Do Multifilament Alloplastic Meshes Increase the Infection Rate?

Analysis of the Polymeric Surface, the Bacteria Adherence, and

the In Vivo Consequences in a Rat Model
U. Klinge,1 K. Junge,1 B. Spellerberg,2 C. Piroth,1 B. Klosterhalfen,3 V. Schumpelick1
1
Department of Surgery Technical University of Aachen, Pauwelsstrasse 30, D-52057 Aachen, Germany

2
Institute of Medical Microbiology, Technical University of Aachen, Pauwelsstrasse 30, D-52057 Aachen, Germany

3
Institute of Pathology, Technical University of Aachen, Pauwelsstrasse 30, D-52057 Aachen, Germany

Received 11 March 2002; revised 20 May 2002; accepted 31 May 2002

Published online 00 Month 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.10449

Abstract: Within the last decade hernia surgery has changed from suture repair to mesh
repair. Biomaterials, and multifilaments in particular, are thought to increase the risk of
infection. Therefore, the aim of this study was to study the influence of the presence of either
a monofilament or a multifilament mesh material on the bacterial infection risk. The filament
surface of a monofilament and a multifilament mesh were calculated on the basis of a
theoretical model. The adherence of Staphylococcus aureus was measured in vitro by fluores-
cence analysis. Additionally, the two mesh materials (8-mm platelets) were implanted subcu-
taneously in Sprague–Dawley rats with daily surveillance for clinical signs of infection. After
7 days the meshes were explanted for histological and microbiological analysis. Calculations
of the mesh surface area revealed a higher level for the multifilament mesh. The extent of
adherent bacteria corresponded to the estimated filament surface in vitro. In vivo, the implan-
tation of meshes in the presence of 5 ⴛ 106 S. aureus did not show an increased infection rate
in rats with either monofilament or multifilament material, compared to the control groups
(mesh implantation without S. aureus contamination). However, after 7 days bacteria were
still detectable in the majority of the implantation sites, and a clinically inapparent intensifi-
cation of local inflammation and fibrosis was induced. The increased surface area of a
multifilament meshes promotes the persistence of bacteria in the implant bed, though this
alone is not sufficient to create a clinically apparent infection. This might explain the
development of mesh-related infections after a delay of several months or even years. In vivo,
the adherence of bacteria to the implant material depends on the surface area, which favors
the use of monofilament materials. © 2002 Wiley Periodicals, Inc. J Biomed Mater Res (Appl Biomater)
63: 765–771, 2002

Keywords: surgical mesh; bacterial infection; adherence; filament surface; monofilament;

multifilament

INTRODUCTION ally, local fluid collections such as hematoma or seroma

Biomaterials are thought to increase the infection rate, though
this could not be proven in several clinical trials.1– 4 Even in
highly sterile alloarthroplastic operations bacterial colonial-
ization has been observed in about 40% of cases.5 Addition-
Correspondence to: Karsten Junge, Department of Surgery, Technical University of
Aachen, Pauwelsstr, 30, 52057 Aachen, Germany (e-mail: karsten.junge@post.rwth-
aachen.de)
Contract grant sponsor: BMBF; contract grant number: 01KS9503/9 for IZKF-
BIOMAT RWTH-Aachen
This study was performed in adherence to the NIH guidelines for the use of
experimental animals and to the guidelines of the Deutsche Tierschutzgesetz.
© 2002 Wiley Periodicals, Inc.
formation can favor the development of a manifest bacterial
infection. The risk might rise in the case of small niches, of
an incomplete ingrowth of tissue, or of markedly hydropho-
bic foreign materials, preventing close contact of the cells to
the foreign-body surface. However, the published experimen-
tal data are overall contradictory. In 1957 Elek and Conen6
evaluated that the amount of staphylococci inducing a local
infection is reduced from 106 to 102 in the presence of silk
sutures. In 1982 Sharp et al.7 analyzed the effect of 16 various
suture materials on the infection rate in mice. They found
lower rates for synthetic or monofilament filaments in com-
parison to natural or multifilament materials. Correspond-
ingly, Blomstedt and Österberg in 19778 saw an increased
765
766 KLINGE ET AL.

