ou

A. Bruhin a, F. Ferreira b, M. Chara Department of Trauma and Visceral Surgery, Luzern, Sb Hospital Pedro Hispano, Matosinhos-Porto, Portugalc Aesthetic Plastic Surgery Institute, Louisville, KY, USA

ospital,

managicommeopen are NPW

Article history: Introduction: Negative Pressure Wound Therapy (NPWT) is widely used in the management of the open

patients but NPWTates were consistentenced by the choiceing eleven specicbdomen. DiscussionNPWT kits with orto tailor its use in a

er the CC BY-NC-NDnses/by-nc-nd/3.0/).

Maintaining an open abdomen by means of temporaryabdominal closure (TAC) is a valuable surgical technique in the

conditions including trauma, damage control, sepsis and re-laparotomy [1]. There is a great degree of heterogeneity in thepatient population and the surgical methods applied regardingmethod of TAC, timing and method of closure. Over the last decade,Negative Pressure Wound Therapy (NPWT) has been recognised asa valid method of TAC [2,3]. Although evidence for NPWT is

* Corresponding author. 101 Hessle Road, Hull HU32BN, UK.

Contents lists available at ScienceDirect

International Jou

ww

International Journal of Surgery 12 (2014) 1105e1114E-mail address: jennifer.smith@smith-nephew.com (J. Smith).1. Introduction management of a wide range of complex abdominal injuries andTemporary abdominal closureEvidence-based recommendationsSystematic review

with septic complications achieved a lower rate of fascial closure than non-septicwith dynamic closure remained the best option to achieve fascial closure. Mortality rand seemed to be related to the underlying medical condition rather than being inuof dressing, Treatment goals for open abdomen were dened prior to developevidence-based recommendations suitable for different stages and grades of open aand conclusion: The most efcient temporary abdominal closure techniques arewithout a dynamic closure procedure. Evidence-based recommendations will helpcomplex treatment pathway for the individual patient. 2014 The Authors. Published by Elsevier Ltd. This is an open access article und

license (http://creativecommons.org/liceKeywords:Negative Pressure Wound Therapy (NPWT)Open abdomen

International Expert Panel and graded according to the quality of supporting evidence. Results: Pooledresults, in non-septic patients showed a 72% fascial closure rate following use of commercial NPWT kitsin the open abdomen. This increased to 82% by the addition of a dynamic closure method. Slightly lowerrates were showed with use of Wittmann Patch (68%) and home-made NPWT (vac-pack) (58%). PatientsReceived 12 June 2014Received in revised form11 August 2014Accepted 19 August 2014Available online 28 August 2014

abdomen despite uncertainty regarding several aspects of usage. An expert panel was convened todevelop evidence-based recommendations describing the use of NPWT in the open abdomen. Methods:A systematic review was carried out to investigate the efcacy of a range of Temporary AbdominalClosure methods including variants of NPWT. Evidence-based recommendations were developed by and Smith & Nephew, Hull, UKe Department of General Surgery, Black Forest Hf University of Freiburg, Germany

h i g h l i g h t s

NPWT is a widely adopted method ofWe present the rst evidence-based re Recommendations divided by grade of Published literature analysed to compa

a r t i c l e i n f ohttp://dx.doi.org/10.1016/j.ijsu.2014.08.3961743-9191/ 2014 The Authors. Published by Elseviernd Therapy in the open abdomen

iker c, J. Smith d, *, N. Runkel e, f

witzerland

Villingen-Schwenningen, Germany

ng the OA but has a weak evidence base.ndations to describe the use of NPWT in OA.bdomen.T with other methods of TAC.

a b s t r a c tof Negative Pressure W

Systematic review and evidence based recommendations for the useOriginal research

journal homepage: wLtd. This is an open access article urnal of Surgery

. journal-surgery.netnder the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

previously published evidence-based recommendations for the useof NPWT in acute and chronic wound management [4e6]. A sys-

intra-abdominal adhesion, infection and stula) each with theirown clinical challenges and prognoses. The absence of wound

Abbreviations

NPWT Negative Pressure Wound TherapyOA open abdomenTAC temporary abdominal closureACS abdominal compartment syndromeEL level of evidenceSTSG split thickness skin grafts

A. Bruhin et al. / International Journal of Surgery 12 (2014) 1105e11141106classication in the majority of published literature to date, lessensthe ability to understand and contextualize the published literaturein relation to real-life clinical scenarios.

The aim of the present study was to develop evidence-basedrecommendations relevant to different grades of OA to clarify thetreatment goals, likelihood of complications, and treatment pro-tocols which may be most appropriate.

2. Methods

This study adopted a combination of a formal evidence-basedmedicine activity (systematic literature review) with a formaltematic review was carried out to evaluate the following researchquestion: for patients requiring open abdomen therapy, willtreatment with NPWT improve fascial closure rates, mortality andstula rates comparedwith other methods of temporary abdominalclosure. Result from the systematic review was used to createevidence-based recommendations relating to the use of NPWT inopen abdomen.

1.1. Classication of the open abdomen (OA)

Reporting of clinical outcomes in OA often takes no account ofthe heterogeneity of the patient populations included within in-dividual studies. A classication system was proposed in Bjorcket al. (2009) [7] (Table 1). The classication system recognizes thatwithin the umbrella clinical indication of OA there exists severalwell-dened sub-indications and wound characteristics (such asemerging at an increasing rate, a consensus on how and when touse NPWT on the open abdomen is lacking.

The international NPWT Expert Panel, 22 expert surgeons fromdifferent surgical disciplines, was established in 2009 and hasconsensus development program.

Table 1Classication of the open abdomen (OA). modied from Bjorck et al., 2009 [7].

Grade Description

1a without adherence between bowel and abdominal wall or xity1b Contaminated OA without adherence/xity2a Clean OA developing adherence/xity2b Contaminated OA developing adherence/xity3 OA complicated by stula formation4 withoutstula

Frozen OA with adherent/xed bowel; unable to close surgically;without stula

4 with stula Frozen OA with adherent/xed bowel; unable to close surgically;with stula2.1. Systematic review

A series of systematic searches were carried out according to thePRISMA guidelines [8]. Searches were limited to studies publishedafter 1996 (when modern formats of NPWT became commerciallyavailable). PubMed was searched using the following terms todescribe the open abdomen: [(open abdomen OR abdominalcompartment syndrome OR laparotomy) NOT review AND [one ofthe following]: (Negative Pressure Wound Therapy OR NPWT ORVacuum assisted OR VAC OR vac pack OR vacuum pack) NOTreview]; (bogota bag OR silo) NOT gastroschisis; (wittmann ORarticial burr). These searches were evaluated separately.Searches were updated in July 2013. Studies reporting patientsrequiring open abdomen therapy for any duration and for anyaetiology were included. Individual literature searches identiedstudies reporting outcomes on specic interventions. All papersreporting the intervention under evaluation were includedregardless of the quality or level of the study design. Duplicateswere removed. Papers reporting paediatric patients, in vivo studiesand case series of fewer than 6 patients were excluded. Opinionpieces or reviews containing no original datawere excluded. Papersdescribing clinical interventions other than those under investi-gation were excluded (labelled irrelevant clinical area; Table 2).Finally, papers reporting major modications to the generallyaccepted application techniques of the interventions under analysiswere excluded. All remaining studies were reviewed. Follow upbeyond fascial closure was not tracked. Papers were supplementedfrom review of bibliographies of the included papers if they weredeemed relevant and had not already been identied. A post hocmodication to the search strategy was to divide the NPWT papersinto different specic NPWT methods. This was due to a realisationthat the precise methods employed appeared to have an impact onoutcomes. Thus papers where NPWT was applied along with amethod of sequential closure, and home-made vac-pack styleNPWT methods were considered separately from commerciallyavailable kits. The decision was made to evaluate these data sepa-rately. Raw data was extracted from all the papers where possible(see Supplementary material) and organized in excel (MS ofce).Where individual data points were missing, the correspondingauthor of the paper was contacted by email. No additional data wasretrieved using this method. The main outcomes under investiga-tion were rates of fascial closure, mortality and stula and wereexpressed as a weighted mean (percentage) in order to minimisebias towards the larger studies. In some cases specic data points(in particular relating to mortality) were inferred from the rate ofsurviving patients and similarly the number of fascial closures wasoccasionally inferred from reporting of the planned ventral herniarate in the surviving population. The rate of fascial closure andstula ratewas therefore calculated on an intent-to-treat basis andnot as a percentage of survivors as commonly reported in theidentied papers. Only the abstracts were reviewed for studies inlanguages other than English and data was included where avail-able. No formal analysis of study bias of individual study designwascarried out because of the very high proportion of non comparativestudies identied (no randomization or concealment, high risk ofselective reporting in the majority of papers, little prospectivereporting of outcomes assessment etc). A high risk of bias wasassumed throughout. In order to evaluate whether outcomesdiffered according to aetiology, studies were broadly divided ac-cording to the presence or absence of sepsis into groups composedof non-septic abdominal aetiologies (ACS, trauma, abdominal aorticaneurism etc), septic aetiologies (diverticulitis, peritonitis,pancreatitis, contaminated abdomens and perforations) and paperswhere these aetiologies were mixed. A minimum of 3 data sets

Table 2Systematic review ow chart. Studies identied from the systematic review were reviewed manually. Those meeting the exclusion criteria were removed. Numbers of papersinclude those where the abstract only was reviewed (in parentheses). The NPWT systematic search identied papers describing basic NPWT as well as NPWT sequentialclosure. The papers describing sequential closure were separated from the papers describing basic NPWT.

