Systematic review and meta-

analysis of the role of somatostatin and its analogues in the treatment of enterocutaneous

fistula. European journal of gastroenterology & hepatology 2011;23:912-22.

13. Rahbour G, Siddiqui MR, Ullah MR, et al. A meta-analysis of outcomes following use of

somatostatin and its analogues for the management of enterocutaneous fistulas. Ann Surg

2012;256:946-54.

14. Lloyd DA, Gabe SM, Windsor AC. Nutrition and management of enterocutaneous fistula.

Br J Surg 2006;93:1045-55.

15. Nightingale JM. The medical management of intestinal failure: methods to reduce the

severity. The Proceedings of the Nutrition Society 2003;62:703-10.

16. Marinis A, Gkiokas G, Argyra E, et al. "Enteroatmospheric fistulae" - gastrointestinal

openings in the open abdomen: a review and recent proposal of a surgical technique. Scand

J Surg 2013;102:61-8.

17. Krpata DM, Stein SL, Eston M, et al. Outcomes of simultaneous large complex abdominal

wall reconstruction and enterocutaneous fistula takedown. AJS 2013;205:354-9.

18. Wind J, van Koperen PJ, Slors JFM, Bemelman WA. Single-stage closure of

enterocutaneous fistula and stomas in the presence of large abdominal wall defects using the

components separation technique. AJS 2009;197:24-9.

19. Dionigi G, Dionigi R, Rovera F, et al. Treatment of high output entero-cutaneous fistulae

associated with large abdominal wall defects: Single center experience. International

Journal of Surgery 2008;6:51-6.

20. Demetriades D. A Technique of Surgical Closure of Complex Intestinal Fistulae in the Open

Abdomen. The Journal of Trauma: Injury, Infection, and Critical Care 2003;55:999-1001.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT21. Girard S, Sideman M, Spain DA. A novel approach to the problem of intestinal fistulization

arising in patients managed with open peritoneal cavities. AJS 2002;184:166-7.

22. Jamshidi R, Schecter WP. Biological dressings for the management of enteric fistulas in the

open abdomen: a preliminary report. Arch Surg 2007;142:793-6.

23. Bhat YM, Banerjee S, Barth BA, et al. Tissue adhesives: cyanoacrylate glue and fibrin

sealant. Gastrointestinal Endoscopy 2013;78:209-15.

24. Subramaniam MH, Liscum KR, Hirshberg A. The floating stoma: a new technique for

controlling exposed fistulae in abdominal trauma. The Journal of Trauma: Injury, Infection,

and Critical Care 2002;53:386-8.

25. Goverman J, Yelon JA, Platz JJ, et al. The "Fistula VAC," a Technique for Management of

Enterocutaneous Fistulae Arising within the Open Abdomen: Report of 5 Cases. The

Journal of Trauma: Injury, Infection, and Critical Care 2006;60:428-31.

26. D'Hondt M, Devriendt D, Van Rooy F, et al. Treatment of small-bowel fistulae in the open

abdomen with topical negative-pressure therapy. AJS 2011;202:e20-e4.

27. Piazza R, Armstrong S, Vanderkolk W, et al. A Modified “Fistula-VAC” Technique:

Management of Multiple Enterocutaneous Fistulas in the Open Abdomen. Plastic and

Reconstructive Surgery 2009;124:453e-5e.

28. Pang TC, Morton J, Pincott S. Novel technique for isolating and dressing enteroatmospheric

fistulae. ANZ J Surg 2012;82:747-9.

29. Al-Khoury G, Kaufman D, Hirshberg A. Improved Control of Exposed Fistula in the Open

Abdomen. Journal of the American College of Surgeons 2008;206:397-8.

30. Layton B, DuBose J, Nichols S, et al. Pacifying the open abdomen with concomitant

intestinal fistula: a novel approach. Am J Surg 2010;199:e48-e50.

31. Rekstad LC, Wasmuth HH, Ystgaard B, et al. Topical negative-pressure therapy for small

bowel leakage in a frozen abdomen. Journal of Trauma and Acute Care Surgery

2013;75:487-91.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT32. Wang G, Ren J, Liu S, et al.. ‘‘Fistula patch’’: Making the treatment of enteroatmospheric

fistulae in the open abdomen easier. Journal of Trauma and Acute Care Surgery

2013;74:1175-7.

33. Ozer MT, Sinan H, Zeybek N, Peker Y. A simple novel technique for enteroatmospheric

fistulae: silicone fistula plug. International wound journal 2014;11 Suppl 1:22-4.

34. Sriussadaporn S, Sriussadaporn S, Kritayakirana K, Pak-art R. Operative Management of

Small Bowel Fistulae Associated with Open Abdomen. Asian Journal of Surgery 2006;29:1-

7.

