small intestinal submucosa for reinforcement of colonic anastomosis

8
ORIGINAL ARTICLE Small intestinal submucosa for reinforcement of colonic anastomosis Jens Hoeppner & Vladan Crnogorac & Goran Marjanovic & Eva Jüttner & Tobias Keck & Hans-Fred Weiser & Ullrich Theodor Hopt Accepted: 8 January 2009 / Published online: 30 January 2009 # Springer-Verlag 2009 Abstract Background Different materials have been evaluated for anastomotic reinforcement to prevent gastrointestinal anas- tomotic leakage. In this experimental study, small intestinal submucosa (SIS) was tested as a sealing for stapled colonic anastomosis in a porcine model. The aims of this study were to determine the macroscopic and microscopic out- comes and to evaluate the safety and feasibility of applying SIS for anastomotic sealing. Materials and methods Circular stapled anastomoses were performed in 18 pigs. Standard anastomosis in the control group (n =8) was compared to an SIS-sealed anastomosis in the study group (n =10). After 30 days, anastomotic segments were examined for macroscopic and microscopic regeneration and their resistance to mechanical stress. Furthermore, animal survival and clinical course were evaluated. Results None of the animals developed anastomotic leak- age, intraabdominal abscess, or peritonitis. Shrinkage of SIS was evident in nine of ten animals. Encapsulation and displacement of the SIS patches were seen in two animals. Quantity of anastomotic granulation tissue and rate of complete mucosal coverage of anastomotic line were increased in SIS-sealed anastomoses without reaching significance. Moreover, no significant differences were found in the rate of survival of the animals, anastomotic stricture formation, intraabdominal adhesions, anastomotic bursting pressure, and microscopic healing parameters of the anastomosis between stapled colonic standard anasto- mosis and anastomosis protected by SIS. Conclusion The results of this study indicate a safe use of SIS for anastomotic reinforcement in a porcine model. Adverse effects like strictures, increased adhesions, and anastomotic abscesses were absent. Promoting effects on colonic wound healing by SIS were microscopically evident. The results argue for a careful clinical evaluation in humans. Keywords Small intestinal submucosa . Anastomosis . Colon . Reinforcement Introduction Anastomotic leaks are a common complication and a major factor of morbidity and mortality in colorectal surgery. The lower the colorectal anastomosis created, the higher the risk of developing a postoperative anastomotic dehiscence due to decreased vascularization and subsequent anastomotic failure. The incidence reaches from 0.51% at the right hemicolon to 823% for low colorectal anastomosis in oncologic resections [16]. Various methods have been Int J Colorectal Dis (2009) 24:543550 DOI 10.1007/s00384-009-0637-y J. Hoeppner : G. Marjanovic : T. Keck : U. T. Hopt Department of Surgery, University of Freiburg, Freiburg, Germany E. Jüttner Department of Pathology, University of Freiburg, Freiburg, Germany V. Crnogorac : H.-F. Weiser Department of Surgery, Diakoniekrankenhaus Rotenburg, Rotenburg, Germany J. Hoeppner (*) Klinik für Allgemein- und Viszeralchirurgie, Chirurgische Klinik, Universitätsklinik Freiburg, Hugstettter Str. 55, 79106 Freiburg, Germany e-mail: [email protected]

Upload: jens-hoeppner

Post on 14-Jul-2016

216 views

Category:

Documents


4 download

TRANSCRIPT

ORIGINAL ARTICLE

Small intestinal submucosa for reinforcementof colonic anastomosis

Jens Hoeppner & Vladan Crnogorac &

Goran Marjanovic & Eva Jüttner & Tobias Keck &

Hans-Fred Weiser & Ullrich Theodor Hopt

Accepted: 8 January 2009 / Published online: 30 January 2009# Springer-Verlag 2009

AbstractBackground Different materials have been evaluated foranastomotic reinforcement to prevent gastrointestinal anas-tomotic leakage. In this experimental study, small intestinalsubmucosa (SIS) was tested as a sealing for stapled colonicanastomosis in a porcine model. The aims of this studywere to determine the macroscopic and microscopic out-comes and to evaluate the safety and feasibility of applyingSIS for anastomotic sealing.Materials and methods Circular stapled anastomoses wereperformed in 18 pigs. Standard anastomosis in the controlgroup (n=8) was compared to an SIS-sealed anastomosis inthe study group (n=10). After 30 days, anastomoticsegments were examined for macroscopic and microscopicregeneration and their resistance to mechanical stress.Furthermore, animal survival and clinical course wereevaluated.