NPWT kits Vac-pack NPWT sequential closure Bogota Bag Articial BurrPapers identied (systematic review) 170 31 21Papers identied from other sources 3 2 0

Reason for Exclusion Duplicates 2 0 0In vivo studies 12 1 0Paediatric 6 5 0Major technical modication 11 3 1Irrelevant clinical area 30 0 6Reviews/comments/letters 16 3 1Case studies (n < 6) 24 5 1

8 40 0

que

ted

A. Bruhin et al. / International Journal of Surgery 12 (2014) 1105e1114 1107were required within each category for pooling to be consideredappropriate.

2.2. Development of recommendations

The recommendations presented in this paper were determinedover a period of 2 years in a series of meetings between the NPWT-Expert Panel (NPWT-EP) members and a broader discussion in aninternational congress where feedback from clinical users wassought.

Recommendations were developed according to a modicationof the SIGN (Scottish Intercollegiate Guidelines Network) classi-cation system [9] (modication described in legend of Table 3).Table 3 describes the classication of the levels of evidence used(L1e4) and the corresponding strength of recommendation thatcan be made from each evidence level (Grade AeD). The expertpanel members were conscious of the potentials of fatal misuse ofNPWT in OA and thus have included some concerns of caution andtwo non-evidence-based recommendations based on good practicepoints (GPP) after extensive discussion within the panel and with a

No relevant outcomes/not available 15Paper not available 3

Papers with relevant data 54 (7)Papers reviewed 29a,b (5)

a One paper [40] contained data relevant to both standard NPWT and NPWT Sesupplemental digital content 1.

b One paper [10] contained information about NPWT kit and vac-pack and is counbroader group of wound experts.

3. Relative efcacy of different TAC methods

The systematic review revealed an evidence base rich in non-comparative retrospective studies but with a comparative lack ofgood quality randomized studies. Nevertheless, the number ofpapers reporting relevant outcomes was high and warranted

Table 3ATranslation of Levels of evidence to graded Recommendations. Adapted from the SIGN mclarify the strength of each evidence-based recommendation (Must for grade A, Should

Recommendation

Grade Terminology Description

A Must At least one meta-analysis, systematic review, or RCT rated as 1A body of evidence consisting principally of studies rated as 1, dirresults

B Should A body of evidence including studies rated as 2, directly applicextrapolated evidence from studies rated as 1 or 1

C May A body of evidence including studies rated as 2, directly applicaExtrapolated evidence from studies rated as 2

D Possible Evidence level 3 or 4; or extrapolated evidence from studies ratedGPP e Good practice point. A basic requirement of application consistenfurther evaluation. The number of relevant papers identied foreach TAC method is shown in Table 2. Rates of fascial closure,mortality and stulisation are shown in Table 4.

The highest rates of primary fascial closure were reportedfollowing use of commercially available NPWT kits in combinationwith a dynamic closure technique e.g. mesh-mediated traction,dynamic retention sutures or ABRA. These techniques all involveattachment of a medical device to the fascial edges and sequentialnarrowing of the wound opening through incremental approxi-mation of the fascial edges. In a non-septic population this treat-ment resulted in a pooled intent-to-treat closure rate of 81%followed by NPWT kits alone (72%), articial Burr (Wittmannpatch) (68%), home-made NPWT (vac-pack) (58%). This isconsistent with another systematic review [2]. There was insuf-cient data presented on non-septic patients treatedwith the BogotaBag to pool results.

It was hypothesized that rates of fascial closure would be lowerfor open abdomen patients with septic aetiologies than with non-septic aetiologies and this was true in all categories where suf-cient data was available for analysis. Addition of a method of dy-namic closure along with an NPWT kit resulted in the highest

8 (1) 8 (1)13b (5) 12a (0) 8 (1) 8 (1)

ntial closure and is counted in both columns. For full data extraction please refer to

in both columns.reported fascial closure rates (74.6%) followed by NPWT kits alone(48%), vac-pack (35%) and Bogota Bag (27%). There were insufcientnumbers of septic patients treated with the Articial Burr (Witt-mann Patch) to calculate pooled fascial closure rates. NPWT,especially in conjunction with a method of dynamic closure ap-pears to be the most suitable TAC method currently available in the

ethod of classication [9]. Modication was made by using specic terminology to for grade B, May for Grade C).

, and directly applicable to the target population; orectly applicable to the target population, and demonstrating overall consistency of

able to the target population, and demonstrating overall consistency of results; or

ble to the target population and demonstrating overall consistency of results; or

as 2t with manufacturers instructions and in agreement with the NPWT-EP.

for stula development. Insufcient information was available onBogota bag or Wittmann Patch in septic patients.

4. Evidence based recommendations

4.1. General treatment goals and recommendations

Table 5Treatment Goals for application of NPWT. Although NPWT is a treatment thatoperates simultaneously through multiple actions, in order that specic recom-mendations could be made and voted on, a single overarching treatment goal wasselected as part of the consensus process. Closed circles indicate the dominant goaland open circles indicate secondary goals within each indication. A cross indicates agoal which is NOT desired. Blank space indicates the goal is not relevant.

Goals Grade

1 2 3a 4withoutstulaa

4 withstulaa

Protection Protect bowel from damagea B B B B BTo splint the wounda B B B B B

Table 3BEvidence Level. Adapted from the SIGN method of classication (9).

Evidence level

Level Description

1 High quality meta-analyses, systematic reviews of RCTs, or RCTs with avery low risk of bias

1 Well-conducted meta-analyses, systematic reviews, or RCTs with a lowrisk of bias

1 Meta-analyses, systematic reviews, or RCTs with a high risk of bias2 High quality systematic reviews of case control or cohort or studies. High

quality case control or cohort studies with a very low risk of confoundingor bias and a high probability that the relationship is causal

2 Well-conducted case control or cohort studies with a low risk ofconfounding or bias and a moderate probability that the relationship iscausal

A. Bruhin et al. / International Journal of Surgery 12 (2014) 1105e11141108septic or contaminated abdomen. These conclusions are consistentwith another systematic review [2].

Mortality in the open abdomen is greatly inuenced by theunderlying medical condition. Mortality rates calculated from thepooled data ranged from 12% to 25% in a non-septic population vs.22%e40% in septic or mixed populations. Studies using vac-pack tomanage patients with septic aetiologies reported the highestoverall mortality rate, at 40%. This was higher than the mortalityrate for patients treated with NPWT kits at 26%. It was not possiblein the systematic review to match the studies in terms of the injuryseverity or other such scores relevant to risk of mortality and so norm conclusions can be drawn with regard to the inuence of TACmethod on mortality however, this difference has also beendemonstrated in a study comparing a new variant of NPWT kitbased dressing with vac-pack [10] and remains to be more thor-oughly investigated.

One past controversy, whether NPWT may be linked to thedevelopment of enterocutaneous stula [11,12], has now largelybeen disproved [13]. Carlson et al. (2013) [13] investigated theincidence of stula formation in NPWT treated open abdomenwounds compared with other methods of TAC in a nation-wideaudit carried out in the UK over an 18 month period. No signi-cant difference in the incidence of stula formation was observed

2 Case control or cohort studies with a high risk of confounding or bias anda signicant risk that the relationship is not causal

3 Non-analytic studies, e.g. case reports, case series, in vivo or in vitrostudies

4 Expert opinionregardless of the method of TAC employed. In another study [14]signicantly fewer stulae were observed in patients treated withNPWT compared with other methods of TAC. The data shown inTable 4 is consistent with these studies and does not show anyevidence of a relationship between use of NPWT and stulaformation.

Table 4Systematic review e relative safety and efcacy of different methods of TAC.

Device/method Condition of abdomen # Patients (#studies) F

NPWT kit Non-septica 337 (7) 7Septic/mixed 941 (22) 4

NPWT kit sequential closure Non-septica 188 (5) 8Septic/mixed 270 (7) 7

Vac-pack Non-septica 551 (8) 5Septic/mixed 238 (5) 3

Articial Burr Non-septic 155 (7) 6Bogota Bagc Mixed 396 (8) 2

a Representing studies composed almost entirely of either non-septic.b Insufcient data in this subset to calculate.c While papers could be divided into non-septic and septic cohorts, this would leave ins

a mixed aetiology.It is likely that a higher degree of stulisation occurs in septic OAcomparedwith non-septic OA (12.1% vs. 3.1% respectively for NPWTkit treated patients) (Table 4). It is uncertain whether the higherincidence of stula in the septic group results directly from thepresence of severe prolonged intra-abdominal sepsis or is related tothe longer duration of TAC observed under septic conditions. Fis-tulisation rates are known to increase with duration of openabdomen [15]and therefore it may be appropriate to label septic OAwounds, especially those undergoing protracted TAC as high-risk

To minimize formation of adhesions B B BTo support the early physiologicalrecovery of the patient

B B

Management To manage wound uid and oedemaa B B B B BTo prevent wound deterioration B B B BManage and divert the stula efuent C C

Closure To provide temporary wound coveruntil fascial closure is possible ordesired

C B B

Extend the window for primaryfascial closure

B C

Encourage the healing of the closedincision after denitive closure.

