In Vivo Biocompatibility of Three Potential Intraperitoneal Implants

Full Paper

1336

In Vivo Biocompatibility of Three PotentialIntraperitoneal Implantsa

Sylvie Defrere,b Melanie Mestagdt,b Raphael Riva, Fabrice Krier,Anne Van Langendonckt, Pierre Drion, Christine Jerome, Brigitte Evrard,Jean-Paul Dehoux, Jean-Michel Foidart, Jacques Donnez*

The intraperitoneal biocompatibility of PDMS, polyHEMA and pEVA was investigated in rats,rabbits and rhesus monkeys. No inflammation was evidenced by hematological analyses andmeasurement of inflammatory markers throughout the experiment and by post-mortemexamination of the pelvic cavity. After 3 or 6 months,histological analysis revealed fibrous tissue encapsulat-ing PDMS and PEVA implants in all species and poly-HEMA implants in rabbits and monkeys. Calciumdeposits were observed inside polyHEMA implants.The intraperitoneal biocompatibility of all 3 polymersmakes them suitable for the design of drug deliverysystems, which may be of great interest for pathologiesconfined to the pelvic cavity.

S. Defrere, A. Van Langendonckt, Prof. J. DonnezUniversite Catholique de Louvain, Institut de Recherche Cliniqueet Experimentale, Department of Gynecology, Avenue Hippocrate10, 1200 Brussels, BelgiumFax: þ32-2-764.95.07; E-mail: [email protected]. Mestagdt, Prof. J.-M. FoidartLaboratory of Tumor and Development Biology, University ofLiege, 4000 Liege, BelgiumR. Riva, Prof. C. JeromeCenter for Education and Research on Macromolecules, Universityof Liege, 4000 Liege, BelgiumF. Krier, Prof. B. EvrardLaboratory of Pharmaceutical Technology, Department ofPharmacy, University of Liege, 4000 Liege, BelgiumP. DrionAnimal Facility-GIGA, University of Liege, 4000 Liege, BelgiumProf. J.-P. DehouxExperimental Surgery Unit, Universite Catholique de Louvain,1200 Brussels, Belgium

a Supporting Information for this article is available from the WileyOnline Library or from the author.

b S. Defrere and M. Mestagdt contributed equally to this work.

Macromol. Biosci. 2011, 11, 1336–1345

� 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonline

Introduction

In recent years, various approaches have been implemen-

ted in an attempt to develop devices to effectively control

the release of drugs. Among the different approaches, active

substances have been incorporated into biostable polymers

as well as biodegradable systems.[1–3] These controlled

delivery systems can be localized in specific regions of

the body, such as vaginal, intrauterine, subcutaneous, or

corneal sites. Although some degree of systemic absorption

is inevitable, with its associated side effects, local delivery

appears to be a more efficient alternative, with limited

adverse effects and increased patient compliance, particu-

larly during long-term treatment.[4]

Biocompatibilty is a key issue to consider when

developing a drug delivery system. To be biocompatible,

an implant should elicit a very low host response in a given

implantation site, essential to prevent harmful effects in

the host, but also maintain the implant function.[5] One of

the factors affecting the biocompatibility of an implant is

protein or other biomolecule adsorption on its surface,

known to trigger an inflammatory response involving

library.com DOI: 10.1002/mabi.201100077

In Vivo Biocompatibility of Three Potential Intraperitoneal Implants

www.mbs-journal.de

recruitment and adhesion of leukocytes, blood clotting,

foreign body reactions, and fibrous encapsulation. Many

properties of a synthetic material have been shown to

influence protein adsorption and affect its behavior in a

biological environment, including its composition, net

charge, wettability and surface Gibbs free energy, as

previously reported.[6]

Several studies have recently demonstrated that the

anatomical site (subcutaneous, under the kidney capsule or

intraperitoneal) of the implant markedly influences host

response to synthetic matrices.[7,8] The objective of the

present study was to specifically evaluate the peritoneal

biocompatibility of three different polymers.

Drug delivery into the pelvic cavity may be of great

interest in the context of pathologies predominantly

confined to the peritoneal cavity, such as endometriosis,[4]

ovarian cancer[9,10] and gastric cancer.[11] The peritoneal

cavity is a specific environment that is quite dynamic and

linked to the immune system. Peritoneum is the most

extensive serous membrane in the body, with an estimated

surface area of >2 m2.[12] It comprises two layers: a loose

connective tissue layer that contains collagen, elastic fibers,

adipocytes and macrophages, and a mesothelial layer that

consists of squamous cells. Peritoneal fluid arises primarily

from two different sources: plasma transudate and ovarian

exudate; other sources include tubal fluid, retrograde

menstruation and macrophage secretions.[13] Peritoneal

fluid contains soluble constituents (steroid hormones,

cytokines, growth factors, angiogenic factors, etc.) in

concentrations different from those found in plasma,[14]

and a variety of free floating cells (such as macrophages,

erythrocytes, mesothelial cells, lymphocytes, natural killer

(NK) cells, eosinophils and mast cells).

In this study, we focused on the peritoneal biocompat-

ibility of three polymers showing varying degrees of

hydrophobicity: polydimethylsiloxane (PDMS) (the most

hydrophobic), poly(ethylene-co-vinyl acetate) (PEVA), and

poly(hydroxyethyl methacrylate) (polyHEMA) (the most

hydrophilic).

