Resorbable crosslinked form stable membrane for use outside the oral cavity

Abstract

The invention relates to resorbable crosslinked form stable membrane which comprises a composite layer of collagen material and inorganic ceramic particles containing 1.5 to 3.5 weight parts of inorganic ceramic for 1 weight part of collagen material, sandwiched between two layers of elastic pretensed collagen material (collagen material that has been stretched such as to be in the linear/elastic region of the stress-strain curve), the collagen material comprising 50-100% (w/w) collagen and 0-50% (w/w) elastin, and has shape and dimensions suitable for use in human tissue regeneration outside the oral cavity in rhinoplasty, postlateral spinal fusion or orbital reconstruction.

Claims

1. A resorbable crosslinked form stable membrane which comprises a composite layer of collagen material and inorganic ceramic particles containing 1.5 to 3.5 weight parts of inorganic ceramic for 1 weight part of collagen material, sandwiched between two layers of elastic pretensed collagen material, wherein the elastic pretensed collagen material is a collagen material that has been stretched such as to be in the linear/elastic region of the stress-strain curve, the collagen material comprising 50-100% (w/w) collagen and 0-50% (w/w) elastin, and has shape and dimensions suitable for use in human tissue regeneration outside the oral cavity in rhinoplasty, postlateral spinal fusion or orbital reconstruction.

2. A resorbable crosslinked form stable membrane according to claim 1 which is selected from the group consisting of: a nasal arch-shaped membrane for rhinoplasty sized in such a way as to fit the desired dimension of the nose, an oval tube membrane for posterolateral spinal fusion with a length such as to cover two or more vertebras and a membrane for orbital fracture reconstruction shaped after identifying the bone ledges apt to support the implant and sized in such a way as to facilitate its insertion into the orbital cavity.

3. A resorbable crosslinked form stable membrane according to claim 2 which is a nasal arch-shaped membrane for rhinoplasty as illustrated in FIG. 1, (4), the length I, the width i and height h of the membrane for rhinoplasty being 40 to 80 mm, 10 to 15 mm and 10 to 15 mm, respectively.

4. A resorbable crosslinked form stable membrane according to claim 3, wherein the thickness of the nasal arch-shaped membrane is 0.5 to 2.5 mm.

5. A resorbable crosslinked form stable membrane according to claim 3, wherein the thickness of the wall of the nasal arch-shaped membrane is 1.0 to 2.0 mm.

6. A resorbable crosslinked form stable membrane according to claim 2 which is a slit oval tube membrane for posterolateral spinal fusion with a length I of 60 to 300 mm, an inner diameter k of 5 to 10 mm and an outer diameter j of 15 to 30 mm.

7. A resorbable crosslinked form stable membrane according to claim 6 wherein the thickness of the membrane is 0.5 to 2.5 mm.

8. A resorbable crosslinked form stable membrane according to claim 1 which is a membrane for orbital fracture reconstruction wherein the length m, width o and height c of the membrane are 30 to 50 mm, 20 to 40 mm and 5 to 25 mm, respectively.

9. A resorbable crosslinked form stable membrane according to claim 8, wherein the thickness of the membrane is 0.5 to 2.5 mm.

10. A resorbable crosslinked form stable membrane according to claim 1, wherein the composite layer of collagen material and inorganic ceramic particles contains 2.0 to 3.0 weight parts of inorganic ceramic for 1 weight part of collagen material.

11. A resorbable crosslinked form stable membrane according to claim 1, wherein the collagen material comprises 70-90% collagen and 10-30% elastin.

12. A resorbable crosslinked form stable membrane according to claim 1, wherein the collagen material is derived from a porcine, bovine or equine peritoneum or pericardium membrane, small intestine mucosa (SIS) or muscle fascie.

13. A resorbable crosslinked form stable membrane according to claim 1, wherein one or both of the layers of the elastic pretensed collagen material includes holes of 5 to 1000 m.

14. A resorbable crosslinked form stable membrane according to claim 1, wherein the inorganic mineral particles have a size of 150 to 500 m.

15. A resorbable crosslinked form stable membrane according to claim 1, wherein the inorganic ceramic is selected from the group consisting of hydroxyapatite or hydroxyapatite bone mineral.