Results None of the animals developed anastomotic leak-age, intraabdominal abscess, or peritonitis. Shrinkage ofSIS was evident in nine of ten animals. Encapsulation anddisplacement of the SIS patches were seen in two animals.Quantity of anastomotic granulation tissue and rate ofcomplete mucosal coverage of anastomotic line wereincreased in SIS-sealed anastomoses without reachingsignificance. Moreover, no significant differences werefound in the rate of survival of the animals, anastomoticstricture formation, intraabdominal adhesions, anastomoticbursting pressure, and microscopic healing parameters ofthe anastomosis between stapled colonic standard anasto-mosis and anastomosis protected by SIS.Conclusion The results of this study indicate a safe use ofSIS for anastomotic reinforcement in a porcine model.Adverse effects like strictures, increased adhesions, andanastomotic abscesses were absent. Promoting effects oncolonic wound healing by SIS were microscopicallyevident. The results argue for a careful clinical evaluationin humans.

Keywords Small intestinal submucosa . Anastomosis .

Colon . Reinforcement

Introduction

Anastomotic leaks are a common complication and a majorfactor of morbidity and mortality in colorectal surgery. Thelower the colorectal anastomosis created, the higher the riskof developing a postoperative anastomotic dehiscence dueto decreased vascularization and subsequent anastomoticfailure. The incidence reaches from 0.5–1% at the righthemicolon to 8–23% for low colorectal anastomosis inoncologic resections [1–6]. Various methods have been

Int J Colorectal Dis (2009) 24:543–550DOI 10.1007/s00384-009-0637-y

J. Hoeppner :G. Marjanovic : T. Keck :U. T. HoptDepartment of Surgery, University of Freiburg,Freiburg, Germany

E. JüttnerDepartment of Pathology, University of Freiburg,Freiburg, Germany

V. Crnogorac :H.-F. WeiserDepartment of Surgery, Diakoniekrankenhaus Rotenburg,Rotenburg, Germany

J. Hoeppner (*)Klinik für Allgemein- und Viszeralchirurgie, Chirurgische Klinik,Universitätsklinik Freiburg,Hugstettter Str. 55,79106 Freiburg, Germanye-mail: [email protected]

proposed to reduce anastomotic failure. Besides preopera-tive bowel preparation, antibiotic prophylaxis, parenteralnutrition, and careful surgical technique in patients withhigh risk for anastomotic leakage, additional methods ofincreasing the stability of colonic anastomoses are required.In these cases, reinforcement of the anastomosis bywrapping it with an artificial or biological graft has beenclaimed to be useful [7]. A promising approach for theprotection of gastrointestinal anastomosis could be anasto-motic sealing with the so-called extracellular matrices.Extracellular matrices are acellular, collagenous resorbablescaffolds of biological origin. The most established materialis “small intestine submucosa” (SIS). SIS is a biodegrad-able, acellular, immunologic inert collagen matrix, which isextracted from the submucosal layer of porcine smallbowel. In recent years, experimental studies have shownconsiderable success for the use of SIS as a tissue graft inblood vessels, bladder, esophagus, biliary duct, smallbowel, and large bowel [8–14]. This study was designedto test the feasibility and safety of SIS in protecting colonicanastomoses in a porcine model and to evaluate theinfluence of SIS on macroscopic, microscopic, and me-chanic parameters of anastomotic healing.

Materials and methods

Animals

Eighteen female German domestic pigs with a medianweight of 30.6 kg (range 24–39.4 kg) were used for thisinvestigation. The study was approved by the Animal Careand Use Committees at the University of VeterinaryMedicine Hannover and the local district government ofLower Saxony in Germany. All procedures in this studywere performed under strict adherence to the GermanAnimal Welfare Law and met the standards set in the“Guide for Care and Use of Laboratory Animals” preparedby the National Academy of Sciences and published by theNational Institutes of Health (NIH Publication No. 86-23,revised 1985).The pigs were weighed before surgicalprocedure and at regular intervals during postoperativeobservation. The animals were fed with a standard diet andwater ad libitum. Streptomycin/penicillin (0.1 ml/kg;Streptocomb®, Fa. Streuli, Swizerland) was given intra-muscularly at the start of anesthesia and the day after theoperation. All surgical procedures were performed undergeneral anesthesia. For premedication, animals receivedintramuscular application of azaperon (2 mg/kg), ketaminehydrochloride (15 mg/kg), and atropine (0.05 mg/kg).Anesthesia was deepened with intravenous application ofatracurium besylate and hypnomidate (10 mg/kg). Subse-quently, the animals were intubated and artificially

ventilated with isoflurane and a mix of oxygen anddinitrogen oxide.