Polysiloxanes like PDMS are biomaterials that have been

used for a long time now: in urology for the treatment

for impotence[15] or incontinence,[16] in ophthalmology for

lens implantations,[17] and in plastic surgery for breast

implants.[18] PDMS has also been utilized for subcuta-

neously controlled drug delivery of levonorgestrel for the

purpose of contraception.[19]

PEVA was initially described as a biomaterial in dental

surgery.[20] It has also been used in subcutaneously

controlled drug delivery systems of nalmefene to treat

alcoholism,[21] and etonogestrel for contraception,[19] as

well as in alternative modes of contraceptive delivery, such

as the vaginal ring.[22] A recent study investigated the

release of two drugs by a PEVA system for the intraoral

treatment of oral lesions, blisters, and fungal diseases.[23]

www.MaterialsViews.com

Macromol. Biosci. 2011

� 2011 WILEY-VCH Verlag Gmb

PolyHEMA is widely used as a biomaterial in a variety of

areas, including eye health care (contact lenses, ocular

implants),[17] dentistry, and controlled drug release,[24] and

tentatively in orthopedic, neurosurgical[25] and cardiovas-

cular[26] applications.

However, to our knowledge, PDMS, polyHEMA and PEVA

implant biocompatibility has never before been studied

long-term (6 months) in the peritoneal cavity. In the present

study, we evaluated the biocompatibility of these three

polymers (PDMS, polyHEMA and PEVA) after synthesizing

implants and placing them intraperitoneally into rats,

rabbits and rhesus monkeys. Blood samples were taken

regularly and inflammation parameters were assessed.

After 3 or 6 months, implants were recovered and

histological analyses were performed.

Experimental Section

Materials for Implant Synthesis

Materials used were PDMS (base, medical grade), tetrapropyl

orthosilicate (cross-linker, medical grade), tin octoate (SnOct2,

catalyst, medical grade), hydroxyethyl methacrylate (HEMA,

Aldrich, distilled before use, non-medical grade), ethylene glycol

dimethacrylate (EGDMA, Aldrich, 0120220, non-medical grade),

ammonium persulfate (APS, Aldrich, batch 07922MC, non-medical

grade), N,N,N0,N0-tetramethyl ethylene diamine (TEMED, Aldrich,

109F0792, non-medical grade) and PEVA (Elvax 3185, Dupont,

batch 70106161, non-medical grade).

Production of PDMS Implants[27]

Preparation of the PDMS Mixture

Typically, 19.4 g of PDMS was placed in a sterile container and kept

at �20 8C for 1 h. Thereafter, 0.5 g of tetrapropyl orthosilicate and

0.1 g of SnOct2 were mixed together in a separate glass container.

This mixture of catalyst and cross-linker was then added to the cold

PDMS under a laminar flow hood to limit contamination. The PDMS

blend was manually mixed for 2 min before being placed under a

vacuum for 5 min in order to remove trapped air bubbles. The PDMS

mixture was finally transferred to a plastic syringe and maintained

at �20 8C.

PDMS Molding

PDMS implants were prepared by cross-linking the PDMS mixture

in a mold at 80 8C. This mold, composed of an iron core covered with

Teflon film, allows preparation of 12 implants of 20 mm in length

and 3 mm in diameter in a row. The PDMS mixture contained in the

syringe was injected into the mold, which was then compressed at

80 8C under a pressure of 30 bar. After 15 min, the mold was cooled

at room temperature and the collected implants were transferred

to a sterile device. The implants were then placed in a Vismara

vacuum oven (VO65) at 950 mbar for 4 h at room temperature to

remove the propanol resulting from the PDMS cross-linking.

, 11, 1336–1345

H & Co. KGaA, Weinheim1337

1338

www.mbs-journal.de

S. Defrere et al.

Production of PEVA Implants

PEVA implants were prepared by extrusion of PEVA pellets in a

micro-extruder (DSM 5 cm3 micro-extruder equipped with a twin-

screw). For this purpose, PEVA pellets were immersed in ethanol in

order to extract butyl hydroxytoluene (BHT). After filtration, they

were dried under a vacuum at room temperature. Thereafter, 8 g of

PEVA was introduced into the twin-screw micro-extruder at 80 8Cat a rotation rate of 100 rpm. After 5 min of mixing, the resulting

PEVA rods were collected and directly placed into sterile water,

primarily to fix their geometry, but also to avoid adsorption of dirt

onto the surface. The rods were then cut into 2-cm implants and

stored in a sterile bag.

Production of polyHEMA Implants[28]

Typically, 5 mL of freshly distilled HEMA containing 0.1% in weight

of EGDMA were transferred to a glass tube. After degassing the

solution by nitrogen bubbling for 5 min, 1.67 mL of APS aqueous

solution (APS concentration¼0.024 mol � L�1) and 7mL of TEMED

were added. After short homogenization, the solution was

transferred to an insulin syringe used as a sterile and disposable

mold. The syringes were placed under a laminar flow hood at room

temperature for 12 h to allow polymerization. The implants were

then collected by applying simple pressure to the syringe piston.

The ends of the implants were cut. The polyHEMA implants were

then washed by repeated immersion in sterile water to remove

unreacted HEMA. After washing 5 times, the implants were placed

in a sterile aqueous solution.

Contact Angle Measurement

Contact angle measurements were obtained by depositing a drop of

water (15ml) on a plane surface of each polymer using the strategy

developed for the synthesis of implants on a GBX DGD fast 160

contact angle meter. Contact angles were evaluated after 30, 60, 90

and 120 s of contact and the experiment was repeated three times.

Sterilization of Implants

Sterilization was performed in accordance with the EC Guide to

Good Manufacturing Practice and Ph. Eur standards (6th edition,

2008 (6.5), point 5.1.1).

PDMS and PEVA implants were sterilized by an ethylene oxide

(EO) procedure, as they do not tolerate any other sterilization

method. This procedure entails the usual three steps: conditioning,

sterilization and aeration.

It begins with an 8-hour conditioning step at 45 8C�5 8C and

relative humidity (RH) controlled between 50% and 80%.

