functional evaluation of the grafted wall with porcine-derived small intestinal submucosa (sis) to a...

8
Functional evaluation of the grafted wall with porcine-derived small intestinal submucosa (SIS) to a stomach defect in rats Tomio Ueno, MD, PhD, a,c,d Sebastian G. de la Fuente, MD, a,c Omar I. Abdel-Wahab, MD, c Toku Takahashi, MD, PhD, a,c Marcia Gottfried, MD, b Mary B. Harris, RVT, c Makoto Tatewaki, MD, PhD, a,c Kenichiro Uemura, MD, PhD, a,c D. Curtis Lawson, MS, c Christopher R. Mantyh, MD, a,c and Theodore N. Pappas, MD, a,c Durham, NC and Yamaguchi, Japan Background. Small intestinal submucosa (SIS) represents a novel bio-scaffolding material that may be used to repair hollow-organ defects. However, it is unclear whether neurophysiologic responses return to SIS-grafted areas in the gut. We evaluated the functional recovery of a stomach defect grafted with the porcine-derived SIS. Methods. Twelve rats had a full-thickness defect created in the stomach. SIS was secured to the gastric wall. After 6 months, muscle strips were harvested from within the grafted area to perform both a histologic and a functional study. Additional full-thickness muscle strips were harvested from the posterior in the same stomach as controls. A dose response curve was obtained with carbachol (CCH) or sodium nitroprusside (SNP). Activation of intrinsic nerves was achieved by electrical field stimulation (EFS). Results. The response to CCH and amplitude in EFS showed tonic contraction in both controls and SIS strips in a concentration-dependent and frequency-dependent manner. The magnitude after each stimulation was significantly lower in SIS strips compared with controls (P .01). However, the contraction ratio of EFS to ED 50 of CCH was not significantly different between the groups. Additionally, SNP produced relaxation in both strips in a concentration-dependent manner. Histologic findings revealed that an insufficient amount of smooth-muscle cells existed in the muscularis propria, whereas compensated growth was observed in the submucosa with nerve regeneration. Conclusions. This study demonstrates that SIS provides a template for nerve migration to the graft in the rodent stomach. Innervations showed a similar distribution to that observed in the controls. The clinical implications of such findings warrant additional investigation. (Surgery 2007;142:376-83.) From the Department of Surgery a and the Department Pathology, b Duke University Medical Center, Durham, NC; the Department of Surgery, Durham VA Medical Center, NC c ; and the Department of Surgery II, Yamaguchi University Graduate School of Medicine, Japan d Recent progress in the field of tissue engineering has led to the development of multiple novel bio- scaffolding materials. Porcine-derived small intesti- nal submucosa (SIS, Surgisis ES; Cook Biotech, LaFayette, Ind) is one such agent that serves as a bioscaffold for the generation of a variety of gastro- intestinal tissues and organs. 1-8 Our previous short- term study with SIS used in the repair of a stomach defect in rats has shown that SIS stimulates regen- eration of native tissue under acidic conditions. 7 We also reported the efficacy of the SIS for the repair of the abdominal wall defects in human. 9 Since the frequency of clinical use of SIS is ex- pected to increase in the future, it is important to determine how well SIS-reorganized tissue func- tions. The aim of this study was to determine whether nerves and smooth muscle migrate to the SIS- grafted area and then become functional by exam- Supported by the Cook Company (makers of SIS). Accepted for publication April 20, 2007. Reprint requests: Tomio Ueno, MD, PhD, Department of Sur- gery II, Yamaguchi University Graduate School of Medicine, 1-1-1 Minami-kogushi, Ube, Yamaguchi, 755-8505 Japan. E-mail: [email protected]. 0039-6060/$ - see front matter © 2007 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2007.04.019 376 SURGERY

Upload: tomio-ueno

Post on 12-Sep-2016

213 views

Category:

Documents


1 download

TRANSCRIPT

Functional evaluation of the graftedwall with porcine-derived smallintestinal submucosa (SIS) to astomach defect in ratsTomio Ueno, MD, PhD,a,c,d Sebastian G. de la Fuente, MD,a,c Omar I. Abdel-Wahab, MD,c

Toku Takahashi, MD, PhD,a,c Marcia Gottfried, MD,b Mary B. Harris, RVT,c

Makoto Tatewaki, MD, PhD,a,c Kenichiro Uemura, MD, PhD,a,c D. Curtis Lawson, MS,c

Christopher R. Mantyh, MD,a,c and Theodore N. Pappas, MD,a,c Durham, NC and Yamaguchi, Japan