Study groups

The animals were divided into two groups, one controlgroup with an unprotected standard anastomosis of thedescending colon and one study group with anastomoses ofthe descending colon sealed by an SIS patch.

Surgical procedure I—anastomosis with or without SISwrapping

After midline laparotomy under sterile conditions andexposure of the peritoneal cavity, the descending colonand rectum were identified and gently exposed. Transectionof the descending colon and adjacent mesentery wasperformed. Fecal contents were carefully removed withiodine gauze. The anastomosis was performed using a 25-mm diameter Premium Plus CEEA Stapler (Covidien,Neustadt, Germany). In the study group, the anastomosiswas wrapped by a 75×25-mm four-layer SIS Patch(Surgisis®; Cook Surgical, Lafayette, IN, USA).The patchwas fixed to the bowel wall with six single-knot sutures 3/0Polysorb (Covidien, Neustadt, Germany; Fig. 1a). In orderto allow some swelling during early healing, 360° anasto-motic sealing was achieved by a 1-cm overlapping of thetwo ends of the SIS patch at the mesenterial side of theanastomosis without fixing the ends to each other. Inthe control group with standard stapler anastomosis at thedescending colon, the anastomosis was anatomically placedin the free abdominal cavity without intentional sealing byany tissue. After physiological re-arrangement of theabdominal organs, the abdominal cavity was closed inlayers with absorbable sutures. The abdominal cavity wasclosed in the same manner in both groups after completionof the anastomosis.

Surgical procedure II—anastomotic healing assessment

Thirty days after surgical intervention, the animals werereanesthetized, and re-laparotomy was performed. Theperitoneal cavity was examined for signs of impairedanastomotic healing, e.g., peritonitis, intraabdominal ab-scess, or fibrinous coverings. The integrity of the anasto-mosis was inspected, and the anastomotic site was checkedfor signs of wound dehiscence, pericolic abscess formation,necrosis, fistulas, and adhesions. Adhesions were examinedin a scoring system from 0 to 3 points. The descendingcolon and adherent organs were removed en bloc forradiological examination, followed by tensile testing andsubsequent detailed macroscopic and microscopic exami-nation of the anastomotic segment.

544 Int J Colorectal Dis (2009) 24:543–550

Radiological examination

Contrast enemas of the descending colon were performedby filling it with water-soluble contrast medium (Gastro-grafin®, Fa. Schering, Berlin, Germany). The diameters ofthe anastomosis and the pre- and postanastomotic segmentswere evaluated by X-ray analysis. X-rays were done ex situat the isolated and transected large bowel segment. Bothends were clamped and the bowel was filled by cannulatingthe proximal segment and filling it with contrast medium.X-rays were digitalized and diameters were measured 2 cmabove and at the level of and 2 cm below the anastomosisusing an image analysis program (Image J, NIH, Bethesda,USA). Ex situ radiological evaluation does not representphysiological conditions; therefore, the findings were judgedas artificial, although the conditions were comparable andthe radiological observations allow the assessment of thedegree of stenosis or dilatation at the anastomotic site. Ananastomotic index (AI) was calculated for each animal.

AI ¼ 2 � anastomotic diameter

proximal diameter þ distal diameter

A straight tube has an AI=0.5. Anastomotic strictureand/or proximal dilatation will result in a decreased AI [15].Additionally, the percentage of reduction in colonic diameterat the anastomotic site was calculated: 100×(1−AI).

Tensile tests

Mechanical testing of the anastomosis was performed bymeasuring the bursting pressure. An 18-Fr catheter wasinserted into both ends of the anastomotic bowel segmentand secured by purse string sutures. Air was insufflatedvia one catheter, and the other one was connected to adigital manometer (GDH 14 AN, Fa. Griesinger, Regen-stauf, Germany). The bowel segment was plunged inwater and the bursting pressure detected by the presenceof ascending air bubbles and the abrupt fall in intra-luminal pressure. The intraluminal pressure in this assaywas limited to 200 mm Hg.

Macroscopic examination

During ex situ preparation, the specimen was always keptmoist by superfusion with 0.9% saline. The descending

Fig. 1 a Stapled end-to-endanastomosis of the descendingcolon wrapped by small intesti-nal submucosa 75×25 mm. Thecircular patch was fixed withseveral single-knot sutures Pol-ysorb 3/0. b Water-soluble ene-ma examination of theanastomosis on day30. Anasto-mosis reinforced with smallintestinal submucosa patch. cEnd-to-end anastomosiswrapped by small intestinalsubmucosa on day 30 with thevascularized patch covering theanstomosis. d Shrunken encap-sulated remnant of the circularsmall intestinal submucosapatch on day 30

Int J Colorectal Dis (2009) 24:543–550 545

colon was cut longitudinally along the mesenteric border.First, the serosal site was checked. Condition and locationof the SIS patch, macroscopic vascularization of the patch,and occurrence of bowel necrosis or fistula were assessed.The specimen was turned around and the mucosal surfaceexamined for epithelial defects, integrity of the staple line,mucosal necrosis, and fistulas.