The next step is the sterilization phase at a temperature of 45 8C,

aiming for RH of 60%� 20%. Contact time is equal to 3 h with an EO

concentration of 730 mg � L�1. This process has been validated using

biological indicators with a sterility assurance level of 10�6.

The final step is the aeration process, which occurs at 40 8C�5 8Cwith fresh air inbleed, and is necessary to desorb EO residues from

products. The implants are then placed in an aeration room for 72 h.

PolyHEMA implants were sterilized by steam at 121 8C for

30 min. This process has been validated by ThermalogTM.



Intraperitoneal Placement of Implants in Rats

All experimental procedures and protocols used in this investiga-

tion were reviewed and approved by the Institutional Animal Care

and Use Committee of the Universite catholique de Louvain. The

‘‘Guide for the Care and Use of Laboratory Animals,’’ prepared by

the Institute of Laboratory Animal Resources, National Research

Council, and published by the National Academy Press, was

followed carefully (Commission on Life Sciences 1996). Forty-eight

10- to 12-week-old female Wistar rats (Faculty of Medicine Animal

House, Universite Catholique de Louvain, Brussels, Belgium) were

used for the present study (16 rats/implant type; 10 rats with 2

implants and 6 control rats without implants).

The rats were anesthetized by intraperitoneal (i.p.) injection of

ketamine (30 mg � kg�1, Anesketin1; Eurovet, Heusden-Zolder,

Belgium) and medetomidine (50mg � kg�1, Domitor1; Pfizer, Cam-

bridge, MA, USA). After surgery, anesthesia was reversed by

injection of atipamezole (0.25 mg � kg�1, Antisedan1; Pfizer).

Buprenorphine (0.1 mg � kg�1, Temgesic1; Schering Plough, Kenil-

worth, NJ, USA) was injected for analgesia.

A 2-cm median laparotomy was performed. Two implants per

rat were fixed to parietal peritoneum with non-absorbable Surgipro

sutures (Tyco Healthcare, Mechelen, Belgium), one on the left side

and one on the right. The abdominal wall and skin were then

sutured with absorbable 3/0 Vicryl (Ethicon, Sommerville, NJ, USA).

The parietal peritoneum of control animals without implants was

sutured with non-absorbable stitches.

Blood samples were collected regularly (3� during the first

week, then 1�/week) from the tail for blood count and serum

biochemistry.

At the end of the experiment, 3 or 6 months after placing the

implants, the rats were euthanized by inhalation of CO2, according

to standard euthanasia guidelines for rodents from the AVMA

Guidelines on Euthanasia (Formerly Report of the AVMA Panel on

Euthanasia) of June 2007 (http://www.avma.org/issues/animal_

welfare/euthanasia.pdf). Eight rats (5 with 2 implants and 3

controls without implants) were euthanized after each time period.

Intraperitoneal Placement of Implants in Rabbits



and Use Committee of the University of Liege. The ‘‘Guide for the

Care and Use of Laboratory Animals,’’ prepared by the Institute of

Laboratory Animal Resources, National Research Council, and

published by the National Academy Press, was followed carefully

(Commission on Life Sciences 1996). Forty-two 4-month-old female

New Zealand white rabbits (Harlan, the Netherlands) were used for

this study (14 rabbits/implant type; 10 rabbits with 2 implants and

4 control rabbits without implants). The biocompatibility of the 3

polymers was evaluated in the peritoneal cavity of rabbits over a

period of 3 or 6 months.

The rabbits were anesthetized by Isoflurane (IsoFlo, Abbott

Animal Health, IL, USA) inhalation. A 3-cm median laparotomy was

performed along the white line of the abdomen. Two implants were

placed in the peritoneal cavity of treated animals and fixed to

parietal peritoneum with non-absorbable Surgipro sutures (one on

the left side and one on the right), while control animals underwent

, 11, 1336–1345

H & Co. KGaA, Weinheim www.MaterialsViews.com


www.mbs-journal.de

the same surgery but without implants. The peritoneum, muscles

and skin were then sutured with absorbable 3/0 Vicryl (Ethicon).

Buprenorphine (0.1 mg � kg�1, Temgesic1) was injected for

analgesia.

Blood samples were collected from the marginal ear vein on day

0, day 4 and then 1� /week till the end of the experiment. After 3 or

6 months, the rabbits were euthanized by i.v. pentobarbital at a

dose of 200 mg � kg�1 20 min after fentanylþdehydrobenzperidol

premedication (0.22 mL � kg�1 of a bolus 25 mgþ1.25 mg �mL�1 IM).

Seven rabbits (5 with 2 implants and 2 controls without implants)

were euthanized after each time period.

Intraperitoneal Placement of Implants in Rhesus

Monkeys



and Use Committee of the Universite Catholique de Louvain and

Belgian legislation on the care and use of laboratory animals.

Three female rhesus macaques (Macaca mulatta) (#124, #220,

and #246) weighing 4.0�0.1 kg, obtained from China via The

Netherlands (Hartelust, Tilburg, The Netherlands), were housed in

the animal facility, where room temperature is maintained at 21 8C,

with 60% humidity and a 12:12 light/dark cycle. They were fed

monkey pellets (Safe1, Augy, France) and seasonal fruit and

vegetables, and had free access to water.

Before any manipulation, the animals were anesthetized with

tiletamine/zolazepam (Zoletil 1001, 6 mg � kg�1; Virbac, Nice,

France) and xylazine (Rompun1 2%, 2 mg � kg�1; Bayer, Sint-

Truiden, Belgium). A 2-cm median laparotomy was performed. Two

implants per monkey were fixed to parietal peritoneum with non-

absorbable sutures, one on the left side and one on the right. After

abdominal closure, the animals received antibiotics for 1 week

(Clamoxyl LA1, Pfizer, Paris, France) and postsurgical pain control

(buprenorphine/Temgesic1 and ibuprofen/Ketofen1; Merial,

Lyon, France) for 3 d.

