AACT

IwCalowe

nA

CFgppn

RbFGCCe

©P

bdominal Wall Hernia Repair:Comparison of Sepramesh and Parietex

omposite Mesh in a Rabbit Hernia Modelimothy W Judge, MD, David M Parker, MD, Robert C Dinsmore, MD, FACS

BACKGROUND: This study compared Parietex composite mesh (PCM) with Sepramesh (SM) in terms of strength oftissue incorporation, adhesion formation, and mesh shrinkage, using an animal model.

STUDY DESIGN: A two-phase, prospective, randomized study using 44 New Zealand white rabbits. Each animalunderwent creation of a standardized ventral hernia defect, followed by repair using either SMor PCM. Half of each group was sacrificed and examined at 1 month, and the remainder at 5months. Outcomes measurements were strength of incorporation (SOI), type and area ofadhesions (AA), and mesh shrinkage.

RESULTS: SOI for PCM was much greater than for SM, both at 1 month (60.8 N versus 42.6 N) and 5months (70.9 N versus 31.5 N). The incidence of bowel adhesions was lower with PCM thanSM, both at 1 month (1 versus 6) and at 5 months (0 versus 4). At 5 months, PCM demon-strated lower AA, both as a percentage of the mesh (5.6% versus 12.8%) and in terms of absolutearea involved (321 mm2 versus 840 mm2). PCM underwent considerably more shrinkage thanSM, at both 1 month (38.2% versus 18.1%) and 5 months (17.4% versus 6.1%).

CONCLUSIONS: PCM demonstrated a substantially stronger SOI, which improved over time, and SOI of SMdecreased. PCM was also superior in terms of adhesion prevention, but underwent considerablymore shrinkage in this experimental model. ( J Am Coll Surg 2007;204:276–281. © 2007 by

the American College of Surgeons)

irtTsWsncf

ratstbutdh

ncisional hernia repair is a common surgical procedure,ith over 10,000 performed in the United States in 2003.omplications associated with incisional hernias are gener-

lly related to bowel incarceration and strangulation, whiched to substantial personal disability and financial burdenn the US economy. The annual, national cost associatedith treating incisional hernias and their complications was

stimated to be over $900 million.1

Studies have demonstrated that closing incisional her-ia defects with mesh results in a lower recurrence rate.2,3

s a result, ventral hernias are being repaired increas-

ompeting Interests Declared: None.unding for this study was provided by the Department of Clinical Investi-ations, under DDEAMC#03-38a. This article represents the personal view-oint of the authors and cannot be construed as a statement of official De-artment of Defense policy. Use of commercial products in this project doesot imply endorsement by the US government.

eceived September 22, 2006; Revised November 6, 2006; Accepted Novem-er 8, 2006.rom the Department of Surgery, Eisenhower Army Medical Center, Ftordon, GA.orrespondence address: Timothy W Judge, MD, Eisenhower Army Medicalenter (MCHF-SCL-GS), Bldg 300, Rm 1C23, Ft Gordon, GA 30905.

smail: timothy.judge@se.amedd.army.mil

2762007 by the American College of Surgeons

ublished by Elsevier Inc.

ngly with prosthetic mesh. Use of prosthetic mesh canesult in serious complications, including wound infec-ion, seroma, mesh extrusion, fistula, and adhesions.here are a variety of mesh materials available to the

urgeon when contemplating incisional hernia repair.hen choosing a mesh product, it is important to con-

ider each material in terms of its strengths and weak-esses. Three of the primary factors influencing this de-ision include strength of incorporation (SOI), adhesionormation, and handling characteristics.

The first use of prosthetic mesh for ventral herniaepair was in the 1960s, when Usher4,5 presented thedvantages of knitted polypropylene mesh (PPM) forhe repair of anterior abdominal wall hernias. PPM istrong, has excellent tissue incorporation, and is rela-ively inexpensive. Unfortunately, it has been found toe associated with a high rate of adhesion formation tonderlying viscera.6 Visceral adhesions can result in in-estinal obstruction, pain, and fistula formation. To ad-ress this issue, a variety of composite mesh productsave been developed, with the goal of preventing adhe-