B B

To promote granulation tissueformation to create good wound bedfor grafting

x x x C B

a Denotes general treatment goals which relate to all grades of OA.NPWT acts via several different modes of action. Specically,NPWT can address several issues relating to the protection of theopen abdomen (e.g. against external damage and bacterial ingress),management of the open abdomen (e.g. uid handling pain man-agement, mobilization, skin protection) and closure of the openabdomen (e.g. extending the window for primary fascial closure orimproving graft take in abdomens closed by STSG). These specic

ascial closure % Mortality (%) Fistula (%) References

2.0 16.5 3.7 [10,18,26,64,77e79]7.5 26.0 12.1 [12e14,22,25,40,58,68e70,80e91]1.9 24.6 6.7 [31e33,35,92]4.6 22.8 5.5 [30,34,36e40]7.7 16.9 6.0 [10,19,20,93e97]5.2 40.5 eb [98e102]8.0 13.7 3.1 [43,44,79,97,103,104,105]6.6 30.3 6.6 [22,106e112]

ufcient data points (

rnalactions can all be converted into specic treatment goals (Table 5).Some treatment goals are relevant to the OA in general and someare relevant only to specic grades (as classied by Bjorck et al. [7]).This is described in more detail below where evidence-based rec-ommendations have been developed that are relevant regardless ofthe grade of OA (Table 6).

4.1.1. General recommendation 1: An interface layer must be usedto protect exposed organs and to avoid adhesions between thebowel and abdominal wall. Table 6; GPP

A key treatment goal for the use of NPWT in grade 1 and 2 OA isto prevent deterioration of the wound to the next grade. To mini-mise the development of abdominal xity (progression to Grade 4)use of a large, fenestrated non-adherent interface layer reducesadhesions between the exposed bowel and the inside of theabdominal wall thus retaining the option of fascial closure. Allcommercial NPWT devices contain a dedicated interface layer aspart of the kit. Application of a non-adherent interface layer canreduce the risk of stula formation (progression to grade 3) [16] asprevention of adhesions leads to lower risk of bowel damage. Theinterface layer should be placed as widely as possible inside theabdomen: in grade 1 and 2 OA, laterally into the paracolic gutters,cranially onto the diaphragm after taking down the falciformeligament and caudally into the pelvic cavity.

Caution: Failure to apply a non-adherent interface layer mayexpose the patient to a signicant risk of stula formation as aresult of potential damage to the bowel during dressing changes.

4.1.2. General recommendation 2: Use of a specialised openabdomen foam-based dressing kit should be used. Table 6; Grade B

Table 6General recommendations for the use of NPWT in OA. The numbered recommen-dations correspond with the order in the main body of the text.

Goal Recommendation and grade (AeD/GPP) References andevidence level (1e4)

To protectbowel fromdamage

1. An interface layer must be used toprotect exposed organs and (wherepossible) avoid adhesions betweenthe bowel and abdominal wall

GPP

To splint thewound

2. Use of a specialised open abdomenfoam-based dressing should be used

B L2: [10,21]L3: [17,18]

To managewound uid

3. NPWT may be used to manageabdominal wound uid

C L2: [22]L3: [23,29,64]

Treatmentvariables

4. Continuous NPWT settings of upto 80 mmHg are recommended

D L3: [28]

A. Bruhin et al. / International JouAll commercial kits contain polyurethane foam in a variety offormats. This foam acts in a similar way in all commercial productsand is able to compress under negative pressure which may lead toimproved preservation of abdominal domain by mediating con-stant medial traction of the abdominal wall [17,18]. Without thissplinting effect, lateral retraction of the abdominal wall over aperiod of days, may result in loss of domain, preventing re-approximation of the fascial edges. Surgical towels used as thewound ller in the vac-pac technique (an off-the-shelf method ofNPWT) [19,20] are known not to shrink under compression andtherefore have a limited ability to splint the wound. This may havean impact on the ability to achieve early fascial closure in damagecontrol indications: Hatch et al. (2011) (EL2) reported that use offoam-based NPWT, but not vac-pack based NPWT was an inde-pendent predictor of early fascial closure [21]. Furthermore pro-spective comparative studies comparing a commercial kit with vac-pack demonstrates signicant increase in fascial closure rate withthe commercial product [10] (EL2). This is in addition to the analysisshown in Table 5 which demonstrates improved outcomes instudies reporting commercial systems as opposed to vac-packdressings (described above). To date no studies have comparedthe clinical efcacy of different commercial NPWT open abdomenspecic products. There are no publications regarding the use ofgauze as a wound ller for TAC.

Caution: Preserve the integrity of the surrounding skin by ac-curate placement of the wound ller within the wound rather thanon top of the skin.

4.1.3. General recommendation 3: NPWT may be used to manageabdominal wound uid. Table 6; Grade C

One major advantage of NPWT over the other TAC techniques isits ability to actively drain uid as a result of the application ofnegative pressure (Table 5) reducing the pooling of sterile orcontaminated uid within the abdominal cavity. Signicantly moreuid can be evacuated from OA wounds treated with NPWT thanwith passive drainage (such as the Bogota bag) with volumes ofaround 800ml being frequently reported [22,23]. The better uidhandling provided by the NPWT dressing may reduce the fre-quency of analgesia-requiring wound dressing changes and canprevent complications secondary to lying in wet beds.

Another potential benet of NPWT is that all evacuated uid isdirected into a waste canister preventing contamination of theward and personal alike. Furthermore, the diversion of uid pro-tects the surrounding skin from maceration and frequent dressingchanges. Measurement of the uid in the canister is easily carriedout and may contribute to more rapid uid replacement andnutritional support. Quick identication of the uid quality such asblood or feces may be observed at an early stage through frequentobservation of the content of the canister.

In an animal model of open abdomen, NPWT was shown tosignicantly reduce intestinal oedema compared with passivedrainage [23]. Although this has not been conrmed clinically inthe open abdomen, an NPWT-mediated reduction in oedema hasbeen observed in other indications [24].

Maintaining the abdomen in a moist environment as a result ofthe NPWT top layer adhesive lm also prevents uncontrolledevaporative uid loss, secondary external bacterial contaminationand reduces heat loss.

4.1.4. General recommendation 4: Continuous NPWT settings of upto 80 mmHg are recommended. Table 6; Grade D

Negative pressures from 75 mmHg [25] to 175 mmHg [26]have been reported. No studies have investigated the most appro-priate level of negative pressure to apply to an open abdomen andinsufcient data was provided in the reviewed studies to compareoutcomes dependent on the level of pressure used (data notshown). Continuous pressure levels of less than e 80 mmHg arerecommended and as low as50mmHg in vulnerable patients (e.g.those with previous anastomosis or dilated small bowel). Theprevalent untested assumption is that the higher the pressurelevels, the higher the potential for damage to the underlying or-gans, hence the tendency towards caution and towards lowerpressures in this series of recommendations. In an animal study(L3) higher levels of NPWT resulted in a measurable reduction inbowel blood ow which correlated with increasing negative pres-sure [27]. Although the clinical impact of this observation is un-certain, this is a potential area for improvement. When choosingthe level of pressure, a balance must be found between potentialdamage to the underlying organs (supporting lower pressures) andthe effective removal of uid (pressures up to 120 mmHg lead togreater uid drainage) [28]. Adoption of a new commerciallyavailable NPWT device resulted in the delivery of no pressure to the

of Surgery 12 (2014) 1105e1114 1109bowel surface, despite effective removal of uid [29].

Intermittent pressure (where pressure settings uctuate be-tween on and off) or variable pressure (where pressure settinguctuate between high and low) are not recommended. Underthese pressure regimes, the ability of NPWT to splint the openabdomen wound is severely compromised.

4.2. Grade-specic goals, recommendations and cautions

In addition to the general treatment goals described above,some treatment goals are only relevant to specic grades of OA asdened by Bjorck et al. [7] (Table 5). These goals are used as thebasis for the specic recommendations described in this sectionshown in Table 7.

weremore likely to have beenmanagedwith NPWTcomparedwith

Table 7Grade-specic recommendations for use of NPWT in Grade 1e4 open abdomen.

OAgrade

Goal Recommendation and grade(AeD)

Supportingreference andevidence level (1e4)

Grade1and2

To provide temporarywound cover untilfascial closure ispossible or desired

5. In experiencedhands NPWT shouldbe used as a rst linetherapy in Grade 1Aand 2A openabdomen wherethere is an optionfor delayed primaryclosure, includingfollowing decom-pression of ACS

B L2: [21,22]L3: [23]

6. Application of asequential dynamicclosure technique,along with NPWT tocounteract retrac-tion and facilitatedelayed primaryclosure should beconsidered

B L1: [40]L2: [31,36]L3: [30e35,38]

To extend thewindow for primaryfascial closure

7.Always use a non-adherent interfacelayer to protectexposed organs and toprevent progression tograde 3 or 4.