Microscopic examination

For histological examination, 2-mm wide strips were cutout at 90° angles to the anastomotic line. The strips werefixed in phosphate-buffered 4% formaldehyde for 4 daysand subsequently embedded in paraffin. The sections werecut in 5-μm slides and stained with hematoxylin and eosin.Mucosal coverage of the anastomotic line, presence ofinflammatory reaction at the layers of the bowel wall, andquantity of granulation tissue were evaluated. Furthermore,microscopic anastomotic presence of abscess, necrosis,fistula, and foreign body reaction as features of impairedanastomotic healing was assessed.

Statistical analysis

All data are expressed as the mean ± standard error of themean (SEM). Statistical analysis was performed using thesoftware Sigmastat® (Systat Software Inc., Chicago, IL,USA). Non-parametric tests were used for the statisticalanalysis in all cases due to the limited size of individualgroups. The Mann–Whitney U test and chi-square test wereused for arbitrarily distributed, independent variables andcomparison of the two groups, respectively. p values of lessthan 0.05 were accepted as significant.

Results

Survival and clinical course

Operating time was significantly longer in operations wherereinforcement of anastomosis by SIS was performedcompared to operations with standard stapled anastomosis(92±1.8 vs. 78±5.6 min, p=0.021). None of the animalsdied or developed clinically relevant disease due toanastomotic leakage or intraabdominal infection. Oneanimal in the control group had signs of large bowel ileuson the tenth postoperative day. Operative explorationshowed a high-grade anastomotic stenosis at the descendingcolon. One animal in the SIS group and two animals in thecontrol group developed superficial wound infectionswithout signs of systemic infection and without impairmentof further clinical course. There was no significantdifference in the 30-day weight gain after the operation in

the two groups (7.79±0.33 kg (+25%) SIS group vs. 6.8±0.57 kg (+24%) control group).

Macroscopic examination

Macroscopic examination of the abdominal cavity at theend of the experiment showed no anastomotic leaks,intraabdominal abscess, fistulas, or necrosis at the anasto-motic site in either group. One animal in the SIS groupshowed a 1-mm diameter ulcer at the mucosal site of thestapler anastomosis without further distinctive features(Table 1). After 30 days, adhesions were frequentlyobserved in both groups. In descending frequency, theadhesive organs were uterus horns, ovaries, and bowelsegments. There were no significant differences in for-mations of adhesions either in quality (p=0.929) or inquantity (p=0.853) in the two groups (Table 2). The SISpatch completely covered the circular anastomosis in six ofthe study group animals. Visible shrinkage of SIS was seenin nine of ten animals. Encapsulation and displacement ofthe SIS patches were evident in two animals (Fig. 1d).Macroscopically distinguishable vascular invasion of theSIS patch was detected in all animals (Fig. 1c; Table 3).

Mechanical testing

In both groups, all anastomoses remained sufficient andleak-proof in a test of bursting pressure with a maximal

Table 1 Macroscopic, microscopic, and radiographic examination ofanastomoses

Study group(SIS)

Controlgroup

p

n=10 n=8

Macroscopic examinationAnastomotic leakage 0/10 0/8 nsAnastomotic ulcer 1/10 0/8 nsIntraabdominal abscess 0/10 0/8 nsPeritonitis 0/10 0/8 nsVisible stenosis 0/10 1/8 nsMicroscopic examinationComplete mucosal coverage 6/10 3/8 nsInflammation mucosa 10/10 8/8 nsInflammation submucosa/mucosa 10/10 8/8 nsAbscess 0/10 0/8 nsNecrosis 0/10 0/8 nsFistula 0/10 0/8 nsGranulation tissue quantitya 1.7 1.38 nsForeign body 1/10 1/10 nsRadiographic examinationFistula 0/10 0/8 nsAnastomotic index 0.78 0.64 ns

a Score 0–2: 0 = minimal; 1 = moderate; 2 = distinctive

546 Int J Colorectal Dis (2009) 24:543–550

intraluminal pressure of 200 mm Hg. There were nosignificant differences in stability and resistance to me-chanical stress.

Radiological examination

No fistulas or leakages were detected at the anastomoticsite. The anastomotic stricture rate showed no significantdifferences between the SIS group and the control group(Fig. 1b). Complete anastomotic stenosis was seen incontrast enema in one of the control group animals. The

anastomotic index was 0.77 in the SIS group and 0.65 inthe control group (p=0.198; Table 1). The percentage ofreduction in colonic diameter at the anastomotic site was22% vs. 35% (p=0.198; Table 2).