iogenesis and allowing adequate tissue in-growth to al-

ISSN 1072-7515/07/$32.00doi:10.1016/j.jamcollsurg.2006.11.003

lrm

csanoii

lsTacsp

wtshc

bfssmos

MTgUsI

attgrm

aTwha

mFRt(tbaFAt

dxmcaitdaePiwRtwss

atu

277Vol. 204, No. 2, February 2007 Judge et al Abdominal Wall Hernia Repair

ow for a strong SOI. Two such mesh products are Sep-amesh (SM; Genzyme Corp) and Parietex compositeesh (PCM; Sofradim Corp).Seprafilm (Genzyme Corp), a sodium hyaluronate/

arboxy-methylcellulose absorbable membrane, has beenhown to be highly effective in reducing both incidencend severity of adhesion formation to midline abdomi-al closures in humans.7,8 Seprafilm has the disadvantagef being somewhat difficult to handle intraoperatively,n addition to having no intrinsic strength or ability toncorporate into the surrounding tissue.

SM is a polypropylene mesh, with sodium hya-uronate/carboxy-methylcellulose bonded to the visceralide of the mesh. The mesh has a weight of 217 g/m2.he goal is to retain the positive characteristics of PPM

nd prevent adhesions to underlying viscera. Recentomparisons of SM with PPM in a rat hernia modelhowed SM to be superior with regard to adhesionrevention.6,9

Parietex composite mesh is a polyester mesh coatedith a collagen hydrogel matrix on the visceral side of

he mesh. This mesh has a weight of 279 g/m2. A recenttudy in rats demonstrated a decreased incidence of ad-esions when compared with PPM. This was compli-ated by an increased infection rate in the PCM group.6

To date, there have been no studies that examinedoth of these materials in a prospective, randomizedashion with respect to their SOI and resistance to adhe-iogenesis. The purpose of this study was to performuch a comparison of SM and PCM using a rabbit herniaodel. The two were compared with regard to strength

f tissue incorporation, adhesion formation, and meshhrinkage.

ETHODShis study was performed in accordance with NIHuidelines, as described in the Guide for the Care andse of Laboratory Animals,10 in an Association for As-

essment and Accreditation of Laboratory Animal Care

Abbreviations and Acronyms

AA � area of adhesionPCM � Parietex composite meshPPM � polypropylene meshSM � SeprameshSOI � strength of incorporation

nternational accredited animal facility and under the n

uspices of the Eisenhower Army Medical Center Insti-utional Animal Care and Use Committee. There werewo study arms, each with 22 subjects. Half of eachroup was sacrificed and examined at 1 month. Theemaining animals were sacrificed and examined at 5onths.The New Zealand white rabbits were quarantined

nd acclimated for 10 days before the operation.hroughout the experiment, animals received food andater ad libitum. Animals were randomized to undergoernia repair with either SM or PCM. There were 11nimals in each experimental arm.

All animals underwent induction of anesthesia with aixture of ketamine hydrochloride (50 mg/kg, Ketaset;

ort Dodge Laboratories Inc) and xylazine (10 mg/kg,ompun; Ben Venue Laboratories) administered IM in

he anterior thigh muscles. Additionally, glycopyrrolate0.01 mg/kg, Robinul; AH Robbins Co) was adminis-ered IM. Supplemental anesthesia with Isoflurane (Ab-ott Laboratories) through an endotracheal tube wasdministered to maintain a surgical plane of anesthesia.eedings were withheld 12 hours before the procedure.ll animals received appropriate analgesics postopera-

ively for pain control.Using sterile surgical technique, a 10-cm midline ab-

ominal incision was made beginning 2 cm below theyphoid process. Bilateral skin flaps were raised. Theidline fascia was incised for a distance of 8 cm, taking

are not to injure the underlying viscera. At the superiornd inferior ends of the incision, lateral cuts were madento the rectus muscle, and the resulting flap was re-racted laterally. This created a standardized 5 � 8-cmefect. The abdominal wall defect was then closed with7 � 10-cm piece of either SM or PCM. The facial

dges were secured to the mesh using interrupted 3-0DS, 1 cm from the edge. Sutures were placed approx-

mately 1 cm apart. This resulted in a standardized repairith 1 cm of mesh underlying the fascia on all margins.abbits possess a tissue layer under the abdominal skin

hat is analogous to the platysma layer in humans. Thisas closed over the mesh with interrupted 3-0 Vicryl

uture. The skin incision was closed with a running,ubcuticular 4-0 Monocryl suture.