inflammatory tissue reaction in the presence of 2 ⫻ 107 S.

aureus8 with multifilament sutures in particular. Furthermore,
in 1979 Österberg and Blomstedt9 demonstrated a signifi-
cantly increased persistence of bacteria in the wound for up to
41 days for multifilament compared to monofilament mate-
rial.
In contrast, in 1973 Edlich and co-workers10,11 investi-
gated both monofilaments and multifilaments in the subcutis
after contamination with 103–106 S. aureus and Escherichia
coli and found no different infection rate within 4 days. This
was confirmed in 1987 by Paterson-Brown et al.12 who com-
Figure 1. Model for the calculation of the surface of multifilaments
pared monofilament and multifilament sutures in pigs after (n ⫽ 7)
inoculation of 105 bacteria E. coli and Bacteroides fragilis for
7 days. Accordingly, in 1981 and 1983 Bucknall and co-
workers13,14 could not find an increased inflammation for
multifilaments after adding S. aureus, but bacteria persisted in
the wound for up to 71 days. Interestingly, Scher et al.15
demonstrated in 1985 that although the number of adherent S.
FR ⫽ 冑 FV
␲ ⴱ FL
.

aureus was markedly reduced for monofilaments compared to

multifilaments, these differences completely disappeared in
the presence of knots.
However, all of these studies used suture materials,
whereas the colonization rate of surgical meshes is still un-
known. Due to the widespread use of mesh material for
surgical hernia repair, the adherence of bacteria in vitro and
the infection risk in vivo of a monofilament and a multifila-
ment mesh after experimental contamination with S. aureus
was analyzed in the present study.
Multiplication with FL results in the filament surface area per
cm2 (FS):
FS ⫽ 2 ⴱ ␲ ⴱ FR ⴱ FL
For multifilaments this calculation underestimates the surface,
which is rather wave like, sometimes with deep cracks. The
number of filaments n laying next to one another make up the
diameter of a corresponding monofilament. This is inversely
correlated to the radius of the single filament rmult (Figure 1).

MATERIAL AND METHODS 1

r mult ⫽ .
n
Mesh Materials
The number of the filaments forming the outer shape of the
The monofilament mesh (Marlex威, Bard) consists of polypro- multifilament are laying on a circle whose perimeter Umult
pylene monofilaments of 19 tex (tex equals to g/cm) in size, can be calculated by
a weight of 95 g/m2, and a pore size of 0.1 to 0.5 mm. The
multifilament (Vypro威, Ethicon, Germany) is a large-pore-
size (3–5 mm) combination of nonabsorbable polypropylene
multifilaments of 6.7 tex (basic fibers 1.3 tex) with absorbable
U mult ⫽ 2 ⴱ ␲ ⴱ 1 ⫺ 冉 冊 1
n
.

polyglactin multifilaments of 8.9 tex (basic fibres 0.2 tex). In

comparison to the monofilament mesh the total weight of the Division by the diameter of a single filament results in the
multifilament is reduced to 55 g/m2 (polypropylene part 27 total number of the outward-lying fibres Z:
g/m2, polyglactin part 28 g/m2).

Calculation of the Filament Surface

Z⫽␲ⴱnⴱ 1⫺ 冉 冊 1
n
.

