Journal of Surgical Research 145, 105–110 (2008)

Adverse Effects Associated with the Use of Porcine Cross-LinkedCollagen Implants in an Experimental Model of Incisional Hernia Repair1

Alexander H. Petter-Puchner, M.D.,*,2 René H. Fortelny, M.D.,† Nadja Walder, C.M.,*Rainer Mittermayr, M.D.,* Wolfgang Öhlinger, M.D.,* Martijn van Griensven, M.D., Ph.D.,*

and Heinz Redl, Ph.D.*

*Ludwig Boltzmann Institute for Traumatology, Vienna, Austria in the Research Center for Traumatology of the Austrian WorkersCompensation Board; †Department of Surgery, Wilhelminenspital der Stadt Wien, Vienna, Austria

Submitted for publication January 5, 2007

doi:10.1016/j.jss.2007.03.090

Background. Porcine cross-linked collagen (PermaCol,PCL; TSL, Aldershot, United Kingdom) h a s been pro-posed as permanent biomaterial in incisional herniarepair. We evaluated the biocompatibility of PCL in anestablished animal model.

Material and methods. In 10 Sprague Dawley rats,two hernias per animal were created in the abdominalwall left and right of the linea alba (1.5 cm in diame-ter), and the peritoneum was spared. The lesions wereleft untreated for 10 days, until incisional herniasdeveloped. These defects were covered with non-perforated (out-of-the-box, n � 12) or perforated (mod-ified; n � 8) PCL (2 � 2 cm). In a first step, 12 non-perforated implants were tested in a short-termobservation period of 17 days. Eight of these non-perforated implants were fibrin sealed (0.3 mL, Tissucol;Baxter, Vienna, Austria), whereas four non-perforatedimplants were sutured with non-resorbable material.In a second step, perforations were added as modifi-cation to PCL to facilitate drainage of fluids, cell in-growth, and transgression of fibrin sealant. All perfo-rated implants were fibrin sealed and included in along-term observation period of 3 months. The obser-vation periods allowed the evaluation of the completedegradation of the fibrin sealant fixation after 2 weeksand of the implant integration in a chronic timeframe.Implant sites were analyzed macroscopically andhistologically.

Results. All PCL samples elicited strong local inflam-mation with signs of foreign body reaction. Integra-

1 Alexander H. Petter-Puchner and René H. Fortelny contributedequally to this study.

2 To whom correspondence and reprint requests should be ad-dressed at the Ludwig Boltzmann Institute for Experimental andClinical Traumatology, Donaueschingenstrasse 13, A-1200, Vienna.

E-mail: alexander.petter@lbitrauma.org.

105

tion of perforated PCL appeared limited after 3months. Three animals had to be euthanized prior tointended time points because of transcutaneous mi-gration of implants.

Conclusions. In an experimental model of incisionalhernia repair, PCL does not integrate well in the ab-dominal wall and shows poor biocompatibility. © 2008

Elsevier Inc. All rights reserved.

Key Words: porcine cross-linked collagen; biocom-patibility; rat model; hernia repair.

INTRODUCTION

During the past few years, several biomeshes (e.g.,derived from natural materials) have been introducedto hernia repair [1, 2]. Potential benefits should includesuperior biocompatibility, reduced adhesion formation,degradability, and decreased risk of infection in con-taminated operation fields. Implants processed fromanimal sources are made from porcine small intestinesubmucosa (Surgisis, SIS; Cook, Bloomington, IN), bo-vine pericard (Tutomesh, TM; Tutogen, Neunkirchen,Germany), or porcine collagen (PermaCol, PCL; TSL,Aldershot, United Kingdom) [3–5]. Human-derived im-plants also belong to this product family, but they arenot commonly used yet and more data on their poten-tial in the future is required [6–8]. Experimental in-vestigations on biomeshes are especially important asadverse effects and unfavorable properties of products,such as SIS, have been demonstrated in laboratorytrials after their introduction to clinics [3, 9]. The sit-uation is similar for cross-linked collagen (PCL) im-plants thus far, as recent experimental data indicatedlimited tissue integration, chronic inflammation, and

proposed modifications of the implant [10]. Clinical

0022-4804/08 $34.00© 2008 Elsevier Inc. All rights reserved.