Background. Small intestinal submucosa (SIS) represents a novel bio-scaffolding material that may beused to repair hollow-organ defects. However, it is unclear whether neurophysiologic responses return toSIS-grafted areas in the gut. We evaluated the functional recovery of a stomach defect grafted with theporcine-derived SIS.Methods. Twelve rats had a full-thickness defect created in the stomach. SIS was secured to the gastricwall. After 6 months, muscle strips were harvested from within the grafted area to perform both ahistologic and a functional study. Additional full-thickness muscle strips were harvested from theposterior in the same stomach as controls. A dose response curve was obtained with carbachol (CCH) orsodium nitroprusside (SNP). Activation of intrinsic nerves was achieved by electrical field stimulation(EFS).Results. The response to CCH and amplitude in EFS showed tonic contraction in both controls andSIS strips in a concentration-dependent and frequency-dependent manner. The magnitude after eachstimulation was significantly lower in SIS strips compared with controls (P � .01). However, thecontraction ratio of EFS to ED50 of CCH was not significantly different between the groups.Additionally, SNP produced relaxation in both strips in a concentration-dependent manner. Histologicfindings revealed that an insufficient amount of smooth-muscle cells existed in the muscularis propria,whereas compensated growth was observed in the submucosa with nerve regeneration.Conclusions. This study demonstrates that SIS provides a template for nerve migration to the graft inthe rodent stomach. Innervations showed a similar distribution to that observed in the controls. Theclinical implications of such findings warrant additional investigation. (Surgery 2007;142:376-83.)

From the Department of Surgerya and the Department Pathology,b Duke University Medical Center, Durham, NC;the Department of Surgery, Durham VA Medical Center, NCc; and the Department of Surgery II, Yamaguchi

University Graduate School of Medicine, Japand

Recent progress in the field of tissue engineeringhas led to the development of multiple novel bio-scaffolding materials. Porcine-derived small intesti-nal submucosa (SIS, Surgisis ES; Cook Biotech,

Supported by the Cook Company (makers of SIS).

Accepted for publication April 20, 2007.

Reprint requests: Tomio Ueno, MD, PhD, Department of Sur-gery II, Yamaguchi University Graduate School of Medicine,1-1-1 Minami-kogushi, Ube, Yamaguchi, 755-8505 Japan. E-mail:[email protected].

0039-6060/$ - see front matter

© 2007 Mosby, Inc. All rights reserved.

doi:10.1016/j.surg.2007.04.019

376 SURGERY

LaFayette, Ind) is one such agent that serves as abioscaffold for the generation of a variety of gastro-intestinal tissues and organs.1-8 Our previous short-term study with SIS used in the repair of a stomachdefect in rats has shown that SIS stimulates regen-eration of native tissue under acidic conditions.7

We also reported the efficacy of the SIS for therepair of the abdominal wall defects in human.9

Since the frequency of clinical use of SIS is ex-pected to increase in the future, it is important todetermine how well SIS-reorganized tissue func-tions.

The aim of this study was to determine whethernerves and smooth muscle migrate to the SIS-

grafted area and then become functional by exam-

Surgery Ueno et al 377Volume 142, Number 3

ining the muscle contractility, the expression ofreceptors, and innervations within the grafted areasas well as histologic evaluation.

MATERIALS AND METHODSStudy design and surgical manipulation. All as-

pects of this research were reviewed and approvedby the Durham VA Medical Center Animal Careand Use Committee and by the Animal Care Com-mittee, Durham, NC.

Twelve male Sprague-Dawley rats (body weight300-350 g; Charles River Laboratories, Raleigh,NC) were housed with free access to water andchow under standard conditions (23°C room tem-perature, 12 h dark–light cycles). Each animal wasrestricted from food except water 18 h before thesurgery. Anesthesia was induced and maintainedwith isoflurane and oxygen. Surgery was performedas previously described.7 An upper midline incisionwas made, and the stomach was identified andgently mobilized with atraumatic forceps. A 1 �1-cm circular full-thickness layer defect was createdon the antrum of the stomach with scissors. Afterhemostasis with electrocautery, a round patch of2-ply SIS (Surgisis ES) was prepared by opposing 2layers and cut to approximately the same size as theexcised portion of the stomach. The SIS was mois-turized with sterile saline for 10 min before implan-tation. To be secured to the gastric wall, stitcheswere taken from the seromuscular layer and placedwithin 1 mm of the edge of the graft with a contin-uous 5-0 polypropylene suture. The skin incisionwas closed in 2 layers. No antibiotics were admin-istered. Animals were checked for signs of distressand administered analgesics as needed. Body weightand food intake were monitored weekly.