Necropsy and specimen examination was conductedt 1 month for half of the animals and at 5 months forhe remaining animals. The abdominal wall was excisedsing a U-shaped incision well away from the mesh her-

ia repair. Photographs were taken of the mesh and any

asdcowmisasto

cr4tCarAa

eiaob

RSb5

omthmhoaggbs0

tpdmadaosva(

dg1(P

FvSm

F(vp

278 Judge et al Abdominal Wall Hernia Repair J Am Coll Surg

ssociated adhesions in vivo. It was noted whether adhe-ions involved bowel, other viscera, or omentum. Afterescribing the adhesions in vivo, the abdominal wall wasompletely excised and the areas of mesh that were freef adhesions were marked with India ink. All adhesionsere then excised sharply, resulting in a topographicap, with uninvolved mesh being marked with India

nk and the involved areas being left unmarked. Eachpecimen was then photographed with a digital camera,nd the resulting images were analyzed with Image PCoftware (version 1.0; Scion Corp). Total mesh area, to-al adhesion area, and degree of mesh shrinkage from itsriginal size were calculated.

Each specimen was then divided in the midline, and 1m of mesh and tissue was excised from each end. Theemaining specimen was then divided transversely into

sections 2 cm in width. These specimen strips werehen tested on an Instron 4502 tensiometer (Instronorp). Testing was performed with a 100-N load cell

nd a crosshead speed of 50 mm/min. Maximum forceequired for disruption was recorded for each sample.verage force of disruption was recorded for eachnimal.

Student’s t-test analysis was used to analyze differ-nces in SOI, total area of adhesions, percentage of meshnvolved with adhesions, and percentage of mesh shrink-ge. Chi-square test was used to compare the incidencef bowel adhesions. A p value � 0.05 was considered toe statistically significant in both statistical tests.

ESULTSOI for PCM was significantly greater than for SM atoth 1 month (60.8 N versus 42.6 N, p � 0.001) and at

igure 1. Strength of tissue incorporation at 30 days (PCM 60.8 Nersus SM 42.6 N, p � 0.001) and 5 months (PCM 70.9 N versusM 31.5 N, p � 0.001) postprocedure. PCM, Parietex compositeesh; SM, Sepramesh.

months (70.9 N versus 31.5 N, p � 0.001) (Fig. 1). S

The majority of adhesions encountered were tomentum. There were omental adhesions in all ani-als and the adhesions encountered in both groups

ypically involved the cephalad edge of the mesh. Ad-esions were also encountered along the edge of theesh, if the mesh was folded back on itself. The ad-

esions were not particularly dense or fibrous in anyf the groups. One month after mesh implant, boweldhesions were encountered only 1 time in the PCMroup, but they were encountered 6 times in the SMroup (p � 0.05). At 5 months, there were no cases ofowel adhesion in the PCM group, but bowel adhe-ions were present 4 times in the SM group (p �.05).The area of adhesions (AA) was measured both in

erms of the total area of mesh involved and as aercentage of mesh area involved. At 1 month, PCMemonstrated a lower AA, both as a percentage of theesh (8.6% versus 9.1%) and in terms of absolute

rea involved (341.4 mm2 versus 482.9 mm2). Thisifference was not statistically significant (p � 0.88nd p � 0.36, respectively). At 5 months, PCM dem-nstrated a statistically significant lower AA, as mea-ured both by percentage of the mesh involved (5.6%ersus 12.8%; p � 0.05) and by absolute area ofdhesions (321 mm2 versus 840 mm2; p � 0.01)Figs. 2, 3).

At tissue harvest, all mesh samples were noted to haveecreased in size. The PCM group had a significantlyreater reduction in the size of the mesh (38.2% versus8.1%; p � 0.001) when compared with the SM groupFig. 4). This finding was also true at 5 months, withCM again demonstrating a larger reduction in size than

igure 2. Area of adhesions as percentage of mesh area at 30 daysPCM 8.6% versus SM 9.1%, p � 0.88) and 5 months (PCM 5.6%ersus SM 12.8%, p � 0.05) postprocedure. PCM, Parietex com-osite mesh; SM, Sepramesh.

M (17.4% versus 6.1%, p � 0.001).