GPP L2: [36]L3: [18,26,42]L4: [41]

To prevent woundprogression

Encourage healing ofthe closed incision

8. Application of inci-sional NPWT on theclosed incision tofacilitate healingshould beconsidered

B L1:[51,52]a

L2: [45,46,48],[53,54]a

L3: [47,49]

Grade To manage and divert 9. NPWT may be used D L3: [56e59,67]

split thickness skingraft.

A. Bruhin et al. / International Journal1110To splint the wound 11. NPWT should beused to enhancesplit skin graft takeat the abdomen.

B L1a: [71e73]L2a: [74e76]L3: [57,59,65]

a Denotes studies carried out on other indications (not open abdomen) whererelevant data have been extrapolated for relevance in the open abdomen.3 the stula efuent to manage output ofan entero-atmospheric stula.

Grade4

To promotegranulation tissueformation to creategood wound bed forgrafting

10. It is possible to useNPWT toencourage granu-lation tissue for-mation to supportsplit closure by

D L3:[12,19,68,69,78,91]GPP good practice point. The numbered recommendations correspond with theorder in the main body of the text.those requiring more prolonged NPWT (73% vs. 49% p < 0.001) andNPWTwas an independent predictor for this early closure [21]. OneEL3 study reported that 43% of trauma OA patients required only asingle episode of NPWT prior to fascial closure [19]. In a retro-spective comparative study signicantly fewer dressing changeswere required in the NPWTgroup compared with other methods ofTAC prior to fascial closure [14].

4.2.1.2. Grade 1 and 2 e recommendation 6: Application of asequential dynamic closure technique, along with NPWT to counteractretraction and facilitate delayed primary closure should be consid-ered. Table 7; Grade B. It is important that fascial closure be tensionfree in order to avoid ischaemia and subsequent necrosis of thefascial and to reduce the risk of ACS from developing. Sequentialdynamic primary closure involves applying small amounts of ten-sion to the fascia sequentially, at each repeat laparotomy in order toeventually reach the goal of primary fascial closure. The develop-ment of zips, clips, resorbable meshes, and Velcro analogues (e.g.the Wittmann patch), human acellular dermal matrix, or the use ofvarious dynamic suturing techniques facilitate this staged approachand can be initiated when patient is stabilised and intra-abdominalinfection/injury is under control. The rate of approximation isdependent on abdominal tension, which is a function of oedemaresolution and abdominal tension.

Application of dynamic suturing [30,31] mesh [32e37] and theABRA system [38,39] in conjunction with NPWT results in anoverall fascial closure rate of 79% (Table 4), the highest closure rateof any closure technique assessed in the systematic review. Pliakoset al. E(L1) [40], Burlew et al. (EL2) [31] and Rasilainen et al. (L2)[36] independently reported statistically signicantly improvedrates of fascial closure in patients who received NPWT in additionto dynamic closure compared with NPWT alone. Rasilainen et al.[36] reported that patients treatedwith NPWT plus meshwere ableto be closed up to 3 weeks following the initial surgery. This meansthat adoption of sequential closure as an adjunct to NPWT may4.2.1. Grade 1 and 2 open abdomen wounds (with no or minimaladherence)

4.2.1.1. Grade 1 and 2 e recommendation 5: In experienced handsNPWT should be used as a rst line therapy in Grade 1 and 2 openabdomen where there is an option for delayed primary closure,including following decompression of ACS; Table 7; Grade B.The key treatment goal in the treatment of Grade 1 and 2 OAwounds is to provide temporary wound cover until fascial closure ispossible or desired (Table 5). In wounds which can be expected toachieve fascial closure relatively easily (such as some ACS andtrauma-derived open abdomen cases), use of NPWT is recom-mended over the other available methods of TAC for a number ofreasons. Firstly, unlike Wittmann patch and other zipper or meshbased techniques NPWT is a suture-less technology and its appli-cation does not damage the fascial tissue [8]. Secondly, unlike theBogota Bag bag approach, NPWT helps to prevent loss of abdominaldomain in addition to providing active uid drainage. Finally, in theperiod immediately following laparotomy, use of NPWT cansignicantly normalize serum lactates [22] and systemic inam-matory mediators [23] compared with passive drainage (such asBogota Bag) indicating that application of NPWT can inuence apatients' physiological stability.

In grade 1A abdominal wounds (trauma, ACS, post-abdominalsurgery) where early and uncomplicated fascial closure is ex-pected, the fewer repeat laparotomies the better before fascialclosure. In a EL2 study, patients closed at the rst repeat laparotomy

of Surgery 12 (2014) 1105e1114increase the window during which fascial closure is possible.

can be used to manage the output and divert it away from the open

lesions. It is therefore important to stop releasing the strong ad-

rnal4.2.1.3. Grade 1 and 2 e recommendation 7: Always use a non-adherent interface layer to protect exposed organs and to preventprogression to grade 3 or 4. Table 7; GPP. Whenmaintaining an openabdomen, there is a window of opportunity within which to ach-ieve primary fascial closure, usually dened as 7e10 days from theoriginal laparotomy, before xity develops and the wound pro-gresses into a Grade 4 abdomen. Treatment of the open abdomen isbi-phasic: typically, grade 1A wounds would require a minimalperiod of TAC leading to early fascial closure (within 7 days).However in many remaining patients and for a variety of reasons,early fascial closure is not possible. An additional treatment goal forthe use of NPWT is to extend thewindow for primary fascial closure[41] (EL4) (Table 5). This would allow more time for resuscitationand sufcient recovery whilst preserving the option of fascialclosure. Miller et al. (2004) [18] reported almost 50% of their suc-cessful primary fascial closures were performed late i.e. after 9days of NPWT and as late as 21 days. Other studies have reportedfascial closure with NPWT as late as 18 [26] and 49 days [42]. Lateclosures have however also been reported for bothWittmann patch(42 days) [43] and to a lesser extent with Bogota Bag (30 days) [44].In the absence of comparative studies it is impossible to statedenitively which TAC method most efciently extends the win-dow for fascial closure in patients unsuitable for early fascialclosure. In addition, the interface may serve as an enhanceddrainage system for uid that may otherwise be retained in thedeep abdominal cavity.

4.2.1.4. Grade 1 and 2 e recommendation 8: Application of incisionalNPWT on the closed incision to facilitate healing should be considered.Table 7; Grade B. Once closure of the fascia and skin have beenachieved it is possible to use NPWT to support healing of the closedlaparotomy incision and prevent complications. A signicantreduction in wound complications including wound dehiscencewas observed compared with standard gauze dressings in acomparative (EL2) study [45]. NPWT has also been used to supporthealing of high risk abdominal incisions in procedures where theabdomen is closed at the index operation [46e48]. In twocomparative retrospective studies (EL2), a signicant reduction inwound complications overall and in particular a reduced incidenceof infection was observed in wounds treated with incisional NPWTcompared with standard dressings [46,48]. The splinting effect ofthe application of NPWT is also believed to aid patient mobility bysupporting the wound as reported in a level 3 study [49]. Earlypatient mobility is thought to be important in reducing duration ofICU stay and improving long-term outcomes [50].

The use of NPWT on closed incisions has been demonstrated togood effect in other clinical indications. Data from these studies canbe extrapolated to provide further support of its likely efcacy inhealing of at risk laparotomy incisions. Several comparativestudies (EL1 [51,52] and EL2 [53,54]) have been reported in otherindications which principally show the reduction in complicationsthat can be achieved through application of NPWT to a closedincision [55].

There are cases when only segments of the abdominal wall canbe closed and others remain dehiscent. In these situations, NPWT isa good choice of wound ller in analogy to other wounds.

4.2.2. Grade 3 open abdomen wounds (complicated by stulaformation)

Intestinal stulae result from anastomotic leakage, traumatizedor ischaemic bowel, pressure ulceration or rupture in case of distalintestinal obstruction. A stula opening directly from the bowelinto an OA is known as an entero-atmospheric stula (EAF). If astula appears before abdominal xation occurs, it is classed as a

A. Bruhin et al. / International JouGrade 3 OA. Surgical repair, by suturing although possible, is rarelyhesions not to risk bowel injuries. At this stage, removal of all non-absorbable meshes is mandatory. Remaining available closure op-tions involve a planned ventral hernia and include skin closure,secondary healing or split skin grafting. The fascia and skin shouldbe closed as far as possible to minimize the remaining surfaceabdominal wound [56e59]. Depending on the viscosity of the s-tula output, pooling beneath the foam may occur. To prevent this,provision of a conduit from the source of the efuent through thedressing to the canister either bymeans of an ostomy bag or a draincan be considered [57]. Use of NPWT can splint the wound toanchor any drainage tubes. This ensures that the drain remains veryclose to the mouth of the stula to maximise uid removal but isprevented from damaging the bowel as a result of becomingdislodged.

Visible stulae can be managed by means of a oating stomawhich involves isolating the stula with an ostomy bag. NPWT isoften the only method available to achieve a secure bag adhesionparticularly in the mobile patient and may simultaneously be usedto manage the adjacent wound [57].