After 6 months, under anesthesia with an intra-muscular injection of xylazine and ketamine (13and 87 mg/kg, respectively), the entire stomachwas excised and 2 full-thickness muscle strips alongwith circular muscle fibers were then taken fromwithin the reorganized area. It should be notedthat great care was taken to harvest the strips fromregenerated stomach strictly within the area delin-eated by a previous polypropylene suture for in vitrostudy. One strip was submitted for a histologic ex-amination, and the other was used to examine thefunction of grafted areas as described previously.10

Additional full-thickness muscle strips were har-vested from the posterior wall of the same stomachas controls.

Histologic evaluation. Tissue specimens werefixed with 10% formalin, embedded in paraffin,sectioned along with the circular muscle, and then

stained with hematoxylin and eosin. To confirm

the regeneration of neural fibers, the specimen wasexamined by immunohistochemistry using antibod-ies against S100 protein, presumably staining theSchwann cells of intramuscular nerve fibers.11-13

Functional evaluation. For use in in vitro study,each strip’s length, width, and wet weight was mea-sured before analysis. Both ends of the muscle stripwere suspended between 2 platinum electrodes in a30-mL organ bath filled with Kreb’s-Henseleitbuffer of the following composition (in mmol/L):NaCl (118), KCl (4.8), CaCl2 (2.5), NaHCO3 (25),KH2PO4 (1.2), MgSO4 (1.2), and glucose (11). ThepH of the buffer solution was maintained at 7.4throughout the experiment by constant bubblingwith 95% O2 and 5% CO2.

Muscle strips were allowed to equilibrate for atleast 60 min before the start of any stimulus. Organbaths were maintained at 37oC. A dose responsecurve was obtained on each muscle strip with car-bamylcholine chloride (carbachol, CCH) (1 � 10�8

to 10�4 M), a muscarinic cholinergic agonist, andsodium nitroprusside (SNP) (1 � 10�7 to 10�4 M).SNP directly stimulates the production of solubleguanylate cyclase in smooth-muscle cells, whichthen induces muscle relaxation.

Activation of the intrinsic nerves was achieved byelectrical field stimulation (EFS) with the followingparameters: 85 V, 0.5-ms duration, 30-s trains at 1,2.5, 5, 10, and 20 Hz, allowing approximately 3 minbetween stimulations. EFS was also applied 20 minafter the administration of atropine sulfate (10�6 M)and guanethidine monosulfate (10�6 M) to exam-ine the nonadrenergic, noncholinergic (NANC)relaxation. Tissues were washed with saline at least3 times and were given a 20-min equilibration pe-riod between each drug treatment or series of EFS.

We routinely prepared fresh stock solutions ofCCH, SNP, atropine, and guanethidine in saline.Each stock solution was then made by further dilu-tion to the concentrations of interest. All chemicalswere purchaed from Sigma (St. Louis, Mo). Changesin mechanical contractility were recorded on apolygraph through isometric transducers (NihonKohden, Tokyo, Japan) and analyzed with a com-puter-assisted system (Power Lab, ADInstruments,Castle Hill, Australia).

Statistical analysis. Data were expressed as themean � the standard error of the mean. Statisticalanalysis was performed by paired t test, unpaired ttest, or ANOVA followed by Bonfferoni. P valuesless than .05 were considered significant.

RESULTSClinical signs and macroscopic findings. All rats

survived and thrived and were healthy at the time

scle (

378 Ueno et al SurgerySeptember 2007

of sacrifice. They showed significant weight gain(preoperation: 321.7 � 6.8 g; postoperation: 732.5 �19.4 g) (P � .01, paired t test). On gross inspection,SIS was not observed and the area of SIS could onlybe established by identification of the remainingpolypropylene sutures. No evidence of diverticularformation and/or shrinkage in the grafted regionwas observed.

Microscopic findings. The defect was completelyclosed, and no histologic evidence of the originalSIS material remained. In most animals, a portionof the wall in the area of the defect was made up ofsmooth muscle (Fig 1). The smooth muscle wasfound in the inner portion of the wall, extendingthrough the areas formerly occupied by the mus-cularis mucosa, the submucosa, and the inner por-tion of the muscularis propria and in closeapposition to the mucosa. Fibrosis occupied theouter wall.

In normal gastric wall, immunostaining for S100protein highlighted the structures of Auerbach’splexus in the muscularis propria and Meissner’splexus in the submucosa, as well as scattered stain-ing throughout the smooth muscle of the muscu-

Fig 1. The SIS grafted area of the defect at 6 months aftmucosa, smooth muscle, and fibrosis with neovascularizmuscle was present in the superficial portion of the wall iis made up of fibrous tissue. Arrow indicates smooth mu

laris propria (Fig 2, C). In the areas of the grafted

defect, neither Auerbach’s nor Meissner’s plexuscould be identified. However, the scattered S100positive cells in the smooth muscle were similar tothose in the normal muscularis propria (Fig 2, Aand B).