Microscopic examination

Histological examination was carried out in all 18 anasto-moses. Table 1 presents data of the scores in the H&Esections. Complete mucosal coverage was evident in six often SIS-reinforced anastomoses and in three of eightstandard stapled anastomoses (p=0.447). All other animalsshowed small mucosal gaps filled with granulation tissue atthe anastomotic line. Inflammation of all bowel wall layerswas clearly visible in both groups. The quantity ofanastomotic granulation tissue was slightly higher in SIS-sealed anastomosis compared to standard anastomosis,without reaching a statistically significant level (p=0.263).One animal in each group showed foreign body granulomaenclosing small fecal contents. No abscess, necrosis, orfistulas were seen at the anastomotic sites in any of theanimals. Besides the higher rate of complete mucosalcoverage in SIS-covered anastomosis and the increasedquantity of granulation tissue, which were both non-significant differences, no obvious differences in woundhealing were observed between standard anastomoses andSIS-covered anastomoses. The SIS patches covering theanastomosis were visible below the serosal layer (Fig. 2).Lymphocytes and macrophages infiltrated the SIS patcheswithout any indication of abscess or infection. Cellular

Table 2 Macroscopic scoring of anastomotic adhesions and radio-graphic evaluation of anastomotic diameter

Animalno.

Percentage reductionin diameter

Adhesion rateI quantitya

Adhesion rateII qualityb

Study group SIS1 28.13 2 22 19.23 3 33 21.65 0 04 22.96 0 05 29.34 0 06 28.68 1 27 29.04 2 28 20.32 1 29 19.27 0 010 4.30 0 0Mean 22.29 0.9 1.1SEM 1.88Control group1 24.60 1 22 27.65 1 13 100.00 1 14 26.16 1 25 17.30 0 06 24.12 0 07 33.76 0 08 30.66 1 2Mean 35.53 0.625 1SEM 9.37

p=0.198 p=0.853 p=0.929

a Adhesion rate I: 0 = no adhesions; 1 = adhesions with one structure;2 = adhesions with two structures; 3 = adhesions with three or morestructuresb Adhesion rate II: 0 = no adhesions; 1 = light adhesions; 2 = fixedadhesions; 3 = solid adhesions, only removable with damage

Table 3 Macroscopic examination of SIS patches 30 days afteranastomotic sealing

Complete SIS coverage of anastomosis 4/10Visible shrinkage 9/10Dislocation 2/10Encapsulation 2/10Macroscopic vascularization 10/10

Fig. 2 Microscopic appearance of the colonic anastomosis covered bySIS at 30 days postsurgery. All layers of the colonic wall can be seenin the areas next to the anastomosis (mucosa—top layer, submucosa—between mucosa and muscle layer and muscle layer). The mucosallayer completely covers the anastomosis. The holes in the middle areartifacts due to the removal of the anastomotic staples. SIS (arrow) islocated beneath the serosal layer (H&E stain, magnification ×10)

Int J Colorectal Dis (2009) 24:543–550 547

infiltration by activated fibroblasts was visible in thecomplete thickness of the four-layer SIS. Furthermore,distinctive neovascularization of the patches was observed(Fig. 3).

Discussion

Reinforcement of gastrointestinal anastomoses by sealingthem with different biological grafts or synthetic materialshas been investigated in different studies. In the past,different studies have been published on free or pedicularomental reinforcement of colonic anastomosis. Somestudies have reported the beneficial effects of omentalreinforcement on rate of leakage [16–18], whereas other

studies have found no effect on frequency of anastomoticleakage [19, 20]. Similar inconsistency of results has to benoted for sealing colonic anastomosis with peritonealpatches [21–24]. Recently, dura mater has been reportedto impair anastomotic healing and increase rate of stricturesin colonic anastomosis [24]. Apart from biological grafts,numerous synthetic materials have been evaluated mostly insingle experimental studies. Dacron [21] and polyglycolicacid mesh [25] did not show any positive effect onanastomotic healing, whereas anastomotic sealing withmarlex mesh reduced the rate of anastomotic insufficiencies[21]. Altogether, the results of these experimental andclinical trials are often inconsistent and mostly notconclusive for definite beneficial effects on anastomotichealing and safety.