106 JOURNAL OF SURGICAL RESEARCH: VOL. 145, NO. 1, MARCH 2008

studies on the reinforcement of the abdominal wallwith PCL had small numbers of patients and came tofavorable conclusions [11–13]. These conclusions werebased on clinical examination only, but no histologywas presented. However, postoperative complicationswhich would have permitted histological analysis werereported in one publication [11]. The rationale for thisstudy was to rule out potential adverse effects of PCLon the basis of sound histology, obtained in an animalmodel of incisional hernia repair.

Collagen is routinely applied as a degradable foil orfleece in different surgical fields, where it is used forsealing purposes and hemostasis [14]. In hernia repair,degradability of an implant might be unfavorable, con-sidering the growing evidence that herniosis presents ageneralized disease, resulting in a systemic weaknessof the connective tissue [15, 16]. Bearing this in mind,permanent mechanical reinforcement of the abdominalwall could be desirable. Porcine cross-linked collagen isdistinctively different from other biomesh materials,because it is non-degradable and is designed as perma-nent matrix for newly formed scar tissue comparable tomost synthetic meshes [10, 17]. The use of a hernia-specific animal model is important to evaluate the rec-ommendations to implant biomeshes, e.g., PCL andSIS, in unfavorable conditions, such as large herniasand/or contaminated wound fields in humans [1, 2, 11,12, 18]. The repair of these defects is associated withsignificant morbidities in an elderly patient population[19]. The primary aim of this study was to elucidatebiocompatibility and tissue integration of out-of-the-box, non-perforated PCL implants as well as modified,perforated ones. The modification included perforationby the investigators. These perforations should facili-tate transgression of fibrin sealant (FS) and cell in-growth and prevent formation of seroma. Fibrin seal-ing of hernia implants is currently gaining increasedinterest based on excellent results for different repairtechniques [20–22]. In this context perforation size hasbeen described as an important parameter for implantdesign and was chosen accordingly [23]. Our studygroup has previously tested degradable biomeshes(e.g., SIS implants) and synthetic meshes (TI-Meshextralight � TMxl and Vypro II � VP) in the sameanimal model and therefore comparable results arenow available for these different concepts of mesh tech-nology [3, 24].

MATERIALS AND METHODS

Male Sprague Dawley rats, weighing 400–450 g, were obtainedfrom the Institut fuer Labortierkunde und-genetik der Medizinis-chen Fakultaet der Universitaet Wien (Himberg, Austria). The cross-linked collagen implant was ordered from the manufacturer (TSL).All reagents used were of analytical grade. The animal protocol wasapproved by the authority of the city government of Vienna beforesurgery and allowed different treatments in one animal. Defects (two

per animal) were randomized to short-term (FS, n � 8 defects, versus

sutures, n � 4 defects) and long-term (only FS fixation, n � 8 defects)periods. Randomizing was done by drawing numbers (the investiga-tor from the institute not involved in the trial).

The groups were defined as follows:

1. Short-term observation period (17 days): Non-perforated PCLimplants sealed with FS (n � 8 defects)

2. Short-term observation period (17 days): Non-perforated PCLimplants sutured (n � 4 defects)

3. Long-term observation period (3 months): Perforated (modified)PCL implants sealed with FS (n � 8 defects), 20 implants in 10 rats(six animals randomized to short-term observation; four animalsrandomized to long-term observation).

The observation periods were chosen to detect early recurrencesafter full degradation of the FS within 2 weeks, as well as to assesslong-term biocompatibility and tissue integration. Non-perforatedimplants were tested to evaluate the incidence of seromas based onour experiences on fluid retention with non-perforated SIS in aprevious trial [3]. Only perforated PCL implants were scheduled forthe long-term observation period because of the hypothesis thatperforations could significantly improve both tissue integration andeffects of sealing.