In vitro contractility. The mean length, width,and tissue wet weight were 1.13 � 0.07 cm, 2.48 �0.15 mm, and 62.7 � 3.9 mg in control strips and1.12 � 0.03 cm, 2.30 � 0.09 mm, and 74.8 � 6.8 mgin grafted strips, respectively. No significant differ-ences were found between the 2 groups in thelength (P � .08, unpaired t test), width (P � .07,unpaired t test), or tissue wet weight (P � .17,unpaired t test).

CCH induced tonic contractions in both normalstomach strips and SIS-regenerated stomach stripsin a concentration-dependent manner (Fig 3).However, the amplitude of the response to CCH inthe SIS-regenerated strips was significantly (P �.01, ANOVA) lower than that in normal stomachstrips. Responses to the concentration of 10�8,10�7, 10�6, 10�5, and 10�4 M CCH in SIS-regener-ated stomach strips were 12.4%, 12.3%, 11.5%,15.8%, and 18.8% of normal stomach strips,

gery. Defect created on the stomach wall was covered by(‘) and infiltration of inflammatory cells (e). Smoothrea of the previous defect. The outer portion of the wall

A, 10�; B, 20�).

er surationn the a

respectively.

Surgery Ueno et al 379Volume 142, Number 3

SNP produced relaxation in both strips in aconcentration-dependent manner (Fig 4). The am-plitude of the response to SNP in the regeneratedstomach strips was also significantly (P � .05,ANOVA) lower than that in the normal strips ex-cept at the concentration of 10�4 M SNP (P � .22,ANOVA) (Fig 4). Responses to the relaxation of10�7, 10�6, 10�5, and 10�4 M SNP in SIS-regener-ated stomach strips were 9.1%, 30.0%, 50.2%, and

Fig 2. S100 staining to both SIS-grafted tissue (A and B)smooth muscle in the area of the previous defect, butpresumably the Schwann cells of intramuscular nerve fibthe smooth muscle cells in SIS-regenerated tissue as obsewhere Meissner’s complex (‘) in the submucosa stains fstains extensively (C).

77.0% of normal stomach strips, respectively.

EFS. Most strips (10 out of 12) in the SIS-regen-erated area showed contraction in response to EFS,although the amplitude was significantly suppressedcompared with controls (P � .01, ANOVA) (Fig 5).Maximum contractile response to EFS in SIS-regen-erated stomach strips was 12.9% of the response innormal stomach strips. In addition, although fewstrips in SIS-regenerated stomach (2 out of 11)experienced relaxation in response to EFS in the

ormal stomach wall (C). S100 staining is positive in then the underlying fibrous tissue. The cells staining are, 10�; B, 20�). Musclaris propria was not replaced by

n a normal stomach mucosa and wall with S100 stainingand Auerbach’s complex (e) in the muscularis propria

and nnot iers (Arved iocally

presence of atropine and guanethidine, all strips in

380 Ueno et al SurgerySeptember 2007

controls exhibited relaxation (Fig 6). As shown inFig 3, the concentration-response curve for the

Fig 3. Response to CCH in SIS-grafted strip and controlstrip. CCH (1 � 10�8 to 10�4 M)-induced contractions incontrol and SIS strips. CCH produced contractions in aconcentration-dependent manner. The amplitude of theresponse to CCH in SIS-regenerated strips was signifi-cantly (P � .01) lower than that in normal strips. Re-sponses to the concentration of 10�8, 10�7, 10�6, 10�5,and 10�4 M CCH in SIS-regenerated stomach strips were12.4%, 12.3%, 11.5%, 15.8%, and 18.8% of normal stom-ach strips, respectively.

