ORIGINAL ARTICLE

Christian Peiper Æ Karsten Junge Æ Uwe Klinge

Eva Strehlau Æ A. Ottinger Æ Volker Schumpelick

Is there a risk of infertility after inguinal mesh repair? Experimentalstudies in the pig and the rabbit

Received: 31 October 2005 / Accepted: 13 November 2005 / Published online: 14 December 2005� Springer-Verlag 2005

Abstract The implantation of a non-absorbable poly-propylene mesh during hernia repair causes chronicforeign body reaction involving the surrounding tissue.In case of inguinal hernia repair using mesh techniques,the spermatic cord is potentially affected by this chronicinflammatory tissue remodeling. This effect has beeninvestigated using standardized animal models (pig andrabbit). Fifteen adult male pigs underwent transinguinalpreperitoneal implantation of a polypropylene mesh.The contralateral side with a Shouldice repair served ascontrol. After 7, 14, 21, 28, and 35 days, three animalswere sacrificed. The spermatic cords were resected andanalyzed histologically. In a second experiment Lich-tenstein repair using the same polypropylene mesh andShouldice repair on the contralateral side was done ineight chinchilla rabbits. Three animals served as con-trols. Three months after operation, the analysis in-cluded testicular size, testicular temperature, andtesticular and spermatic cord perfusion. We added his-tological evaluation of the foreign body reaction and thespermatogenesis using the Johnsen score. In the pig, weobserved a certain foreign body reaction with diffuseinfiltrating inflammatory cells after mesh implantation.Venous thrombosis of the spermatic veins occurred infive of 15 cases. One animal presented focal fibrinoidnecrosis of the deferent duct wall. The side of Shouldicerepair showed only minor postoperative changes. In therabbit, we also observed a typical foreign body reaction

at the interface between mesh and surrounding tissue,which was not detectable after Shouldice repair. Themesh repair led to a decrease of arterial perfusion, tes-ticular temperature, and the rate of seminiferus tubuleswith regular spermatogenesis classified as Johnsen 10(Lichtenstein: 48.1%, Shouldice: 63.8%, controls:65.8%). Testicular volume increased about 10% aftereach operation. The implantation of a polypropylenemesh in the inguinal region induces major response ofthe structures of the spermatic cord. This may have aninfluence also on spermatogenesis. Due to this a strictindication for implantation of a prosthetic mesh duringinguinal hernia repair is recommended.

Keywords Inguinal mesh repair Æ Spermatic cord ÆForeign body reaction Æ Spermatogenesis Æ Testicularperfusion

Introduction

Modern concepts for the treatment of inguinal herniasalways include prosthetic mesh repairs. Some surgeonsuse the mesh repair exclusively [1]. Reported results arepromising concerning recurrence rates [2] and return tophysical activity [3]. On the other hand, there is a certainrisk of complications that are difficult to detect. Alter-ations of fertility may be recorded only in very largeseries, if observable. We know that prosthetic meshescause chronic inflammatory changes of the surroundingtissue [4]. Due to the close contact between mesh and thestructures of the spermatic cord after inguinal mesh re-pair, these changes may also alter the reproductivestructures in male patients. Case reports have beenpublished about spermatic granuloma [5] and sperma-toceles [6] after Lichtenstein mesh implantation. Espe-cially after mesh implantation in younger patients, thesefindings may be of great importance. In the animalexperimental models of the pig and the rabbit weinvestigated the interaction between mesh and theadjacent spermatic cord concerning the extent of

C. Peiper (&)Surgical Clinic, Evangelisches Krankenhaus,Pferdebachstr. 27, 58455 Witten, GermanyE-mail: ch.peiper@dwr.deTel.: +49-2302-1752461Fax: +49-2302-1752076

K. Junge Æ U. Klinge Æ E. Strehlau Æ V. SchumpelickSurgical Clinic, Rhenish-Westfalian Technical University,Aachen, Germany

A. OttingerJoint Institute for Surgical Research,Russian State Medical University, Moscow, Russia

Hernia (2006) 10: 7–12DOI 10.1007/s10029-005-0055-1

inflammatory changes and their relation to the usedmaterial, the integrity of the deferent duct, and theinfluence on the testicular function.

Transinguinal preperitoneal mesh prosthesis in the pig

Materials and methods

The experiments were officially approved by the DistrictPresident of Nordrhein-Westfalen (AZ 23.203.2-Ac18,16/97). All animals received humane care in accor-dance with the requirements of the German Tier-schutzgesetz, §8 Abs. 1. All operations were carried outunder general anesthesia and under aseptic and sterilesurgical conditions.