Creation of Hernia

The anesthetic and surgical procedure has been described in de-tail elsewhere [3]. Briefly, rats were anesthetized with an intramus-cular injection of 100 mg/kg bodyweight (BW) Ketavet (Ketaminhy-drochloride 100 mg/mL; Pharmacia, Erlangen, Germany) and 16mg/kg BW Rompun (Xylazine-Hydrochloride; Bayer, Leverkusen,Germany). An abdominal midline skin incision was made and twolesions of 1.5 cm in diameter were dissected with a scalpel in theright and left abdominal rectus muscles. The defects were created 1.5cm below the rib cage and 1.5 cm laterally to the linea alba and theperitoneum was spared. Subsequently, the skin incision was closedwith non-resorbable suture material (Synthofil 4-O; Ethicon, Norder-stedt, Germany). One milliliter of physiological saline was adminis-tered subcutaneously to avoid dehydration. Animals were singlecaged and allowed to recover for 10 days. This interval ensuredtreatment of defects free of acute inflammatory response as observedin a previous study [3].

Treatment of Hernia

Ten days after primary surgery, rats were re-anesthetized asdescribed above. The skin incision was reopened and the defects weredebrided. Absence of defect closure as well as free accessibility to theperitoneum was verified. PCL was processed to 2 � 2 cm implants.Modified implants were produced in a standardized manner, inwhich three rows of three holes (each 3 mm in diameter) werepinched with a pliers at equal distance (5 mm). In all fibrin-sealedPCL implants, the surgical field was not manipulated for 90 s afterapplication to allow FS not only to clot, but to start cross-linking. Allsurgical procedures were conducted under sterile conditions.

Short-Term Observation (17 days)

Eight implants were sealed with 0.3 mL Tissuecol (Baxter, Vi-enna, Austria), using the Duploject (Baxter) mixing needle and fourimplants were sutured with four single stitches of non-resorbablesuture in the corners (Synthofil; Ethicon). In the non-perforatedimplants randomized to sealing, one corner was lifted with a fineforceps and FS was applied between PCL and the abdominal wall.

Long-Term Observation (3 months)

In these modified samples in which perforations allowed its trans-

gression, FS was applied directly on the implant.

107PETTER-PUCHNER ET AL.: PORCINE CROSS-LINKED COLLAGEN IMPLANTS

The skin incision was closed as described and 1 mL of physiolog-ical saline was administered subcutaneously.

Postoperative Care

Rats were kept isolated for 17 days and then caged in pairs for therest of the scheduled long-term observation period. Because of ourexperiences with SIS [3], rats were checked daily for signs of infec-tion, seroma formation, or abscess formation. Analgesic treatment inthe form of the intramuscular application of Temgesic (2 mg/kg BW;Merck, Vienna, Austria) was routinely supplied once daily for 3 dayspostoperatively after primary and re-operation. In case of an occur-rence of adverse effects, an experienced veterinarian was consultedto decide additional treatment or early euthanasia. Additional treat-ment included restart of analgesic therapy, daily application of localantiseptics, and wound dressing.

Autopsy

Animals were scheduled to be sacrificed on the 17th day (short-term observation period, non-perforated implants) or after 3 months(perforated implants) postoperatively. The animals were sacrificed indeep anesthesia by an i.v. injection of 1 mL Thiopental (1 g; Sandoz,Kundl, Austria). The skin incision was reopened and the macroscopicstatus of the defect site was evaluated.

Macroscopy

Seroma formation, signs of local inflammation, and tissue inte-gration were scored independently in situ by two investigators(A.H.P., N.W.). In case of discrepancies between investigators, thebetter score was accepted. The score used an A (no), B (modest), andC (severe alteration) scale and is based on the good findings obtainedwith synthetic meshes and negative findings with SIS in the samemodel [3, 24].

Seroma Formation

No seroma was scored as A; a seroma (encapsulation with fluid)closely adjacent to the implant was scored as B, and massive “bubble-

FIG. 1. At autopsy, PCL was not degraded but was often found tobe rolled up and dislocated, regardless of fixation technique. Manip-ulation during explantation indicated a reduced strength of the PCLimplants. (Color version of figure is available online.)

like” seroma formation as found with SIS was scored as C [3].