16. A method of regenerating human tissue outside of the oral cavity comprising applying the resorbable crosslinked form stable membrane of claim 1 to a surgical site in a human patient undergoing rhinoplasty, postlateral spinal fusion or orbital reconstruction.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be described in further detail hereinafter with reference to illustrative examples of preferred embodiments of the invention and the accompanying drawing figures, in which:

(2) FIG. 1 represents typical shapes and typical dimensions of resorbable crosslinked form stable membranes: For use in the oral cavity according to the invention of EP-A1-3175869. Those membranes may be flat (1), (1), U-shaped straight (2), (2) or U-shaped curved (3), (3) corresponding to the alveolar spaces of 1 to 3 teeth (incisors, canine, premolar or molars) situated at the front, in the left-hand side or right-hand side curvature or at the rear of the denture. The size of the anterior products is similar to that of the posterior products, the radius of the curvature being such as to conform to the alveolar ridge. Typical dimensions are a=5-20 mm, b=8-20 mm, c=6-10 mm, d=25-40 mm, e=15 mm, f=20-40 mm. For use in human tissue regeneration outside the oral cavity according to the present invention: the arch-shaped membrane for rhinoplasty (4), the slit rectangular or oval tube membrane for posterolateral spinal fusion (5), (5), respectively, and a membrane for orbital fracture reconstruction (6) and (6) (front view and plan view, respectively.

(3) Typical dimensions of membrane for rhinoplasty (4) are: 1=40-80 mm, i=10-15 mm and h=10-15 mm. The rostral ridge of the device may come in different shapes.

(4) Typical dimensions of a membrane for posterolateral spinal fusion (5 and 5) are: 1=60-300 mm, j=15-30 mm and k=5-10 mm. The shape of the posterolateral spinal fusion device may be rectangular (5) or oval (5).

(5) (6) and (6) show a front view and a plan view of a representative membrane for orbital fracture reconstruction. Typical dimensions are: m=30-50 mm, o=20-40 mm and c=5-25 mm.

(6) FIG. 2 is a schematic view of equipment suitable for enabling the tensioning of the polymer layers prior to their assembling into a flat or U-shaped form stable membrane prepared according to the invention.

(7) FIG. 3 represents the assembly of a flat form stable membrane, wherein (1) is a steel plate, (2) is a compressed polyurethane sponge, (3) is a polyamide net, (4) is a layer of elastic pretensed collagen and (5) is a crosslinked hydroxyapatite-collagen plate.

(8) FIG. 4 represents the variation of the force as a function of the strain in a 3-point bending analysis test for the resorbable form stable membrane of the invention crosslinked by EDC/NHS or DHT in comparison to the PLA membrane Resorb-X (KLS Martin).

(9) FIG. 5 represents the stress-strain curves of a few commercially available, wet and sterile collagen materials that could be used in the layers of elastic pretensed collagen material of the resorbable crosslinked form stable membranes according to the invention, namely porcine peritoneum derived Geistlich Bio-Gide collagen membrane (Geistlich Pharma AG), porcine pericardium derived Jason collagen membrane (aap Biomaterials/Botiss) and porcine SIS derived Dynamatrix collagen membrane (Cook Biotech Inc.), and a collagen material derived from muscle fascie. In each of those stress-curves there is a toe region characterized by large strains upon minimal values of stress, a linear or elastic region characterized by a linear increase in strain per unit stress and a failure region characterized by rupture of polymeric fibres. In the stress-stain curves represented in this figure, the elastic modulus (or Young's modulus, i.e. the slope of the linear region of the stress-strain curve) is about 8 MPa for the Geistlich Bio-Gide membrane, about 64 MPa for the Jason membrane, about 54 MPa for the Dynamatrix membrane and about 56 MPa for the collagen material derived from muscle fascie.

(10) FIG. 6 is a column diagram of the % of human gingival fibroblasts that have adhered to the membrane after incubation for 24 hours at 37 C. for Geistlich Bio-Gide collagen membrane, a prototype of the resorbable form stable membrane of the invention crosslinked by DHT (FRM) and the Cystoplast PTFE membrane (Keystone Dental).

(11) The following examples illustrate the invention without limiting its scope.

EXAMPLE 1 PREPARATION OF THE RAW MATERIALS

(12) Preparation of Hydroxyapatite Fine Particles Having a Size of 250 to 400 m (A)

(13) Hydroxyapatite bone mineral fine particles were produced from cortical or cancellous bone as described in Examples 1 to 4 of U.S. Pat. No. 5,417,975, using an additional sieving step between 250 and 400 m.