35. Stephensen B, Brown J, Lambrianides A. A novel method for managing enterocutaneous

fistulae in the open abdomen using a pedicle flap. J Surg Case Rep 2012;2012:5.

36. Al-Nahawi M. A rare approach to entero-atmospheric fistula. In: Am J Case Rep; 2013:476-

80.

37. Clark CE, Walker T, Bacon L, et al. Staged suspension of an enteroatmospheric fistula: a

novel surgical approach. Am Surg 2013;79:E139-40.

38. Di Saverio S, Villani S, Biscardi A, et al. Open Abdomen With Concomitant

Enteroatmospheric Fistula: Validation, Refinements, and Adjuncts to a Novel Approach.

The Journal of Trauma: Injury, Infection, and Critical Care 2011;71:760-2.

39. Schecter WP, HIRSHBERG A, Chang DS, et al. Enteric Fistulas: Principles of

Management. ACS 2009;209:484-91.

40. Terzi C, Egeli T, Canda AE, Arslan NC. Management of enteroatmospheric fistulae.

International wound journal 2014;11 Suppl 1:17-21.

.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1. Fistula Management Techniques

Technique Technique description PROs CONs Estimated cost

of the equipment

Technical difficulty

Baby Bottle Nipple Diversion

A baby bottle nipple is placed over the fistula with a ring of colostomy paste. A Foley catheter is placed inside the fistula hole and pulled through the nipple tip. Everything is then kept in place with a commercial NPWT dressing.

- Feasible even in case of multiple fistulae

- Quick and easy to apply

- Not applicable in case of fistulae larger than the teat $$ Low

NPWT A simple, hand-made NPWT dressing is placed over the open abdomen.

- Quickest and easiest to apply

- Atraumatic

- Does not provide effective fistula effluent diversion

$$ Low

Baby Bottle Nipple VAC

A baby bottle nipple is placed over the fistula with a ring of colostomy paste. A Malecot catheter is pulled through the nipple tip. Everything is then kept in place with a commercial NPWT dressing.

- Feasible even in case of multiple fistulae

- Quick and easy to apply

- Teat easily displaced during nursing maneuvers $$ Low

Floating Stoma A plastic silo placed over the open abdomen and openings are created to fit the fistula. The edges of these openings are sutured to the margins of the fistula with a continuous polypropylene suture. A stoma bag is then placed over this “floating stoma”.

- Adjustable to the dimension of the fistula

- Good seal

- Very traumatic (running sutures can lacerate bowel wall)

$ High

Tube VAC Fistula is intubated with a Malecot catheter and the surface of the open abdomen is covered with petroleum-impregnated gauge. The wound is then covered with a polyurethane sponge and the Malecot catheters are pulled through it. Everything is then kept in place with a commercial NPWT dressing.

- Feasible even in case of multiple fistulae

- Quick and easy to apply

- Risk of enlargement of fistula (there is no stop to catheter movements)

- Not applicable to fistulae larger than the catheter

$$ Low

Fistula VAC A sheet of Xeroform dressing is placed to protect the wound bed, with a hole cut for the enteric opening. The VAC sponge is placed over the Xeroform layer. An adherent polyurethane drape is eventually placed over the sponge and a hole is cut in this drape directly over the fistula, to allow

- Quick and easy to apply

- Atraumatic

- Risk of pooling of enteric contents under sponge

- Large area around the fistula exposed to bowel content

$$ Low

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

placement of the appliance of an ostomy bag for collecting the fistula effluent.

VAC Chimney A chimney is created with a white sponge dressing. A plastic tube, shorter than the chimney, is then placed into the chimney itself, to avoid its collapse. Everything is then kept in place with a commercial NPWT dressing, placing the connector to the pump directly over the chimney.

- Allows progressive abdominal wall closure

- Application of aspiration directly over the fistula

- Risk of pooling of enteric contents under the sponge

- Complex tand time consuming

$$ High

Biological Dressing Human Acellular Dermal Matrix or Cadaveric Split-thickness Skin Graft) are applied over small fistulae or deserosalization. Fibrin glue can improve the outcome, keeping them in place and effectively sealing the fistula. Cyanoacrylates may be beneficial as an adjunct to primary suturing.

- When successful provides immediate closure of the fistula

- Expensive - Not applicable if

ongoing peritonitis/sepsis $$$ High

Primary Suture A usual suture is used to seal the fistula. Adjunct of Cyanoacrylates and/or fibrin glue can improve outcome.

- When successful provides immediate closure of the fistula

- Quick

- Traumatic - Risk of

enlargement of fistula

$ Medium

Cyanoacrylates – Fibrin Glue

Fibrin glue can improve the outcome in adjunct to biological dressings or primary suture. Cyanoacrylates may be beneficial as an adjunct to primary suturing.