Local InflammationNo detectable inflammation (defined as unfavorable inflammation

with pus and debris) was scored as A; minimal amounts of debris andpus were scored as B, and abscess formation was scored as C.

Tissue IntegrationGood integration (tissue ingrowth and vascularization visible to

the naked eye) was scored as A; an implant only partly integrated(less than 50% of surface area) was scored as B, whereas no detect-able integration (e.g., no tissue penetration through perforationholes, edges of the implant not attached to abdominal wall) wasscored as C. If the local situation did not allow scoring of a param-eter, it was rated X (not definable).

Histology

After macroscopic evaluation, all samples were fixed in 10% buff-ered formaldehyde solution (Merck) and embedded in paraffin. Five-micron sections were stained with H&E. Blinded analysis and grad-ing for the following parameters were performed (W.Ö.):

● Macrophages● Lymphocytes and plasma cells● Foreign body reaction (as defined as prolonged neutrophil re-

sponse, foreign body giant cells, and necrosis)● Neovascularization through the interstices of the implant

Histological Grading

Histological grading was performed as follows: 0 � no; 1 � mod-erate; 2 � strong; and 3 � maximum alteration in comparison totissue of native rats [24].

RESULTS

Macroscopy

General Observation

Macroscopic assessment was difficult due to complexlocal findings. All implants were covered by a smeary

TABLE 1

Macroscopical Findings

Parameter

NonperforatedPCL sealed; 17d

n � 8

NonperforatedPCL sutured; 17d

n � 4

PerforatedPCL; 3 mo

n � 8

Seroma X X XInflammation 1A/7C C CTissue Integr. 1B/7C C 4B/4eE

Notes. Short-term observation period. Seroma formation: All im-plants were scored X due to the complex local findings, regardless offixation technique or observation period ( see also “local inflamma-tion”). Local inflammation: One sealed implant was scored A, allother 11 implants were scored C. The local findings were character-ized by substantial retention of pus (Fig. 1). Tissue Integration: Onesealed sample (scored as A for inflammation) was scored as B; allother PCL implants were scored C due to the lack of integration.

Long-term observation period (eE � early euthanasia). Seromaformation: X (see above). Local inflammation: All PCL implants thatcould be scored after 3 months showed clear signs of pronounced localinflammation and were scored C. Tissue Integration: Tissue integra-tion was scored B in the four sealed PCL implants, which had notbeen harvested prior to the intended date due to transcutaneous

migration. No implant was scored A.

108 JOURNAL OF SURGICAL RESEARCH: VOL. 145, NO. 1, MARCH 2008

layer of debris (Fig. 1) and became vulnerable to ma-nipulation with fine tools used for debridement andexplantation. Results of the macroscopic scoring aresummarized in detail in Table 1.

Early Euthanasia

Four perforated (of 8) and one original PCL implant(of 12) (in three animals) migrated transcutaneously,causing severe wound infection. The affected animalshad to be euthanized to prevent intolerable suffering.These perforated implants were harvested in thefourth week postoperatively, the non-perforated oneafter 7 days (Figs. 2 and 3).

FIG. 2. Five PCL implants migrated transcutaneously andcaused severe local wound infections. These defects were treatedwith antiseptic wound dressing and analgesics. However, progres-sion of these findings necessitated early euthanasia in the affectedanimals. (Color version of figure is available online.)

FIG. 3. Superficial parts of migrated PCL were firm to the touch,but parts that remained adjacent to the abdominal wall were sovulnerable that sutures used for fixation cut through the implant(black arrow; compare to Fig. 1). The white arrow points out themargin of the chronically inflamed wound. (Color version of figure is

available online.)

Histology

Short- and Long-Term Observation Periods

Histology confirmed the macroscopic observation of apronounced tissue reaction to the PCL implants. Allimplants invariably caused a local foreign body reac-tion with substantial inflammation and necrosis, re-gardless of fixation method or observation period. PCLappeared to be inpenetrable and was not degraded. Thefollowing parameters were graded as 3 (maximum al-teration) compared to native tissue in all samples: mac-rophages, lymphocytes and plasma cells, foreign bodyreaction.