For monofilaments the length of the applied filaments per cm2 Assuming that only half of the circumference, together with
(FL) results from the mesh weight (g/cm2) divided by the an additional segment whose size correlates inversely to the
filament size (tex equals to g/cm). The corresponding volume number of filaments, is forming the outer shape, this portion
(FV) is given by the ratio of the mesh weight to the specific AFil can be calculated by
gravity of the polymer (polypropylene 0.9 g/cm3, polyglactin
1.2 g/cm3). The filament radius (FR, cm) can be calculated
due to the definition of the filament volume, which is given
by multiplication of the filament cross section with the fila-
A Fil ⫽
冉 180⬚ ⫹
360⬚
n
.
冊
ment length: 360⬚
 MESH AND INFECTION
The approximated surface Omult of a multifilament with fila-
ments of a size of 1/n of a monofilament is given by multi-
plication with the total number of outward laying filaments
1 without 100 ␮l dissolved S. aureus (ATCC 25923). The skin
O mult ⫽ 2 ⴱ ␲ ⴱ ⴱ Z ⴱ AFil .
n
If n tends toward ⬁, both (1-1/n) and 360°/n tend to 0, after
transformation leading to Omult ⫽ ␲2 for a filament with r ⫽
1. The comparison of this multifilament surface with a mono-
filament surface results in
Omult ␲2 ␲
lim ⫽ ⫽ .
Or⫽1 2 ⴱ ␲ 2
n3⬁
To conclude, in comparison to a monofilament, the use of a
multifilament leads to an increase of the surface of at least the
factor of ␲/2 ⫽ 1.57 or 57%. Thus the surface of the mesh
made of multifilaments resulting from the weight and the
filament size is multiplied by 1.6 to reflect the increase of
filament surface.
In vitro Bacterial Adherence
The adherence of bacteria to the mesh material in vitro was
measured with the use of sterile circular samples of the
monofilament and multifilament mesh (8 mm in size) incu-
bated with fluorescently labeled S. aureus (ATCC 25923). S.
aureus was labeled with Fluorescein Isothiocyanate (FITC)
(Sigma, Deisenhofen, Germany) at a concentration of 1
mg/ml in 0.05-M sodium carbonate, 0.1-M sodium chloride
for 20 min at room temperature. After two washings twice,
bacteria were adjusted to a density of 105, 106, and 107
bacteria/ml and incubated with the mesh material for 30 min
at 37 °C in Dulbecco’s phosphate-buffered saline (DPBS).
Nonadherent bacteria were removed by washing twice in
(DPBS), and adherent bacteria were quantified in a fluores-
cence counter (Titertek Fluoroskan II, Flow Laboratories).
The number of adherent bacteria was calculated according to
a standard curve that was generated by the addition of a
defined number of fluorescently labelled bacteria to the mesh
platelet. Each measurement was repeated nine times.
Animal Experiment
Seventy-two Sprague-Dawley rats (250 to 300 g) were used
in this study. The animals were housed under conditions of
constant light and temperature and received a complete diet
of rat food and water ad libidum throughout the entire study,
which was performed according to the rules of the “Deutsche
Tierschutzgesetz” (AZ 23.203.2 AC 18 31/96) and to the NIH
guidelines for the use of laboratory animals. The animals
were randomly divided into four groups (group 1, monofila-
ment mesh; group 2, multifilament mesh; a ⫽ without; b ⫽
with 5 ⫻ 106 bacteria). The animals were anaesthetized by
isoflurane and i.m. injection of 0.4-mg/100-g KG 2% xylazin-
767
hydrochloride with 10% ketamine. After disinfection of the
skin two vertical incisions of 1-cm length were made 2 cm
paravertebrally to both sides. After blunt dissection in the
subcutaneous tissue a mesh platelet was implanted with or
was closed by staplers and a spray dressing was added. No
antibiotic treatment was given before or during the experi-
ments. In the following days the rats were surveyed daily and
clinical signs of infection were recorded. At Day 7 the ani-
mals were sacrificed and the wound excised. Half of the
samples were used for semiquantitative microbiology and
histological analysis, respectively. To detect bacteria in the
explanted mesh platelets, samples were cultured in 1 ml of
Tryptic soy broth at 37 °C for 48 h and subcultured on blood
agar plates. Identification of S. aureus colonies was per-
formed according to standard microbiological techniques
(colony morphology, catalase, coagulase).
Morphological Study
Specimens were studied by light microscopy (LM). For LM
tissue specimens were fixed in 10% formalin, embedded in
paraffin, and sections were stained with haematoxylin and
eosin (H&E), as well as Gram. The morphometric evaluation
consisted of a quantitative analysis of the inflammatory re-
action and the soft-tissue reaction. Partial volumes (PV) of
tissues were counted in 10 fields of 5 HE slides at a grid of 10
points (100 ⫻; area 0.1 mm2) within the macrointerface of
0 –300 ␮m. Parameters measured were the percentage share
of the area covered by an inflammatory infiltrate [partial
volume (PV) %], fat tissue (PV%), vessels (PV%), or by
connective tissue (PV%). Additionally, the percentage share
of macrophages (%), granulocytes (%), giant cells (%), lym-
phocytes (%), and fibroblasts (%) was counted within the
macrointerface of 0 –150 ␮m (400 ⫻; area 625 ␮m2).
Statistics
Statistical analysis was carried out with the use of the Statis-
tical Package for Social Sciences (SPSS) software. All func-
tional and morphological results were analyzed for statistical
significance using an analysis of variance, followed by inde-
pendent t- tests in case of significant differences. In case of
ordinal scaled variables a chi2 test was performed. p values ⬍
.05 were considered to be significant.
RESULTS
Calculation of Mesh Surface
The calculation of the surface revealed that though the mul-
tifilament mesh was considerably reduced to 57%, compared
to the monofilament, the construction of multifilaments was
followed by an increase to 135% of surface area (Table I).
The calculation of the total surface of the subfilaments
showed an increase to 205% for the multifilament mesh. Both
768 KLINGE ET AL.

