JOURNAL OF SURGICAL RESEARCH 68, 126–132 (1997)ARTICLE NO. JR964979

Effects of Physical Barriers in Prevention of Adhesions:An Incisional Hernia Model in Rats1

AHMET ALPONAT, M.D.,* SATISH R. LAKSHMINARASAPPA, MBBS,* MING TEH, M.D.,†ANDREA RAJNAKOVA, M.D.,* SHABBIR MOOCHHALA, PH.D.,* PETER M. Y. GOH, M.D.,*

AND STEVEN T. F. CHAN, PH.D.*,2

*Department of Surgery and †Department of Pathology, National University Hospital, Singapore

Submitted for publication July 22, 1996

sponsible for about 60% of bowel obstructions andBackground: Adhesion formation between viscera 20% of all infertility with substantial costs associ-

and mesh is almost inevitable following incisional ated with adhesiolysis and hospitalizations [2–4].hernia repair with prosthetic mesh. Such adhesions Foreign materials such as prosthetic mesh whichmay lead to intestinal obstruction and enterocuta- are used to bridge large ventral abdominal incisionalneous fistulae formation and make further laparoto- hernias represent a strong stimulus for the develop-mies extremely difficult. Sodium carboxymethylcel- ment of permanent adhesions. Following repair of anlulose (SCMC) and Interceed TC7 (oxidized regener- incisional hernia with a mesh, any further operationated cellulose) as physical barriers have been shown in the abdomen may be extremely difficult due toto be effective in reducing postoperative adhesions. adhesion formation between the mesh and viscera.Materials and methods: To evaluate the effects of

However, a second repair may be required in 10 toSCMC and Interceed TC7, we used an incisional her-12% of the cases due to reherniation [5, 6]. Moreover,nia model in rats. A ventral abdominal defect (15 1reoperation may be needed due to complications re-25 mm) was created in each of 36 male rats whichlated to previous hernia repair such as intestinalwere then divided into three equal groups. In Groupobstruction and fistulae formation or to other disor-I (control) the defect was repaired with polypropyl-ders necessitating surgery in the abdominal cavity.ene mesh (PPM) only; in Group II the defect was re-

Various attempts have been made to prevent adhe-paired after a layer of Interceed TC7 was laid oversion formation following abdominal surgery. Thesethe viscera with Interceed TC7-covered PPM; ininclude the use of corticosteroids, nonsteroidal anti-Group III the defect was repaired after a layer of

SCMC was laid over the viscera with SCMC-coated inflammatory drugs, dextran, anticoagulants, tissuePPM. Six of the animals from each group were sacri- plasminogen activators, and physical barriers [7–9].ficed at Postoperative Day 7 and the adhesions were However, in most instances initial enthusiasm fadedscored. The remaining 6 were sacrificed at Day 30 when clinical use and further experimental studiesand histological evaluation was made in addition to failed to show the efficacy of these agents. More re-the adhesion score. Results: Animals in the SCMC- cently, promising results with the use of sodium car-treated group developed significantly less adhesions boxymethylcellulose (SCMC), a water-soluble, ab-(P Å 0.0002) compared with control and Interceed sorbable, substitute polysaccharide [10–16], andTC7-treated groups. However, histological analysis Interceed TC7, a biocompatible and absorbable oxi-revealed poor fibroblast proliferation with impaired dized regenerated cellulose in a knit pattern, havewound healing in the SCMC group. Conclusion:

been reported in animal and clinical studies [17–21].SCMC prevented adhesion formation but seriouslyIn this study, we used an experimentally createdimpaired wound healing, and Interceed TC7 was in-

abdominal wall defect model in rats. The defect waseffective in preventing adhesion in this model. q 1997repaired with polypropylene mesh (PPM) in all ani-Academic Pressmals studied. The effects of physical barriers, a ly-ophilized form of SCMC as coating on the mesh and

INTRODUCTION Interceed TC7 as a cover on the mesh, were thenSixty to 95% of patients develop adhesions follow- evaluated on adhesion formation.

ing major abdominal surgery [1]. Adhesions are re-

MATERIALS AND METHODS1 This study was supported by a grant (RP 3950343) from the Na-tional Medical Research Council, Singapore.