DNasTpmPavsre

imftawtsb

whpaWso5hf9

Ts

pmPtt2mS

buwtappa

F1p

TMG

1

5

*A

F3mc

279Vol. 204, No. 2, February 2007 Judge et al Abdominal Wall Hernia Repair

ISCUSSIONumerous studies have been conducted to investigate

dhesion formation to prosthetic mesh materials and thetrength of tissue incorporation of these materials.11-15

he ideal prosthetic mesh provides strength, flexibility,ermanence by incorporation of host tissue, and incitesinimal inflammatory response to underlying viscera.PM provides excellent SOI, but is too reactive by itselfnd, as a result, forms dense adhesions to underlyingiscera.13 The resultant morbidities of fistulization, ob-truction, fertility problems, pain, and difficulties witheoperation are a source of considerable personal andconomic hardship.16

The goal of this study was to compare PCM with SM,n an effort to determine the superior product. Our pri-

ary areas of interest were SOI and rate of adhesionormation. We expected both types of mesh to be effec-ive in preventing adhesions, as compared with PPM,nd that was the case at 1 month. The area of adhesionsas not statistically significant. Even at 1 month, PCM

ended toward fewer adhesions overall and demon-trated a substantial difference in terms of adhesions toowel.At 5 months, differences between the two products

ere statistically significant. PCM resulted in fewer ad-esions than SM. This was true when evaluating theercentage of mesh involved, raw surface area involved,nd also when specifically considering bowel adhesions.

hat is perhaps most interesting is that PCM showed atable adhesion profile, or even a slight improvementver time (341 mm2 versus 321 mm2, and 8.6% versus.6% at 1 and 5 months, respectively). SM, on the otherand, showed an increasing tendency toward adhesionormation over time (482.9 mm2 versus 840 mm2, and

igure 3. Area of adhesions in square millimeters at 30 days (PCM41 mm2 versus SM 482 mm2, p � 0.36) and 5 months (PCM 321m2, SM 840 mm2, p � 0.01) postprocedure. PCM, Parietex

omposite mesh; SM, Sepramesh.

.1% versus 12.8% at 1 and 5 months, respectively). s

his suggests that even after incorporation, SM demon-trates an adhesiogenic effect, but PCM does not.

There was also a dramatic difference between theroducts in terms of SOI. At 1 month, the time whenost patients would begin to resume normal activities,CM demonstrated an �43% stronger integration intohe native tissues. In addition, with PCM the SOI con-inued to improve by almost 17%, but SM showed a6% reduction in SOI. The end result was that at 5onths, the SOI for PCM was more than twice that of

M (Table 1).The most likely explanation for the disparity in SOI

etween PCM and SM is the difference in type of weavesed to form the two meshes. PCM is a woven polyesterith a three-dimensional weave and a larger pore size

han SM. The differences in three-dimensional structurere easily seen on electron microscopy (Fig. 5). We hy-othesized that the increase in surface area and largerore size is likely to aid in the process of incorporation byllowing better tissue ingrowth. Previous studies also

igure 4. Decrease in mesh size at 30 days (PCM 38.2% versus SM8.1%, p � 0.001) and 5 months (PCM 17.4% versus SM 6.1%,� 0.001). PCM, Parietex composite mesh; SM, Sepramesh.

able 1. Comparison of Sepramesh and Parietex Compositeesh at 1 and 5 Monthsroup SM PCM

MonthSOI (n)* 42.6 60.8AA (%) 9.1 8.6AA (mm2) 482 341Shrinkage (%)* 18.1 38.2MonthsSOI (n)* 31.5 70.9AA (%)* 12.8 5.6AA (mm2)* 840 321Shrinkage (%)* 6.1 17.4

p � 0.05.A, area of adhesions; PCM, Parietex composite mesh; SM, Sepramesh; SOI,

trength of incorporation.

st

stpdsicmslhtidth

wt5

ASAADC

Am

Cip

R

1

1

1

tex c

280 Judge et al Abdominal Wall Hernia Repair J Am Coll Surg

uggest that polyester can have improved tissue integra-ion when compared with polypropylene.17

Although both materials demonstrated considerablehrinkage, this was much more dramatic with PCMhan with SM. Others have noted increased shrinkage inolyester meshes as well.17,18 We suspect this correlatesirectly to increased weight and tissue ingrowth, withubsequent contracture during the healing process. Thiss supported by the fact that both materials showed de-reased shrinkage at 5 months, as compared with 1onth. Presumably, the inflammatory response and tis-

ue remodeling was such that the initial shrinkage re-axed over time. The definitive cause for this observationas yet to be determined. It is also not clear whether thisendency toward shrinkage is clinically relevant. Thisncreased tendency toward shrinkage might represent aisadvantage, in that it can result in increased tension onhe repair. If so, it did not manifest clinically because wead no failures of repair in either group.In this study, PCM showed a clear advantage over SM

ith regard to SOI and adhesion formation. This advan-age was present at 1 month and even more dramatic atmonths.

uthor Contributionstudy conception and design: Judge, Dinsmorecquisition of data: Judge, Parker, Dinsmorenalysis and interpretation of data: Judge, Dinsmorerafting of manuscript: Judgeritical revision: Judge, Parker, Dinsmore

cknowledgment: We express great appreciation to the

Figure 5. Electron microscopy image of Parie

embers of the Dwight David Eisenhower Army Medical

enter Department of Clinical Investigation Laboratory An-mal Support Services for their invaluable assistance in com-leting this project.