(14) Alternatively, hydroxyapatite bone mineral fine particles were produced by grinding Geistlich Bio-Oss Small Granules (available from Geistlich Pharma AG, CH-6110, Switzerland) by careful impactation using a pistol and an additional sieving step between 250 and 400 m.

(15) The hydroxyapatite bone mineral fine particles having a size of 250 to 400 m prepared above (A) were stored in glass bottles until use.

(16) Preparation of Collagen Fibres (B)

(17) As described in Example of EP-B1-1676592, peritoneal membranes from young pigs were completely freed from flesh and grease by mechanical means, washed under running water and treated with 2% NaOH solution for 12 hours. The membranes were then washed under running water and acidified with 0.5% HCl. After the material had been acidified through its entire thickness (for about 15 minutes) the material was washed with water until a pH of 3.5 was obtained. The material was then shrunk with 7% saline solution, neutralised with 1% NaHCO.sub.3 solution and washed under running water. The material was then dehydrated with acetone and degreased with n-hexane and dried using ethanol ether. 22 cm pieces of the collagen membranes thus obtained were cut by hand using scissors.

(18) Alternatively, 22 cm pieces of the Geistlich Bio-Gide membrane (available from Geistlich Pharma AG) were cut by hand using scissors.

(19) 1 g of the 22 cm pieces of the collagen membranes obtained above was mixed with 200 ml of dry ice and mixed in a knife mill (Retsch Grindomix) at 5000 rpm until no blockage occurred. The speed was then increased to 6000, 7000, 9000 and 10000 rpm for 20 to 30 seconds, each time adding 50 ml of dry ice.

(20) The dry ice was evaporated and the collagen fibres thus obtained (B) were stored in Minigrip plastic wraps until further use.

(21) Preparation of Cutting Mill Collagen Fibre Segments (C)

(22) The 22 cm collagen fibre pieces obtained above were cut in a cutting mill with a 0.8 mm sieve at 1500 rpm, giving a sieved fraction of cutting mill collagen fibre segments (C).

(23) Preparation of a Collagen Fibre Glue (D)

(24) The sieved fraction of cutting mill collagen fibre segments (C) was mixed in water to obtain a solution of 3%, the pH was set to 3.5 by adding phosphoric acid H.sub.3PO.sub.4 and the suspension was high pressure homogenized at 1500-2000 bar, this being repeated 3 to 5 times.

(25) The resulting slurry was neutralized to about pH 7 by adding a sodium hydroxide solution NaOH and gelled overnight at 4 C. The collagen was concentrated by lyophilisation at 10 C. and 0.310 mbar after freezing for 4 hours at 40 C. and homogenized by knife milling.

(26) The collagen fibre glue (D) was prepared from the slurry obtained as a 2-10% solution in phosphate buffered saline, pH 7.4 by heating to 60 C. until no further particles were visible.

EXAMPLE 2 PREPARATION OF AN OPTIONALLY CROSSLINKED HYDROXYAPATITE/COLLAGEN PLATE (E)

(27) 4 g of collagen fibres (B) and 6 g of cutting mill collagen fibre segments (C) prepared in Example 1 were mixed with 140 g of phosphate buffered saline and shaked in a cocktail mixer. In another example, collagen fibres were substituted completely by cutting mill collagen fibre segments.

(28) 20 to 30 g hydroxyapatite fine particles (A) prepared in Example 1 were added and mixed by hand.

(29) 34.14 g of this mixture were centrifuged at 7000g (7000 times the acceleration of gravity) for 2 minutes.

(30) The pellet was poured between two polyamide-nets (of pore size 21 m and a total of 17% of open structure) in a flat rectangular form of 812 cm and the matter was condensed by removing excess water with a laboratory spoon. The plates obtained were compressed at a pressure of 1-1.7 kPa and dried in a vacuum oven at 30 C./50 mbar for 2 hours, then at 30 C./10 mbar for 8 hours. The polyamide-nets were removed.

(31) Optional Crosslinking of the Hydroxyapatite-Collagen Plate

(32) To facilitate handling of the hydroxyapatite-collagen plate, the latter was crosslinked chemically or by dehydrothermal treatment (DHT).

(33) Chemical cross-linking of the collagen with EDC/NHS was performed, leading to an increase of overall stability of the hydroxyapatite-collagen plate plates. The dried plates were then cross-linked in 10-400 mM EDC and 13-520 mM NHS in 0.1 M MES (2-(N-morpholino)-ethanesulfonic acid) and 40% ethanol at pH 5.5 for 2 hours at room temperature.

(34) The reaction was stopped by incubating the prototypes twice in 0.1 M Na.sub.2HPO.sub.4 buffer at pH 9.5 for an hour. Polar residuals were removed by incubating the prototypes for 1 hour in a 1 M sodium chloride solution and twice for an hour in a 2 mol/1 sodium chloride solution. The chemically crosslinked prototypes were washed a total of 8 times for 30-60 minutes in distilled water, then dehydrated by immersion in ethanol for 15 minutes a total of 5 times. Drying was then performed by carrying out three times diethylether treatment for 5 minutes and subsequent drying at 10 mbar and 40 C. for 30 minutes, or by lyophilisation (freezing below 10 C. and drying by conventional lyophilisation treatment).

(35) Alternatively, cross-linking was performed by dehydrothermal treatment (DHT) at 0.1-10 mbar and 80-120 C. for 1-4 days. In this case no subsequent drying method was necessary.

EXAMPLE 3 PREPARATION OF A RESORBABLE CROSSLINKED FORM STABLE MEMBRANE (M) BY ASSEMBLING AND GLUING TWO ELASTIC PRETENSED COLLAGEN LAYERS ON THE TWO OPPOSITE FACES OF THE HYDROXYAPATITE/COLLAGEN PLATES (E)

(36) The following description will be better understood by referring to FIGS. 2 and 3. The assembly of a flat or U-shaped prototype requires the use of fixed or bendable frames enabling the tensioning of the layers of collagen material.

(37) Forming of Flat or U-Shaped Prototypes (F)

(38) FIG. 2 is a schematic view of equipment suitable for enabling the tensioning of the layers of collagen material prior to their assembling into a flat or U-formed form stable membrane of the invention.

(39) That equipment consists of a frame (a), which can be made of any suitable material, e.g. steel or aluminum. The main purpose for the frame is to anchor the springs (b), which tension the two wet collagen layers (c). The hydroxyapatite/collagen plate (E) was positioned in between the two collagen layers (c).

(40) If a U-shaped resorbable crosslinked form stable membrane is desired, a negative form (e) for bending the collagen plate (E) and frames with hinges (f) are used, thus leading to U-shaped straight prototypes.

(41) Collagen material layers of unsterile Geistlich Bio-Gide Collagen layers were pretensed by elongating or stretching 40 to 100% of initial length through tensioning each spring by 2-3 N, such as to be in the linear region of the stress-curve of the collagen material. Within this linear region, the elastic modulus is highest and therefore the highest stiffness is achieved

(42) Due to the viscoelastic nature of collagenous tissues, wet and tensioned materials were kept for approximately 30 minutes in tensioned state. Due to the relaxing of the pretensed collagen membrane, the springs were tensioned again to 1-3 N, such as to be in the linear region of the stress-curve of the collagen material.

(43) Two round pieces of collagen with a diameter of 10 cm cut from unsterile Geistlich Bio-Gide collagen membrane were used, one of which was punctured with a needle drum containing 50 needles per cm.sup.2 with a shaft diameter of 0.88 mm. Those two round pieces of collagen were wetted and tensioned in a radial manner by 12 springs each tensioned to 1-3 N, leading to an elongation of 40-100% of the initial size of the collagen pieces.

(44) Upon completion of this step, the hydroxyapatite/collagen plates (E) were wetted on both faces with the collagen fibre glue (C) and then, the hydroxyapatite/collagen plate was placed between the two elastic pretensed collagen layers. The central bar (e) as well as the hinges (f) are necessary to produce U-shaped prototypes (see below).

(45) The elastic pretensed membranes were placed on a heating plate and prewarmed to 40 C.

(46) The cross-linked Bio-Oss plate (E) obtained in Example 2 was shortly submerged in prewarmed collagen fibre glue (D) and placed between the two elastic pretensed collagen membranes.

(47) Polyamide nets, as well as sponges (of thickness 5 cm, density of approx. 20-25 mg/cm.sup.3, containing interconnected pores, made of polyurethane), were placed on both sides, compressed by 50-95% leading to compression pressures of up to 120 kPa.

(48) See FIG. 3, which represents the assembly of a flat form stable membrane, wherein (1) is a steel plate, (2) is a compressed polyurethane sponge, (3) is a polyamide net, (4) is a layer of elastic pretensed collagen and (5) is a crosslinked hydroxyapatite-collagen plate.

(49) Subsequently, the construct was dried in a vacuum oven at 40 C. with a steady decrease in air pressure down to 10 mbar for a total of 32 hours.

(50) Forming of U-Shaped Prototypes

(51) The skilled person will readily adapt the apparatus of FIGS. 2 and 3 and the method described above to the forming of U-shaped prototypes straight or curved, by bending the construct over an appropriate negative form and replacing one of the sponges by a thinner polyurethane sponge or a fibre free paper towel.

(52) Cross-Linking of Flat or U-Shaped Prototypes (G)

(53) Flat or U-shaped prototypes (F) were cut into the desired dimensions using scissors or a small circular saw. The prototypes were then crosslinked chemically or by dehydrothermal treatment (DHT).

(54) Chemical crosslinking was performed in 0.1 M MES buffer at pH 5.5 and an ethanol content of 40 Vol-% at concentration of EDC and NHS of 10 to 400 mM and 13 to 520 mM respectively. The prototype concentration in the cross-linking solution was 10%. To enable homogenous cross-linking, plates were initially treated under vacuum (<40 mbar) and the cross-linking reaction was carried out at 4 C. for 2 hours, all buffers being precooled to this temperature.

(55) The reaction was stopped by incubating the prototypes twice in 0.1 M Na.sub.2HPO.sub.4 buffer at pH 9.5 for an hour. Polar residuals were removed by incubating the prototypes for 1 hour in a 1 M NaCl solution and twice for an hour in a 2 M NaCl solution. Prototypes were washed a total of 8 times for 30-60 minutes in distilled water. Dehydration and drying was then performed by carrying out 5 times ethanol treatment for 15 minutes and three times diethylether treatment for 5 minutes and subsequent drying at 10 mbar and 40 C. overnight or until the product was completely dry, or by conventional lyophilisation (freezing below 10 C. and drying by conventional lyophilisation treatment) of the not by solvent treated product. Alternatively, cross-linking was performed by dehydrothermal treatment (DHT) at 0.1-10 mbar at 80-160 C. for 1-4 days. In this case no subsequent drying method was necessary.

(56) Prototypes obtained by the above described methods are wetted in saline or PBS within an hour or two. To allow wetting within 10 min, prototypes are rewetted in distilled water for approximately 1 to 2 hours. At this time the perforation of one or two sides with the above described needle drum is possible too. Sodium chloride is applied by incubating the prototypes three times for 40 min in a 200 g/1 NaCl solution. The sodium chloride is precipitated as described below (H).

(57) Drying of Cross-Linked Flat or U-Shaped Prototypes (H)

(58) The crosslinked prototypes were dehydrated by immersion in ethanol for 15 minutes a total of 5 times. They were then dried by either solvent drying (three times diethylether treatment for 5 minutes and subsequent drying at 10 mbar and 40 C.) or conventional lyophilisation (freezing below 10 C. and drying by conventional lyophilisation treatment).

(59) The thickness of the crosslinked form stable membrane of the different prototypes in wet state was from 1.0 to 2.0 mm, for most of them from 1.2 to 1.8 mm. The dried prototypes were optionally sterilized by x-ray-irradiation at 27-33 kGy.

EXAMPLE 4 PROPERTIES OF THE RESORBABLE CROSSLINKED FORM STABLE MEMBRANE

(60) The following characteristics of the resorbable cross-linked form stable membrane obtained in Example 3 were determined: (1) Wettability in PBS, (2) Mechanical strength, (3) Enzymatic degradation using collagenase from Clostridium histolyticum and (4) Cell adhesion (5) Measurement of the elongation of the elastic pretensed collagen material layers (6) Measurement of the thickness of the collagen-hydroxyapatite plates and final prototypes

(61) (1) Wettability in PBS

(62) The time of complete wetting in PBS (Phosphate buffer saline) as assessed visually was observed to be between 5 and 10 minutes for the different prototypes of the resorbable crosslinked form stable membrane, that time depending mainly on the treatment with sodium chloride prior to dehydration with ethanol and drying.