2 To whom correspondence should be addressed at Department of Animal experiments were conducted in the Animal Holding Unitof the National University of Singapore in accordance with usualSurgery, National University Hospital, 5 Lower Kent Ridge Road,

119704 Singapore. Fax: 65 7778427. guidelines.

1260022-4804/97 $25.00Copyright q 1997 by Academic PressAll rights of reproduction in any form reserved.

AID JSR 4979 / 6n1a$$$$$1 05-01-97 14:39:48 srga

127ALPONAT ET AL.: PREVENTION OF ADHESIONS USING PHYSICAL BARRIERS

TABLE 2TABLE 1

Adhesion Scores in Different Groups of AnimalsGrading of Adhesions

Extent Control Interceed TC7 SCMCScore (n Å 12) (n Å 12) (n Å 10)(% surface of mesh involvement) Adhesion score

None 0 0 0 0 91 2 2 0£25 1

£50 2 2 0 0 03 2 3 0£75 3

ú75 4 4 8 7 1

Median 4 4 0(range) (1–4) (1–4) (0–4)*

An abdominal wall defect was created in rats and subsequentlyrepaired by a knitted monofilament PPM (Surgipromesh, Autosuture * P Å 0.0002.Co., Norwalk, USA). A total number of 36 male Sprague–Dawleyrats (250–320 g) were fasted overnight prior to surgery and dividedinto three main groups. In Group I (control) the defect was repaired

on the underlying viscera to prevent probable adhesions to the su-with PPM only; in Group II the defect was repaired after a layer oftures fixing the mesh to the abdominal wall. The skin was closedInterceed TC7 was laid over the abdominal viscera with Interceedwith continuous sutures of polypropylene 3-0.TC7-covered PPM; in Group III the defect was repaired after a layer

Six of the animals (subgroup A) from each of the three main groupsof lyophilized SCMC was laid over the abdominal viscera with SCMC-were sacrificed at 7 days following mesh repair and the remaining 6coated PPM. Each group was then divided into two subgroups (A(subgroup B) at 30 days using an overdose of ether. The subcutaneousand B), the former (A) were sacrificed at 7 days and adhesions wereplane was initially exposed and the scar tissue over the mesh wasscored, the latter group (B) were sacrificed at 30 days and an evalua-examined. Then the inner surface of the mesh and abdominal cavitytion of wound healing was made in addition to the adhesion score.were inspected via a U-shaped incision (see Figs. 1 and 2). Adhesions

Preparation of SCMC-coated PPM. A 1.6% aqueous solution of between the mesh and intraabdominal viscera were scored by anSCMC of viscosity 1500 to 3000 cps (Sigma Chemical Co., St. Louis, independent investigator, using the modified classification devisedMO) was autoclaved and cooled to 377C. Polypropylene mesh sized by Diamond, with a scale of 0 to 4 with regard to percentage of15 1 25 mm was placed in 60-mm tissue culture dishes and 10 ml adhesions involving the prosthetic mesh surface [12] (Table 1).solution of SCMC was pipetted over the mesh. The SCMC solution

Histological analysis. One to two blocks of abdominal wall con-containing mesh inside was allowed to gel at 47C for 12 hr and thentaining mesh were sampled and sections were cut at 4 mm eachcooled to 0707C for 3 hr. The frozen SCMC solution was then lyophi-and stained with hematoxylin and eosin (H&E). Masson’s trichromelized (Edwards Super Modulyo vacuum system) for 3 days at 0607Cstains were used to demonstrate collagen tissue.and 08 mbarr to a dry spongelike wafer-containing mesh. In addi-

Statistical analysis. Comparisons of adhesion scores betweention, some plain SCMC wafers were prepared with no embeddedPostoperative Day 7 (subgroup A) and Day 30 (subgroup B) werePPM [15]. The final amount of SCMC wafer used in each animal wasanalyzed by Mann–Whitney U test. Kruskal–Wallis test was useda lyophilized form of 5 ml of 1.6% SCMC gel.for statistical analysis between control, Interceed TC7, and SCMCPreparation of Interceed TC 7-covered PPM. PPM of the samegroups. P õ 0.05 was considered significant.size (15 1 25 mm) was covered by slightly larger tailored Interceed

TC7 (Johnson & Johnson Patient Care Inc., New Brunswick, NJ)and then fixed to the PPM at each corner by separate 4-0 polypropyl- RESULTSene sutures (Surgipro, Autosuture Co.) to prevent displacement.