Nevertheless, wrapping of gastrointestinal anastomosesby materials which mechanically seal the anastomosis andpromote processes of anastomotic wound healing throughtheir own biological activity appears to be a promisingapproach for the prevention of anastomotic leakage. Itseemed obvious and reasonable to propose SIS, which isalready in human clinical use for hernia repair [26, 27] andfor anal fistula repair [28–31], as a biomaterial which couldbe able to improve the healing of gastrointestinal anasto-mosis. It has been shown that SIS provokes a host responsefor neoangiogenesis, tissue regeneration, and restoration ofstructure and function that is specific to the implantationsite [32]. Different proteins responsible for cellular migra-tion and attachment, like fibronectin and heparin sulfateproteoglycan, and different growth factors, like FGF,VEGF, and TGF-β, have been identified in SIS [32–34].

This is the first study to examine the reinforcement ofcolonic anastomoses by SIS sealing. Besides the feasibilityand safety of SIS sealing, the macroscopic and microscopicparameters of anastomotic healing and host reaction toanastomotic sealing by SIS were examined. Side effects ofreinforcement with SIS patches like infectious complica-tions, intraabdominal adhesions, or anastomotic stricturewere addressed. In contrast to other materials like Dacron[21] and polyglycolic acid mesh (PGA) [25], which havebeen used for anastomotic protection, no peritonitis,fistulas, leakages, and anastomotic abscess could be foundin SIS-protected anastomosis. Moreover, we found noincreased rate of anastomotic strictures or intraabdominaladhesions compared to standard anastomoses. SIS did notcause relevant luminal obstruction, which has been reportedfor anastomotic protection with other biomaterials likeDacron, PGA, and dura mater [20, 24, 25]. Concerningadhesions, other materials which have been evaluated foranastomotic reinforcement caused relevant increase ofanastomotic adhesions [24]. SIS did not affect the forma-tion of intraabdominal adhesions in our study. Anastomoticstrength, measured by evaluation of bursting pressure, was

Fig. 3 a Host tissue response to SIS sealing 30 days postsurgery(serosal side on the left, abdominal side on the right side of thepicture). Neovascularization (vessels filled with blood cells) as wellas a cellular infiltrate composed of lymphocytes and macrophagescan be observed together with activated fibroblasts (H&E stain,magnification ×100). b Higher magnification of the cellular infiltrateof macrophages and lymphocytes. Small capillaries can be seen at theperiphery (H&E stain, magnification ×200)

548 Int J Colorectal Dis (2009) 24:543–550

not impaired in SIS-protected anastomoses. In eachspecimen, supraphysiological intraluminal pressure valuesof >200 mg Hg were reached in mechanical testing.Encapsulation and displacement of SIS patches initiallycompletely covering the anastomosis were phenomenawhich were recognized in two of ten animals in the presentstudy. The reason for this reaction may be an early breakageof the fixation sutures and a subsequent retraction anddisplacement of the SIS patches. Some shrinkage of the SISpatch covering the anastomosis was noted in nearly allanimals. In another two animals, the non-displaced SISpatch only partially covered the anastomosis in itscircumference. Probably, the use of patches longer than75 mm with a larger overlap of the two ends of the SISwould reduce the rate of partial anastomotic coverage.Anastomotic safety was not reduced through these facts. Noinfluence on complication rate could be found in thepresence of shrinkage, encapsulation, and displacement ofthe SIS patches.

Microscopically, our study demonstrated a migration offibroblasts into the SIS implant and showed markedneovascularization of SIS patches within 30 days.Concerning regeneration of the bowel layer at the anasto-motic site, the rate of complete mucosal coverage of theanastomotic line and the quantity of anastomotic granula-tion tissue were increased in SIS-sealed anastomosis. Thesefindings suggest that there are interactions between the SISand the host tissue which lead to cytotaxis and incorpora-tion of the extracellular matrix. Moreover, regeneration ofthe bowel wall at the anastomotic site seems to be promotedby SIS. Adverse effects like formation of low-gradelymphoma of the bowel wall at the anastomotic site, ashas been reported in anastomotic reinforcement by Dacron,were not seen [21].

Limitations of this study include a lack of long-termobservations since colonic anastomoses were observed for aperiod of 30 days in our study. Especially contraction ofSIS which has been reported in other studies [8, 13] couldcause bowel obstruction on the long term if gastrointestinalanastomoses are wrapped by SIS. In view of the completeabsence of luminal narrowing in SIS-protected anastomosisafter 30 days, it is a strong indicator that SIS is unlikely tocause anastomotic stricture in observation periods greaterthan 30 days. This speculation is underlined by observa-tions in which SIS was used for anastomotic reinforcementof gastrojejunostomy in laparoscopic Roux-en-Y gastricbypass and did not cause significant increase in rate ofstenosis compared to standard anastomosis with an medianfollow-up of 87 days [35]. Concerning the transfer of ourfindings to clinical application, we are aware of thedifferences in porcine and human colonic healing. Evenso, our study could demonstrate the safe use of SIS foranastomotic reinforcement in a large animal model with

comparable colonic anatomy. Whether SIS can actuallyreduce the incidence of anastomotic leakage in humans wasnot the subject of the present safety observation and has tobe evaluated in further studies.

