Journal of Investigative Surgery, 14:259± 265, 2001

Copyright °c 2001 Taylor & Francis0894-1939/01 $12.00 + .00

Original Research

Ureteral Segment Replacement Usinga Circumferential Small-Intestinal

Submucosa Xenogenic Graft

Jamison S. Jaffe,

Phillip C. Ginsberg,

Stephen J. Yanoshak,

Louis E. Costa Jr.,

Francis N. Ogbolu,

Christopher P. Moyer,

Charlotte H. Greene,

Leonard H. Finkelstein,

and Richard C. Harkaway

Department of Surgery,Division of Urology andDepartment of BiomedicalSciences, Philadelphia College ofOsteopathic Medicine,Philadelphia, Pennsylvania, andAlbert Einstein Medical Center,Philadelphia, Pennsylvania, USA

ABSTRACT We wished to determine whether small-intestinal submu-cosa (SIS) will epithelialize when used as a ureteral replacement material.An 11-mm segment of native ureter was excised from eight New ZealandWhite rabbits and replaced with an 11-mm porcine SIS graft, which wascircumferentially wrapped around a ureteral stent. The SIS ureteral graftswere harvested at 11 days or 35 days postimplantation and examinedgrossly and by standard light microscopy techniques. Partial epithelial-ization with the ingrowth of urothelium, smooth muscle cells, and bloodvessels was observed in the grafts harvested at 11 days postimplantation.The SIS ureteral grafts examined at 35 days postimplantation showedadditional restructuring of the smooth muscle cell layer and more or-ganized epithelialization in comparison to the SIS graft examined at 11days. After 35 days of regenerative healing, elements of all three lay-ers of the native ureter were observed within the collagen matrix of theSIS graft. No signi� cant complications were observed, but all subjects(8/8) demonstrated mild intra-abdominal adhesions. Mild collecting sys-tem dilatations were observed in 4/4 (100%) of the animals harvested at35 days and in 0/4 (0%) of the animals harvested at 11 days. We have thisdemonstrated in this preliminary study that SIS xenografts will epithelial-ize when used as a ureteral replacement material. The repair mechanismof these ureteral grafts occurred through a regenerative healing processrather than by scar formation. With further studies, this material mayprove to be a useful treatment option in patients with ureteral injuries.

KEYWORDS biocompatible materials, regeneration, small-intestinal mucosa,ureter

Received 28 June 2000;accepted 15 August 2000.

Address correspondence to Richard CHarkaway, MD, 5401 Old York Road,Klein Building Suite 500, Philadelphia,PA 19141, USA. E-mail: Harks99@aol.com

259

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

Ureteral injury may arise secondary to trauma,retroperitoneal � brosis, radiotherapy, neo-plasia, or previous surgery [1, 2]. Histor-

ically, surgical repair of long-segment ureteral in-jury has been a urologic challenge. Options currentlyavailable for ureteral repair include vascularizedbowel or bladder � aps, transureteroureterostomy, orprimary reanastomosis with psoas bladder hitch [3].To date, various prosthetic materials such as Vital-lium, tantalum, silicone, Te� on, and polytetra� u-oroethylene have been studied as possible ureteralsubstitutes without success [4–8]. These syntheticsubstitutes have been plagued with complicationsof stricture, migration, rejection, encrustation, andinfection [6, 9–12].

Small-intestinal submucosa (SIS), a biodegrad-able, collagen-based, nonimmunogenic material har-vested from the submucosal layer of intestine, hasbeen shown to have regenerative capabilities in vari-ous tissues [13–16]. SIS has been well studied in theurinary bladder and has been shown to demonstratecontractile activity and exhibit functional choliner-gic and purinergic innervation similar to the nativerat urinary bladder [17]. As a result of the structuralsimilarity between the bladder and the ureter, we feltthat the porcine SIS xenograft could potentially serveas a ureteral substitute in the surgical repair of long-segment ureteral injury. We designed a prospectivepreliminary study to determine whether porcine SISwill epithelialize when used as a ureteral replacementmaterial.

MATERIALS AND METHODS

Small Intestinal SubmucosaPreparation and Circumferential

Graft Preparation

The preparation of the SIS graft is a modi� cationof the protocol designed by Badylak et al. [13]. Theintestinal tract from a porcine model (150–250 lb)from the pyloric sphincter to the anus was obtainedfrom an abettoir in the fresh state. The entire jejunumwas thoroughly � ushed and immediately placed innormal saline solution, refrigerated, and retainedbrie� y in this manner until the graft was to be pre-pared.