Fifteen uncastrated adult male pigs underwent aunilateral transinguinal preperitoneal implantation of apolypropylene mesh following the technique describedby Schumpelick and Arlt [7]. After identification of theinguinal ligament, the inguinal canal was opened bydividing the external oblique fascia. We resected thecremasteric muscle and secured the spermatic cord by aloop. The transversal fascia was then divided, therebycreating a hernial defect. The polypropylene mesh wastailored to an appropriate size of 10·15 cm with a lateralslash of 4 cm to allow the cord to pass through. Afterplacement of the mesh into the preperitoneal space, itwas sutured to Cooper’s ligament, laterally to theinguinal ligament and medially to the internal obliqueabdominal musculature, to prevent migration usinginterrupted non-absorbable sutures. Then we closed theslash of the mesh again using two interrupted sutures.The floor of the inguinal canal was enforced by aShouldice plasty [8]. Hereby contact between mesh andspermatic cord was limited to the deep inguinal ring.Then the spermatic cord was placed on top of the repairin its normal anatomic position. The external obliqueaponeurosis was closed with a running suture recreatingthe superficial inguinal ring. Re-adaptation of the sub-cutaneous tissue and skin closure terminated theoperation.

Afterwards, the contralateral inguinal region wasprepared. Here a simple Shouldice plasty using non-absorbable polyester sutures was performed followingthe same extensive preparation. Hereby every animalserved as its own control.

After 7, 14, 21, 28 and 35 days, three animals weresacrificed. The spermatic cords were dissected off thefloor of the canal. They were excised including the deepinguinal ring. Representative cross-sections were ob-tained at this area and analyzed histologically usinghematoxylin and eosin stains.

Results

All animals recovered soon from the operation and hadan uneventful postoperative course. At the time of

explantation of the spermatic cord, the deep inguinalring was macroscopically swollen and thick after meshimplantation. The histological examination of the sper-matic cord close to the deep inguinal ring revealed acertain foreign body reaction at the side of meshimplantation with a progress over the time. We observedtypical foreign body giant cells and a granuloma body ofmature epitheloid cell macrophages at the interface be-tween mesh fibers and host tissue. Moreover, we saw adiffuse infiltration of inflammatory cells in all thesecases, in particular polymorphous nuclear granulocytes.Obviously, this infiltration started at the interface be-tween mesh and spermatic cord. Over time, it developedtowards the deferent duct. The outer zone consisted ofsmaller amounts of t-cells and a connective tissue cap-sula rich of collagen fibers in all animals. Five of 15specimen presented venous thrombosis of the veins ofthe spermatic cord due to a secondary vasculitis. Older,partly organized thrombi from the clotting type could beobserved (Fig. 1). The development of these changes didnot depend on the postoperative time and occurred after14, 21 (two cases), 28, and 35 days. Additionally, oneanimal showed focal fibrinoid necrosis of the wall of thedeferent duct following mesh implantation (Fig. 1). Theside of Shouldice repair presented only minor postop-erative changes, especially no venous thrombosis of thespermatic cord or affection of the ductus deferens wall(Fig. 2)

Lichtenstein repair in the rabbit

Materials and methods

The experiments were officially approved by the DistrictPresident of Nordrhein-Westfalen (AZ 50.203.2-AC 18,3/01). All animals received humane care in accordancewith the requirements of the German Tierschutzgesetz,

Fig. 1 Diffuse granulocytes infiltration of the spermatic cord5 weeks after transinguinal preperitoneal polypropylene mesh inthe pig. The picture shows also fibrinoid necrosis of the deferentduct (left) and thrombosis of the spermatic veins (right). Thespecimen was taken from the deep inguinal ring (H.�E. stains, ·40)

8

§8 Abs. 1. All operations were carried out under generalanesthesia and under aseptic and sterile surgical condi-tions.

The animals were fixed in the supine position. Uni-lateral Lichtenstein repair [9] was done in eight Chin-chilla rabbits. The animals received implantation of asmall porous, heavy weight polypropylene mesh. Afteropening the inguinal canal, the spermatic cord was se-cured by a loop. We preserved the cremasteric muscle inall cases. The Lichtenstein repair was performed withfixation of the mesh at the inguinal ligament and theinternal oblique muscle using interrupted polyester su-tures. The lateral end of the mesh was slit into two tailsand closed against the lateral aspect of the internalinguinal ring using also polyester sutures.

The Shouldice repair was carried out on the contra-lateral side using running polyester sutures. The otherthree rabbits served as controls.