- When successful provides immediate closure of the fistula

- Atraumatic - Applicable as

adjunct to primary suture or biological dressings

- Not applicable to large fistulae

- Often fails to close the fistula

$ Low

Pedicle flaps Different types of pedicle flaps are described in literature. They should be designed and adapted according to the specific anatomy of every single case.

- When successful allows quick reconstruction of abdominal wall

- Not applicable if ongoing peritonitis/sepsis

- Technically demanding

$ Very High

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Legend of costs ranges: $=low cost range (approximately <250 USD), $$=intermediate cost range (approx 250-1500 USD), $$$=high cost range (approx >1500 USD)

Fistula Plug Vicryl suture is passed through the center of a

circular silicone layer and then tied to a rubber band, with a bridge of foam-covered aluminium. The silicone plug is then rolled and inserted into the fistula hole and the plug is hung on the bridge.The bridge is fixed to the abdominal wall. When the fistula is closed, the tread is cut off and the silicon plug is discharged at the time of defecation.

- Tight sealing of fistula output

- Promotes fistula closure

- Complex - Difficult nursing

care - Risk of bowel

obstruction $ High

Fistula Patch Sutures are placed through the center of a gel lamellar circle and are fixed to a drain tube placed across the fistula outside the bowel. The gel lamellar is then folded and pushed into the fistula, where it will unwrap, acting like a patch inside the bowel. The fistula patch remains in situ until the definitive fistula takedown operation.

- Tight sealing of fistula output

- Not tested in case of ongoing peritonitis, nor in case of multiple fistulae

- Must remain in situ until definitive fistula takedown

$ High

Fistula Suspension Fistula edges are gently mobilized and tacked to the lateral aspect of the dermis. This allows the fistula to maturate into an ostomy. Negative pressure is applied to the granulating open abdomen.

- Artificial creation of a fistula tract

- Allows reconstruction of abdominal wall

- Applicable only if adjacent well vascularized dermis

- Traumatic to fistula edges

$ High

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 2. EAF Classification

Anatomic location Proximal Stomach, duodenum, jejunum, proximal ileus

Distal Distal ileus, colon

Output volume Low < 200 mL/24h

Moderate 200 – 500 mL/24h

High > 500 mL/24h

Location inside the open abdomen

Superficial Drains on top of a granulating abdominal wound

Deep Drains intestinal content inside the peritoneal cavity

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 3. Take Home Messages 1. An Enteroatmospheric fistula (EAF) is a major surgical complication that offers an extremely challenging range of critical care burdens including the full spectrum of surgical, metabolic, nutritional and nursing issues. 2. Because spontaneous closure is extremely rare, the acute fistula management should aim to completely divert fistula output, protect the surrounding viscera and allow the clean granulation of exposed bowel, thus promoting the fistula to become chronic and controlled; fistula takedown and abdominal wall reconstruction should be delayed optimally by at least 8 – 12 months. 3. Significant heterogeneity in the surgical techniques described for EAF management remains and there is no single approach that has proven to be ideal in every circumstance: a rationale approach to EAF management is needed. 4. The Baby Bottle Nipple Diversion approach offers the best chance of diverting the fistula output in case of multiple fistulae, either when located in close proximity or far from each other.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFigure Legends

Figure 1. From upper-left to lower-right, (A)(B) application of colostomy paste underlining the

nipple, (C) the complex nipple tube is held over the fistula hole with the aid of the paste layer, (D)

the NWPT drape is positioned and the nipple tube is extended through the exterior of the drape.

Figure 2. From upper-left to lower-right; (A) the fistula hole is visible over the frozen viscera, (B)

the baby nipple is cut and appropriately shaped to allow the passage of a Pezzer tube or a Foley

catheter, (C) the Foley catheter may be inserted into the fistula hole through the nipple tip and the

nipple itself is positioned around the fistula to drain the fistula leakage, (D) final result of the

exterior isolated EAF after negative pressure application over the remaining abdominal viscera.

Figure 3. Patient 2, Complete Baby Bottle Nipple Diversion dressing for multiple fistulas.

Figure 4 Patient 1, at follow up three years later, (A) frontal, (B) angled, and (C) lateral views.

Figure 5. Patient 1; three years follow-up, CT scan showing the evolution of the abdomen and the

large herniation of intra-abdominal organs. Only a thin skin layer is covering the viscera.

Figure 6. Patient 2; intraoperative appearance demonstrating good granulation over the exposed

bowel and an optimum diversion of bowel content.

Figure 7. Flowchart summarizing the various options available for EAF treatment.

Video Legends

Video 1. Patient 2, Baby Bottle Nipple Diversion technique. Video 2. Patient 1, CT Video 3. Patient 2, Baby Bottle Nipple Diversion technique, a strategy for multiple fistulas

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT