Cartilage repair with chondrocytes in fibrin hydrogel and MPEG ...

Knee Surg Sports Traumatol Arthrosc (2008) 16:690–698 DOI 10.1007/s00167-008-0522-1

EXPERIMENTAL STUDY

Cartilage repair with chondrocytes in fibrin hydrogel and MPEG polylactide scaffold: an in vivo study in goats Martin Lind Æ Allan Larsen Æ Christian Clausen Æ Kurt Osther Æ Hanne Everland

Received: 23 September 2007 / Accepted: 4 March 2008 / Published online: 17 April 2008 Ó Springer-Verlag 2008

Abstract Polylactic acid polymers have been used extensively as biomaterials and have shown promising properties for cartilage tissue engineering. Numerous scaffold materials exist and the optimal scaffold needs to be identified. We have tried to assess the possibilities for cartilage repair by the use of two different scaffold techniques; autologous chondrocytes in a fibrin hydrogel and a novel MPEG-PLGA scaffold, where autologous chondrocytes are immobilized within the MPEG-PLGA scaffold by a fibrin hydrogel. Twenty adult goats were used for the study. A 6 mm circular full-thickness cartilage defect was created in both medial femoral condyles. The defects were randomized to the following four treatment groups. (1) Empty defect (control). (2) Subchondral drilling (control). (3) Fibrin hydrogel with autologous chondrocytes. (4) Fibrin hydrogel/chondrocyte solution in a MPEG-PLGA porous scaffold. Animals were followed for 4 month. Eight defects in each treatment group completed the study. ICRS macroscopic scoring (0–12). Indentation test was performed to assess stiffness of repair tissue. Histological analyses was performed using O’Driscoll and Pineda cartilage scores as well as percentage tissue filling of the defects. The MPEG-PLGA/chondrocytes scaffold was the M. Lind (&) Sportstrauma Clinic, University Hospital of Aarhus, Tage Hansens Gade 2, 8000 Aarhus, Denmark e-mail: [email protected] A. Larsen Artros Clinic Aalborg, Aalborg, Denmark C. Clausen K. Osther Interface Biotech A/S, Denmark, Hørsholm, Denmark H. Everland Coloplast Research A/S, Humlebæk, Denmark

123

superior treatment modality based on the macroscopic surface score, histological scores and defect filling. The mechanical test demonstrated no difference between treatment groups. The MPEG-PLGA/chondrocyte composite demonstrated significantly better cartilage repair response than empty defects, osteochondral drilling and fibrin hydrogel with chondrocytes. The novel MPEGPLGA scaffold in combination with chondrocytes need further studies with respect to longer follow-up times. Keywords Cartilage repair MPEG-PLGA Fibrin Hydrogel Animal model Goat Microfracture

Introduction The present clinical focus of cartilage tissue engineering is to develop methods with consistent regenerative potential of forming hyaline cartilage and surgically feasible methods that can be used with arthroscopic techniques. Porous scaffolds in combination with cultured chondrocytes have presently demonstrated good ability to regenerate hyaline cartilage both experimentally and clinically [12, 14]. Several porous biomaterials in combination with chondrocytes have been investigated for cartilage repair such as polylactide polymers, collagen mesh/gel, hyaluronic acid fibermesh, chitosan and several others [7, 8, 28]. Presently only collagen and hyaluronic acid scaffold have been tested clinically with reasonable follow-up [1, 12]. The polyglycolic acid (PGA), polylactide acid (PLA), and their copolymers poly(lactic-co-glycolic) (PLGA) has previously been used in cartilage engineering. These polymers are attractive scaffold materials for tissue engineering, because of their biodegradability and their approved status with the FDA. Porous three-dimensional

Knee Surg Sports Traumatol Arthrosc (2008) 16:690–698

PLGA scaffolds seeded with cultured marrow stromal cells have been transplanted into large defects in rabbit knees. Defects transplanted with PLGA scaffold and cells demonstrated smooth white tissue macroscopically and hyaline-like cartilage on histology at 12 weeks after the transplantation [24]. Only limited data exist with polylactide scaffolds and chondrocytes in large animal models. One study in goats tested a polylactide scaffold mesh as a membrane to be sutured over a cartilage defect and with chondrocytes injected under the membrane. A periostal membrane served as control. After 3 months follow-up the periosteum-covered defects had the best tissue repair compared to polylactide mesh, but both demonstrated abnormal repair tissue [27]. However porous scaffolds are difficult to handle during arthroscopic application. Hydrogel-based chondrocyte suspensions on the other hand can be injected into cartilage defects and are thus feasible for arthroscopic application. Injectable scaffold materials such as PLGA microspheres, collagen hydrogel and fibrin hydrogel, have been used for cartilage repair [17]. The main limitation for the use of injectable scaffolds such as fibrin hydrogels in cartilage repair procedures is the mechanical properties that typically are softer than solid scaffolds, e.g., PLGA. Kang et al investigated PLGA microspheres combined with autologous chondrocytes as injectable scaffold for cartilage tissue engineering. The injectable composite was able to form differentiated cartilage subcutaneously and to regenerate hyaline cartilage in cartilage defects better than empty defects and cells without polymer [9, 10]. Fibrin hydrogel has been used in a number of studies, where different formulations have been combined with chondrocytes [4, 26]. In an equine study, chondrocytes were combined with fibrin hydrogel and implanted into trochlea defects. Generally cartilage repair tissue was observed in the treated groups and increased percentage of collagen type II and glycosaminoglycan content were observed [6]. Preclinically both the PLGA microspheres and fibrinogen have shown to be suitable polymers for injectable matrices when combined with cultured chondrocytes. In order to demonstrate feasibility for human clinical use a potential scaffold candidate needs to be tested in full thickness cartilage lesions in weighbearing areas and in a large animal model in order to mimic the clinical situation. The present study aims to investigate the cartilage repair potential of both a fibrin based hydrogel combined with chondrocytes and a non injectable novel methoxypolyethyleneglycol (MPEG)-PLGA scaffold/chondrocyte/fibrin selfadhesive composite in a full thickness cartilage defect model in goats. The novel MPEG-PLGA scaffold used has not been described previously. This scaffold has extensive hydrophilic capabilities, which are a potential advantage in cell-based cartilage repair.

691

We hypothesize that both the hydrogel and MPEGPLGA scaffold in combination with autologous chondrocytes will generate at better cartilage repair response than empty defects and defects with subchondral drilling.

Materials and methods Design Twenty adult goats were used for the study. At primary surgery a 6 mm full-thickness circular defect was created in both right and left medial femoral condyles. The cartilage tissue removed is used as biopsies for autologous chondrocyte cultures. Four weeks later the defects were reexposed at a second surgery. The defects were cleaned and randomized to the following four treatment groups. (1) Empty defect (control). (2) Subchondral drilling (control). (3) Fibrin hydrogel with autologeous chondrocytes. (4) Fibrin hydrogel/chondrocyte solution in a MPEG-PLGA porous scaffold. Ten defects in each treatment group were included in the study. Animals were followed for 4 months. Cell culture Each cartilage biopsy was washed extensively in PBS, and subsequently diced into small explants (*1 to 2 mm3) with a sterile scalpel. Explants were placed in culture flasks containing an optimized growth medium; DMEM/F12 containing L-ascorbic acid 2-phosphate (75 lg/mL), 20% (v/v) FBS, fungizone (2.4 lg/mL) and gentamicin (10 mg/ mL). No collagenase treatment was used in order to liberate the chondrocytes from the extracellular matrix. The culture system used in this study promotes the migration of chondrocytes from the cartilage tissue to the plastic surface of the culture flask, due to the high amounts of growth factors in the serum. Growth medium was changed after 24 h in order to remove non-adherent cells from the culture and cells were subsequently fed continuously by replacing the growth medium every 3–4 days until 70–80% confluence. Explants were cultured in a humidified atmosphere of 5% CO2 at 37°C. After one week migration from the cartilage explants to the plastic surface was observed. After 3–4 weeks in culture, ex vivo expanded chondrocytes of passage 1–2 were detached by trypsin/EDTA treatment and suspended in fibrinogen solution (50 mg/mL). Fibrinogen was dissolved at 37°C in DMEM/F12 containing L-ascorbic acid 2-phosphate (75 lg/mL) and gentamicin (10 mg/mL). Dulbecco’smodified Eagle’s Medium (DMEM/F12) supplemented with GlutaMAX-1TM (L-alanyl-L-glutamin), fungizone, gentamicin, foetal bovine serum (FBS), phosphate-buffered saline

123

692

(PBS) and trypsin/EDTA were all purchased from GIBCO Invitrogen, Denmark. Bovine fibrinogen, bovine thrombin, calcium chloride (CaCl2), Toluidine Blue O and L-ascorbic acid 2-phosphate were purchased from Sigma–Aldrich, Brøndby, Denmark. Scaffold fabrication 4.00 g Methoxypolyethyleneglycol-block-co-poly(lactideco-glycolide) (MPEG-PLGA) (50:50 LA:GA) was dissolved on 100 ml 1,4 dioxane overnight at 50°C. Seven millilitre of polymer solution was poured into a precooled aluminium mould 7.3 9 7.3 cm2 and placed in a freeze dryer at -20°C. The drying program was 1 h at -20°C and 1 h at +30°C. The scaffolds were afterward dried overnight in a vacuum dessicator at room temperature. The scaffolds were sterilized in 100% EtOH, then dried and packed into aluminized PET-bags. The avg. porosity of the scaffolds was above 90% and the thickness 0.8 mm. The SEM picture in Fig. 1 shows the interconnectivity of a porous MPEG-PLGA scaffold (Fig. 1). Mechanical testing The mechanical test was performed on the repair tissue and the cartilage tissue surrounding the defect area with a TAXT. Plus Texture Analyser (Stable Micro Systems, Surrey, UK). The load was applied to the tissue using a stainless probe with an areal of 0.184 mm2. Each sample was fixed in a position for appropriate surface indentation. After attaining a deformation depth of 0.05 mm the load (N) was recorded. Each sample was evaluated in at least three places around the actual area (repair cartilage or surrounding cartilage).


Histological analysis The specimens were fixed in 10% paraformaldehyde solution for 48 h, decalcified and embedded in paraffin. Sections (10 lm) were stained with haematoxylin/eosin (H&E) and safranin O. A minimum of four sections from the central portion of defects were subjected to semiquantitative histological analyses. Tissue fill was measured using grid-point counting technique. The expected surface level of the intact cartilage over the defects was marked and point counting was done in the area between this marking and the subchondral bone surface. Points to hit repair tissue in comparison to total points in defects were measured. General cartilage repair response was measured by two scoring systems, O’Driscoll and Pineda histological scores [16, 19]. The O’Driscoll score ranges from 0 to 24 and is a semiquatitatively, comprehensive cartilage scoring system including a large number of both cellular -and tissue parameters. The Pineda Score ranges from 12 to 0 and is also a semiquantitative cartilage scoring system; however, more simplified than the O’Driscoll score, including fewer parameters. There was only one histological investigator (ML) that was blinded to treatment groups. In vitro evaluation of MPEG-PLGA after application of chondrocytes in fibrin hydrogel To test the cellular anchoring in the scaffold structure immediately after application; chondrocytes were isolated and cultured under the same conditions as described in the cell culture section. 10 9 106 chondrocytes were resuspended in 0.5 mL fibrinogen solution (50 mg/mL) and applied onto the MPEG-PLGA scaffolds in 6 well plates (Nunc, Roskilde, Denmark) together with a thrombin/ CaCl2 solution (1:1 v/v). Chondrocytes were applied drop wise to the scaffolds and the volume needed to saturate the 0.1 cm3 scaffolds was recorded. Five different volumes were tested (20, 40, 60, 80 and 100 lL). After application of chondrocytes/fibrin hydrogel the scaffolds were left for 2 min and the appropriate volume for saturating the 0.1 cm3 MPEG-PLGA scaffolds was determined. In vitro test of chondrogenic phenotype in MPEG-PLGA and fibrin hydrogel

Fig. 1 Scanning electron microscopy picture of the MPEG-PLGA scaffold

123

The MPEG-PLGA scaffold system and the fibrin hydrogel have previously been optimized in our laboratory, to ensure the viability, chondrogenic phenotype, and migration of the applied articular chondrocytes. For characterizing the MPEG-PLGA scaffold system chondrocytes were isolated and cultured under the same conditions as described in the cell culture section. 10 9 106


chondrocytes were resuspended in 1 mL fibrinogen solution (50 mg/mL) and applied onto the MPEG-PLGA scaffolds (1 9 106 cells/cm2) together with a thrombin/CaCl2 solution. Chondrocyte-loaded MPEG-PLGA scaffolds of 1 cm2 were placed in wells (12 well plates, NUNC), containing 2 mL growth medium as described above. For characterizing the fibrin hydrogel chondrocytes were isolated and cultured under the same conditions as in the cell culture section. 10 9 106 chondrocytes were resuspended in 1 mL fibrinogen solution (50 mg/mL) and combined with a thrombin/CaCl2 solution. The fibrin hydrogel/chondrocyte scaffold was placed in wells (12 well plates, NUNC), containing 2 mL growth medium as described above. After 4 weeks incubation (37°C, 5% CO2), the scaffold constructs were processed for histology (Safranin O and haematoxylin–eosin staining) and gene expression analysis of chondrogenic marker genes (collagen type II, Sox 9 and aggrecan). Glyceralaldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal standard. Macroscopic scoring The ICRS macroscopic scoring was used to assess the macroscopic appearance of the repair tissue. This score ranges from 0 to 12. The score included semiquantitative scales on defect filling, integration to native cartilage and repair tissue surface topography [18]. Operation The goats were positioned at the back with their legs fixated. The operations were performed through a miniarthrotomy with an average length of 3 cm (ranging from 2.5 to 4 cm). To create the 6 mm cartilage defect in the femoral condyle we used OATS chisel instruments to cut through the cartilage to the subchondral bone. The cartilage in the defects was subsequently removed by scalpel and curettes. Care were taken not to affect the subchondral bone to an extent that bleeding occurred. All cartilage defects were made in the weight bearing part of the medial femoral condyle. After creating the defects the removed cartilage served as cartilage biopsies and were send for chondrocyte culturing. The joint capsule, subcutaneous tissue and skin was closed with Vicryl 3-0. The skin was sprayed with tissue protection tar to prevent infection. No bandage was used to cover the incision. At the second operation 4 weeks later the cartilage defects were re-exposed and the four treatment methods were randomly allocated between sides of the hind legs and between animal. Subchondral drilling was performed by drilling down to 3 mm depth with a 1 mm drill with 1–2 mm between drillholes. This method was used to

693

simulate the biological healing response seen with the clinical cartilage surgery ‘‘Microfracture’’. The paste-like fibrin/chondrocyte suspension was gently injected with 0.2 ml suspension in the defect containing 1 9 106 chondrocytes. The MPEG-PLGA/chondrocyte scaffold containing 1 9 106 chondrocytes was gently pressed into the defect and secured with fibrin glue. The goats received standard injections twice a day with NSAID to prevent/treat any postoperative pain. After the operation the goats were fully weight bearing since they were not capable of partly weight bear. The goats were fed on a standard diet plus allowed to eat grass when being outside. Postoperatively the goats were kept indoors in cages to limit mobility and limit the risk of infection. Statistics Data are presented as mean and standard deviation. Comparison between the four groups was done by one-way ANOVA and comparison between control and test groups were performed by Holm–Sidak multiple comparison test. P values \0.05 were considered significant.

Results Four goats were lost during the observation period leaving eight condyles with each treatment modality for analyses. One goat was terminated after 1 month after second surgery due to septic arthritis in one of the operated knees. Two goats were terminated due to mamma infection during the four months follow-up period and one goat died due to unknown causes in the field. The excluded defects due animal termination were evenly distributed between treatment groups, leaving eight defects in each treatment group. In vitro testing The cell suspension volume needed to saturate the MPEGPLGA (0.1 cm3) was determined to 80 lL/0.1 cm3. After 2 min the scaffolds were removed from Plate 1 and the plastic surface was examined with light microscopy. No cells were observed on the surface, demonstrating that the entire cell population in an 80 lL volume was absorbed by the hydrophilic MPEG-PLGA scaffold. In average 12.000 (SD 8000) cells (1.2% of the applied cell population) were observed in the medium obtained from Plate 2 after 10 min. After 4 weeks of culture cells were homogenously dispersed throughout the MPEG-PLGA scaffold construct and the fibrin hydrogel. Light microscopy analysis of H&E stained sections demonstrated a high fraction of vital cells ([80%). In addition safranin O staining demonstrated that

123

694


Fig. 2 Hematoxylin–eosin and Safranin O stained histological sections of fibrin hydrogel and MPEG-PLGA scaffold with cells after 4 weeks of culture. Note the even distribution of viable cells. In the Safranin O stained sections the positive staining around the cell indicates proteoglycan synthesis by the chondrocytes

0.042 (0.024) N, which was 10% of the tissue in the repair sites (Table 1). Histology (Figs. 4, 5)

Fig. 3 Expression of chondrogenic marker genes in chondrocytes by RT-PCR after 4 weeks of culture in the fibrin hydrogel and MPEGPLGA scaffold. The gene expression analysis demonstrate expression of matrix proteins Aggregan (Ag) and Collagen II (Col (II)) and the cartilage differentiation factor (Sox-9). A housekeeping enzyme GADPH is shown as reference level of gene expression

chondrocytes cultured within this scaffold system produced chondrogenic matrix proteins (Fig. 2). The gene expression analysis demonstrated expression of all three classical chondrogenic markers in cells derived from the fibrin hydrogel and MPEG-PLGA scaffold constructs (Fig. 3). Macroscopic appearance The ICRS score demonstrated significantly improved scores for all test groups with microfracture being equal to fibrin/chondrocytes and the MPEG-PLGA/chondrocyte scaffold being the superior treatment modality. Generally repair tissue tended to be thicker and with a more cartilage like appearance in margins of the defects (Table 1).

Repair tissue filling in defects demonstrated the following. In the control group with empty defects no or very limited repair tissue was found. In the subchondral drilling and fibrin/cell group repair tissue was generally less than 50% of defect thickness and with less cartilage like cell architecture and organization. In the subchondral drilling group subchondral bone changes were seen reflecting the healing and partial healing of the subchondral drilling holes. There were signs of minor subchondral destruction in the empty defect group where no repair tissue existed to protect the subchondral bone. Both Pineda and O’Driscoll score reflected the above with significant better scores in the MPEG-PLGA/chondrocyte treated defects when compared to both empty and subchondral drilling and fibrin hydrogel groups (Table 1). In MPEG PLGA group most defects had almost complete defect filling and good border integration. Cellular organization was still immature without the normal columnar organization but cells had chondrocytic appearance and Safranin O staining demonstrated high content of proteoglycans in the cell matrix.

Mechanical test Discussion The mechanical test demonstrated no difference between treatment groups. However all groups were significantly stiffer than the uninjured cartilage which had a stiffness of

123

Several studies have shown that new cartilage can be engineered in vivo by transplanting chondrocytes seeded


695

Table 1 Histological and biomechanical data Empty defect

Subchondral drilling

Fibrin + Chondrocytes

MPEG-PLGA + Chondrocytes

ICRS macroscopic score (0–12)

3.1 (2.7)

5.4 (1,7)

5.8 (1.8)

7.9* (2.2)

Stiffness Indentation test (N)

0.45 (0.91)

0.65 (0.96)

0.32 (0.41)

0.38 (0.43)

Histology defect fill (%)

12.5 (13.1)

30.5* (19.1)

40.6* (27.9)

80.6** (29.1)

Histology O’Driscoll score (0–24)

3.0 (2.8)

11.5* (8.7)

11.3* (5.2)

20.6** (2.7)

Histology Pineda Score (12–0)

11.1 (3.0)

8.7 (2.7)

8.1 (3.5)

2.0** (1.6)

All quantitative data macroscopic scoring, stiffness test, and histology are presented in the the table. Data are expressed as mean and standard deviation in brackets * Significantly different from control group (Empty defect) ** Significantly different from control groups (Empty defect) and (subchondral drilling)

Fig. 4 Histology sections of all four study groups stained with both Safranin O and HE stainings. Top row illustrates overview of the defects at low magnification with H&E staining. Middel row illustrates the repair tissue/native cartilage interface at higher magnification with H&E staining. Bottom row illustrates overview of the defects at low magnification with Safranin O staining. The Safranin O stains proteoglycan in hyaline cartilage orange. For the 1209 slides the arrow indicates the interface between repair tissue and native cartilage

into a three-dimensional scaffold [3, 11, 21, 22]. Most of these scaffolds are solid preformed structures often made by fibrous meshes or porous foam. When using these scaffolds, the surgeon normally has to make a large incision to achieve access to the damaged cartilage, since the scaffolds need to be either sutured or otherwise fixed to the normal cartilage. When using an injectable scaffold like a hydrogel the surgeon gets the advantage of being able to use an arthroscopic technique. This reduces the morbidity related to the operation and accelerates rehabilitation. The use of scaffolds has several advantages over the first generation ACI technology: homogeneity of the cells within the scaffold, no cell leakage, easier surgery and less tedious. Our in vitro data of both the MPEG-PLGA/chondrocyte constructs and the fibrin/chondrocyte constructs demonstrates that chondrocytes cultured in both types of scaffold materials, survive within the structures. In addition the gene expression analysis demonstrated that the cells express genes related to the chondrogenic phenotype after culture in the scaffolds. From the cell adherence tests we

conclude that the cellular anchoring in the MPEG-PLGA scaffold system is relatively effective and a dramatically cell loose prior to implantation is avoided partly by the hydrophilic characteristic of MPEG-PLGA and partly by the fibrin hydrogel in which the cells are applied to the scaffold. In our study both the macro- and microscopic scoring results demonstrated that the most cartilage like tissue repair tissue was obtained by the MPEG-PLGA/chondrocyte construct. The fibrin/chondrocyte composite was only able to regenerate tissue to the same level as subchondral drilling. However, at 4 months the cartilage repair tissue was not in the normal range (11–12 for the ICRS score) in any of the treatment groups, which was to be expected at this relatively early time point. Mechanical testing revealed increased stiffness of all treatment groups compared to normal cartilage. This could indicate that the relative thin layer of cartilage repair tissue in the defects results in the subchondral bone to contribute significantly to the stiffness values. This study represents cartilage repair response at an early time point. But this early repair tissue typically

123

696


Fig. 5 Histology sections of four representative sections from all 4 study groups. Slides are overview pictures of the defects at low magnification with H&E staining. Arrows indicates the interface between repair tissue and native cartilage

represent later repair responses, especially when cell based therapies are used where the early cellular responses are crucial for the later tissue maturation [2]. The fibrin/ chondrocyte hydrogel composite is an injectable formulation for cartilage repair due to the hydrogel structure. In our study this construct did not stimulate cartilage repair sufficiently. We speculate that chondrocyte containment at the defect site was too limited to initiate a proper cartilage repair response in the fibrin hydrogel/chondrocyte group. Applying cells in gel probably would require more restrictive rehabilitation protocols than can be achieved in a goat model. We did see minor destruction of subchondral bone in the empty defect group, which we ascribe to the lack off repair tissue thereby exposing the subchandral bone during weight loading. The cartilage repair in the MPEG-PLGA/chondrocyte group was better than the hydrogel technique with good macroscopic appearance, good filling of the defect and high histology scores. However, a sheat shaped scaffold cannot be used as an injectable cartilage repair product. Probably a specialized arthroscopic instrumentation can be developed using circular scaffold discs that can be applied arthroscopically through a specialized tube [13]. The present study represents a challenging model for cartilage repair. The defects were positioned in the weight bearing part of the medial condyles and the animals were allowed free activity indoor for the first 2 weeks and outdoors thereafter. Compared to human cartilage repair rehabilitation protocols, this regimen is very challenging. Typically non-weight bearing would be recommended for

123

6 weeks as well as controlled knee activity. In an animal model restricted rehabilitation is difficult to control and with potentially ethical problems. The relatively poor biological healing response in the subchondral drilling and fibrin/cell groups could therefore be a result of the free rehabilitation of the animals, since fibrin/cell construct has limited mechanical strength to withstand manipulation and the same problem could be an issue for the immature healing tissue from bone marrow after subchondral drilling. Polylactide scaffolds in combination with chondrocytes have previously demonstrated good ability to support cartilage repair. However these studies have mainly been performed using osteochondral defects in small animals [24]. In these models the intrinsic healing capacity of cartilage is high and the potential of healing response from bone marrow cells is possible source of bias. Also osteochondral defect healing is not as clinically relevant as healing from full thickness cartilage defects [21, 29]. Studies in large animal models with polylactic acid based scaffold have been performed. A biphasic PLGA scaffold without cells containing both a chondral and an osseous phase was tested in goat femoral defects. The scaffolds partly demonstrated hyaline cartilage formation and bone remodeling in osseous phase [15]. In a study more similar to ours using a porcine model, cartilage defects were filled with a poly-L,D-lactic acid scaffold with a without chondrocytes. The defects with chondrocytes did not demonstrate a better cartilage repair response than scaffold alone. Remnants of the relatively stable poly-L,D-lactic acid scaffold was seen after 3 months observation [20]. No


previous studies have tested both a hydrogel with chondrocytes and polylactic acid based 3D scaffold in the goat defect model so our results can not be compared to previous studies. Fibrin glue has been tested a carrier for chondrocytes and cartilage matrix for implantation in cartilage defects. In a porcine femoral defect model a combination of chondrocytes frozen cartilage chips and fibrin glue demonstrated the best cartilage repair response compared to chips alone, cells alone and fibrin glue alone [17]. Also fibrin glue have been used for incorporation of xenografted chondrocytes applied in a goat defect model demonstrating intact matrix production by the xenografted chondrocytes within the fibrinhydrogel after in vivo implantation [25]. A potential important parameter related to the MPEGPLGA/chondrocyte construct used in this study, is the fact that chondrocytes are not cultured in the MPEG-PLGA scaffold in vitro prior to implantation. Instead chondrocytes are applied to the MPEG-PLGA structure directly in the defect area. In recent described methods the cells are cultured within the scaffold for a period of time before implantation. A recent study has demonstrated cell-densities from 0.5 to 10% of the seeded at the time of implantation [5]. With the novel MPEG-PLGA/chondrocyte method presented in this study the cell density is better controlled at surgery, which might ensure a more optimal starting point for cartilage repair. The present study is limited to investigating an early cartilage repair response. We are therefore unable to answer whether the favorable repair response seen in the present study will develop into durable cartilage repair tissue at later time points. Another problem with the animal model is the lack of ability to perform specific rehabilitation regimens. More controlled weight loading and movement might positively influence a cartilage repair response. This could result in poor repair response with drilling technique, were a specific rehabilitation is thought to be important for optimal healing [23].

Conclusion The MPEG-PLGA/chondrocyte construct used in this study demonstrated promising early cartilage repair in femoral condyle defect goat model. Our study failed to demonstrate that a hydrogel composed of fibrin and chondrocytes could generate a clinically acceptable repair tissue better than subchondral drilling. Acknowledgments The present study was supported by The Danish Reumatism Association, The Obel Foundation, Denmark. The Orthopedic Research Foundation at Aalborg Hospital, Denmark and Interface Biotech, Denmark.

697 Conflict of interest statement Allan Larsen and Martin Lind have received funds as scientific consultants for Interface Biotech. Interface Biotech did not collect data, interpret data or were involved in the writing of the manuscript.

References 1. Bentley G, Biant LC, Carrington RW, Akmal M, Goldberg A, Williams AM, Skinner JA, Pringle J (2003) A prospective, randomised comparison of autologous chondrocyte implantation versus mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg Br 85:223–230 2. Breinan HA, Minas T, Hsu HP, Nehrer S, Sledge CB, Spector M (1997) Effect of cultured autologous chondrocytes on repair of chondral defects in a canine model. J Bone Joint Surg Am 79:1439–1451 3. Dorotka R, Windberger U, Macfelda K, Bindreiter U, Toma C, Nehrer S (2005) Repair of articular cartilage defects treated by microfracture and a three-dimensional collagen matrix. Biomaterials 26:3617–3629 4. Fortier LA, Mohammed HO, Lust G, Nixon AJ (2002) Insulinlike growth factor-I enhances cell-based repair of articular cartilage. J Bone Joint Surg Br 84:276–288 5. Gigante A, Bevilacqua C, Ricevuto A, Mattioli-Belmonte M, Greco F (2007) Membrane-seeded autologous chondrocytes: cell viability and characterization at surgery. Knee Surg Sports Traumatol Arthrosc 15:88–92 6. Hendrickson DA, Nixon AJ, Grande DA, Todhunter RJ, Minor RM, Erb H, Lust G (1994) Chondrocyte-fibrin matrix transplants for resurfacing extensive articular cartilage defects. J Orthop Res 12:485–497 7. Hoemann CD, Sun J, Legare A, McKee MD, Buschmann MD (2005) Tissue engineering of cartilage using an injectable and adhesive chitosan-based cell-delivery vehicle. Osteoarthritis Cartil 13:318–329 8. Hunziker EB (2002) Articular cartilage repair: basic science and clinical progress. A review of the current status and prospects. Osteoarthritis Cartil 10:432–463 9. Kang SW, Jeon O, Kim BS (2005) Poly(lactic-co-glycolic acid) microspheres as an injectable scaffold for cartilage tissue engineering. Tissue Eng 11:438–447 10. Kang SW, Yoon JR, Lee JS, Kim HJ, Lim HW, Lim HC, Park JH, Kim BS (2006) The use of poly (lactic-co-glycolic acid) microspheres as injectable cell carriers for cartilage regeneration in rabbit knees. J Biomater Sci Polym Ed 17:925–939 11. Lee CR, Grodzinsky AJ, Hsu HP, Spector M (2003) Effects of a cultured autologous chondrocyte-seeded type II collagen scaffold on the healing of a chondral defect in a canine model. J Orthop Res 21:272–281 12. Marcacci M, Berruto M, Brocchetta D, Delcogliano A, Ghinelli D, Gobbi A, Kon E, Pederzini L, Rosa D, Sacchetti GL, Stefani G, Zanasi S (2005) Articular cartilage engineering with Hyalograft C: 3-year clinical results. Clin Orthop Relat Res 435:96–105 13. Marcacci M, Zaffagnini S, Kon E, Visani A, Iacono F, Loreti I (2002) Arthroscopic autologous chondrocyte transplantation: technical note. Knee Surg Sports Traumatol Arthrosc 10:154–159 14. Marlovits S, Zeller P, Singer P, Resinger C, Vecsei V (2006) Cartilage repair: generations of autologous chondrocyte transplantation. Eur J Radiol 57:24–31 15. Niederauer GG, Slivka MA, Leatherbury NC, Korvick DL, Harroff HH, Ehler WC, Dunn CJ, Kieswetter K (2000) Evaluation of multiphase implants for repair of focal osteochondral defects in goats. Biomaterials 21:2561–2574