Remote stulae are not suitable for direct stomal isolation. Oneoption is to convert a remote entero-atmospheric stula into anentero-cutaneous stula by performing a separate incision anddrainage. In abdominal wounds which have not yet developedxity, this may allow progression of the wound as described ingrade 1 and 2 aiming at early primary fascial closure. An alternativestrategy is to treat the entire wound with NPWT, allowing thewound to granulate and contract. This leads to closure of the tractbut results in a grade 4 OA wound which is incapable of fascialclosure as well as being both time-consuming and costly. In thesewounds, spontaneous closure of the entero-atmospheric stula hasbeen observed in 55% [58], and 8% [60] of wounds as well as beingobserved in a handful of case studies [59,61e64]. Once output iscontrolled, and a granulating bed has been achieved, it may bepossible to make conservative efforts to close by skin grafting[57,59,65]. This may reduce the risk of further entero-atmosphericstula observed as a result of prolonged maintenance of an openabdomen [12,66].

NPWTmay also protect skin from the stula output thus helpingto preserve the integrity of peri-wound skin [56,67].

4.2.3. Grade 4 open abdomen wounds (frozen abdomen)Prolonged inability to close the fascia will ultimately result in a

xed abdomen in which granulation tissue forms over the surfaceand in amongst the internal organs, classed as a Grade 4 OA. Oncethe abdominal wound has progressed to Grade 4, adhesions be-tween the bowel loops and the xity of the intestines to theabdominal wall are so advanced, that attempts to release the ad-hesions by physical manipulation is highly likely to result in majorsuccessful. Segmental resection can be attempted although caremust be taken with the site of anastomosis during all subsequentprocedures. If this is not possible then a controlled stoma can beconsidered as described below.

4.2.2.1. Grade 3 e recommendation 9: NPWT may be used to manageoutput of an entero-atmospheric stula. Table 7; grade D. No existingTAC methods are ideal in the treatment of a grade 3 OA (with s-tula) although NPWT has been used to good effect in these woundsand may be described as the most best option currently available ifused with caution. The principle goal in grade 3 wounds whenapplying NPWT is to manage and divert the stula output to pre-vent the spread of intra-abdominal sepsis [56e58] (Table 5). NPWT

of Surgery 12 (2014) 1105e1114 1111requiring a skin graft. The planned ventral hernia approach

rnalrequires denite operative reconstruction somemonths later and isusually deferred for a minimum of 6 months.

4.2.3.1. Grade 4 e recommendation 10: It is possible to use NPWT toencourage granulation tissue formation to support split closure bysplit thickness skin graft. Table 7; Grade D. The principle goal forapplication of NPWT in grade 4 OA, now that fascial closure is nolonger possible, is to promote granulation tissue formation tocreate a suitable surface for rapid grafting (Table 5). The degree ofgraft-take depends on the quality of the recipient wound bed, thusonce granulation tissue has covered the surface of the exposedorgans and a frozen abdominal wound is present, NPWT may beused to encourage rapid granulation tissue growth [68,69]. Underthese circumstances the degree of adherence between the visceraand the abdominal wall is too far advanced to permit application ofthe non-adherent interface layer. Physical manipulation of the ad-hesions should be avoided at this stage. The granulation tissue andunderlying organs are delicate and must be protected by the use ofa wound contact layer (e.g. non-adherent silicon adherent layer)placed between the NPWT wound ller and the wound bed. Thewound contact layer prevents growth of newly formed granulationtissue into the foam NPWT wound ller which may result indamage to the underlying tissue when removed, during dressingchanges [68]. The promotion of granulation tissue by NPWT iswidely reported in other indications however direct extrapolationof this data to the specic indication of open abdomen is notappropriate given the additional risks posed by having vulnerablebowel structures in close proximity to the granulating bed. A cor-relation has been reported between the number of re-explorationsafter initiation of NPWT and stula formation [70]. This reinforcesthe need for extreme caution during dressing changes and the useof the interface layer to minimize the impact of dressing changes.

4.2.3.2. Grade 4 e recommendation 11: NPWT should be used toenhance split skin graft take in the abdomen. Table 7; Grade B.Once granulation tissue has formed over the surface of the bowel,leaving an abdominal wound that can't be closed by primary fascialclosure, a skin graft can be performed as soon as clinically appro-priate (e.g. on resolution of sepsis). In large defects, split skin graft isthe only option to rapidly close the abdomen. NPWT is shown topromote a signicantly higher degree of graft take in a variety ofnon-abdominal indications in several EL1 RCTs [71e73] and EL2comparative cohort studies [74e76] The use of NPWT to bolsterSTSG has been systematically reviewed in more detail elsewhere[4]. Several case reports (EL3) of using NPWT to bolster skin graftson open abdomen wounds have been reported [57,59,65] sug-gesting its positive effect in these wounds also. There is a highdegree of relevance when comparing data from other indicationsand the use of NPWT to bolster a skin graft in a planned ventralhernia approach to abdominal closure. The results of the EL1 andEL2 studies can therefore be extrapolated to support the use ofNPWT to bolster skin grafts in the open abdomen.

5. Conclusion

In the clinical setting, each patient with open abdomen is amajor challenge for the physicians and nurses. As each case iscomplex and unique, NPWT is a key technique to provide an indi-vidualised approach because NPWTcan provide different actions atdifferent stages of treatment: e.g. wound and uid management,facilitating primary fascial closure and splinting of skin grafts. Asystematic review of the literature has demonstrated greater rate ofprimary fascial closure in patients treated with NPWT, especiallywhere NPWT is used along-side a dynamic closure technique,

A. Bruhin et al. / International Jou1112compared with other methods of TAC. Evidence-basedrecommendations presented here, specify the use of NPWTthroughout the treatment pathway of the open abdomen andprovide greater clarity around the appropriate use of NPWT.

Ethical approval

N/A (no clinical results presented).

Funding

S&N provided funding for travel and accommodation for allface-to-face expert panel meetings, provided support regarding thepublished literature in NPWT, provided support during the draftingof treatment goals and recommendations (author JS) and providedmedical writing services in preparation of this manuscript (JS). Therecommendations reect the independent and unbiased views ofthe panel and the consensus derived during the project and thecontent of the manuscript has not been unfairly inuenced bySmith & Nephew.

The NPWT-Expert Panel was supported by Smith & Nephew.

Author contribution

Authors AB, FF, MC and NR participated in all expert panelmeetings relating to this project, identied, discussed and agreedall recommendations through a combination of clinical knowledgeand review of available evidence. All authors contributed tomanuscript preparation. Author JS provided methodologicalexpertise and conducted the systematic review and providededitorial assistance with regard to manuscript preparation.

Contributors: Thanks to the wider NPWT expert Panel (NPWT-EP): J.M. Francos Martnez, S. Vig, C. Caravaggi, C. Dowsett, P. Rome,L. Berg, H. Birke-Sorensen, M. Depoorter, R. Dunn, F. Duteille, G.Grudzien, D. Hudson, S. Ichioka, R. Ingemansson, S. Jeffery, E. Krug,C. Lee, M.Malmsjo for their contribution to the early discussionswhich formed the original basis for this manuscript. Thanks toHussein Dharma and Alan Rossington who provided support withrelation to data management and analysis.

Conict of interest

Authors MC, AB, FF and NR have undertaken consultancy workfor Smith & Nephew (S&N) in the area of NPWT and have receivedhonoraria from S&N. MC has served as a fact witness in legal tes-timony for S&N. Author JS is an employee of S&N.

Disclaimer

These materials are provided for educational use only and donot imply that the authors have endorsed Smith & Nephew'sproducts in any way or that the techniques being used are endorsedor recommended by Smith & Nephew.

Appendix A. Supplementary material

Supplementary material related to this article can be found athttp://dx.doi.org/10.1016/j.ijsu.2014.08.396.

References

[1] D.J. Worhunsky, G. Magee, D.A. Spain, Challenges in the management of theopen abdomen, ICU Dir. 4 (1) (2012) 33e38 [Internet].

[2] A.J. Quyn, C. Johnston, D. Hall, A. Chambers, N. Arapova, S. Ogston, et al., The

of Surgery 12 (2014) 1105e1114open abdomen and temporary abdominal closure systems e historical evo-lution and systematic review, Colorectal Dis. 14 (8) (2012) e429ee438.

rnal of Surgery 12 (2014) 1105e1114 1113[3] D.J. Roberts, D.A. Zygun, J. Grendar, et al., Negative-pressure wound therapyfor critically ill adults with open abdominal wounds: a systematic review,J. Trauma Acute Care Surg. 73 (3) (2012 Sep) 629e639.

[4] E. Krug, L. Berg, C. Lee, et al., Evidence-based recommendations for the use ofnegative pressure wound therapy in traumatic wounds and reconstructivesurgery: steps towards an international consensus, Injury (42 Suppl. 1)(2011) S1e12.

[5] H. Birke-Sorensen, M. Malmsjo, P. Rome, et al., Evidence-based recommen-dations for negative pressure wound therapy: treatment variables (pressurelevels, wound ller and contact layer) e steps towards an internationalconsensus, J. Plast. Reconstr. Aesthet. Surg. (64 Suppl. l) (2011) S1eS16.