Long-Term Observation Period

Perforations were insufficient to provide adequatedrainage of fluids due to obstruction and coverage bycell debris, which had appeared as a smeary layer atmacroscopic assessment. Neovascularization was notdetectable through implant interstices: 0 (Fig. 4).

DISCUSSION

Implant design is an important technical factor inthe development of recurrences and chronic pain asso-ciated with hernia repair. Macroporosity, light weight,and polyfilament structure improve tissue integrationand biocompatibility in synthetic meshes and deliverconvincing clinical results in incisonal and inguinal

FIG. 4. Histology verified the poor biocompatibility of PCL inthis setting. The histological picture was characterized by prolongedinflammation, excessive neutrophil response, and accumulation offoreign body giant cells. Large areas of necrosis around PCL weredetected. The white arrow points to PCL, which was not penetratedby cells after 3 months, while the black arrow marks the area ofmaximum inflammation and necrosis. H&E staining; magnification:�25. (Color version of figure is available online.)

hernia repair when combined with FS for fixation [23,

109PETTER-PUCHNER ET AL.: PORCINE CROSS-LINKED COLLAGEN IMPLANTS

25–27]. To our knowledge, the relevance of these char-acteristics, as well as the specific aspect of degradabil-ity, has not yet been elucidated in biomeshes. PCL waschosen because collagen is successfully used in a vari-ety of surgical fields and its design as permanent ma-trix seemed especially appropriate for hernia repair. Ina recent study, Macleod et al. reported minor chronicinflammation occurred and limited tissue integration[10]. The need for modification of the implant wassuggested by the authors. Based on these suggestionsfrom literature on PCL and our own investigations onboth biological and synthetic implant materials, theinvestigators decided to perforate the implants as po-tentially significant modification to the commercialproduct [10].

The rationale for creating perforations was to yieldbetter transgression of FS for fixation and to providedrainage to prevent seroma formation. However, PCLcaused severe local adverse effects related to poor bio-compatibility in this setting as verified histologically.The worse results in our model compared to Macleod’sfindings could be the consequence of the placement ofPCL in subcutaneous pockets without fixation in hismodel, in which the implant was not permanently inclose contact with the abdominal wall [10]. To ourmind, these discrepancies emphasize the importance ofanimal models that are specifically designed to simu-late the intended clinical use. A negative influence ofFS or sutures on tissue integration and inflammationcan be ruled out based on our previous studies [3, 24].

PCL triggered an acute inflammatory response withprogressive deterioration of the local situation duringthe whole observation period. These unfavorable prop-erties of PCL had detrimental consequences. First, im-plants were often rolled up regardless of fixation tech-nique (� dislocation, Fig. 1) at autopsy, which couldtranslate to an increased risk of recurrence of the her-nia in a clinical setting. It is likely that the sutures,which still appeared in place at autopsy, cut throughPCL due to the weakening of the collagen biomaterialunder permanent attack by inflammatory cells. Sec-ond, the PCL migrated transcutaneously, causing localwound infections leading to early euthanasia in threeanimals (Fig. 2). This finding occurred in 50% of allperforated implants and in one non-perforated im-plant. We assume that the perforations had no impacton implant migration, because signs of local infectionwere also detected in four non-perforated implantsaround the seventh to tenth day. These animals wereeuthanized on the 17th day postoperatively as sched-uled, but, in the animals randomized to long-term ob-servation, the similar symptoms aggravated continu-ously until euthanasia was necessitated after 1 month.Antibiotics were not delivered because they are notstandard therapy in clinical hernia repair [28] and

could unpredictably alter the outcome. Three-month

trials on non-perforated PCL were waived, because wehypothesized that poor biocompatibility of PCL in thissetting was responsible for the adverse effects encoun-tered [29, 30].