123

698 16. O’Driscoll SW, Marx RG, Beaton DE, Miura Y, Gallay SH, Fitzsimmons JS (2001) Validation of a simple histological– histochemical cartilage scoring system. Tissue Eng 7:313–320 17. Peretti GM, Xu JW, Bonassar LJ, Kirchhoff CH, Yaremchuk MJ, Randolph MA (2006) Review of injectable cartilage engineering using fibrin gel in mice and swine models. Tissue Eng 12:1151– 1168 18. Peterson L, Minas T, Brittberg M, Nilsson A, Sjogren-Jansson E, Lindahl A (2000) Two- to 9-year outcome after autologous chondrocyte transplantation of the knee. Clin Orthop 374:212– 234 19. Pineda S, Pollack A, Stevenson S, Goldberg V, Caplan A (1992) A semiquantitative scale for histologic grading of articular cartilage repair. Acta Anat (Basel) 143:335–340 20. Pulliainen O, Vasara AI, Hyttinen MM, Tiitu V, Valonen P, Kellomaki M, Jurvelin JS, Peterson L, Lindahl A, Kiviranta I, Lammi MJ (2007) Poly-L-D-lactic acid scaffold in the repair of porcine knee cartilage lesions. Tissue Eng 13:1347–1355 21. Rudert M, Wilms U, Hoberg M, Wirth CJ (2005) Cell-based treatment of osteochondral defects in the rabbit knee with natural and synthetic matrices: cellular seeding determines the outcome. Arch Orthop Trauma Surg 125:598–608 22. Solchaga LA, Temenoff JS, Gao J, Mikos AG, Caplan AI, Goldberg VM (2005) Repair of osteochondral defects with hyaluronan- and polyester-based scaffolds. Osteoarthritis Cartil 13:297–309 23. Steadman JR, Rodkey WG, Rodrigo JJ (2001) Microfracture: surgical technique and rehabilitation to treat chondral defects. Clin Orthop 391:362–369

123

Knee Surg Sports Traumatol Arthrosc (2008) 16:690–698 24. Uematsu K, Hattori K, Ishimoto Y, Yamauchi J, Habata T, Takakura Y, Ohgushi H, Fukuchi T, Sato M (2005) Cartilage regeneration using mesenchymal stem cells and a three-dimensional poly-lactic-glycolic acid (PLGA) scaffold. Biomaterials 26:4273–4279 25. van Susante JL, Buma P, Homminga GN, van den Berg WB, Veth RP (1998) Chondrocyte-seeded hydroxyapatite for repair of large articular cartilage defects. A pilot study in the goat. Biomaterials 19:2367–2374 26. van Susante JL, Buma P, Schuman L, Homminga GN, van den Berg WB, Veth RP (1999) Resurfacing potential of heterologous chondrocytes suspended in fibrin glue in large full-thickness defects of femoral articular cartilage: an experimental study in the goat. Biomaterials 20:1167–1175 27. Vasara AI, Hyttinen MM, Lammi MJ, Lammi PE, Langsjo TK, Lindahl A, Peterson L, Kellomaki M, Konttinen YT, Helminen HJ, Kiviranta I (2004) Subchondral bone reaction associated with chondral defect and attempted cartilage repair in goats. Calcif Tissue Int 74:107–114 28. Wakitani S, Goto T, Young RG, Mansour JM, Goldberg VM, Caplan AI (1998) Repair of large full-thickness articular cartilage defects with allograft articular chondrocytes embedded in a collagen gel. Tissue Eng 4:429–444 29. Yan H, Yu C (2007) Repair of full-thickness cartilage defects with cells of different origin in a rabbit model. Arthroscopy 23:178–187