TABLE I. Calculation of the Surface of a Monofilament and Multifilament Mesh (% Compared to Monofilament Mesh)

Monofilament Mesh Multifilament Mesh

Polymer Polypropylene (PP) Polypropylene (PP) and polyglactin (PG)
Weight (g/m2) 95 55 (57%)
Type Monofilament Multifilament (45 subfilaments: 5 PP, 40 PG)
Filament size (tex ⫽ g/1000 m) 18.9 PP 6.7 (subfilaments 1.3)
PG 8.9 (subfilaments 0.2)
Length (cm/cm2) 50 71
Filament radius (cm⫺3) 8.2 PP 4.9
PG 4.9
Subfilament radius (cm⫺3) – PP 2.9
PG 2.3
Filament surface per cm2 mesh, calculated 258 PP 122
for monofilaments (mm2) PG 96
total 218 (84%)
Filament surface, corrected for multifilaments – 349 (135%)
(factor 1.6) (mm2)
Surface on the basis of the subfilaments – PP 294
(mm2) PG 235
total 528 (205%)

meshes revealed a considerable length of used filaments of 50
cm or 71 cm per cm2 of mesh.
Bacterial Adherence In Vitro
For all bacteria concentrations (105–107) the proportion of
adherent bacteria was significantly increased (p ⬍ .01) for the
multifilament mesh, even appearing pronounced for the high-
est concentrations of S. aureus (Figure 2). For the monofil-
ament mesh, after the washing procedure 1.3–2% of the
bacteria were still detectable, whereas 2.6 –3.7% were still
detectable for the multifilament mesh.
Contamination of the Mesh In Vivo
Clinical manifest infections could not be observed in any
group, either with or without the presence of bacteria.
Whereas without contamination with S. aureus no animal
developed an infection, after application of 5 ⫻ 106 bacteria
four animals with the monofilament mesh (11%) and three
animals with the multifilament mesh (8%) showed a suppu-
ration around the implant, without any significant difference
between the two mesh types (Table II).
However, after inoculation of bacteria in almost all
wounds bacteria were still detectable after 7 days. Most often
they were observed within 24 h by immediate culturing of the
supernatant. In some further cases they could be detected
after 24 hours of culturing. In total, bacteria were seen in 78%
of the monofilament meshes, and in 94% of multifilament
meshes. (p ⬎ .05).
Tissue Response
The tissue response to both meshes has been published be-
fore.16,17 Briefly, Marlex威 indicates a typical chronic foreign-
body reaction within the interface of the recipient tissues.
Furthermore, the tissue response is accompanied by a marked
formation of connective tissue appearing as scar plate around
the mesh. In contrast to Marlex威, the Vypro威 mesh reveals a
significantly decreased inflammation and chronic foreign
body reaction with a significantly reduced connective tissue
formation.