Evaluation of testicular changes was done 3 monthslater. We determined the volume of the testicles byultrasound investigation and calculation using the for-mula of a rotating ellipsoid [10]. Testicular temperaturewas measured by a digital thermometer (Voltcraft 300 k)with needle sensor.

Arterial perfusion of the spermatic cord and the tes-ticles was investigated using the IC-VIEW system(PULSION Medical Systems AG, Munich, Germany).Skin and subcutaneous tissue was removed from thelower abdomen and the scrotal skin was removed com-pletely. A bolus of indocyanin green (ICG, 0.5 mg/kg)was injected intravenously into the ear vein. Excitationlight was provided by an IC-VIEW camera-mountednear an infrared light source (780 nm, NIR-light). ICG-derived fluorescence was detected by a digital camerausing the super nightshot mode and transferred to acomputer for further quantification of perfusion relatedfluorescence intensity (IC-CALC). We calculated the

difference between the intensity before and the maxi-mum intensity after injection as an index for maximumperfusion. This parameter was recorded at the spermaticcord and at the testicles.

Spermatogenesis as the main testicular function wasestimated histologically using the Johnsen Score [11].Hereby, complete spermatogenesis with many sperma-tozoa is rated at ten (Table 1).

We excised the spermatic cords including the deepinguinal ring and obtained representative cross-sectionsat this area. Histological analysis followed using hema-toxylin and eosin stains with focus on the inflammatoryresponse to the prosthetic mesh.

Global statistical analysis was carried out using theKruskal–Wallis test. For statistics among the groups weused the Mann–Whitney test.

Results

In the postoperative period, we observed an increasedtesticular volume (Fig. 3) with no difference between thekinds of repair (P=0.53) and a decreased testiculartemperature compared to the controls (Fig. 4). Here alsothe differences among the operations were very small(P=0.246).

A cause for these findings may be the reduced arterialperfusion after any repair (P<0.05). In comparison tothe controls, this decrease of the spermatic cord perfu-

Fig. 2 Specimen of the spermatic cord of the pig 5 weeks afterShouldice repair serving as a control. Only minor postoperativechanges are present. The wall of the ductus deferens is vital (H.�E.stains, ·40)

Table 1 Johnsen score for qualitative evaluation of spermatogen-esis [11]

1 No cells in tubular section2 No germs but Sertoli cells are present3 Spermatogonia are the only germ cells present4 Only few spermatocytes (<5) and no

spermatids or spermatozoa present5 No spermatozoa, no spermatids but several

or many spermatocytes present6 No spermatozoa and only few spermatids (<5–10) present7 No spermatozoa but many spermatids present8 Only few spermatozoa (<5–10) present in section9 Many spermatozoa present but germinal

epithelium disorganized with marked sloughingor obliteration of lumen

10 Complete spermatogenesis with manyspermatozoa (late spermatids)

0

1

2

3

4

Lichtenstein Shouldice Control

ml

Fig. 3 Testicular volume of rabbits 3 months after inguinal herniarepair

9

sion was more obvious after Lichtenstein repair(P=0.03, Fig. 5) than after the Shouldice operation.The testicular perfusion also was significantly reducedafter Lichtenstein mesh implantation (P<0.05, Fig. 6).

Evaluation of spermatogenesis revealed a certaindecrease of Johnsen ten scores in seminiferous tubulesafter Lichtenstein repair. Following Shouldice repair,the testicles showed regular spermatogenesis (P=0.20,Fig. 7).

During histological evaluation of the surroundingtissue, we observed a characteristic foreign body reac-tion to the mesh (Fig. 8), which we did not find after theShouldice repairs. Destruction of the vas deferens wallor venous thrombosis was not detected. The duct waspatent in all specimens. Obviously, the preserved cre-masteric muscle protected the structures of the spermaticcord in this model (Fig. 8).

Discussion

The effects of the long-term implantation of a meshbioprosthesis on the surrounding soft tissue duringhernia repair are numerous. In the recipient tissues evenyears after implantation, a persisting inflammatoryproliferative foreign-body reaction with increased cellturnover is described [4]. Typical signs are inflammatorycells and numerous macrophages at the interface, alsoeven after years [12]. This inflammatory response to themesh implantation is reported not only after incisionalhernia repair, but also following bioprosthesis implan-tation during inguinal hernia operation. Trabucchireported about similar findings in human biopsies 7 days

to 9 years after inguinal dacron mesh implantation. Heobserved a foreign-body giant cell layer around thefibers and the presence of macrophages in an interme-diate layer [13].