A critical review of existing clinical literature showsthat postoperative complications resembling our find-ings have also been described by Parker et al. in aseries of nine patients in which PCL was implanted forincisional hernia repair [11]. In one patient, transcu-taneous exposure of the implant was observed; in asecond patient, abscedation of the implant occurred butwas blamed on suture erosion [11]. Other trials onincisional hernia repair using PCL in humans haveyielded favorable results in small cohorts of patients,but none of these hernia-specific studies provided anyhistological proof of the suggested safety and integra-tion [12, 13]. In contrast, histology was obtained in areport on PCL interposition after trapeziectomy [29].This trial had to be terminated, because local foreignbody reaction necessitated re-operation and removal ofimplants [29]. Additional doubts about the biocompat-ibility of PCL have been raised by different studygroups of ophthalmologists, describing their experi-ence with PCL as orbital floor implant [30, 31]. PCL ledto chronic, granulomatous inflammation, which washistologically verified after exploratory surgery andremoval of the implant [30]. The available histologicaldata reveal similar pathomorphological features ofPCL implantation regardless of species or location.However, we are aware of the limitations of our study,considering the small group size and the observationperiods, which should be extended in future protocols.More efforts are also needed to provide histologicaldata of human-derived biomeshes (e.g., Alloderm®) toassess whether they elicit less foreign body reactionsthan animal materials [6–8]. Despite its limitations,this study allows the indirect comparison of a non-degradable biomesh (PCL) and a degradable biomesh(SIS), as well as macroporous, synthetic meshes (TMxland VP) in the same model [3, 24]. The results obtainedfrom these experimental studies using the same ani-mal model demonstrated that two distinctively differ-ent biomeshes, derived from animal sources (e.g., PCLand SIS), are inferior to synthetic implants with re-gards to clinically relevant parameters, such as bio-compatibility, tissue integration, foreign body reaction,and prolonged inflammatory response.

CONCLUSION

In summary, we report that the implantation of PCLresulted in severe adverse effects in experimental her-nia repair. Inflammation, infection, and implant mi-gration due to poor biocompatibility were observed dur-

ing a subacute and chronic observation period.

110 JOURNAL OF SURGICAL RESEARCH: VOL. 145, NO. 1, MARCH 2008

REFERENCES1. Liyanage SH, Purohit GS, Frye JN, et al. Anterior abdominal

wall reconstruction with a Permacol implant. J Plast ReconstrAnesthet Surg 2006;59:553.

2. Helton WS, Fisichella PM, Berger R, et al. Short-term outcomeswith small intestinal submucosa for ventral abdominal hernia.Arch Surg 2005;140:549.

3. Petter-Puchner AH, Fortelny RH, Mittermayr R, et al. Adverseeffects of porcine small intestine submucosa implants in exper-imental ventral hernia repair. Surg Endosc 2006;20:942.

4. van Tuil C, Saxena AK, Willital GH. Experience with manage-ment of anterior abdominal wall defects using bovine pericard.Hernia 2006;10:41.

5. Ko R, Kazacos EA, Snyder S, et al. Tensile strength comparisonof small intestinal submucosa body wall repair. J Surg Res2006;135:9.

6. Ringley CD, Bochkarev V, Ahmed SI, et al. Laparoscopic hiatalhernia repair with human acellular dermal matrix patch: Ourinitial experience. Am J Surg 2006;192:767.

7. Kim H, Bruen K, Vargo D. Acellular dermal matrix in themanagement of high-risk abdominal wall defects. Am J Surg2006;192:705.

8. Schuster R, Singh J, Safadi BY, et al. The use of acellulardermal matrix for contaminated abdominal wall defects: Woundstatus predicts success. Am J Surg 2006;192:594.

9. Poulose BK, Scholz S, Moore DE, et al. Physiologic properties ofsmall intestine submucosa. J Surg Res 2005;123:262.

10. Macleod TM, Williams G, Sanders R, et al. Histological evalu-ation of Permacol as a subcutaneous implant over a 20-weekperiod in the rat model. Br J Plast Surg 2005;58:518.