Fig 4. Response to SNP in SIS-grafted strip and controlstrip. SNP (1 � 10�7 to 10�4 M)-induced relaxations incontrol and SIS strips. SNP produced relaxations in aconcentration-dependent manner. The amplitude of theresponse to SNP in SIS-regenerated strips was signifi-cantly (P � .01) lower than that in normal strips exceptat the concentration of 10�4 M (P � 0.222). Responses tothe relaxation of 10�7, 10�6, 10�5, and 10�4 M SNP inSIS-regenerated stomach strips were 9.1%, 30.0%, 50.2%,and 77.0% of normal stomach strips, respectively.

muscarinic cholinergic agonist indicated that the

ED50 for producing contraction in stomach stripsfrom control rats was approximately 1 mmol/L.When the EFS-evoked contraction was standard-ized as a percent of contraction to 1-mmol/L CCHas described in the previous studies,14,15 percentmagnitude of contractions in SIS-regeneratedstomach strips in response to EFS showed no sig-nificant difference compared with normal stomach

Fig 5. Contraction in response to EFS (85 V, 0.5 ms, 1-20Hz, 30-s trains). EFS produced contractions in a frequen-cy-dependent manner. The amplitude was significantlysuppressed compared with control (P � .01). Responsesto the electrical stimulation of 1, 2.5, 5, 10, and 20 Hz inSIS-regenerated stomach strips were 2.6%, 6.1%, 6.2%,9.2%, and 12.9% of normal stomach strips, respectively.

Fig 6. NANC relaxation in response to EFS (85 V, 0.5 ms,1-20 Hz, 30-s trains). EFS produced NANC relaxations ina frequency-dependent manner. The amplitude was sig-nificantly suppressed compared with control (P � .01).Responses to the electrical stimulation of 1, 2.5, 5, 10,and 20 Hz in SIS-regenerated stomach strips were 0%,0%, 9.3%, 12.5%, and 7.0% of normal stomach strips,respectively.

strips (P � .76, .80, .47, .60, and .64 for stimulation

Surgery Ueno et al 381Volume 142, Number 3

of 1, 2.5, 5, 10, and 20 Hz, respectively, unpaired ttest) (Fig 7).

DISCUSSIONSIS is a xenogenic membrane harvested from

the porcine small intestine in which the tunicamucosa is removed from the inner surface, and theserosa and tunica muscularis are removed from theouter surface. According to the literature,16-21 whenimplanted in vivo, SIS induces cellular responsesthat recapitulate embryogenesis and tissue regen-eration in that appropriate tissue structure andfunction are restored with minimal scar formation.

Recently, SIS has been used experimentally toreplace arteries and veins with success22-24 and hasbeen shown excellent results to serve as a scaffoldwhen used to repair the urinary tract and blad-der.14,15,17,20,25-30 Several authors have evaluatedthe function of SIS-regenerated tissue related tobladder augmentation.14,15,20,25,26 Vaught et al14

have reported the functional characteristics of SIS-induced tissue in the setting of bladder augmenta-tion in 4 rats 11 months after surgery. Theyexamined the regeneration of muscarinic, puriner-gic, and �-adrenergic receptors with CCH, �, �-meth-ylene ATP, and isoproterenol, respectively. Theywitnessed 39 � 4% recovery of contraction in re-sponse to CCH. The maximum contractile re-sponse to EFS in SIS-regenerated bladder strips was46 � 6% of the response in normal bladder strips.They determined that contractions in SIS-regener-ated bladder were significantly smaller, despite asimilar quantity of smooth muscle observed in histo-

Fig 7. Normalization of frequency response curve ofcontrol and SIS strips as percent of contraction to1-�mol/L CCH. Percent magnitude of contractions inresponse to EFS showed no significant difference be-tween SIS-regenerated strips and normal strips. Thesedata suggest that the amounts of innervation in the SIS-regenerated stomach and normal stomach were propor-tional.

logic examination. However, when each EFS-evoked

response was normalized by being expressed as apercent of the ED50 CCH-induced contraction, theamplitude of the EFS-evoked contractions in SIS-regenerated bladder and normal bladder wereequal. They speculated that one possibility for dif-ferences in agonist-induced contractile responseswas decreased receptor density and/or less-effi-cient receptor-contraction coupling in the SIS-re-generated bladder. Our findings that recovery ofresponse to SNP in SIS-regenerated stomach stripswas superior to CCH-induced contraction and thatrecovery of NANC relaxation in SIS-regeneratedtissue was inferior to that of SNP are consistent withthis hypothesis because SNP-induced relaxationdoes not require receptors, whereas a response toCCH, atropine, and guanethidine is mediated byneural receptors.

Kropp et al15 also examined the function ofSIS-induced tissue in 8 dogs 15 months after blad-der augmentation. They showed the magnitude ofthe response to CCH in regenerated bladder stripswas 43 � 6% of that in normal bladder strips andthat maximum contractile response to EFS in re-generated bladder strips was 32 � 13% of that innormal bladder strips. They also demonstrated thepresence of calcitonin gene-related peptide andsubstance P immunoreactive nerve fibers, whichare thought to be neurotransmitters of primaryafferent nerve fibers in the bladder. Based on theirresults they concluded that SIS was capable of gen-erating neurotransmitter mediated contractionand relaxation as in normal tissue and that SIS hadpromising potential for use as bladder augmenta-tion material.