Similar changes have been found within the spermaticcord after mesh inguinal hernia repair in the animalmodel [14]. Here the aforementioned changes togetherwith a fibrotic formation were observed after6–12 months. We could show with our results that theinflammatory changes are also appearing during theshort-term observation of 1–12 weeks. This correlateswell with the results of Beets et al. [15], who foundincreasing foreign-body giant-cell reaction to polypro-pylene mesh in the preperitoneal position until the thirdweek after implantation in the pig. Afterwards the re-sponse gradually decreased, until at 6 months, it per-sisted at half the maximal level at 3 weeks.

Involvement of testicular veins is also alreadydescribed in literature. LeBlanc found testicular venouscongestion after mesh implantation in the pig [16]. Ourresults confirmed this observation. Moreover, we sawvenous thrombosis within the spermatic cord. Thisobservation has not been described in literature before.Perhaps this spermatic thrombosis represents the causefor the inflammatory changes. More likely, however, isthe explanation that the venous thrombosis is one of theresults of the foreign body reaction. Up to now, theclinical relevance of this observation was unclear.

Therefore we conducted additional experimentalinvestigations to unveil the effect of this inflammatoryreaction on the testicular function. In the rabbit model,we observed less evident inflammatory changes. Perhapsthis reduced reaction depends on the duration of the

30

31

32

33

34

Lichtenstein Shouldice Control

° C

Fig. 4 Testicular temperature of rabbits 3 months after inguinalhernia repair

0

20

40

60

80

100

120

Lichtenstein Shouldice Control

au

Fig. 5 Difference between the fluorescence intensity before and themaximum intensity after ICG injection at the spermatic cords ofrabbits 3 months after inguinal hernia repair

0

20

40

60

80

100

120

Lichtenstein Shouldice Control

au

Fig. 6 Difference between the fluorescence intensity before and themaximum intensity after ICG injection at the testicles of rabbits3 months after inguinal hernia repair

0

20

40

60

80

Lichtenstein Shouldice Control

Fig. 7 Rate of Johnsen 10 seminiferous tubules 3 months afteringuinal hernia repair

10

postoperative period. Nevertheless we found a signifi-cant influence on testicular perfusion and function. Inthe postoperative phase, testicular temperature andperfusion were reduced after any repair, with a strongereffect following the Lichtenstein operation. Spermato-genesis also showed a certain reaction on the mesh. Themesh repair reduced the amount of regular spermato-genesis classified as Johnsen 10 in comparison to theShouldice repair and the controls. This difference,however, was without statistical significance. The chan-ges mentioned might by of no large clinical relevance inunilateral repair, but in bilateral cases, they must bebrought under consideration. The influence on humoralconditions will be topic of further investigations.

One explanation for the morphological and func-tional changes observed in our studies may be adhesionsbetween the mesh and the structures of the spermaticcord, as a result of the foreign body reaction. Theseadhesions were described by Fitzgibbons et al. [17] in thepig. They found adhesions between the mesh andstructures of the spermatic cord even after intraperito-neal placement of the mesh. LeBlanc et al. [16] placed aheavy weight polypropylene mesh into the preperitonealspace and also observed severe adhesions to the sper-matic cord, 30 days after implantation. Ninety daysafter operation, adhesions to the spermatic vessels andto the spermatic cord as well as venous congestion of thetestis were described.

One additional aspect is the protection of the struc-tures of the spermatic cord by the cremasteric muscle.Our data show integration of the structures of thespermatic cord into the inflammatory soft tissue re-sponse on the prosthetic mesh after cremasteric resec-tion, as we performed it in the pig. The structures wereprotected from the inflammation, if the cremastericmuscle was spared, as we did it in the rabbit model.

There are also reports about reactions of inguinalprosthetic mesh in man. Wingenbach et al. [18] reported

long-lasting pain during copulation in 3.9% of all casesafter laparoscopic hernia repair. Langenbach et al. [19]found painful ejaculation in 10%, 12 weeks after lapa-roscopic repair, which correlated with the kind of mesh.Hetzer et al. [6] reported spermatoceles requiring oper-ation after Lichtenstein repairs in 0.8%. Silich andMcSherry [5] reported a case of a spermatic granulomarequiring operation, 2 years after mesh repair of aninguinal hernia. Shin et al. [20] reported a total of 14cases of azoospermia secondary to inguinal vasalobstruction related to previous polypropylene meshherniorrhaphy. Summarizing these publications and ourexperimental results, we suggest some negative influenceon the structures of the spermatic cord by the foreignbody reaction to the prosthetic mesh. Therefore, wecaution against the implantation of polypropylene meshinto the inguinal region without a good indication.