In summary, we could demonstrate in the present studythe feasibility and safety of the use of SIS patches foranastomotic reinforcement at the descending colon in aporcine model. Promoting effects of SIS on colonic woundhealing were microscopically evident. Adverse effects ofanastomotic reinforcement like strictures, increased forma-tion of adhesions, and anastomotic abscesses were absent.The results of this study indicate the safe use of SIS foranastomotic reinforcement in large animals and warrantmore extensive clinical investigation.

References

1. Law WL, Chu KW (2004) Anterior resection for rectal cancerwith mesorectal excision: a prospective evaluation of 622 patients.Ann Surg 240(2):260–268 Aug

2. Law WI, Chu KW, Ho JW, Chan CW (2000) Risk factors foranastomotic leakage after low anterior resection with totalmesorectal excision. Am J Surg 179(2):92–96 Feb

3. Hainsworth PJ, Egan MJ, Cunliffe WJ (1997) Evaluation of apolicy of total mesorectal excision for rectal and rectosigmoidcancers. Br J Surg 84(5):652–656 May

4. Bokey EL, Chapuis PH, Fung C, Hughes WJ, Koorey SG, BrewerD, Newland RC (1995) Postoperative morbidity and mortalityfollowing resection of the colon and rectum for cancer. Dis ColonRectum 38(5):480–486 discussion 486–7, May

5. Nesbakken A, Nygaard K, Lunde OC, Blucher J, Gjertsen O,Dullerud R (2005) Anastomotic leak following mesorectalexcision for rectal cancer: true incidence and diagnostic chal-lenges. Colorectal Dis 7(6):576–581 Nov

6. Graf W, Glimelius B, Bergstrom R, Pahlman L (1991) Compli-cations after double and single stapling in rectal surgery. Eur JSurg 157(9):543–547 Sep

7. Marescaux JF, Aprahamian M, Mutter D, Loza E, Wilhelm M,Sonzini P, Damge C (1991) Prevention of anastomosis leakage: anartificial connective tissue. Br J Surg 78(4):440–444 Apr

8. Chen MK, Badylak SF (2001) Small bowel tissue engineeringusing small intestinal submucosa as a scaffold. J Surg Res 99(2):352–358 Aug

9. Ueno T, Oga A, Takahashi T, Pappas TN (2007) Small intestinalsubmucosa (SIS) in the repair of a cecal wound in unpreparedbowel in rats. J Gastrointest Surg 11(7):918 –9 22 Jul

10. Kropp BP, Rippy MK, Badylak SF, Adams MC, Keating MA,Rink RC, Thor KB (1996) Regenerative urinary bladder augmen-tation using small intestinal submucosa: urodynamic and histo-pathologic assessment in long-term canine bladder augmentations.J Urol 155(6):2098–2104 Jun

11. Badylak SF, Lantz GC, Coffey A, Geddes LA (1989) Smallintestinal submucosa as a large diameter vascular graft in the dog.J Surg Res 47(1):74–80 Jul

12. Rosen M, Ponsky J, Petras R, Fanning A, Brody F, Duperier F(2002) Small intestinal submucosa as bioscaffold for biliary tractregeneration. Surgery 132(3):480–486 Sep

13. Demirbilek S, Kanmaz T, Ozardali I, Edali MN, Yucesan S (2003)Using porcine small intestinal submucosa in intestinal regenera-tion. Pediatr Surg Int 19(8):588–592 Oct

Int J Colorectal Dis (2009) 24:543–550 549

14. Badylak S, Meurling S, Chen M, Spievack A, Simmons-Byrd A(2000) Resorbable bioscaffold for esophageal repair in a dogmodel. J Pediatr Surg 35(7):1097–1103 Jul

15. Jansson OK, Zilling TL, Walther BS (1991) Healing of colonicanastomoses: comparative experimental study of glued, manuallysutured, and stapled anastomoses. Dis Colon Rectum 34(7):557–562 Jul

16. Smith SR, Swift I, Gompertz H, Baker WN (1988) Abdomino-perineal and anterior resection of the rectum with retrocolicomentoplasty and no drainage. Br J Surg 75(10):1012–1015

17. Lanter B, Mason RA (1979) Use of omental pedicle graft toprotect low anterior colonic anastomosis. Dis Colon Rectum 22(7):448–451