Preparation of the graft was carried out in an asep-tic manner with attention to avoid accidental con-tamination of the graft material. An intact and highlyvascularized segment of jejunum was used to createthe graft. The luminal opening of the jejunum wasmaintained patent while it was incised longitudinallyto create a sheet of material. The material was thenoriented with the serosal side exposed. The upperlayer was removed by grasping the layer at a cornerof the sample and teasing it away from the serosallayer of the submucosa, allowing the two layers tobe separated. These layers, composed of the tunicaserosa and tunica muscularis externa, are not utilizedfor the � nal preparation. The remaining tissue wasinverted so that the mucosal layer was exposed. Themucosal layer was removed with gentle traction usinga brushing motion of a scalpel handle. The approx-imately 80- to 100-¹m-thick SIS graft is relativelytranslucent and consists of the remaining submu-cosa, stratum compactum, and muscularis mucosaof the tunica mucosa.

An 11-mm segment of SIS material was used tocreate the graft. An Intramedic PE number 60 polye-thylene tubing with an inside diameter of 0.03 inchesand an outside diameter of 0.048 inches was used asa ureteral stent. The SIS sheet was wrapped aroundthe ureteral stent with the mucosal side adjacent tothe stent and sewn in place using a continuous stringof sutures (6-0 Vicryl) along the longitudinal seam.The completed grafts were kept on the stents, placedin gentamicin/saline solution, and refrigerated untilsurgical implantation.

Surgical Implantation

Anesthesia was induced using sodium pentobarbi-tal (30 mg/kg iv), and ceftriaxone (50 mg/kg im) wasdelivered. An 11-mm segment of native left proximalureter was excised, which is approximately one-� fthof the total length of the rabbit ureter or equivalentto 6 cm in the adult human male. The ureteral stentwith the SIS graft in place was positioned within theureter to the level of the renal pelvis proximally andthe bladder distally. The graft was then sewn in placeusing six 7-0 prolene interrupted sutures at the proxi-mal and distal margins. Prolene was used to aid in lo-cating the graft at the time of harvest. The stents were

260 J. S. JAFFE ET AL.

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

left in for the entire length of the study to ensure ad-equate drainage of the collecting system. The rabbitswere evaluated by the veterinarian on a daily basisfor signs of infection or postoperative discomfort.

The rabbits were sacri� ced with concentratedsodium pentobarbital (2 mg for the � rst 10 lb plus1 mg per each additional 10 lb iv) at 11 or 35 dayspostoperatively. A 20-mm segment around and in-cluding the ureteral graft site was resected along withan 11-mm segment of native right ureter to use as acontrol. The SIS ureteral graft was measured at thetime of removal from the proximal to distal site ofanastomoses to measure any shrinkage that mighthave occurred.

Histological Analysis Protocol

All specimens were � xed in 10% buffered forma-lin and processed for light microscopy. Sections ofparaf� n-embedded tissue were cut at 6 ¹m on a ro-tary microtome and routinely stained with hema-toxylin and eosin stains. Representative slides of theoriginal SIS graft, 11-day and 35-day ureteral SISgraft, and native right ureter were reviewed underlight microscopy.

RESULTS

Gross Results

Eight New Zealand White domestic rabbits wereentered into the study. All eight animals remainedviable until sacri� ce at day 11 (4/8) or 35 (4/8).

The SIS grafts were all intact at harvest. No dila-tion, perforation, or shrinkage of the grafts was ob-served in any of the rabbits in the study. All of therabbits at both the 11- and 35-day harvests demon-strated a mild degree of intra-abdominal adhesions.These adhesions were not severe enough to createcomplications such as bowel or urinary obstruction.The four (100%) rabbits harvested at 35 days demon-strated a mild dilatation of the left renal collectingsystem, while the rabbits harvested at 11 days did notdemonstrate any gross evidence of collecting systemdilatation. None of the rabbits displayed any signsof systemic infection or postoperative discomfort.

Histological Results

The original SIS graft used in this study demon-strated a well-vascularized matrix of loose, collage-nous, connective tissue (Figure 1).

At 11 days postimplantation, urothelial cells wereobserved lining the luminal surface of the graft(Figure 2). The graft also contained smooth muscle� bers re� ecting a moderate degree of organization.Adipose tissue was shown to be surrounding the ad-ventitial layer of the ureteral graft. Blood vessels werenoted throughout the entire portion of the SIS graftat 11 days.