11. Parker DM, Armstrong PJ, Frizzi JD, et al. Porcine dermalcollagen (Permacol) for abdominal wall reconstruction. CurrSurg 2006;63:255.

12. Armellino MF, De Stefano G, Scardi F, et al. [Use of Permacolin complicated incisional hernia]. Chir Ital 2006;58:627.

13. Catena F, Ansaloni L, Gazzotti F, et al. Use of porcine dermalcollagen graft (Permacol) for hernia repair in contaminatedfields. Hernia 2007;11:57.

14. Cappabianca P, Esposito F, Cavallo LM, et al. Use of equinecollagen foil as dura mater substitute in endoscopic endonasaltranssphenoidal surgery. Surg Neurol 2006;65:144.

15. Jansen PL, Mertens PP, Klinge U, et al. The biology of herniaformation. Surgery 2004;136:1.

16. Klinge U, Si ZY, Zheng H, et al. Collagen I/III and matrixmetalloproteinases (MMP) 1 and 13 in the fascia of patients

with incisional hernias. J Invest Surg 2001;14:47.

17. Soiderer EE, Lantz GC, Kazacos EA, et al. Morphologic study ofthree collagen materials for body wall repair. J Surg Res 2004;118:161.

18. Catena F, Ansaloni L, Gazzotti F, et al. Use of porcine dermalcollagen graft (Permacol) for hernia repair in contaminatedfields. Hernia 2006; Epub ahead of print.

19. Dunne JR, Malone DL, Tracy JK, et al. Abdominal wall hernias:Risk factors for infection and resource utilization. J Surg Res2003;111:78.

20. Novik B, Hagedorn S, Mork UB, et al. Fibrin glue for securingthe mesh in laparoscopic totally extraperitoneal inguinal her-nia repair: A study with a 40-month prospective follow-up pe-riod. Surg Endosc 2006;20:462.

21. Benizri EI, Rahili A, Avallone S, et al. Open inguinal herniarepair by plug and patch: The value of fibrin sealant fixation.Hernia 2006;10:389.

22. Olmi S, Erba L, Bertolini A, et al. [Use of fibrin glue (Tissucol)for mesh fixation in laparoscopic transabdominal hernia re-pair]. Chir Ital 2005;57:753.

23. Schmidbauer S, Ladurner R, Hallfeldt KK, et al. Heavy-weightversus low-weight polypropylene meshes for open sublay meshrepair of incisional hernia. Eur J Med Res 2005;10:247.

24. Petter-Puchner AH, Fortelny R, Mittermayr R, et al. Fibrinsealing versus stapling of hernia meshes in an onlay model inthe rat. Hernia 2005;9:322.

25. Lovisetto F, Zonta S, Rota E, et al. Use of human fibrin glue(Tissucol) versus staples for mesh fixation in laparoscopictransabdominal preperitoneal hernioplasty: A prospective, ran-domized study. Ann Surg 2007;245:222.

26. Schug-Pass C, Tamme C, Tannapfel A, et al. A lightweightpolypropylene mesh (TiMesh) for laparoscopic intraperitonealrepair of abdominal wall hernias: Comparison of biocompatibil-ity with the DualMesh in an experimental study using theporcine model. Surg Endosc 2006;20:402.

27. Rosch R, Junge K, Quester R, et al. Vypro II mesh in herniarepair: Impact of polyglactin on long-term incorporation in rats.Eur Surg Res 2003;35:445.

28. Sanchez-Manuel FJ, Seco-Gil JL. Antibiotic prophylaxis forhernia repair. Cochrane Database Syst Rev 2003;CD003769.

29. Belcher HJ, Zic R. Adverse effect of porcine collagen interposi-tion after trapeziectomy: a comparative study. J Hand Surg [Br]2001;26:159.

30. Cheung D, Brown L, Sampath R. Localized inferior orbitalfibrosis associated with porcine dermal collagen xenograft or-bital floor implant. Ophthal Plast Reconstr Surg 2004;20:257.

31. Alwitry A, Burns SJ, Abercrombie LC. Orbital implant expo-sure treatment with porcine dermal collagen patching. Orbit

2006;25:253.