After 3 months, the monkeys were euthanized by anesthesia

followed by i.v. injection of embutramide/mebezonium/tetracaine

hydrochloride (0.3 mL � kg�1, T611; Eurovet).

Inflammation Evaluation

Inflammation was evaluated by regular hematological analyses

and measurement of inflammatory markers (C-reactive protein

(CRP) and fibrinogen) in the blood of animals throughout the

experiment, and by post-mortem examination of the peritoneal

cavity.

In rats, total blood count and differential leukocytes were

determined using an MS9 cell counter (MS9-3; Melet Schloesing

Laboratoires, Osny, France). Serum CRP assays were performed on

the BNII nephelometer (Dade Behring, Marburg, Germany) and

fibrinogen assays on the Sysmex analyzer CA-7000 (Dade Behring).

In rabbits, CRP and fibrinogen measurements and blood cell

count were carried out at the Collard Laboratory (Liege, Belgium)

(according to ISO 45001 and ISO 15189 norms).

After 3 or 6 months, the animals were euthanized, and a

laparotomy and post-mortem examination were performed.

Implants were macroscopically examined for signs of encapsula-




tion and removed for histological analysis. The implantation site,

peritoneal organs and peritoneal cavity were carefully inspected

for typical signs of inflammation, infection or injury.

Histological Analysis

Implants were recovered by laparotomy. PDMS and polyHEMA

implants were immediately fixed in 4% buffered formaldehyde and

embedded in paraffin. PEVA implants were embedded in optimal

cutting temperature compound (Tissue-Tek1; Sakura Finetek

Europe B.V., Zoeterwoude, The Netherlands), frozen in liquid

nitrogen-cooled isopentane, and stored at �80 8C. Embedded

implants were then cut into sections of 5mm and slides were

stained with hematoxylin–eosin. Ten sections per implant were

carefully examined under a microscope for histological analysis.

Statistical Analysis

Statistical analysis was applied to compare CRP, fibrinogen and

hematological data between animals with and without implants.

The non-parametric two-tailed Mann-Whitney U test was used,

with p< 0.5 considered statistically significant. Statistical analysis

was performed using SPSS 14.0 software (SPSS Inc., Chicago, IL).

Results

Inflammation Evaluation

All the animals were found to be in good health, remaining

active with no change in their weight or appearance.

Inflammatory Markers: CRP and Fibrinogen

In all rat blood samples, from animals with or without

PDMS, polyHEMA or PEVA implants, CRP concentrations

were always below the detection limit (<0.015 mg �dL�1)

throughout the analysis. In rabbits, CRP levels ranged

between 0.1 and 0.6 mg �dL�1, but remained under the

positive reference value (0.9 mg �dL�1) in all cases (Table 1).

As reported in Table 2, in all rat samples, fibrinogen

concentrations were around 1–2 g � L�1, which are physio-

logical levels.[29,30] No difference was detected between

control rats and rats with implants with any of the

polymers (PDMS, polyHEMA or PEVA). In rabbits, fibrinogen

concentrations remained under the reference value

(4 g � L�1) throughout the entire experiment (from 0 to

6 months), in both control rabbits and those with implants

made of any of the three polymers (see Table 2).

Hematological Analysis

Basic hematological values were unaltered between rats

with implants and those without, for all 3 polymers tested

(Table 3a) and for each implantation period (data not

shown). All values were within the normal reference range

for Wistar rats. Similarly, in rabbits, the presence of polymer

implants did not modify the blood leukocyte count and all

, 11, 1336–1345


Table 1. CRP concentrations in rabbits (mg �dL�1; mean� standard deviation).

Control PDMS PDMS Control polyHEMA PolyHEMA Control PEVA PEVA

Day 0 0.1� 0 0.1� 0 0.1� 0 0.14� 0.03 0.11� 0.01 0.16� 0.10

Day 3 0.21� 0.05 0.21� 0.06 0.25� 0.06 0.48� 0.14 0.18� 0.03 0.21� 0.14

1 month 0.18� 0.11 0.29� 0.07 0.15� 0.03 0.30� 0.14 0.16� 0.04 0.26� 0.14

3 months 0.20� 0.04 0.17� 0.07 0.20� 0.09 0.21� 0.07 0.11� 0.01 0.17� 0.02

6 months 0.11� 0.01 0.13� 0.03 0.19� 0.11 0.22� 0.15 0.19� 0.09 0.21� 0.14

Table 2. Fibrinogen concentrations in rats and rabbits (g � L�1; mean� standard deviation).


Rats Day 0 1.49� 0.46 1.70� 0.40 1.67� 0.35 1.85� 0.25 0.99� 0.33 1.08� 0.21

Day 3 1.68� 0.71 1.63� 0.67 1.98� 0.80 2.05� 0.59 1.28� 0.29 1.16� 0.36

1 month 1.38� 0.31 0.90� 0.33 1.54� 0.15 1.06� 0.37 1.33� 0.22 0.97� 0.36

3 months 0.97� 0.34 0.88� 0.30 1.74� 0.53 1.74� 0.60 1.18� 0.31 1.14� 0.33

6 months 1.05� 0.40 0.92� 0.28 1.86� 0.56 1.62� 0.54 1.22� 0.29 1.10� 0.34

Rabbits Day 0 1.15� 0.16 1.55� 0.23 1.45� 0.35 1.55� 0.13 1.80� 0.42 1.84� 0.33

Day 3 1.59� 0.46 1.43� 0.15 1.35� 0.21 1.78� 0.31 1.75� 0.21 1.54� 0.51

1 month 1.95� 0.49 1.54� 0.70 1.6� 0.14 1.54� 0.24 1.60� 0.14 1.70� 0.35

3 months 1.90� 0.28 1.68� 0.12 1.65� 0.07 1.40� 0.44 1.90� 0.14 1.64� 0.66

6 months 1.95� 0.35 2.25� 0.53 1.70� 0.17 1.85� 0.37 2.10� 0.11 1.80� 0.43

Table 3a. Hematological data on rats after 6 months.