[6] S. Vig, C. Dowsett, L. Berg, et al., Evidence-based recommendations for theuse of negative pressure wound therapy in chronic wounds: steps towardsan international consensus, J. Tissue Viability (20 Suppl. 1) (2011) S1e18.

[7] M. Bjorck, A. Bruhin, M. Cheatham, et al., Classicationeimportant step toimprove management of patients with an open abdomen, World J. Surg. 33(6) (2009) 1154e1157.

[8] A. Liberati, D.G. Altman, J. Tetzlaff, et al., The PRISMA statement for reportingsystematic reviews and meta-analyses of studies that evaluate healthcareinterventions: explanation and elaboration, BMJ 339 (2009) b2700eb2700.

[9] Scottish Intercollegiate Guidelines Network. Sign 50-A Guideline DevelopersHandbook. pp. 1e111. Available from: http://www.sign.ac.uk/pdf/sign50.pdf.

[10] M.L. Cheatham, D. Demetriades, T.C. Fabian, et al., Prospective study exam-ining clinical outcomes associated with a negative pressure wound therapysystem and Barker's vacuum packing technique, World J. Surg. 37 (9) (2013)2018e2030.

[11] S.L. Trevelyan, G.L. Carlson, Is TNP in the open abdomen safe and effective?J. Wound Care 18 (1) (2009) 24e25.

[12] M. Rao, D. Burke, P.J. Finan, et al., The use of vacuum-assisted closure ofabdominal wounds: a word of caution, Colorectal Dis. 9 (3) (2007) 266e268.

[13] G.L. Carlson, H. Patrick, A.I. Amin, et al., Management of the open abdomen: anational study of clinical outcome and safety of negative pressure woundtherapy, Ann. Surg. 257 (6) (2013) 1154e1159.

[14] I. Pliakos, T.S. Papavramidis, N. Michalopoulos, et al., The value of vacuum-assisted closure in septic patients treated with laparostomy, Am. Surg. 78(9) (2012) 957e961.

[15] R.S. Miller, J.A. Morris, J.J. Diaz, et al., Complications after 344 damage-controlopen celiotomies, J. Trauma Inj. Infect. Crit. Care 59 (6) (2005) 1365e1374.

[16] J.J. Dubose, J.B. Lundy, Enterocutaneous stulas in the setting of trauma andcritical illness, 1 (212) (2010) 182e189.

[17] E. Anesater, M. Roupe, P. Robertsson, et al., The inuence on woundcontraction and uid evacuation of a rigid disc inserted to protect exposedorgans during negative pressure wound therapy, Int. Wound J. 8 (4) (2011)393e399.

[18] P.R. Miller, J.W. Meredith, J.C. Johnson, et al., Prospective evaluation ofvacuum-assisted fascial closure after open abdomen, Ann. Surg. 239 (5)(2004) 608e616.

[19] D.E. Barker, J.M. Green, R. Maxwell, et al., Experience with vacuum-packtemporary abdominal wound closure in 258 trauma and general andvascular surgical patients, J. Am. Coll. Surg. 204 (5) (2007) 784e792.

[20] P.H. Navsaria, M. Bunting, J. Omoshoro-Jones, et al., Temporary closure ofopen abdominal wounds by the modied sandwich-vacuum pack technique,Br. J. Surg. 90 (6) (2003) 718e722.

[21] Q.M. Hatch, L.M. Osterhout, A. Ashraf, et al., Current use of damage-controllaparotomy, closure rates, and predictors of early fascial closure at the rsttake-back, J. Trauma 70 (6) (2011) 1429e1436.

[22] S. Batacchi, S. Matano, A. Nella, et al., Vacuum-assisted closure device en-hances recovery of critically ill patients following emergency surgical pro-cedures, Crit. Care 13 (6) (2009) R194.

[23] B.D. Kubiak, S.P. Albert, L. Gatto, et al., Peritoneal negative pressure therapyprevents multiple organ injury in a chronic porcine sepsis and ischemia/reperfusion model, Shock 34 (5) (2010) 525e534.

[24] S.R. Young, S. Hampton, R. Martin, Non invasive assessment of negativepressure wound therapy using high frequency diagnostic ultrasound:oedema reduction and new tissue accumulation, Int. Wound J. 10 (2013)383e388.

[25] D. Wondberg, H.J. Larusson, U. Metzger, et al., Treatment of the openabdomen with the commercially available vacuum-assisted closure systemin patients with abdominal sepsis: low primary closure rate, World J. Surg.32 (12) (2008) 2724e2729.

[26] J.W. Suliburk, D.N. Ware, Z. Balogh, et al., Vacuum-assisted wound closureachieves early fascial closure of open abdomens after severe trauma,J. Trauma 55 (6) (2003) 1155e1160 discussion 1160e1.

[27] S. Lindstedt, J. Hansson, J. Hlebowicz, Comparative study of the microvas-cular blood ow in the intestinal wall during conventional negative pressurewound therapy and negative pressure wound therapy using parafn gauzeover the intestines in laparostomy, Int. Wound J. 9 (2) (2012) 150e155.

[28] S. Lindstedt, M. Malmsjo, J. Hansson, et al., Pressure transduction and uidevacuation during conventional negative pressure wound therapy of theopen abdomen and NPWT using a protective disc over the intestines, BMCSurg. 12 (1) (2012) 4.

[29] T. Bjarnason, A. Montgomery, J. Hlebowicz, et al., Pressure at the bowelsurface during topical negative pressure therapy of the open abdomen: an

A. Bruhin et al. / International Jouexperimental study in a porcine model, World J. Surg. 35 (4) (2011)917e923.[30] R. Kafka-Ritsch, M. Zitt, N. Schorn, et al., Open abdomen treatment withdynamic sutures and topical negative pressure resulting in a high primaryfascia closure rate, World J. Surg. 36 (8) (2012) 1765e1771.

[31] C.C. Burlew, E.E. Moore, W.L. Bif, et al., One hundred percent fascialapproximation can be achieved in the postinjury open abdomen with asequential closure protocol, J. Trauma Acute Care Surg. 72 (1) (2012)235e241.

[32] K. Sorelius, A. Wanhainen, S. Acosta, et al., Open abdomen treatment afteraortic aneurysm repair with vacuum-assisted wound closure and mesh-mediated fascial traction, Eur. J. Vasc. Endovasc. Surg. (2013) 1e7.

[33] S. Acosta, T. Bjarnason, U. Petersson, et al., Multicentre prospective study offascial closure rate after open abdomen with vacuum and mesh-mediatedfascial traction, Br. J. Surg. 98 (5) (2011) 735e743.

[34] U. Petersson, S. Acosta, M. Bjorck, Vacuum-assisted wound closure andmesh-mediated fascial tractionea novel technique for late closure of theopen abdomen, World J. Surg. 31 (11) (2007) 2133e2137.

[35] A. Seternes, H.O. Myhre, T. Dahl, Early results after treatment of openabdomen after aortic surgery with mesh traction and vacuum-assistedwound closure, Eur. J. Vasc. Endovasc. Surg. 40 (1) (2010) 60e64.

[36] S.K. Rasilainen, P.J. Mentula, A.K. Leppaniemi, Vacuum and mesh-mediatedfascial traction for primary closure of the open abdomen in critically illsurgical patients, Br. J. Surg. 99 (12) (2012) 1725e1732.

[37] J. Kleif, R. Fabricius, C.A. Bertelsen, et al., Promising results after vacuum-assisted wound closure and mesh-mediated fascial traction, Dan. Med. J.59 (9) (2012 Sep) A4495.

[38] F.J. Verdam, D.E. Dolmans, M.J. Loos, et al., Delayed primary closure of theseptic open abdomen with a dynamic closure system, World J. Surg. 35 (10)(2011) 2348e2355.

[39] E. Salman, F. Yetisir, M. Aksoy, et al., Use of dynamic wound closure system inconjunction with vacuum-assisted closure therapy in delayed closure ofopen abdomen, Hernia (2012) 0e5.

[40] I. Pliakos, T.S. Papavramidis, N. Mihalopoulos, et al., Vacuum-assisted closurein severe abdominal sepsis with or without retention sutured sequentialfascial closure: a clinical trial, Surgery 148 (5) (2010) 947e953.

[41] W.P. Schecter, R.R. Ivatury, M.F. Rotondo, et al., Open abdomen after traumaand abdominal sepsis: a strategy for management, J. Am. Coll. Surg. 203 (3)(2006) 390e396.

[42] P.R. Miller, J.T. Thompson, B.J. Faler, et al., Late fascial closure in lieu ofventral hernia: the next step in open abdomen management, J. Trauma 53(5) (2002) 843e849.

[43] B.H. Tieu, S.D. Cho, N. Luem, et al., The use of the Wittmann Patch facilitates ahigh rate of fascial closure in severely injured trauma patients and criticallyill emergency surgery patients, J. Trauma 65 (4) (2008) 865e870.