At 35 days postimplantation, smooth muscle� bers were more highly organized. They were dis-tributed in a circular and continuous fashion in var-ious locations of the graft (Figure 3). The mucosallayer was characterized by a continuous and notice-ably thickened urothelium. At day 35, a signi� cantdegree of vascularity was observed throughout theremodeling graft. After 35 days of regenerative heal-ing, elements of all three layers of the native ureter

FIGURE 1 Small-intestinal submucosa: hematoxylin andeosin (H & E), 200£. Initial appearance of SIS section priorto forming the graft. A small vessel distributed within looseconnective tissue matrix is indicated by arrow.

URETERAL REPLACEMENT WITH A SIS GRAFT 261

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

FIGURE 2 SIS ureteral graft at 11 days postsurgery: H&E,100£. Urothelium (U) and smooth muscle �bers (SM) aredemonstrated.

FIGURE 3 SIS ureteral graft at 35 days postsurgery: H&E, 200£. Smooth muscle �bers (SM) appear to be distributed in a circularand continuous manner. A pronounced urothelium (U!) is observed.

(urothelium, muscularis, and adventitia) (Figure 4)were observed within the graft.

COMMENTS

This study demonstrates that SIS ureteral graftsmay have potential in the surgical repair of ureteralinjuries. The criteria for an ideal ureteral substi-tute developed by Baum et al. [18] and summarizedby Dahms et al. [19] are presented in Table 1. Wedid not attempt to determine during this prelimi-nary study whether normal innervation or adequatepharmacologic response along with peristaltic activ-ity synchronous with host ureter occurred. As a re-sult of the stents remaining throughout the entirestudy period, we were also unable to comment on thegrafts’ ability to allow for free transport of urine or ifstricture formation occurred at the anastomotic site.These four issues will be addressed in a future studybut were not the focus of this study. Although thiswas a preliminary study concerned primarily with de-terming whether porcine SIS will epithelialize whenused as a ureteral replacement material, we were able

262 J. S. JAFFE ET AL.

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

FIGURE 4 Normal ureter: H&E, 100£. The structural orga-nization of the native ureter can be compared to the remod-eling SIS ureteral graft at 35 days.

to demonstrate that the SIS ureteral grafts adequatelyful� lled the remaining six criteria to be consideredan ideal ureteral substitute.

Our � ndings were similar to those of Kropp et al.with regard to the natural progression of regenera-

TABLE 1 Criteria for the ideal ureteral substitute

² Close histologic resemblance to normal ureter² Peristaltic activity synchronous with host ureter² Adequate blood supply² Free transport of urine² Impermeable, nonabsorptive lining² No immunologic reaction² No stone formation² Technical ease² No stricture formation at the anastomotic site² Normal innervation and adequate pharmacologic

response

Note. Developed by Baum et al. [18] and summarized by Dahms et al.[19].

tive healing that occurred with the SIS grafts usedas a bladder wall substitute, but the ureteral graft re-generation occurred at a more rapid rate [20]. Allthree layers of the ureter were observed within theSIS graft, along with vascularity at the 11-day inter-val. At day 35, these same layers were observed butthe graft was more structurally similar to the nativeureter than the graft was at day 11. One possibilitythat may explain the difference in the accelerated rateof regeneration in the ureter versus the bladder maybe the size difference between these two organs. Thewall of the ureter is thinner and contains fewer layersof smooth muscle compared to the wall of the blad-der. This difference may allow for faster ingrowth ofnative tissue into the ureteral grafts and result in anoverall faster rate of regeneration.

Dahms et al. demonstrated ingrowth of severallayers of urothelium and smooth muscle � bers at10 weeks in an acellular matrix graft that was used asan experimental ureteral replacement in rats [19]. Weobserved these same results at 11 days. This increasedrate of regeneration observed in the SIS grafts as com-pared to other collagen-based materials has been pre-viously reported [21]. SIS is composed of cytokines,structural proteins, glycoproteins, and proteoglycansthat create a three-dimensional structure, which pro-vides the necessary environment to promote cell mi-gration, cell-to-extracellular matrix interaction, andcell growth and differentiation [22]. Also, Brightmanet al. have demonstrated the presence of transform-ing growth factor-beta along with other unidenti� edgrowth factors in SIS material [23]. These growthfactors along with the three-dimensional structureof the SIS may allow for faster regenerative healingwhen compared to other collagen-based materials.