Operative procedure. Under light ether anesthesia each animal Two of the animals in the SCMC group died; nec-was shaved with an electrical clipper and the abdomen painted with

ropsy did not reveal the cause of the deaths. In theiodine solution. A 4-cm midline incision was made and subcutaneousanimals sacrificed on the 7th postoperative day, SCMCdissection was carried out to expose at least 30 1 40 mm of anterior

abdominal wall. The abdomen was entered at the linea alba 10 mm sponges were found to be gellike over the mesh. Fur-distal to the xiphoid and a full thickness layer of the anterior abdomi- thermore, small amounts of free-floating SCMC werenal wall 15 1 25 mm in size was excised. After precise hemostasis found in the peritoneal cavity, though not in the ani-of cut edges by pressure, PPM, SCMC-coated PPM, and Interceed

mals sacrificed at 30 days. Interceed TC7 was not de-TC7-covered PPM were sutured to the inner surface of the abdominaltectable at Postoperative Day 7.wall, deep to the peritoneum in control, Interceed TC7-treated and

SCMC-treated groups, respectively. The PPM was fixed to the defect The results of adhesion scoring are summarized inwith six separate sutures of 4-0 polypropylene in each animal. Each Table 2. All control animals developed some degree ofsuture run parallel to the edge of the PPM and took 3 to 4 mm bite adhesions to the mesh indicating that adhesions areof the mesh, and the space of about 5 to 6 mm was left between the

inevitable in the presence of PPM (Fig. 1). There wassutures. Prior to application of coated meshes, a plain SCMC spongeor Interceed TC7 sized slightly larger than the mesh was laid down no difference in adhesion scores between subgroups A

FIG. 1. Grade 4 adhesions (control group). Note dense adhesions between PPM and intraabdominal viscus.FIG. 2. Grade 0 adhesions (SCMC group). The entire surface of polypropylene mesh (PPM) is free of adhesions. Arrows indicate

herniation. Note the omentum herniating between sutures. The suture in the upper corner was pulled out intentionally to show the absenceof host tissue incorporation.

FIG. 3. One month after the introduction of polypropylene mesh (PPM) (control group). Note that there is good wound healing. Thepresence of PPM has excited a foreign body reaction (arrows) (H&E stain, 1001) (f, fibrosis; m, muscle).

FIG. 4. One month after the introduction of Interceed TC7-covered polypropylene mesh (PPM). There is good wound healing with well-formed fibrosis (f) at the wound site. In addition, aggregates of foamy histiocytes (arrows) are present but this does not seem to impair thehealing process (H&E stain, 1001) (m, muscle).

AID JSR 4979 / 6n1a$$$$$2 05-01-97 14:39:48 srga

128 JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 2, MARCH 1997

AID JSR 4979 / 6n1a$$4979 05-01-97 14:39:48 srga

129ALPONAT ET AL.: PREVENTION OF ADHESIONS USING PHYSICAL BARRIERS

AID JSR 4979 / 6n1a$$4979 05-01-97 14:39:48 srga

130 JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 2, MARCH 1997

and B in any of the three main groups (P Å 0.33, 0.36, When a defect is made in the peritoneum, the entireand 0.31 for control, Interceed TC7, and SCMC groups, surface becomes epithelialized in 5 to 8 days and afterrespectively). When the 7th and 30th day postoperative this period adhesions are unlikely to form [8]. There-adhesion scores were evaluated together, adhesion for- fore, a physical barrier between traumatized surfacesmation was significantly less in SCMC-treated animals can be effective if it can remain in place during thiscompared to control and Interceed TC7 groups (P Å early period. In the present study, there were no differ-0.0002). Interceed TC7-treated animals developed the ences in adhesion scores in any of the groups on Postop-most extensive and dense adhesions and their scores erative Days 7 and 30.did not differ significantly from the control group. In the present study, SCMC prevented adhesion for-