TABLE II. Clinical Infections and Bacteria Detection in Cultures

after Explantation of Mesh Samples Wounds with 5 ⴛ 106 S.
aureus (n ⴝ 36 rats, 7 days)

Monofilament mesh Multifilament

With Without With Without
S. aureus S. aureus S. aureus S. aureus
Clinical infection 4/36 0/36 3/36 0/36
Bacteria in cultures
Figure 2. Bacterial adherence in vitro (S. aureus) to monofilament
(48 h) 14/18 – 17/18 –
and multifilament mesh samples
 MESH AND INFECTION 769

TABLE III. Tissue response to monofilament and multifilament precipitation with metallic salts or antibacterial drugs.20,21
meshes after contamination with S. aureus in rats after 7 days Recently this has been realized for the PTFE with its postu-
[partial volume (%) or relative number of cells (%) with SD]
lated elevated infection rate.22 Nevertheless, for this material
Monofilament Multifilament Brown et al.23 could demonstrate that the adherence of S.
(Marlex威) (Vypro威) aureus is reduced compared to polypropylene, probably due
With Without With Without to its high hydrophobia. However, this mesh usually has to be
S. aureus S. aureus S. aureus S. aureus removed in case of infection.24,25
Despite the implantation of meshes in the presence of 5 ⫻
PV% inflammatory
cells 54 (9) 26 (4) 58 (13) 19 (4)
106 S. aureus, the experiments did not show an increased
PV% connective clinically manifest infection rate in rats, with neither mono-
tissue 18 (4) 43 (6) 17 (4) 22 (4) filament nor multifilament material. After 7 days in almost all
PV% fat tissue 21 (4) 25 (3) 25 (8) 49 (8) animal tissue samples bacteria were still detectable. Although
PV% vessels 3 (2) 9 (2) 2 (1) 18 (3) the performed animal experiment over 7 days could not
Macrophages % 38 (7) 43 (4) 35 (9) 30 (7) clearly demonstrate long-term clinical consequences, these
Granulocytes % 44 (11) 8 (3) 49 (14) 2 (1) data confirm the persistence of bacteria in wounds in the
Lymphocytes % 8 (3) 4 (2) 9 (2) 6 (0) presence of foreign bodies as seen by Österberg and Blom-
Giant cells % 0 (0) 2 (2) 0 (0) 4 (2) stedt9 for up to 41 days. Furthermore, in a retrieval study of
Fibroblasts % 18 (6) 36 (3) 16 (5) 11 (2) human explanted meshes, Klosterhalfen et al.26 saw a bacte-
rial colonialization in almost one third of the samples, some-
times even after years of implantation and without any ap-
However, in the case of experimental mesh infection, both pearance of a clinical infection. These persisting bacteria, in
meshes show a comparable inflammatory reaction within the most cases protected by biofilm formation (e.g., Staphylococ-
interstitium (Table III). The inflammation basically is defined cus epidermidis) might explain the increasing number of
by a dense infiltrate of polymorphonuclear granulocytes, oc- published late abscesses following mesh implantation with a
casionally forming microabscesses in the interface and the delay of up to 39 months.25,27–30 Accordingly, Leber et al.31
surrounding recipient tissues. Finally, due to the nature of an described the clinical manifestation of infections after inci-
acute and suppurative infection, the interstitial reaction is sional hernia repair with meshes with a mean delay of 6
substantially altered by oedema formation, small areas with months. Interestingly, they found significantly more infec-
signs of bleeding and focal fibrinoid necrosis. Moreover, S. tions with a multifilament polyester mesh in comparison to
aureus can easily be detected in the standard H&E stain and, monofilament polypropylene meshes.31
in particular, in specific Gram stain, indicating smaller colo- Histologically, the presence of bacteria induced a clini-
nies of the germs within the infected tissues (Figure 3). cally inapparent intensification of the local inflammation and
fibrosis, diminishing the favorable influence of the material
reduction on the tissue response. It is still unclear how often
DISCUSSION an insidious infection has to be taken into account in clinical
practice. However, the surface of the filaments is assumed not
Although there are several reasons for an increased infectious only to effect the adherence of bacteria but should influence
potential for biomaterials, the evaluation of the related risk is the tissue response as well. Thus, for the further improvement
difficult, as is evident from the high number of clinical of biocompatibility a construction of meshes of monofila-
reports failing to prove an elevated infection rate. Generally, ments seems to be favorable.
the ability of bacteria to adhere to the foreign body depends The large surface of meshes promotes the persistence of
on the polymer and its surface. bacteria in the implant bed. Though multifilament materials
The present calculations of the mesh surface revealed an might additionally favor the adherence of bacteria, either by
increased surface area for the multifilament mesh, though it the increased surface or by additional niches, a persistence of
has markedly reduced weight. The theoretical suggestions bacteria has to be taken into account in any case. The pres-
indicate that the surface of a multifilament is increased by at ence of bacteria alone seems not to be sufficient to create a
least 1⁄2 ␲ or a factor of 1.57 in comparison to monofilaments. clinically apparent infection, as confirmed by published in-
The doubled extent of adherent bacteria in vitro corresponds fection rates of 0.1%–3%.1,32,33 However, the fact that there
to the estimated increase of the filament surface for the is sometimes a long delay before manifestation might prevent
multifilament mesh to ⬎135% of the surface of the monofil- the consideration of all infections in the reported clinical
ament. This is in accordance to investigations of Shuhaiber et studies. Although the present clinical data resemble those of
al. in 1989,18 who found an increased bacterial adherence for mesh-free techniques, Tylor and O’Dwyer30 reported that
multifilament polyester in comparison to monofilament almost 20% of British surgeons have already seen late infec-
polypropylene. In contrast, in 1984 Chu and Williams19 pub- tions after mesh surgery. Thus, the construction of a low-
lished that the extent of bacterial adherence largely depends weight large-pore-size mesh of monofilaments seems to be
on the physical properties of the filaments rather than the advantageous, particularly in the case of the potentially con-
surface. Experimentally the adherence can be decreased by taminated wounds of many incisional hernias.
770 KLINGE ET AL.