WBC (103 �mm�3) 8.2� 0.9 5.9� 2.2 7.3� 1.1 7.2� 1.4 4.8� 4.6 4.3� 2.5

Diff

Lymphocytes (%) 81.1� 2.4 80.3� 5.7 85.7� 1.4 86.9� 5.4 81.5� 9.1 82.8� 3.3

Monocytes (%) 7.2� 0.9 8.3� 2.0 6.0� 0.5 5.6� 1.8 8.3� 2.4 7.6� 1.3

Granulocytes (%) 11.7� 1.5 11.4� 3.7 8.3� 1.1 7.5� 3.6 10.2� 6.9 9.4� 2.1

RBC (106 �mm�3) 7.5� 0.6 7.0� 0.6 6.7� 1.0 6.5� 1.4 5.2� 1.6 5.8� 1.4

HCT (%) 45.3� 2.7 42.8� 3.7 37.7� 6.2 36.7� 8.7 28.5� 8.6 33.0� 8.7

HGB (g � dL�1) 14.0� 0.3 13.5� 1.3 12.6� 2.1 12.4� 2.6 9.6� 3.0 11.5� 2.8

PLT (103 �mm�3) 771.0� 97.1 675.4� 127.6 712.0� 332.9 672.8� 220.2 737.3� 235.5 417.0� 401.9

WBC: total white blood cell count; Diff: differential white blood cell count; RBC: red blood cell count; HCT: hematocrit; HGB: hemoglobin;

PLT: platelet count.

1340

www.mbs-journal.de

S. Defrere et al.

values were found to be within reference intervals

(Table 3b).

Peritoneal Cavity Examination

Visual post-mortem examination of the peritoneal cavity

did not reveal any signs of inflammation, infection or injury

in any animals (rats, rabbits or monkeys) with or without

PDMS, polyHEMA or PEVA implants for either 3 or 6 months

(Figure 1).



Implant Retrieval and Macroscopic Aspect

As illustrated in Figure 1, macroscopic examination of the

pelvic cavity showed some implants to be fixed to parietal

peritoneum with non-absorbable suture (as originally

placed) (Figure 1B), but also some free implants within

the peritoneal cavity (Figure 1C, 1E and 1H), and implants

enclosed by surrounding adipose tissue, sutured (Figure 1B)

or not (Figure 1I) to the peritoneum. In some cases, implants

, 11, 1336–1345


Table 3b. Hematological data on rabbits after 6 months.

Control PDMS PDMS Control polyHEMA PolyHEMA Control EVA EVA

WBC (103 �mm�3) 5.9� 1.9 4.5� 0.9 5.2� 0.6 4.3� 1.0 4.2� 1.7 4.2� 1.1

Diff

Lymphocytes (%) 69.5� 7.8 47.5� 11.1 50.0� 2.4 59.8� 10.9 32.5� 12.0 37.8� 14.9

Monocytes (%) 6.0� 1.4 4.5� 1.0 2.5� 0.7 2.2� 0.2 8.0� 3.5 7.8� 4.7

Granulocytes (%) 24.5� 9.2 48.0� 10.7 47.5� 24.8 38.0� 11.1 59.5� 20.5 54.4� 13.8

RBC (106 �mm�3) 5.5� 0.6 5.7� 0.2 4.8� 0.1 4.9� 0.4 5.3� 0.4 5.0� 0.2

HCT (%) 36.0� 4.2 36.6� 1.1 32.7� 0.7 32.0� 3.4 33.5� 2.1 32.2� 1.1

HGB (g � dL�1) 11.8� 1.2 12.2� 0.3 11.2� 0.1 10.9� 1.3 11.2� 0.8 10.7� 0.4

PLT (103 �mm�3) 328.0� 16.1 372.6� 39.7 326.5� 44.6 300.0� 127.0 205.0� 40.0 324.6� 36.3

WBC: total white blood cell count; Diff: differential white blood cell count; RBC: red blood cell count; HCT: hematocrit; HGB: hemoglobin;

PLT: platelet count.


www.mbs-journal.de

were recovered stuck together (Figure 1C, 1D, 1H and 1I). The

macroscopic aspect of PDMS, polyHEMA and PEVA implants

placed in the peritoneal cavity of rats, rabbits and monkeys

for a period of 3 or 6 months is summarized in Table 4.

Macroscopic examination also revealed a thin white

layer (Figure 1D) on the surface of all PDMS and PEVA

implants after 3 or 6 months in the peritoneal cavity of all

the animals. This layer was histologically identified as

fibrocellular tissue surrounding the implants, as described

below. This white layer was not macroscopically visible in

rats on the surface of polyHEMA implants.

Histological Analysis

All sections were examined by an experienced pathologist.

Hematoxylin-eosin staining of the implants and their

surrounding encapsulation (if present) showed the extent

of inflammation and fibrosis produced by the polymers.

Histologically, the implants appeared to be biocompatible,

as we did not observe any signs of major inflammatory

response with any of the polymers (PDMS, polyHEMA or

PEVA) after 3 or 6 months.