EFERENCES

1. HCUP nationwide inpatient sample. Rockville, MD: Agencyfor Healthcare Research and Quality; 2003.

2. Luijendijk RW, Hop WC, van den Tol MP, et al. A comparisonof suture repair with mesh repair for incisional hernia. N EnglJ Med 2000;343:392–398.

3. Morris-Stiff GJ, Hughes LE. The outcomes of nonabsorbablemesh placed within the abdominal cavity: literature review andclinical experience. J Am Coll Surg 1998;186:352–367.

4. Usher FC. A new plastic prosthesis for repairing tissue defects ofthe chest and abdominal wall. Am J Surg 1959;97:629–633.

5. Usher FC. The repair of incisional and inguinal hernias. SurgGynecol Obstet 1970;131:525–530.

6. van’t RM, de Vos van Steenwijk PJ, Bonthuis F, et al. Preventionof adhesion to prosthetic mesh: comparison of different barriersusing an incisional hernia model. Ann Surg 2003;237:123–128.

7. Becker JM, Dayton MT, Fazio VW, et al. Prevention of postop-erative abdominal adhesions by a sodium hyaluronate-basedbioresorbable membrane: a prospective, randomized, double-blind multicenter study. J Am Coll Surg 1996;183:297–306.

8. Diamond MP. Reduction of adhesions after uterine myomec-tomy by Seprafilm membrane (HAL-F): a blinded, prospective,randomized, multicenter clinical study. Seprafilm AdhesionStudy Group. Fertil Steril 1996;66:904–910.

9. Felemovicius I, Bonsack ME, Hagerman G, Delaney JP. Preven-tion of adhesions to polypropylene mesh. J Am Coll Surg 2004;198:543–548.

0. Institute for Laboratory Animal Research. Guide for the careand use of laboratory animals. Washington, DC: National Acad-emies Press; 1996.

1. Bellon JM, Contreras LA, Bujan J, et al. Effect of relaparotomythrough previously integrated polypropylene and polytetrafluo-roethylene experimental implants in the abdominal wall. J AmColl Surg 1999;188:466–472.

2. Dinsmore RC, Calton WC Jr. Prevention of adhesions to polypro-

omposite mesh (left) and Sepramesh (right).

pylene mesh in a rabbit model. Am Surg 1999;65:383–387.

1

1

1

1

1

1

281Vol. 204, No. 2, February 2007 Judge et al Abdominal Wall Hernia Repair

3. Dinsmore RC, Calton WC Jr, Harvey SB, Blaney MW. Preven-tion of adhesions to polypropylene mesh in a traumatized bowelmodel. J Am Coll Surg 2000;191:131–136.

4. Johnson EK, Hoyt CH, Dinsmore RC. Abdominal wall herniarepair: a long-term comparison of Sepramesh and Dualmesh ina rabbit hernia model. Am Surg 2004;70:657–661.

5. Young RM, Gustafson R, Dinsmore RC. Sepramesh vs. Du-almesh for abdominal wall hernia repairs in a rabbit model. Curr

Surg 2004;61:77–79.

6. Ray NF, Larsen JW Jr, Stillman RJ, Jacobs RJ. Economic impactof hospitalizations for lower abdominal adhesiolysis in theUnited States in 1988. Surg Gynecol Obstet 1993;176:271–276.

7. Gonzalez R, Ramshaw BJ. Comparison of tissue integrationbetween polyester and polypropylene prostheses in the preperi-toneal space. Am Surg 2003;69:471–476.

8. Coda A, Bendavid R, Botto-Micca F, et al. Structural alterations

of prosthetic meshes in humans. Hernia 2003;7:29–34.

JACS CME-1 PROGRAM1.0 credit is earned for completing both questions for each article.Completion of all four articles (8 questions) earns 4 CME-1 creditseach month.

www.jacscme.facs