[44] J.G. Hadeed, G.W. Staman, H.S. Sariol, et al., Delayed primary closure indamage control laparotomy: the value of the Wittmann Patch, Am. Surg. 73(1) (2007) 10e12.

[45] A. Conde-Green, T.L. Chung, L.H. Holton, et al., Incisional negative-pressurewound therapy versus conventional dressings following abdominal wallreconstruction: a comparative study, Ann. Plast. Surg. 71 (4) (2012)394e397.

[46] D. Vargo, Negative pressure wound therapy in the prevention of woundinfection in high risk abdominal wound closures, Am. J. Surg. 204 (6) (2012)1021e1023.

[47] M. Lopez-Cano, M. Armengol-Carrasco, Use of vacuum-assisted closure inopen incisional hernia repair: a novel approach to prevent seroma formation,Hernia 17 (1) (2011) 129e131.

[48] A.U. Blackham, J.P. Farrah, T.P. McCoy, et al., Prevention of surgical site in-fections in high-risk patients with laparotomy incisions using negative-pressure therapy, Am. J. Surg. 205 (6) (2013) 647e654.

[49] M. Dutton, K. Curtis, Well-wound therapy: use of NPWT to prevent lapa-rotomy breakdown, J. Wound Care 21 (8) (2012) 386e388.

[50] E.E. Vasilevskis, E.W. Ely, T. Speroff, et al., Reducing iatrogenic risks: ICU-acquired delirium and weaknessecrossing the quality chasm, Chest 138(5) (2010) 1224e1233.

[51] J.P. Stannard, D.A. Volgas, G. McGwin, et al., Incisional negative pressurewound therapy after high-risk lower extremity fractures, J. Orthop. Trauma26 (1) (2012) 37e42.

[52] J.P. Stannard, J.T. Robinson, E.R. Anderson, et al., Negative pressure woundtherapy to treat hematomas and surgical incisions following high-energytrauma, J. Trauma 60 (6) (2006) 1301e1306.

[53] R.N. Reddix, X.I. Leng, J. Woodall, et al., The effect of incisional negativepressure therapy on wound complications after acetabular fracture surgery,J. Surg. Orthop. Adv. 19 (2) (2010) 91e97.

[54] G.W. Schmedes, C.A. Banks, B.T. Malin, et al., Massive ap donor sites and therole of negative pressure wound therapy, Otolaryngol. Head Neck Surg. 147(6) (2012) 1049e1053.

[55] D. Masden, J. Goldstein, M. Endara, et al., Negative pressure wound therapyfor at-risk surgical closures in patients with multiple comorbidities: a pro-spective randomized controlled study, Ann. Surg. 255 (6) (2012) 1043e1047.

[56] G. Al-Khoury, D. Kaufman, A. Hirshberg, Improved control of exposed stulain the open abdomen, J. Am. Coll. Surg. 206 (2) (2008) 397e398.

[57] J. Goverman, J. Yelon, J.J. Platz, et al., The Fistula VAC, a technique formanagement of enterocutaneous stulae arising within the open abdomen:

report of 5 cases, J. Trauma 60 (2) (2006) 428e431.

A. Bruhin et al. / International Journal of Surgery 12 (2014) 1105e11141114[58] S. Stremitzer, A. Dal Borgo, T. Wild, et al., Successful bridging treatment andhealing of enteric stulae by vacuum-assisted closure (VAC) therapy andtargeted drainage in patients with open abdomen, Int. J. Colorectal Dis. 26 (5)(2011) 661e666.

[59] A. Marinis, G. Gkiokas, G. Anastasopoulos, et al., Surgical techniques for themanagement of enteroatmospheric stulae, Surg. Infect. Larchmt 10 (1)(2009) 47e52.

[60] J.M. Draus, S. Huss, N.J. Harty, et al., Enterocutaneous stula: are treatmentsimproving? Surgery 140 (4) (2006) 570e576.

[61] T. Banasiewicz, M. Borejsza-Wysocki, W. Meissner, et al., Topical negativepressure as a safe and helpful treatment in patients with large abdominalwounds with multiple stulae, Colorectal Dis. 12 (11) (2010) 1166e1167.

[62] K. Boulanger, V. Lemaire, D. Jacquemin, Vacuum-assisted closure of enter-ocutaneous stula, Acta Chir. Belg. 107 (6) (2007) 703e705.

[63] M. D'Hondt, A. D'Haeninck, L. Dedrye, et al., Can vacuum-assisted closure andinstillation therapy (VAC-Instill therapy) play a role in the treatment of theinfected open abdomen? Tech. Coloproctol. 15 (1) (2011) 75e77.

[64] P. Navsaria, A. Nicol, D. Hudson, et al., Negative pressure wound therapymanagement of the open abdomen following trauma: a prospective studyand systematic review, World J. Emerg. Surg. 8 (1) (2013) 4 (Available from:).

[65] M.C. Byrnes, A. Riggle, G. Beilman, et al., A novel technique to skin graftabdominal wall wounds surrounding enterocutaneous stulas, Surg. Infect.Larchmt 11 (6) (2010) 505e509.

[66] J.E. Fischer, A cautionary note: the use of vacuum-assisted closure systems inthe treatment of gastrointestinal cutaneous stula may be associated withhigher mortality from subsequent stula development, Am. J. Surg. 196 (1)(2008) 1e2.

[67] C. Cro, K.J. George, J. Donnelly, S.T. Irwin, K.R. Gardiner, Vacuum assistedclosure system in the management of enterocutaneous stulae, Postgrad.Med. J. 78 (920) (2002) 364e365.

[68] I. Shaikh, A. Ballard-Wilson, S. Yalamarthi, et al., Use of topical negativepressure in assisted abdominal closure does not lead to high incidence ofenteric stulae, Colorectal Dis. 12 (9) (2010) 931e934.

[69] A.I. Amin, I.A. Shaikh, Topical negative pressure in managing severe perito-nitis: a positive contribution? World J. Gastroenterol. 15 (27) (2009)3394e3397.

[70] M. Schmelzle, I. Alldinger, H. Matthaei, et al., Long-term vacuum-assistedclosure in open abdomen due to secondary peritonitis: a retrospectiveevaluation of a selected group of patients, Dig. Surg. 27 (4) (2010) 272e278.

[71] J.D. Vuerstaek, T. Vainas, J. Wuite, et al., State-of-the-art treatment of chronicleg ulcers: a randomized controlled trial comparing vacuum-assisted closure(V.A.C.) with modern wound dressings, J. Vasc. Surg. 44 (5) (2006)1029e1037.

[72] E. Moisidis, T. Heath, C. Boorer, et al., A prospective, blinded, randomized,controlled clinical trial of topical negative pressure use in skin grafting, Plast.Reconstr. Surg. 114 (4) (2004) 917e922.

[73] S. Llanos, S. Danilla, C. Barraza, et al., Effectiveness of negative pressureclosure in the integration of split thickness skin grafts: a randomized,double-masked, controlled trial, Ann. Surg. 244 (5) (2006) 700e705.

[74] A. Korber, T. Franckson, S. Grabbe, et al., Vacuum assisted closure deviceimproves the take of mesh grafts in chronic leg ulcer patients, Dermatology216 (3) (2008) 250e256.

[75] L.A. Scherer, S. Shiver, M. Chang, et al., The vacuum assisted closure device: amethod of securing skin grafts and improving graft survival, Arch. Surg. 137(8) (2002) 930e933.

[76] D.M. Vidrine, S. Kaler, E.L. Rosenthal, A comparison of negative-pressuredressings versus Bolster and splinting of the radial forearm donor site,Otolaryngol. Head Neck Surg. 133 (3) (2005) 403e406.

[77] G.B. Garner, D.N. Ware, C.S. Cocanour, J.H. Duke, B.A. Mckinley, et al., Vac-uum-assisted wound closure provides early fascial reapproximation intrauma patients with open abdomens, Am. J. Surg. 182 (2001) 630e638.

[78] P. Stone, S.M. Hass, S.K. Flaherty, et al., Vacuum-assisted fascial closure forpatients with abdominal trauma, J. Trauma Inj. Infect. Crit. Care 57 (5) (2004)1082e1086.

[79] J. Weinberg, R.L. George, R.L. Grifn, et al., Closing the open abdomen:improved success with Wittmann Patch staged abdominal closure, J. Trauma65 (2) (2008) 345e348.

[80] Z. Serclova, O. Ryska, P. Dytrych, J. Marvan, K. Marx, E. Konecna, A. Mikskova,F. Antos, Fascial closure of the abdominal wall by dynamic suture aftertopical negative pressure laparostomy treatment of severe peri-tonitiseresults of a prospective randomized study, Rozhl. Chir. 91 (1) (2012)26e31.

[81] A. Perathoner, A. Klaus, G. Mhlmann, et al., Damage control with abdominalvacuum therapy (VAC) to manage perforated diverticulitis with advancedgeneralized peritonitisea proof of concept, Int. J. Colorectal Dis. 25 (6) (2010)767e774.

[82] J. Olejnik, I. Sedlak, I. Brychta, I. Tibensky, Vacuum supported laparostomy ean effective treatment of intraabdominal infection, Bratisl. Lek. Listy 108 (7)(2007) 320e323.