PDMS and EVA Implants

In rats and rabbits, thin encapsulating fibrocellular tissue

was observed around all the PDMS and EVA implants (after

3 or 6 months’ implantation) (Figure 2A-D and Figure 3).

This capsular tissue showed low to moderate cellularity.

Fibroblasts were predominant in the cell population, with

some inflammatory cells and a mesothelial layer on both

sides (implant side and peritoneal cavity side). We did not

observe any multinucleated foreign body giant cells

(FBGCs) in this capsular tissue, but FBGCs were detected

around suture. Macroscopic examination showed some

implants to be simply surrounded by encapsulating




fibrocellular tissue, while in other cases, this tissue was

adherent to adipose tissue (Figure 2B and 2D).

In monkeys, only one PDMS implant and both PEVA

implants were recovered after 3 months. These implants

were encapsulated in fibrocellular tissue and adherent to

peritoneal tissue (Figure 4A and B). This capsular tissue

showed moderate cellularity. A mesothelial cell layer was

observed on the implant side, and so fibroblasts were

predominant in the cell population, with some inflamma-

tory cells. We did not observe any FBGCs in this capsular

tissue, but detected FBGCs around suture.

PolyHEMA Implants

In rats, there were no signs of fibrous encapsulation around

polyHEMA implants, some parts of the surface were free of

cells, while other parts were covered with cells (without any

conjunctive tissue) (Figure 2E). After 3 months, calcium

deposits were identified in 2 implants (20%). After 6 months,

calcium deposits were observed in all 10 implants

(Figure 2F).

In rabbits, thin encapsulating fibrocellular tissue was

observed around all the implants after 3 and 6 months, as

seen with PDMS and PEVA implants (Figure 3).

In monkeys, two implants were recovered after 3

months, encapsulated in fibrocellular tissue (Figure 4C).

This capsular tissue showed low to moderate cellularity.

Fibroblasts were predominant in the cell population, with

some inflammatory cells. Calcium deposits were observed

in both implants.

Discussion

In the present study, we synthesized and evaluated the

biocompatibility of three polymers (PDMS, polyHEMA and

, 11, 1336–1345


Figure 1. Macroscopic aspect of PDMS implants recovered after6 months from the peritoneal cavity of rats (A-E) and rabbits (F-I).A: Peritoneal cavity of a control rat, with non-absorbable suturesvisible (arrows) on parietal peritoneum. B: Implant (arrow) fixedto parietal peritoneum, as initially placed, and bound in part toadipose tissue. C: Two implants stuck together. D: The sameimplants as C, removed from the pelvic cavity. A thin white layeris visible on the surface of the implants. E: Implant free in theperitoneal cavity. F: Peritoneal cavity of a control rabbit. G:Peritoneal cavity of a rabbit with 2 PDMS implants. H: Twoimplants stuck together but free in the peritoneal cavity. I:Two implants stuck together and bound to adipose tissue.

1342

www.mbs-journal.de

S. Defrere et al.

PEVA) showing different degrees of hydrophobicity. Their

hydrophobicity was quantified by measurement of the

contact angle of a water drop on a plane surface of each

(Table 5). As expected, PDMS was the most hydrophobic

polymer. Indeed, the value of the contact angle was higher

than with the other polymers, which indicated a weaker

affinity for water. On the other hand, polyHEMA exhibited

the lowest contact angle, and was thus the least hydro-

phobic, whereas PEVA showed a value between the two.



The different implants prepared for this study had an

identical shape (cylinder of 2 cm in length with a diameter

of 3 mm) and were particularly flexible in order to avoid

injury when placed into the intraperitoneal cavity of rats,

rabbits and rhesus monkeys for a period of 3 or 6 months.

Regular hematological analyses and measurement of

inflammatory markers (CRP and fibrinogen) in the blood of

live animals, as well as post-mortem peritoneal examina-

tion (after 3 or 6 months), did not evidence any signs of

inflammation. Histological analysis showed that fibrous

encapsulation developed around PDMS and EVA implants

in all three species, and around polyHEMA implants only in

rabbits and monkeys. In rats, cell coverage of polyHEMA

implant surfaces was limited. Calcium deposits were

observed inside polyHEMA implants in rats and monkeys.

Formation of a fibrous capsule is commonly observed

after implantation of biomaterials. This process, known as

foreign body reaction, is triggered by non-specific adsorp-

tion of proteins on the implant surface shortly after

biomaterial implantation, and involves a complex cellular

process, as described by Liu et al.[31] This protein coating

then triggers neutrophil followed by macrophage recruit-

ment. Macrophages either digest and eliminate the implant

considered to be a foreign body, or fuse into MNGCs to

enclose it. The appearance of these giant cells signals the

recruitment of fibroblasts synthesizing the collagen coat-

ing, which will encapsulate the biomaterial within 2 to

4 weeks of implantation.[31]

However, the anatomical site of the implant has also

been shown to influence host response to polymeric

materials.[7,8] The intraperitoneal cavity is a very specific

environment, but polymer biocompatibility has been

poorly investigated in this area, despite the fact that

intraperitoneal drug delivery through a specially adapted

system could well prove useful for pathologies confined to

the peritoneal cavity.

In the present study, we did not observe a typical foreign

body reaction, as we did not detect any FBGCs in the fibrous

capsule around the implant, but only around suture.

Our findings with PDMS implants are consistent with

those of an earlier study.[32] Indeed, in Belanger’s study, in

vivo biocompatibility of PDMS was investigated after

intraperitoneal implantation in a rat model for 1, 2, 6, 9, and

12 weeks. The results showed the percentages of peripheral

blood T cells to be similar to those in the control group for

each implantation time. Histologically, PDMS triggered a

tissue healing cascade, characterized by a mild inflamma-

tory reaction in the first two weeks, followed by a more

discrete chronic inflammatory phase from 6 to 12 weeks

post implantation. Our study showed similar results after 3

and 6 months, demonstrating, for the first time, that no

further major evolution takes place beyond 3 months.