Figure 3. Tissue reaction of both investigated mesh modifications after contamination with S. aureus
at Day; (A) Low power view of an infected Vypro姞 mesh in the subcutaneous space; (B), correspond-
ingly, an infected Marlex姞 mesh (both H&E,25⫻); (C) and (D) reveal the interface of Vypro姞 and Marlex姞
after infection with S. aureus with a marked inflammatory reaction in the background forming small
micro-abscesses in the surface area of the Vypro姞 mesh (both H&E, 400⫻); (E) detail of (C) (note the
staphylococci and the dense inflammatory infiltrate in the interface of the mesh) (Gram, 630⫻); (F)
oedematous area with colonies of S. aureus (Gram, 630⫻). [Color figure can be viewed in the online
issue, which is available at www.interscience.wiley.com.]

REFERENCES
1. Avtan L, Avci C, Bulut T, Fourtanier G. Mesh infections after
laparoscopic inguinal hernia repair. Surg Laparosc Endosc
1997;7:192–195.
2. Gilbert AI, Felton LL. Infection in inguinal hernia repair con-
sidering biomaterials and antibiotics. Surg- Gynecol- Obstet
1993;177:126 –130.
3. Thill RH, Hopkins WM. The use of Mersilene mesh in adult
inguinal and femoral hernia repairs: A comparison with classic
techniques. Am Surg 1994;60:553–556.
4. Sitzmann JV, McFadden DW. The internal retention repair of
massive ventral hernia. Am Surg 1989;55:719 –723.
5. Deysine M. Pathophysiology, prevention, and management of
prosthetic infections in hernia surgery. Surg Clin North Am
1998;78:1105–1115.
6. Elek S, Conen PE. The virulence of staphylococcus pyogeneus
for man. Br J Exp Path 1957;38:573.
7. Sharp WV, Belden TA, King PH, Teague PC. Suture resistence
to infection. Surgery 1982;91:61– 63.
8. Blomstedt B, Österberg B. Suture materials and wound infec-
tion. Acta Chir Scand 1977;144:269 –274.
9. Österberg B, Blomstedt B. Effect of suture materials on bacte-
rial survival in infected wounds. Acta Chir Scand 1979;145:
431– 434.
10. Edlich R, Panek PH, Rodeheaver GT, Turnbull VC, Kurtz LD,
Edgerton MT. Physical and chemical configuration of sutures in
the development of surgical infection. Ann Surg 1973;177:679 –
688.
11. Edlich RF, Panke PH, Rodehaver GT, Kurtz L, Edgerton M.
Surgical sutures and infection: A biomaterial evaluation.
J Biomed Mater Res Symp 1974;5:115–126.
12. Paterson-Brown S, Cheslyn-Curtis S, Biglin J, Dye J, Easmon
C, Durdley H. Suture materials in contaminated wounds: A
detailed comparison of a new suture with those currently in use.
Br J Surg 1987;74:734 –735.
 MESH AND INFECTION
13. Bucknall TE. Abdominal wound closure: Choice of suture. J R
Soc Med 1981;74:580 –585.
14. Bucknall TE, Teare L, Ellis H. The choice of a suture to close
abdominal incisions. Eur Surg Res 1983;15:59 – 66.
15. Scher KS, Bernstein JM, Jones CW. Infectivity of vascular
sutures. Am Surg 1985;51:577–579.