We are strongly expecting that the SIS will havegrowing usefulness in gastroplasty to repair a rela-tively large defect in the stomach, such as after theresection of greater sized gastrointestinal stromaltumors, to successfully prevent the deformity of thestomach contour. In this study, CCH was also usedto infer functionally the presence of a muscarinicreceptor. Response to SNP implies a possible involve-ment of the nitric oxide pathway in the nonadrener-gic, noncholinergic (NANC) relaxation. EFS tech-niques, which possibly indicate the regeneration ofnerve terminal stimulated by ACH, were applied todemonstrate contractility through activation of in-trinsic nerves within the SIS-regenerated tissuesand the host stomach wall. We observed less re-sponse to CCH and EFS in the regenerated stom-ach wall at the time of 6 months after surgery.Microscopically, there was not enough regrowth ofsmooth-muscle cells in the proper muscle layer. Instudies of rodent bladder, Kropp et al25 reported

that at 3 months small myofibers with morphologic

382 Ueno et al SurgerySeptember 2007

and staining properties of smooth muscle werepresent and scattered throughout and that at 24and 48 weeks postaugmentation, all 3 layers of thenormal rat bladder were present and grossly andmicroscopically indistinguishable from the normalrat bladder. In another article using dogs,20 theyobserved a similar pattern of histologic findings atthe different postoperative periods. Our recentstudy applying the SIS to the defect of the stomachwall showed that the granulation tissue replaced allthe layers of the gastric wall, including the submu-cosa, muscularis propria, and subserosa 3 weeksafter the implantation.7 After 6 months, however,we found the regeneration of the smooth-musclecells in this series, which implies that the tissue re-modeling with SIS may be time-dependent. One pos-sible explanation for the poor response in the SIS-regenerated stomach is having fewer muscle fibersavailable for contraction. The return of the full con-tractile function of the SIS-regenerated stomachmay require a much longer period than 6 months.The attenuated contractile force observed in thisexperiment may represent a partial return of func-tion on a continuum of the healing process.

Another possible explanation for the differencesof the response between the SIS-regenerated blad-der and stomach is the material that was used ineach study. Other investigators applied original lab-oratory-made SIS, whereas we used a commerciallyavailable SIS. Original noncommercial SIS used byBadylak et al22 has been shown to contain angio-genic growth factors such as basic fibroblast growthfactor (bFGF) and vascular endothelial cell growthfactor (VEGF).21,31 However, we cannot determinewhat role differences in growth factors may havecontributed to our results because we did not ex-amine what growth factors are contained in com-mercially available SIS.

Another reason why the SIS-regenerated stom-ach exhibited poor functional activity and histo-logic recovery while the SIS-regenerated bladderhas been more successful is because of the differ-ences in the target organ. Despite increased clinicalinterest into the application of the prosthetic ma-terials in the gastrointestinal tract, a few studies todate have assessed the use of SIS in the esophagus,1

the small intestine,3 the biliary tract,5 the stomach,7

and the colon.32 Badylak et al1 examined the re-modeling events that occur when an extracellularmatrix scaffold derived from either small intestineor urinary bladder is used as a resorbable scaffoldfor repair of esophageal defects in a dog model.They reported that the xenogenic scaffolds usedfor repair of patch defects were reabsorbed com-

pletely within 30 to 60 days and showed replace-

ment by skeletal muscle, organized collagenousconnective tissue, and a complete and intact squa-mous epithelium. Chen et al3 evaluated the feasi-bility of using SIS as a scaffold for small bowelregeneration in an in situ xenograft model in dogs.Histologic studies revealed complete absence ofthe SIS by 3 months, and the layers of the remod-eled wall contained a mucosal epithelial layer, vary-ing amounts of smooth muscle, sheets of collagen,and a serosal covering at 6 months and beyond.Rosen et al5 applied the use of SIS to the biliarytract and reported that at 5 months SIS was com-pletely replaced with native collagen with a cover-ing biliary epithelium that was normal caliber forcanine common bile duct. Likewise, in a recentstudy by the current authors of this study, when SISwas applied to a rodent colon,32 the defect wascompletely replaced by the normal constituents(mucosa, muscle, and nerve cells) of the bowelwall.