Our data evidenced calcification of polyHEMA in rats and

monkeys, as previously reported in the literature. However,

, 11, 1336–1345


Table 4. Macroscopic observations of removed implants.

Species Experiment

duration

PDMS PolyHEMA PEVA

Rats 3 months 2 implants fixed to parietal

peritoneum, as initially

placed

6 implants fixed to parietal


placed


peritoneum, as initially placed

5 implants free in the PC 4 implants free in the PC 4 implants free in the PC

4 implants bound to perito-

neal tissue

6 months 2 implants fixed to parietal


placed



placed

7 implants free in the PC

2 implants free in the PC 6 implants free in the PC 3 implants bound to peritoneal

tissue

2� 2 implants stuck together

3 implants bound to

peritoneal tissue

Rabbits 3 months 4 implants free in the PC 6 implants free in the PC 6 implants free in the PC

4 implants stuck together

(2� 2) free in the PC





2 implants bound to adipose

tissue


tissue

6 months 2 implants free in the PC 4 implants free in the PC 4 implants free in the PC








tissue


tissue


tissue

Rhesus 3 months 1 implant lost 1 implant fixed to parietal


placed


peritoneum, as initially placed

1 implant bound to perito-

neal tissue

1 implant bound to perito-

neal tissue


www.mbs-journal.de

we were also able to prove that polyHEMA calcification is

species-dependent, as we did not observe calcification in

rabbits.

Indeed, there is compelling evidence that HEMA-based

hydrogels can undergo dystrophic calcification when

placed inside living tissues, or even in the absence of any

biological agents (in aqueous solutions of calcium ions and

phosphate ions).[33] Biomaterial-associated calcification

involves deposition of calcium phosphate phases (CaP)

onto and within the biomaterial itself. Depending on the

particular field of application, calcification is either actively

pursued or avoided. For example, ocular applications of

hydrogels require the use of transparent hydrogels, so

calcification must be prevented. On the contrary, in dental




implants or orthopedic applications, deposition of CaP

on biomaterials is beneficial.[33] In the context of intra-

peritoneal drug delivery, calcification of implants may be

questionable.

The results of our in vivo study evidence the intraper-

itoneal biocompatibility of the three tested polymers

(PDMS, polyHEMA and PEVA). These substances would

thus be suitable for the design of intraperitoneal drug

delivery systems. However, fibrocellular encapsulation

around implants may limit the in vivo performance of

such medical devices by altering drug delivery. In order to

evaluate the effect of encapsulation on specific drug release,

further studies should be conducted with drug-loaded

implants placed inside the peritoneal cavity.

, 11, 1336–1345


Figure 2. Histological analysis of implants recovered from theperitoneal cavity of rats. A and B: PDMS implants. C and D: PEVAimplants. Thin encapsulating fibrocellular tissue was observedaround all PDMS and PEVA implants (after 3 or 6 months’implantation). Some implants were simply surrounded by encap-sulating fibrocellular tissue (A and C), while in other cases, thistissue was adherent to adipose tissue (B and D). E and F: Poly-HEMA implant, with calcium deposits (arrow) visible in F. Originalmagnification: A, B, C and E: 200�. D and F: 100�.

Figure 3. Histological analysis of implants recovered from theperitoneal cavity of rabbits. Thin encapsulating fibrocellular tis-sue was observed around all 3 types of implants (after 3 or6 months’ implantation). PDMS implants are shown as a repre-sentative example. A and B: Implants adherent to tissue (originalmagnification: 100�). C: Implant free in the peritoneal cavitycovered with fibrocellular tissue (original magnification: 40�).D: Fibrocellular tissue (original magnification: 200�).

Table 5. Measurement of the contact angle of a water drop on aplane surface of each of the three polymers.

Polymer Contact angle

PDMS 92.28� 4.1

PEVA 78.98� 1.4

PolyHEMA 41.98� 1.5

1344

www.mbs-journal.de

S. Defrere et al.

Conclusion

In conclusion, we have successfully demonstrated the

biocompatibility of three polymers (PDMS, polyHEMA and

PEVA) implanted into the peritoneal cavity of rats, rabbits

and monkeys. Despite the presence of fibrous tissue

encapsulating the implants, there was no evidence of

inflammation over a 6-month period. Levels of inflamma-

Figure 4. Histological analysis of implants recovered from the peritoneimplant. Both are encapsulated in fibrocellular tissue and adherenfibrocellular tissue, with visible calcium deposits (arrow). Original m



tory markers (CRP and fibrinogen), hematological cell

counts and the macroscopic aspect of the pelvic cavity

were similar in animals with and without polymer

implants. The peritoneal cavity is a specific environment

that is quite dynamic and linked to the immune system.

Drug delivery to the pelvic cavity may be of great interest

al cavity of rhesus monkeys after 3 months. A: PDMS implant. B: PEVAt to peritoneal tissue. C: PolyHEMA implant encapsulated in thinagnification: 100�.

, 11, 1336–1345



www.mbs-journal.de

for pathologies predominantly confined to the peritoneal

cavity. These polymers could well prove suitable for the

design of effective intraperitoneal drug delivery systems.