Acknowledgements These experiments were supported by theDeutsche Forschungsgemeinschaft (AZ KON 709/2002, PE 718/4-1)

References

1. Bittner R, Schmedt CG, Schwarz J, Kraft K, Leibl BJ (2002)Laparoscopic transperitoneal procedure for routine repair ofgroin hernia. Br J Surg 89:1062–1066

2. Wara P, Bay-nielsen M, Juul P, Bendix J, Kehlet H (2005)Prospective nationwide analysis of laparoscopic versus Lich-tenstein repair of inguinal hernia. Br J Surg 92:1277–1281

3. Fenoglio ME, Bermas HR, Haun WE, Moore JT (2005)Inguinal hernia repair: results using an open preperitoneal ap-proach. Hernia 9:160–161

4. Klosterhalfen B, Klinge U, Hermanns B, Schumpelick V (2000)Pathology of traditional surgical nets for hernia repair afterlong-term implantation in humans. Chirurg 71:43–51

5. Silich RC, McSherry CK (1996) Spermatic granuloma. Anuncommon complication of the tension-free repair. Surg En-dosc 10:537–539

6. Hetzer FH, Hotz T, Steinke W, Schlumpf R, Decurtins M,Largiader F (1999) Gold standard for inguinal hernia repair:Shouldice or Lichtenstein? Hernia 3:117–120

7. Schumpelick V, Arlt G (1996) Transinguinal preperitonealmesh-plasty in inguinal hernia using local anesthesia. Chirurg67:419–424

8. Shouldice EE (1945) Surgical treatment of hernia. Ontar MedRev 4:43

9. Lichtenstein IL, Shulman AG, Amid PK, Montllor MM (1989)The tension-free hernioplasty. Am J Surg 157:188–193

10. Peiper Ch, Ponschek N, Truong S, Schumpelick V (2000)Ultrasound-based volumetric evaluation of fluid retention afteringuinal hernia repair. Surg Endosc 14:666–669

11. Johnsen SG (1970) Testicular biopsy score count—a methodfor registration of spermatogenesis in human testes: normalvalues and results in 335 hypogonadal males. Hormones 1:2–25

12. Klinge U, Klosterhalfen B, Muller M, Schumpelick V (1999)Foreign body reaction to meshes used for the repair ofabdominal wall hernias. Eur J Surg 165:665–673

13. Trabucchi EE, Corsi FR, Mainardi C, Cellerino P, Allevi R,Foschi DA (1998) Tissue response to polyester mesh for herniarepair: an ultramicroscopic study in man. Hernia 2:107–112

14. Uzzo RG, Lemack GE, Morrissey KP, Goldstein M (1999) Theeffects of mesh bioprosthesis on the spermatic cord structures: apreliminary report in a canine model. J Urol 161:1344–1349

Fig. 8 Inflammatory changes of the spermatic cord 3 months afterpolypropylene mesh implantation in the rabbit (mesh fibers at thetop, deferent duct left below, H.+E., ·100)

11

15. Beets GL, van Mameren H, Go PMNYH (1998) Long-termforeign-body reaction to preperitoneal polypropylene mesh inthe pig. Hernia 2:153–155

16. Leblanc KA, Booth WV, Whitaker JM, Baker D (1998) In vivostudy of meshes implanted over the inguinal ring and externaliliac vessels in uncastrated pigs. Surg Endosc 12:247–251

17. Fitzgibbons RJ Jr, Salerno GM, Filipi CJ, Hunter WJ, WatsonP (1994) A laparoscopic intraperitoneal onlay mesh techniquefor the repair of an indirect inguinal hernia. Ann Surg219:144–156

18. Wingenbach O, Waleczek H, Kozianka J (2004) Laparoscopichernioplasty by transabdominal preperitoneal approach.Analysis and review in 267 cases. Zentralbl Chir 129:369–373

19. Langenbach M, Schmidt J, Lazika M, Zirngibl H (2003)Urological symptoms after laparoscopic hernia repair.Reduction with a variant of polypropylene mesh. Urologe42:375–381

20. Shin D, Lipshultz LI, Goldstein M, Barme GA, Fuchs EF,Nagler HM, McCallum SW, Niederberger CS, Schoor RA,Brugh VM 3rd, Honig SC (2005) Herniorrhaphy with poly-propylene mesh causing inguinal vasal obstruction: a pre-ventable cause of obstructive azoospermia. Ann Surg241:553–558

12