SIS has been previously demonstrated to be anunsuccessful ureteral replacement material in a studyby Shalhav et al. [24]. At 12 weeks, the graft mate-rial had a stricture formation rate of 100% in the sixpigs studied. This seems discouraging toward the useof SIS as a ureteral replacement, but strictures werealso observed in 100% of the animals in the controlgroup in this study as well. This leads one to believethat the inherent mechanism of stricture formationmay be attributed more to the laparoscopic proce-dure performed than to the SIS ureteral grafts.

URETERAL REPLACEMENT WITH A SIS GRAFT 263

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

Atala et al. have also developed a method of cre-ating new urothelial tissue in vivo [25]. This pro-cess involved removing urothelial and smooth mus-cle cells from the bladder and attaching them to abiodegradable polymer scaffolding, which was usedsuccessfully as a urethral, ureteral, and bladder sub-stitute [26–28]. After 6 weeks, the ureteral implantsdisplayed normal cellular organization consisting ofa urothelial lined lumen overlying submucosal tissueand a layer of smooth muscle [27].

The complications observed in our study weremild intra-abdominal adhesions in 8/8 (100%) andmild left renal collecting system dilatations in 4/4(100%) of the animals harvested at 35 days. The ad-hesions and collecting system dilatations were not se-vere enough to be detected prior to necropsy. We at-tribute the collecting system dilatations secondary toinadequate drainage provided by the ureteral stentsused in this study. Calculi were not observed. Dahmset al. also reported no incidence of calculi formationbut did report moderate adhesions, stent migration,and varying degrees of hydroureteronephrosis [19].However, evidence of encrustation, graft migration,postoperative urinary leakage, rejection, or infectionat the graft site was not found in either their studyor ours. One would not expect to observe complica-tions such as calculi formation or encrustation in astudy carried out for a 35-day period. These compli-cations will be better observed in a future study thatfollows the animals for a more signi� cant postoper-ative period.

Our next step is to build on the basic knowledgethat we have developed from this preliminary study.We are extending this study to the canine model.The SIS grafts will be harvested at 7-day intervals andstudied by both light and electron microscopy in or-der to better understand the natural progression andrate of regenerative healing that occurs in these SISureteral grafts. We will determine if normal innerva-tion and adequate pharmacologic response occur inthe ureteric grafts, as well as if synchronous peristalticactivity between the graft and host ureter can be ob-served. In addition, the incidence of postoperativecomplications will be more closely monitored sincethe study design incorporates a longer postoperativeperiod.

CONCLUSIONS

With the limited follow-up period in this prelimi-nary study, we were unable to demonstrate the long-term functionality of the SIS ureteral grafts, but wehave demonstrated that these xenografts will epithe-lialize when used as a ureteral replacement material.The complications observed with other ureteral pros-theses materials were not encountered with the SISgrafts, and the ureteral grafts healed through a regen-erative healing process rather than by scar formation.With further studies, this material may prove to be auseful treatment option in those patients with long-segment ureteral injuries.

REFERENCES

1. Peters JJ, Shingleton WB, Morgan D, Allen B, Fowler JE Jr.Neointimized Gore-Tex graft for ureteral prosthesis in thedog. Urology. 1996;48:379–382.

2. Gangai MP, Agee RE, Spence CR. Surgical injury to theureter. Urology. 1976;8:22–27.

3. Sabanegh ES, Downey JR, Sago AL. Long-segmentureteral replacement with expanded polytetra�uoroethy-lene grafts. Urology. 1996;48:312–316.

4. Lord JW Jr, Eckel JH. The use of Vitallium tubes in theurinary tract of the dog. J Urol. 1942;48:412–417.

5. Labash S. Experiences with tantalum tubes in the reim-plantation of the ureters into the sigmoid in dogs andhumans. J Urol. 1947;57:1010–1016.

6. Schreiber B, Homann W, Mlynek M, Mellin P. Ureteral re-placement with a new prosthesis. Trans Am Soc Artif In-tern Organs. 1979;25:61–65.

7. Ulm AH, Krauss L. Total unilateral Te�on ureteral substi-tutes in the dog. J Urol. 1960;83:575–582.

8. Baltaci S, Ozer G, Ozer E, Soygur T, Besalti O, Anafarta K.Failure of ureteral replacement with gore-tex tube grafts.Urology. 1998;51:400–403.