At surgery, on macroscopic evaluation the PPM was mation in 90% of the animals. The tensile strengthfound to be firmly adherent to the subcutaneous planes of the wound was not studied; however, the histologicwith healthy-looking scar tissue on the external surface analysis suggested that wound healing was severelyin both control and Interceed TC7 groups on the 30th impaired in SCMC-treated animals compared to bothpostoperative day. However, SCMC-coated PPM control and Interceed TC7-treated animals. On exami-showed no adherence to the subcutaneous planes and nation, on Postoperative Day 30 in the SCMC-treatedno fibrotic scar tissue was seen covering the external animals host tissue was not incorporated in the mesh.surface. Four of five animals in this group developed The animals in the control group and Interceed TC7herniation between sutures (Fig. 2). group had rapid incorporation of host tissue in the PPM

Histological analysis of H&E and Masson’s tri- and all developed well-formed scar tissue over thechrome-stained specimens revealed good fibroblast pro- mesh. Histological examination revealed that thereliferation and collagen accumulation in the wound site were sheets of foamy histiocytes in the vicinity ofresulting in well-formed scar tissue in the control and SCMC-coated PPM. Smaller aggregation of foamy his-Interceed TC7-treated animals (Figs. 3 and 4). In the tiocytes was also present in those animals treated withSCMC group, epithelialization of the inner surface of Interceed TC7 but did not appear to interfere with fi-the mesh was complete in 90% of animals without evi- broblast proliferation and fibrosis.dence of adhesion formation. However, there were Wound healing is a highly regulated sequence of cel-sheets of foamy histiocytes and impaired fibroblast pro- lular events leading to reconstitution of the integrityliferation in the wound site resulting in poor and/or of a tissue following injury. Laying down of collagenabsent collagen accumulation implying poor wound is regulated by the interaction of fibroblasts with thehealing (Figs. 5a and 5b), with resultant failure of mesh surrounding extracellular matrix and by cytokines.repair for the ventral hernias in the SCMC group. This study suggests that the presence of sheets of

foamy histiocytes is associated with poor wound heal-DISCUSSION ing. Whether SCMC inhibits the release of fibroblast-

stimulating cytokines from platelets, endothelial cells,The mechanism of action by which SCMC reducesand inflammatory cells or whether SCMC induces theadhesion formation is not clear. It has been found thatrelease of rapidly acting inhibitors of fibroblast growthSCMC when implanted intraperitoneally attracts fluidfrom macrophages remains speculative.in its surrounding and thereby prevents serosa to se-

Although no toxic effect of SCMC has been re-rosa or serosa to peritoneal contact—the so-called ‘‘hy-ported, in the present study, 2 of the 12 rats fromdroflotation effect.’’ In addition, there is evidence tothe SCMC group died postoperatively of unknownsuggest that SCMC coats intraperitoneal surfaces andcauses. Diamond et al. have reported similar deathsreduces the direct apposition of traumatized struc-in high dosage of SCMC-treated rabbits [13].tures—the so-called ‘‘siliconizing effect’’ [10, 11].

Wuster et al. have shown in rats that 12 ml of 1%The other proposed mechanism of action of SCMCSCMC is the optimal dose needed to prevent initialis through its effect on fibroblastic and cellular activi-adhesions. The same dose could prevent reformationties [15, 22]. Hemadeh et al. have shown that SCMCof adhesions after surgical lysis [14]. In the presentin combination with vitamin E supplementation wasstudy, a lyophilized form of 5 ml of 1.6% SCMC geleffective in preventing adhesions [22]. They specu-was used in each animal and this dose was found tolated that this combination might affect fibroblastsbe adequate to prevent adhesions to the mesh.through different pathways and show synergistic ef-