18. Agnifili A, Schietroma M, Carloni A, Mattucci S, Caterino G,Lygidakis NJ (2004) The value of omentoplasty in protectingcolorectal anastomosis from leakage. A prospective randomizedstudy in 126 patients. Hepatogastroenterology 51(60):1694–1697

19. Tocchi A, Mazzoni G, Lepre L, Costa G, Liotta G, Agostini N(2000) Prospective evaluation of omentoplasty in preventingleakage of colorectal anastomosis. Dis Colon Rectum 43(7):951–955

20. Merad F, Hay JM, Fingerhut A, Flamant Y, Molkhou JM, LabordeY (1998) Omentoplasty in the prevention of anastomotic leakageafter colonic or rectal resection: a prospective randomized study in712 patients. Ann Surg 227(2):179–186

21. Gulati SM, Thusoo TK, Kakar A, Iyenger B, Pandey KK (1982)Comparative study of free omental, peritoneal, Dacron velour, andMarlex mesh reinforcement of large-bowel anastomosis: anexperimental study. Dis Colon Rectum 25(6):517 –5 21 Sep

22. Devine J (1946) Preliminary report of a method for prevention ofleakage of intestinal anastomosis: an experimental study. SurgGynaecol Obstet 82:475

23. Dilek ON, Bakir B, Dilek FH, Demirel H, Yiğit MF (1996)Protection of intestinal anastomoses in septic environment withperitoneal graft and polyglycolic acid mesh: an experimentalstudy. Acta Chir Belg 96(6):261–265 Nov–Dec

24. Eryilmaz R, Samuk M, Tortum OB, Akcakaya A, Sahin M,Goksel S (2007) The role of dura mater and free peritoneal graft inthe reinforcement of colon anastomosis. J Invest Surg 20(1):15–21Jan–Feb

25. Henne-Bruns D, Kreischer HP, Schmiegelow P, Kremer B (1990)Reinforcement of colon anastomoses with polyglycolic acid mesh:an experimental study. Eur Surg Res 22(4):224–230

26. Helton WS, Fisichella PM, Berger R, Horgan S, Espat NJ,Abcarian H (2005) Short-term outcomes with small intestinalsubmucosa for ventral abdominal hernia. Arch Surg 140(6):549–560 discussion 560–2, Jun

27. Franklin ME Jr, Gonzalez JJ Jr, Glass JL (2004) Use of porcinesmall intestinal submucosa as a prosthetic device for laparoscopicrepair of hernias in contaminated fields: 2-year follow-up. Hernia8(3):186–189 Epub 2004 Feb 26, Aug

28. Champagne BJ, O’Connor LM, Ferguson M, Orangio GR,Schertzer ME, Armstrong DN (2006) Efficacy of anal fistulaplug in closure of cryptoglandular fistulas: long-term follow-up.Dis Colon Rectum 49(12):1817–1821 Dec

29. O’Connor L, Champagne BJ, Ferguson MA, Orangio GR,Schertzer ME, Armstrong DN (2006) Efficacy of anal fistulaplug in closure of Crohn’s anorectal fistulas. Dis Colon Rectum49(10):1569 – 1573 Oct

30. Schwandner O, Stadler F, Dietl O, Wirsching RP, Fuerst A (2008)Initial experience on efficacy in closure of cryptoglandular andCrohn’s transsphincteric fistulas by the use of the anal fistula plug.Int J Colorectal Dis 23(3):319–324 Epub 2007 Nov 22, Mar

31. Johnson EK, Gaw JU, Armstrong DN (2006) Efficacy of analfistula plug vs. fibrin glue in closure of anorectal fistulas. DisColon Rectum 49(3):371–376 Mar

32. Hodde J (2002) Naturally occurring scaffolds for soft tissue repairand regeneration. Tissue Eng 8(2):295–308 Apr

33. Voytik-Harbin SL, Brightman AO, Kraine MR, Waisner B,Badylak SF (1997) Identification of extractable growth factorsfrom small intestinal submucosa. J Cell Biochem 67(4):478–491Dec 15

34. Hodde JP, Record RD, Liang HA, Badylak SF (2001) Vascularendothelial growth factor in porcine-derived extracellular matrix.Endothelium 8(1):11–24

35. Kini S, Gagner M, de Csepel J, Gentileschi P, Dakin G (2001) Abiodegradable membrane from porcine intestinal submucosa toreinforce the gastrojejunostomy in laparoscopic Roux-en-Ygastric bypass: preliminary report. Obes Surg 11(4):469–473Aug

550 Int J Colorectal Dis (2009) 24:543–550