9. Donovan MG, Barrett DM. Ureteral prosthesis. SeminUrol. 1984;2:158–166.

10. Tremann JA, Tenchoff H. Long-term ureteral replacementprosthesis. Urology. 1978;11:347–351.

11. Wayenknecht LV, Aurert J, Sausse A. Ureteral replace-ment with synthetic prosthesis in dogs. Eur Surg Res.1972;4:131–139.

12. Boxer RJ, Johnson SF. Ureteral substitution. J Urol.1978;12:269–278.

13. Badylak SF, Lantz GC, Coffey AC, Geddes LA. Small in-testinal submucosa as a large diameter vascular graft inthe dog. J Surg Res. 1989;47:74–80.

14. Lantz GC, Badylak SF, Hiles MC, Coffey AC, GeddesLA, Kokini K, Sandusky GE, Morff MJ. Small intestinalsubmucosa as a vascular graft: A review. J Invest Surg.1993;6:297–310.

264 J. S. JAFFE ET AL.

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.

Original Research

15. Ezzell C. Pig intestine yields versatile tissue graft. Sci News.1992;141:246–247.

16. Kropp BP, Rippy MK, Badylak SF, Adams MC, Keating MA,Rink RC, Thor KB. Regenerative urinary bladder augmen-tation using small intestinal submucosa: urodynamic andhistopathologic assessment in long-term canine bladderaugmentation. J Urol. 1996;155:2098–2104.

17. Vaught JD, Kropp BP, Sawyer BD, Rippy MK, Badylak SF,Shannon HE, Thor KB. Detrusor smooth muscle regener-ation in the rat using porcine small intestinal submucosagrafts: Functional innervation and receptor expression. JUrol. 1996;155:374–378.

18. Baum N, Mobley DF, Carlton CE. Ureteral replacements.Urology. 1975;5:165–171.

19. Dahms SE, Piechota HJ, Numnes L, Dahiya R, Lue TF,Tanagho EA. Free ureteral replacement in rats: Regener-ation of ureteral wall components in the acellular matrixgraft. Urology. 1997;50:818–825.

20. Kropp BP, Eppley BL, Prevel CD, Rippy MK, Harruff RC,Badylak SF, Adams MC, Rink RC, Keating MA. Experimen-tal assessment of small intestinal submucosa as a bladderwall substitute. Urology. 1995;46:396–399.

21. Kropp BP. Small intestinal submucosa for bladder aug-mentation: A review of preclinical studies. World J Urol.1998;16:262–267.

22. Badylak, SF. Speculation (with a little evidence) for theroles of cell proliferation, differentiation, neovasculariza-

tion, and environmental stressors in SIS-induced remodel-ing. Proceedings, 1st SIS Symposium, Orlando, FL, 11–12December 1996.

23. Brightman AO, Waisner B, Klich DL, Voytik-Harbin SL. Ex-traction of a biologically inactive form of transforminggrowth factor B from small intestinal submucosa. Pro-ceedings, 1st SIS Symposium, Orlando, FL, 11–12 Decem-ber 1996.

24. Shalhav AL, Elbahnasy AM, Bercowsky E, Kovacs G,Brewer A, Maxwell KL, McDougall EM, Clayman RV.Laparoscopic replacement of urinary tract segments us-ing biodegradable materials in a large-animal model. JEndourol. 1999;13:241–244.

25. Atala A, Vacanti J, Peters C. Formation of urothelial struc-tures in vivo from dissociated cells attached to biodegrad-able polymer scaffolds in vitro. J Urol. 1992;148:658–662.

26. Cilento B, Retik A, Atala A. Urethral reconstruction usinga polymer scaffolds seeded with urothelial and smoothmuscle cells. J Urol. 1996;155:5.

27. Yoo J, Satar N, Retik A, Atala A. Ureteral replace-ment using biodegradable polymer scaffolds seeded withurothelial and smooth muscle cells. J Urol. 1995;153:375A.

28. Yoo JJ, Meng J, Oberpenning F. Bladder augmentationusing allogenic bladder submucosa seeded with cells.Urology. 1998;51:221–225.

URETERAL REPLACEMENT WITH A SIS GRAFT 265

J In

vest

Sur

g D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Q

UT

Que

ensl

and

Uni

vers

ity o

f T

ech

on 1

1/01

/14

For

pers

onal

use

onl

y.