Hori et al33 reported similar results using anacellular collagen scaffold graft reinforced withpolyglycolic acid applied to a 4 � 4-cm gastricresection in dogs 16 weeks after the operation.They observed regeneration consisting of a muco-sal layer and a submucosal layer with a thin musclelayer just beneath the mucosal layer. However, nocontraction in response to 105 M ACH was ob-served in their tissue-engineered stomach wall.They offered two explanations for the decreasedfunctional activity: (1) the histologically observedsmooth-muscle cells were insufficiently differenti-ated for contraction, and (2) the surrounding con-nective tissue was too thick and tight to allow thesmooth-muscle cells to contract. In our series, wealso cannot ignore the effect of fibrosis as a cause ofpoor muscle contraction.

In conclusion, we identified a different type ofwound healing with contraction and relaxation re-sponse using SIS in the previous stomach defect.Furthermore, the contraction ratio of EFS to ED50

of CCH was not significantly different between thegroups, which implies that the amounts of innerva-tion in the SIS-regenerated stomach and the nor-mal stomach were proportional. The attenuation ofabsolute contractile response may be from havingless muscle and, perhaps, a more primitive neuro-logic structural organization, and from fibrosis withchronic inflammation observed in the outer por-tion of the SIS-grafted tissue at 6 months aftersurgery. Thus, SIS has a potential to be both a novelphysical and a physiologic scaffolding agent in thegut. Additional evaluation will hopefully allow foruse of SIS in clinical application to gastric wall

defects.

Surgery Ueno et al 383Volume 142, Number 3

The authors are indebted to Gerald Olson, DVM, MS,and Rick Leary, RVT, for their skilled technical assistanceat Durham VA animal facility.

REFERENCES1. Badylak S, Meurling S, Chen M, Spievack A, Simmons-Byrd

A. Resorbable bioscaffold for esophageal repair in a dogmodel. J Pediatr Surg 2000;35:1097-103.

2. Kajitani M, Wadia Y, Xie H, Hinds MT, Shalaby SW, SwartzKR, et al. Use of a new elastin patch and glue for repair ofa major duodenal injury. ASAIO J 2000;46:409-14.

3. Chen MK, Badylak SF. Small bowel tissue engineering usingsmall intestinal submucosa as a scaffold. J Surg Res2001;99:352-8.

4. Kini S, Gagner M, de Csepel J, Gentileschi P, Dakin G. Abiodegradable membrane from porcine intestinal submu-cosa to reinforce the gastrojejunostomy in laparoscopicRoux-en-Y gastric bypass: preliminary report. Obes Surg2001;11:469-73.

5. Rosen M, Ponsky J, Petras R, Fanning A, Brody F, DuperierF. Small intestinal submucosa as a bioscaffold for biliarytract regeneration. Surgery 2002;132:480-6.

6. Schultz DJ, Brasel KJ, Spinelli KS, Rasmussen J, Weigelt JA.Porcine small intestine submucosa as a treatment for en-terocutaneous fistulas. J Am Coll Surg 2002;194:541-3.

7. de la Fuente SG, Gottfried MR, Lawson DC, Harris MB,Mantyh CR, Pappas TN. Evaluation of porcine-derived smallintestine submucosa as a biodegradable graft for gastroin-testinal healing. J Gastrointest Surg 2003;7:96-101.

8. Oelschlager BK, Barreca M, Chang L, Pellegrini CA. Theuse of small intestine submucosa in the repair of paraesoph-ageal hernias: initial observations of a new technique. Am JSurg 2003;186:4-8.

9. Ueno T, Pickett LC, de la Fuente SG, Lawson DC, PappasTN. Clinical application of porcine small intestinal submu-cosa in the management of infected or potentially contam-inated abdominal defects. J Gastrointest Surg 2004;8:109-12.

10. Takahashi T, Nakamura K, Itoh H, Sima AA, Owyang C.Impaired expression of nitric oxide synthase in the gastricmyenteric plexus of spontaneously diabetic rats. Gastroen-terology 1997;113:1535-44.

11. Miampamba M, Yang H, Sharkey KA, Tache Y. IntracisternalTRH analog induces Fos expression in gastric myentericneurons and glia in conscious rats. Am J Physiol GastrointestLiver Physiol 2001;280:G979-91.

12. Gonzalez-Martinez T, Perez-Pinera P, Diaz-Esnal B, Vega JA.S-100 proteins in the human peripheral nervous system.Microsc Res Tech 2003;60:633-8.

13. Castagna C, Viglietti-Panzica C, Carlo Panzica G. ProteinS100 immunoreactivity in glial cells and neurons of theJapanese quail brain. J Chem Neuroanat 2003;25:195-212.

14. Vaught JD, Kropp BP, Sawyer BD, Rippy MK, Badylak SF,Shannon HE, et al. Detrusor regeneration in the rat usingporcine small intestinal submucosal grafts: functional inner-vation and receptor expression. J Urol 1996;155:374-8.