Interceed TC7 has been shown to prevent or atfect. Ryan and Sax have observed marked suppres-least to provide a graded reduction in adhesion for-sion of serosal fibroblast growth in SCMC-treated ani-mation after abdominopelvic surgery [17–21]. How-mals in a cecal abrasion model. However, in the inever, in other studies it has been shown to be ineffec-vitro component of their study they were unable totive or even cause de novo adhesion formation [23,show significant changes in fibroblast growth and col-

lagen formation [15]. 24]. Best et al. have shown that Interceed TC7 was

FIG. 5. (a) There are sheets of foamy histiocytes but no fibrosis at the wound site 1 month after the introduction of SCMC-coatedpolypropylene mesh (PPM) (H&E stain, 1501) (m, muscle). (b) Note the absence of collagen fibers along the wound site. Arrow indicates aforeign body giant cell reacting to the SCMC (Masson’s trichrome stain, 2501) (m, muscle).

AID JSR 4979 / 6n1a$$$$$2 05-01-97 14:39:48 srga

131ALPONAT ET AL.: PREVENTION OF ADHESIONS USING PHYSICAL BARRIERS

AID JSR 4979 / 6n1a$$4979 05-01-97 14:39:48 srga

132 JOURNAL OF SURGICAL RESEARCH: VOL. 68, NO. 2, MARCH 1997

12. Diamond, M. P., Linsky, C. B., Cunningham, T. J., Constantine,ineffective in the presence of foreign bodies includingB., diZerega, G. J., and DeCherney, A. H. A model for side wallabsorbable sutures [25]. It may be that the mesh actsadhesions in the raabbits: Reduction by an absorbable barrier.as a foreign body and causes lasting severe tissue Microsurgery 8: 197, 1987.

insult which remains after absorption of Interceed 13. Diamond, M. P., DeCherney, A. H., Linsky, C. B., Cunningham,TC7, thereby rendering it ineffective in the present T., and Constantre, B. Assessment of carboxymethylcellulose

and %32 Dextrom 76 for prevention of adhesions in a rabbitmodel.uterine horn model. Int. J. Fertil. 33(4): 278, 1988.In conclusion, Interceed TC7 was found to be inef-

14. Wuster, S. H., Bonet, V., Mayberry, A., Hadolinott, M., Wil-fective in preventing adhesion in this model. SCMCliams, T., and Chassdry, I. H. Intraperitoneal sodium carboxy-significantly reduced adhesion formations and en- methylcellulose administration prevents reformation of perito-

abled the entire inner prosthetic surface to reepithe- neal adhesions following surgical lysis. J. Surg. Res. 59: 97,lialize in 7 days but it significantly impaired abdomi- 1995.nal wound healing, most probably by affecting fi- 15. Ryan, C. K., and Sax, H. C. Evaluation of a carboxymethylcellu-

lose sponge for prevention of postoperative adhesions. Am. J.broblastic activity. Further studies are needed toSurg. 169: 154, 1995.determine the precise mode of action of SCMC on

16. Becker, J. M., Dayton, M. T., Fazio, V. W., Beck, D. E., Stryker,adhesion formation and its effects on fibroblastic ac-S. J., Wexner, S. D., Wolff, B. G., Roberts, P. L., Smith, L. E.,tivity. Moore, M., and Beart, R. W. Sodium hyaluronate-based biore-sorbable membrane (HAL-F) in the prevention of postoperativeabdominal adhesion: A prospective randomised double-blindedREFERENCESmulticentre study. Br. J. Surg. 82(Suppl. 1), 1995.

17. Diamond, M. P., Cunningham, T. J., Linsky, C. B., and DeCher-1. Weibel, M. A., and Majno, G. Peritoneal adhesions and theirney, A. H. Laparoscopic application of Interceed (TC 7) in therelation to abdominal surgery. Am. J. Surg. 126: 345, 1973.pig. J. Gynecol. Surg. 5(2): 145, 1989.2. Ellis, H. The causes and prevention of intestinal adhesion. Br.

18. Interceed (TC 7) Adhesion Barrier Study Groups. PreventionJ. Surg. 69: 241, 1982.of postsurgical adhesions by Interceed (TC7), an absorbable ad-3. Soules, M. R., Dennis, L., Boserge, A., and Moore, D. E. Thehesion barrier: A prospective, randomized multicenter clinicalprevention of postoperative adhesions: An animal study com-study. Fertil. Steril. 51(6): 933, 1989.paring barrier methods with dextran 70. Am. J. Obstet. Gynecol.