Acknowledgements: The authors thank Olivier Van Kerk (Depart-ment of Gynecology, Universite catholique de Louvain), ErikaKonradowski, Marie Dehuy (Laboratory of Tumor and Develop-ment Biology, University of Liege), Guillermo Araujo (AnimalFacility, Universite catholique de Louvain), and Luc Duwez(Animal Facility-GIGA, University of Liege) for their technicalassistance. We thank Mira Hryniuk, B.A., for reviewing the Englishlanguage of the manuscript. We also thank the Department ofAnatomopathology (Cliniques Universitaires St Luc, Brussels,Belgium) for specimen embedding and hematoxylin–eosin stain-ing and the department of Clinical Biology (Cliniques Universi-taires St Luc) for CRP and fibrinogen assays. The present study wassupported by a grant from ‘la Region Wallonne’ (Belgium).

Received: February 23, 2011; Revised: May 26, 2011; Publishedonline: August 5, 2011; DOI: 10.1002/mabi.201100077

Keywords: biocompatibility; implant; polydimethylsiloxane(PDMS); poly(ethylene-co-vinyl acetate) (PEVA); poly(hydroxy-ethyl methacrylate) (polyHEMA)

[1] M. D. Blanco, R. M. Trigo, C. Teijon, C. Gomez, J. M. Teijon,Biomaterials 1998, 19, 861.

[2] C. Gomez, M. D. Blanco, M. V. Bernardo, R. L. Sastre, J. M. Teijon,J. Pharm. Pharmacol. 1998, 50, 703.

[3] W. E. Roorda, H. E. Bodde, A. G. De Boer, H. E. Junginger,Pharmacol. Weekblad Sci. Ed. 1986, 8, 165.

[4] F. Petraglia, S. Luisi, Gynecol. Endocrinol. 2007, 23, 662.[5] K. Burczak, E. Gamian, A. Kochman, Biomaterials 1996, 17,

2351.[6] M. Pradny, P. Petrovicky, V. Fronkova, J. Vacık, K. Smetana, Jr.,

J. Mater. Sci., Mater. Med. 2002, 13, 107.[7] D. Dufrane, M. Steenberghe, R. M. Goebbels, A. Saliez, Y. Guiot,

P. Gianello, Biomaterials 2006, 27, 3201.




[8] J. B. Mendes, P. P. Campos, M. A. Ferreira, Y. S. Bakhle, S. P.Andrade, J. Biomed. Mater. Res. B: Appl. Biomater. 2007, 83,408.

[9] E. A. Ho, V. Vassileva, C. Allen, M. Piquette-Miller, J. ControlledRelease 2005, 104, 181.

[10] V. Vassileva, J. Grant, R. De Souza, C. Allen, M. Piquette-Miller,Cancer Chemother. Pharmacol. 2007, 60, 907.

[11] P. H. Sugarbaker, Scand. J. Surg. 2006, 95, 270.[12] G. S. Dizerega, K. E. Rodgers, ‘‘The peritoneum’’, Springer, New

York 1992.[13] E. Oral, D. L. Olive, A. Arici, Hum. Reprod. Update 1996, 2, 385.[14] P. R. Koninckx, S. H. Kennedy, D. H. Barlow, Hum. Reprod.

Update 1998, 4, 741.[15] G. E. Beheri, Plast. Reconstr. Surg. 1966, 38, 92.[16] F. B. Scott, W. E. Bradley, G. W. Timm, Urology 1973, 1, 252.[17] T. Amzallag, J. Pynson, J. Fr. Ophtalmol. 2007, 30, 757.[18] L. Garrido, V. L. Young, Magn. Reson. Med. 1999, 42, 436.[19] P. V. Shastri, Contraception 2002, 65, 9.[20] K. Matsumura, S. H. Hyon, N. Nakajima, C. Peng, S. Tsutsumi,

J. Biomed. Mater. Res. 2000, 50, 512.[21] L. C. Costantini, S. R. Kleppner, J. McDonough, M. R. Azar,

R. Patel, Int. J. Pharm. 2004, 283, 35.[22] F. J. Roumen, T. O. Dieben, Contraception 1999, 59, 59.[23] A. Ramadevi, T. Padmavathy, G. Stigall, D. Paquette,

S. Kalachandra, J. Mater. Sci., Mater. Med. 2008, 19, 721.[24] T. L. Tsou, S. T. Tang, Y. C. Huang, J. R. Wu, J. J. Young, H. J.

Wang, J. Mater. Sci., Mater. Med. 2005, 16, 95.[25] A. Hejcl, P. Lesny, M. Pradny, J. Michalek, P. Jendelova, J. Stulık,

E. Sykova, Physiol. Res. 2008, 57 Suppl 3, S121.[26] L. R. Madden, D. J. Mortisen, E. M. Sussman, S. K. Dupras, J. A.

Fugate, J. L. Cuy, K. D. Hauch, M. A. Laflamme, C. E. Murry, B. D.Ratner, Proc. Natl. Acad. Sci. USA 2010, 107, 15211.

[27] A. D. Woolfson, R. K. Malcolm, R. J. Gallagher, J. ControlledRelease 2003, 91, 465.

[28] W. E. Henninck, C. F. van Nostrum, Adv. Drug Delivery Rev.2002, 54, 13.

[29] A. Larsson, J. Bjork, C. Lundberg, Vet. Immunol. Immuno-pathol. 1997, 59, 163.

[30] J. A. Palma, Experientia 1976, 32, 1481.[31] L. Liu, G. Chen, T. Chao, B. D. Ratner, E. H. Sage, S. Jiang,

J. Biomater. Sci., Polym. Ed. 2008, 19, 821.[32] M. C. Belanger, Y. Marois, J. Biomed. Mater. Res. 2001, 58, 467.[33] Z. Zainuddin, D. J. Hill, T. V. Chirila, A. K. Whittaker, A. Kemp,

Biomacromolecules 2006, 7, 1758.

, 11, 1336–1345


In Vivo Biocompatibility of Three Potential Intraperitoneal Implants

Documents