15. Kropp BP, Sawyer BD, Shannon HE, Rippy MK, Badylak SF,Adams MC, et al. Characterization of small intestinal sub-mucosa regenerated canine detrusor: assessment of reinner-vation, in vitro compliance and contractility. J Urol1996;156:599-607.

16. Lantz GC, Badylak SF, Hiles MC, Coffey AC, Geddes LA,Kokini K, et al. Small intestinal submucosa as a vascular

graft: a review. J Invest Surg 1993;6:297-310.

17. Knapp PM, Lingeman JE, Siegel YI, Badylak SF, Demeter RJ.Biocompatibility of small-intestinal submucosa in urinarytract as augmentation cystoplasty graft and injectable sus-pension. J Endourol 1994;8:125-30.

18. Badylak SF, Tullius R, Kokini K, Shelbourne KD, KlootwykT, Voytik SL, et al. The use of xenogeneic small intestinalsubmucosa as a biomaterial for Achilles tendon repair in adog model. J Biomed Mater Res 1995;29:977-85.

19. Prevel CD, Eppley BL, Summerlin DJ, Jackson JR, McCartyM, Badylak SF. Small intestinal submucosa: utilization forrepair of rodent abdominal wall defects. Ann Plast Surg1995;35:374-80.

20. Kropp BP, Rippy MK, Badylak SF, Adams MC, Keating MA,Rink RC, et al. Regenerative urinary bladder augmentationusing small intestinal submucosa: urodynamic and his-topathologic assessment in long-term canine bladder aug-mentations. J Urol 1996;155:2098-104.

21. Voytik-Harbin SL, Brightman AO, Kraine MR, Waisner B,Badylak SF. Identification of extractable growth factorsfrom small intestinal submucosa. J Cell Biochem 1997;67:478-91.

22. Badylak SF, Lantz GC, Coffey A, Geddes LA. Small intestinalsubmucosa as a large diameter vascular graft in the dog.J Surg Res 1989;47:74-80.

23. Badylak SF, Lantz GC, Jeffries M. Prevention of reperfusioninjury in surgically induced gastric dilatation-volvulus indogs. Am J Vet Res 1990;51:294-9.

24. Sandusky GE, Jr., Badylak SF, Morff RJ, Johnson WD, LantzG. Histologic findings after in vivo placement of small in-testine submucosal vascular grafts and saphenous vein graftsin the carotid artery in dogs. Am J Pathol 1992;140:317-24.

25. Kropp BP, Eppley BL, Prevel CD, Rippy MK, Harruff RC,Badylak SF, et al. Experimental assessment of small intesti-nal submucosa as a bladder wall substitute. Urology1995;46:396-400.

26. Kropp BP, Badylak S, Thor KB. Regenerative bladder aug-mentation: a review of the initial preclinical studies withporcine small intestinal submucosa. Adv Exp Med Biol1995;385:229-35.

27. Liatsikos EN, Dinlenc CZ, Kapoor R, Alexianu M, YohannesP, Anderson AE, et al. Laparoscopic ureteral reconstructionwith small intestinal submucosa. J Endourol 2001;15:217-20.

28. Jaffe JS, Ginsberg PC, Yanoshak SJ, Costa LE, Jr, Ogbolu FN,Moyer CP, et al. Ureteral segment replacement using acircumferential small-intestinal submucosa xenogenic graft.J Invest Surg 2001;14:259-65.

29. Paterson RF, Lifshitz DA, Beck SD, Siqueira TM, Jr, ChengL, Lingeman JE, et al. Multilayered small intestinal submu-cosa is inferior to autologous bowel for laparoscopic blad-der augmentation. J Urol 2002;168:2253-7.

30. Sofer M, Rowe E, Forder DM, Denstedt JD. Ureteral seg-mental replacement using multilayer porcine small-intesti-nal submucosa. J Endourol 2002;16:27-31.

31. Voytik-Harbin SL, Brightman AO, Waisner BZ, Robinson JP,Lamar CH. Small intestinal submucosa: a tissue-derived ex-tracellular matrix that promotes tissue-specific growth anddifferentiation of cells in vitro. Tissue Eng 1998;4:157-73.

32. Ueno T, Oga A, Takahashi T, Pappas TN. Small intestinalsubmucosa (SIS) in the repair of a cecal wound in unpre-pared bowel in rats. J Gastrointest Surg. In press.

33. Hori Y, Nakamura T, Kimura D, Kaino K, Kurokawa Y,Satomi S, et al. Functional analysis of the tissue-engineered

stomach wall. Artif Organs 2002;26:868-72.