19. Diamond, M. P., Linsky, C. B., Cunningham, T. J., Kamp, L.,143(7): 829, 1982.Pines, E., DeCherney, A. H., and diZerga, G. S. Synergistic ef-4. Ray, N. F., Larser, J. W., Stillian, R. J., and Jacobs, R. J. Eco- fects of Interceed (TC 7) and heparin in reducing adhesion for-nomic impact of hospitalization for lower abdominal adhesio- mation in the rabbit uterine horn model. Fertil. Steril. 55(2):lysis in the US in 1988. Surg. Gynecol. Obstet. 176: 271, 1993. 389, 1991.

5. Bauer, S. S., Salky, B. A., Gelent, I. M., and Kreel, I. Repair of 20. Sekibes, K., and the Obstetric and Gynaecology Adhesion Pre-large abdominal wall defects with expanded polytetrafluroeth- vention Committee. Use of Interceed (TC 7) absorbable adhe-ylene (PTFE). Ann. Surg. 206(6): 765, 1987. sion barrier to reduce postoperative adhesion reformation in

6. Larson, G. M., and Harrower, H. W. Plastic mesh repair of inci- infertility and endometriosis surgery. Obstet. Gynecol. 79(4):sional hernia. Am. J. Surg. 135: 559, 1978. 518, 1992.

7. Vipond, M. N., Whawell, A. A., Scott-Coombes, D., Thompson, 21. Nair, J. O., Pulley, D., Scanlan, K., Hinshaw, J. R., and Lanza-J. N., and Dudley, A. F. H. Experimental adhesion prophylaxis fame, R. J. Reduction of postoperetive adhesions to Marlexwith recombinant tissue plasminogen activator. Ann. R. Coll. mesh using experimental adhesion barriers in rats. J. Laparo-Surg. Engl. 76: 412–415, 1994. endosc. Surg. 3(2): 187, 1993.

8. diZerega, G. S. Contemporary adhesion prevention. Fertil. 22. Hemadeh, O., Chilukuri, S., Bonet, V., Hussein, S., and Chaw-Steril. 612: 219, 1994. diy, I. H. Prevention of peritoneal adhesions by administration

of carboxymethylcellulose and oral vitamin E. Surgery 114(5):9. Jennifer, K., Chowdhury, S. M., Sawhney, A. S., Pathok, C. P.,907, 1993.Dunn, R. C., and Hubbell, J. A. Prevention of postoperative ad-

hesions in the rat by in situ photopolymerization of bioabsorb- 23. Pangidas, K., and Tulands, T. Effects of Rigeris lactate, In-able hydrogel barriers. Obstet. Gynecol. 83(1): 59, 1994. terceed (TC7) and Gore-Tex surgical membrane on postopera-

tive surgical adhesion formation. Fertil. Steril. 57(1): 199, 1992.10. Elkins, T. E., Burns, R. S., Ritter, J. L., Ling, F. W., Alokas,R. A., Hause, C. A., and Malinak, L. R. Adhesion prevention by 24. Haney, A. F., and Doty, E. Murine peritoneal injury and de novosolution of sodium carboxymethylcellulose in the rat I. Fertil. adhesion formation caused by oxidised regenerated celluloseSteril. 41(6): 926, 1984. (INTERCEED TC 7) but not expanded polytetrafluoroethylene

(Gore-Tex surgical membrane). Fertil. Steril. 57(1): 202, 1992.11. Elkins, T. E., Ling, F. W., Alokas, R. A., Abdella, T. N., Homsey,C. A., and Malinak, L. R. Adhesion prevention by solutions of 25. Best, C. L., Rittenhause, D., and Sueldo, C. E. A comparison of

TC 7 and 32% Dextran for prevention of postoperative adhe-sodium carboxymethylcellulose in the rat II. Fertil. Steril. 41(6):929, 1984. sions in hamster. Obstet. Gynecol. 78: 858, 1991.

AID JSR 4979 / 6n1a$$$$$3 05-01-97 14:39:48 srga