[83] P. Oetting, B. Rau, P.M. Schlag, Abdominal vacuum device with openabdomen, Chirurg 77 (7) (2006) 588e593.

[84] J. Horwood, F. Akbar, A. Maw, Initial experience of laparostomy with im-mediate vacuum therapy in patients with severe peritonitis, Ann. R. Coll.

Surg. Engl. 91 (8) (2009) 681e687.[85] P. Padalino, G. Dionigi, G. Minoja, et al., Fascia-to-fascia closure withabdominal topical negative pressure for severe abdominal infections: pre-liminary results in a department of general surgery and intensive care unit,Surg. Infect. 11 (6) (2010) 523e552.

[86] T. Wild, S. Stremitzer, A. Budzanowski, et al., Abdominal dressing e a newmethod of treatment for open abdomen following secondary peritonitis,Zentralbl. Chir. 129 (S) (2004) 20e23.

[87] L. Perez Domnguez, H. Pardellas Rivera, N. Caceres Alvarado, et al., Vacuumassisted closure in open abdomen and deferred closure: experience in 23patients, Cir. Esp. 90 (8) (2012 Oct) 506e512.

[88] A. Caro, C. Olona, A. Jimenez, et al., Treatment of the open abdomen withtopical negative pressure therapy: a retrospective study of 46 cases, Int.Wound J. 8 (3) (2011) 274e279.

[89] D. Perez, S. Wildi, N. Demartines, Prospective evaluation of vacuum-assistedclosure in abdominal compartment syndrome and severe abdominal sepsis,J. Am. Coll. Surg. 205 (4) (2007) 586e592.

[90] J.P. Arigon, O. Chapuis, E. Sarrazin, F. Pons, A. Bouix, R. Jancovici, Managingthe open abdomen with vacuum-assisted closure therapy: retrospectiveevaluation of 22 patients, J. Chir. Paris 145 (3) (2008) 252e261.

[91] L. Labler, J. Zwingmann, D. Mayer, et al., V.A.C. abdominal dressing system,Eur. J. Trauma 31 (5) (2005) 488e494.

[92] C.C. Cothren, E.E. Moore, J.L. Johnson, et al., One hundred percent fascialapproximation with sequential abdominal closure of the open abdomen, Am.J. Surg. 192 (2) (2006) 238e242.

[93] C.B. Ross, L. Irwin, K. Mukherjee, et al., Vacuum-pack temporary abdominalwound management with delayed-closure for the management of rupturedabdominal aortic aneurysm and other abdominal vascular catastrophes:absence of graft infection in long term survivors, Am. Surg. 75 (2007)565e571.

[94] J. Jiang, Y. Dai, M. Zhu, et al., Clinical application of vacuum pack system fortemporary abdominal closure, Zhonghua Wei Chang. Wai Ke Za Zhi 9 (1)(2006) 50e52.

[95] M. Chavarria-Aguilar, W.T. Cockerham, D.E. Barker, et al., Management ofdestructive bowel injury in the open abdomen, J. Trauma 56 (3) (2004)560e564.

[96] T.K. Bee, M. Croce, L.J. Magnotti, et al., Temporary abdominal closure tech-niques: a prospective randomized trial comparing polyglactin 910 mesh andvacuum-assisted closure, J. Trauma 65 (2) (2008) 337e342.

[97] J.W. Smith, R.N. Garrison, P.J. Matheson, et al., Direct peritoneal resuscitationaccelerates primary abdominal wall closure after damage control surgery,J. Am. Coll. Surg. 210 (5) (2010) 658e664, 664e7.

[98] L. Lopez-Quintero, G. Evaristo-Mendez, F. Fuentes-Flores, et al., Treatment ofopen abdomen in patients with abdominal sepsis using the vacuum packsystem, Cir. Cir. 78 (4) (2010) 322e326.

[99] S. Prichayudh, S. Sriussadaporn, P. Samorn, et al., Management of openabdomen with an absorbable mesh closure, Surg. Today 41 (1) (2011) 72e78.

[100] C. Stonerock, R. Bynoe, M. Yost, J. Nottingham, Use of a vacuum-assisteddevice to facilitate abdominal closure, Am. Surg. 69 (12) (2003) 1030e1034.

[101] H. Ozguc, E. Paksoy, E. Ozturk, Temporary abdominal closure with the vac-uum pack technique: a 5-year experience, Acta Chir. Belg. 108 (4) (2008)414e419.

[102] W. Brock, D. Barker, R. Burns, Temporary closure of open abdominal woundse the vacuum pack, Am. Surg. 61 (1) (1995 Jan) 30e35.

[103] J. Cipolla, S.P. Stawicki, W.S. Hoff, et al., A proposed algorithm for managingthe open abdomen, Am. Surg. 71 (3) (2007) 202e207.

[104] C. Aprahamian, D. Wittmann, J. Bergstein, E. Quebbeman, Temporaryabdominal closure (TAC) for planned relaparotomy (etappenlavage) intrauma, J. Trauma 30 (6) (1990 Jun) 719e723.

[105] M. Keramati, A. Srivastava, S. Sakabu, et al., The Wittmann Patch s a tem-porary abdominal closure device after decompressive celiotomy forabdominal compartment syndrome following burn, Burns 34 (4) (2008)493e497.

[106] L.N. Tremblay, D.V. Feliciano, J. Schmidt, R.A. Cava, K.M. Tchorz, W.L. Ingram,et al. Skin only or silo closure in the critically ill patient with an openabdomen, 182 (2001) 670e675.

[107] A. Brox-Jimenez, D. Daz-Gome z, P. Parra-Membrives, D. Martnez-Baena,M.Marquez-Mu~noz, J. Lorente-Herce, et al., A vacuum assisted closure systemin complex wounds: a retrospective study, Cir. Esp. 87 (5) (2010) 312e317.

[108] T.R. Howdieshell, C.D. Proctor, E. Sternberg, J.I. Cue, J.S. Mondy, M.L. Hawkins,Temporary abdominal closure followed by denitive abdominal wallreconstruction of the open abdomen, Am. J. Surg. 188 (3) (2004) 301e306.

[109] C. Manterola, J. Moraga, S. Urrutia, Contained laparostomy with a Bogota bag.results of case series, Cir. Esp. 89 (6) (2011) 379e385.

[110] P.J. Offner, L. de Souza, E.E. Moore, et al., Avoidance of abdominal compart-ment syndrome in damage-control laparotomy after trauma, Arch. Surg. 136(6) (2001) 676e681.

[111] D. Gonll, F.N. Koksoy, O. Demiray, S.G. Ozkan, T. Ycel, O. Ycel, Lapa-rostomy in patients with severe secondary peritonitis, Ulus. Travma AcilCerrahi Derg 15 (1) (2009 Jan) 52e57.

[112] J.A. Garca I~niguez, C.F. Orozco, M.I. Muci~no Hernandez, et al., Complicationsof the management of secondary peritonitis with contained-open abdomen.Comparison of the Bogota's bag vs polypropylene mesh, Rev. Gastroenterol.Mex. 69 (3) (2004) 147e155.

Systematic review and evidence based recommendations for the use of Negative Pressure Wound Therapy in the open abdomen1 Introduction1.1 Classification of the open abdomen (OA)

2 Methods2.1 Systematic review2.2 Development of recommendations

3 Relative efficacy of different TAC methods4 Evidence based recommendations4.1 General treatment goals and recommendations4.1.1 General recommendation 1: An interface layer must be used to protect exposed organs and to avoid adhesions between th ...4.1.2 General recommendation 2: Use of a specialised open abdomen foam-based dressing kit should be used. Table 6; Grade B4.1.3 General recommendation 3: NPWT may be used to manage abdominal wound fluid. Table 6; Grade C4.1.4 General recommendation 4: Continuous NPWT settings of up to 80 mmHg are recommended. Table 6; Grade D

4.2 Grade-specific goals, recommendations and cautions4.2.1 Grade 1 and 2 open abdomen wounds (with no or minimal adherence)4.2.1.1 Grade 1 and 2 recommendation 5: In experienced hands NPWT should be used as a first line therapy in Grade 1 and 2 ...4.2.1.2 Grade 1 and 2 recommendation 6: Application of a sequential dynamic closure technique, along with NPWT to counter ...4.2.1.3 Grade 1 and 2 recommendation 7: Always use a non-adherent interface layer to protect exposed organs and to preven ...4.2.1.4 Grade 1 and 2 recommendation 8: Application of incisional NPWT on the closed incision to facilitate healing shoul ...

4.2.2 Grade 3 open abdomen wounds (complicated by fistula formation)4.2.2.1 Grade 3 recommendation 9: NPWT may be used to manage output of an entero-atmospheric fistula. Table 7; grade D

4.2.3 Grade 4 open abdomen wounds (frozen abdomen)4.2.3.1 Grade 4 recommendation 10: It is possible to use NPWT to encourage granulation tissue formation to support split ...4.2.3.2 Grade 4 recommendation 11: NPWT should be used to enhance split skin graft take in the abdomen. Table 7; Grade B

5 ConclusionEthical approvalFundingAuthor contributionConflict of interestDisclaimerAppendix A. Supplementary materialReferences