16. Klinge U, Klosterhalfen B, Muller M, Schumpelick V. Foreign
body reaction to meshes used for the repair of abdominal wall
hernias. Eur J Surg 1999;165:665– 673.
17. Klosterhalfen B, Klinge U, Schumpelick V. Functional and
morphological evaluation of different polypropylene-mesh
modifications for abdominal wall repair. Biomaterials 1998;19:
2235–2246.
18. Shuhaiber H, Chugh T, Burns G. In vitro adherence of bacteria
to sutures in cardiac surgery. J Cardiovasc Surg Torino 1989;
30:749 –753.
19. Chu CC, Williams DF. Effects of physical configuration and
chemical structure of suture materials on bacterial adhesion. A
possible link to wound infection. Am J Surg 1984;147(2):197–
204.
20. Farrah SR, Erdos GW. The production of antibacterial tubing,
sutures, and bandages by in situ precipitation of metallic salts.
Can J Microbiol 1991;37:445– 449.
21. Troy MG, Dong QS, Dobrin PB, Hecht D. Do topical antibiotics
provide improved prophylaxis against bacterial growth in the
presence of polypropylene mesh? Am J Surg 1996;171:391–
393.
22. Carbajo MA, et al. Laparoscopic treatment of ventral abdominal
wall hernias: Preliminary results in 100 patients. Jsls 2000;4:
141–145
771
23. Brown GL, Richardson JD, Malangoni MA, Tobin GR, Acker-
man D, Polk HC, Jr. Comparison of prosthetic materials for
abdominal wall reconstruction in the presence of contamination
and infection. Ann Surg 1985;201:705–711.
24. Amid P. Classification of biomaterials and their related compli-
cations in abdominal wall hernia surgery. Hernia 1997;1:5– 8.
25. Slim K, Pezet D, Le Roux S, Chipponi J. Mesh infection after
laparoscopic herniorrhaphy. Eur J Surg 1996;162:247–248.
26. Klosterhalfen B, Klinge U, Hermanns B, Schumpelick V. Pa-
thology of traditional surgical nets for hernia repair after long-
term implantation in humans. Chirurg 2000;71:43–51.
27. Foschi D, Corsi F, Cellerino P, Trabucchi A, Trabucchi E. Late
rejection of the mesh after laparoscopic hernia repair. Surg
Endosc 1998;12:455– 457.
28. Coda A, Micca F, Bossotti M. Reoperations for chronic infec-
tions following prosthetic hernia repair. Hernia 1998;2:163–
167.
29. Hofbauer C, Andersen PV, Juul P, Qvist N. Late mesh rejection
as a complication to transabdominal preperitoneal laparoscopic
hernia repair. Surg Endosc 1998;12:1164 –1165.
30. Tylor S, O’Dwyer P. Chronic groin sepsis following tension-
free inguinal hernioplasty. Br J Surg 1999;86:562–565.
31. Leber GE, Garb JL, Alexander AI, Reed WP. Long-term com-
plications associated with prosthetic repair of incisional hernias.
Arch Surg 1998;133:378 –382.
32. Loh A, JS Rajkuma, LM South. Anatomical repair of large
incisional hernias. Ann R Coll Surg Engl 1992;74:100 –105.
33. Kumar S. Chronic groin sepsis following tension-free inguinal
hernioplasty. Br J Surg 1999;86:1482.