(63) (2) Mechanical Strength

(64) The form stability of the membrane of the invention was assessed by a 3-point uniaxial bending test which is similar to the methods described in EN ISO 178 and ASTM D6272-10, the membrane of the invention being submerged in PBS at a pH of 7.4 and a temperature of 37 C.

(65) This test was considered most useful, because every form stable membrane designed to mechanically stabilize a bony defect at a non-containing site will experience bending moments. Therefore, 3- or 4-point bending can be used as a test to characterize the used materials and additionally to compare different products with e.g. different thicknesses. For material characterization, the bending modulus is the most suitable parameter. However, to compare different products which have different thicknesses, the maximal force after 8-10 mm of indentation is more relevant and therefore used, to characterize the product.

(66) In the 3-point uniaxial bending test used, the specimens were cut to a size of 5013 mm and incubated in PBS at 37 C. until complete wetting as visually observed. Mechanical testing was conducted at 5 mm per minutes in a 3-point bending apparatus with a support span width of 26 mm and a radius of 5 mm of each supporting structure. The bending module was calculated within 1 and 5% bending strain. The resulting maximal forces were read out after lowering the central indenter between 8 and 10 mm.

(67) The test was performed for a membrane of the invention of thickness 1.5 mm crosslinked by EDC/NHS, a membrane of the invention of thickness 1.6 mm crosslinked by DHT and the PLA membrane Resorb-X from KLS Martin of thickness 0.137 mm.

(68) FIG. 4, which represents the variation of the force as a function of the strain for those membranes, shows that the mechanical stability of membrane of the invention crosslinked by EDC/NHS (about 0.65 N for 8 mm strain) or crosslinked by DHT (about 0.40 N for 8 mm strain) is substantially superior to that of the PLA membrane Resorb-X (about 0.10 N for 8 mm strain). The membrane of the invention will thus better stabilize the bony defect at a non-containing site.

(69) (3) Enzymatic Degradation Test Using Collagenase from Clostridium histolyticum

(70) In the human body collagens are degraded by human tissue matrix-metalloproteinase (MMP), cathepsins and putatively by some serine proteinases. Best studied are the MMPs where collagenases (notably MMP-1, MMP-8, MMP-13 and MMP-18) are the most important enzymes for direct collagen degradation (Lauer-Fields et al. 2002 Matrix metalloproteinases and collagen catabolism in BiopolymersPeptide Science Section and Song et al. 2006 Matrix metalloproteinase dependent and independent collagen degradation in Frontiers in Bioscience).

(71) Collagenase capability to degrade collagen tissues and membranes depends on the substrate flexibility and collagen type, MMP active sites and MMP exosites. Collagenases align at the triple helical collagen, unwind it and subsequently cleave it (Song et al. 2006, see above).

(72) With the view of overcoming differences in degradation between the different collagen types, collagenase degradation of collagen is often assessed using collagenase from Clostridium histolyticum which has a high catalytic speed (Kadler et al. 2007 Collagen at a glance in J Cell Sci). Generally, a natural collagen product degrades faster than a chemically cross-linked collagen product.

(73) In this test the collagen products (samples of the resorbable cross-linked form stable membrane at 1 mg/ml collagen) were incubated at 37 C. with 50 units/ml from Clostridium histolyticum (one unit being defined as liberating peptides from collagen from bovine Achilles tendon equivalent in ninhydrin color to 1.0 micromole of leucine in 5 hours at pH 7.4 at 37 C. in the presence of calcium ions) in a calcium containing Tries-buffer and the degradation of the collagen matrix was measured visually and with the DC Protein Assay from Bio-Rad Laboratories (Hercules, USA, Order Nor. 500-0116) using Collagen Type I as reference material. The collagen concentration was determined using a microwellplate spectrometer (Infinite M200, available from Tecan).

(74) All prototypes of the resorbable crosslinked form stable membrane of the invention showed at least 10% collagen degradation (as assessed by DC Protein assay using collagen type I as standard.) after 4 hours, the rate of collagen degradation (lower than for the Geistlich Bio-Gide membrane) being dependent on the crosslinking conditions used.

(75) (4) Cell Adhesion

(76) Cell adhesion to different membranes was assessed by first seeding 8 mm membrane punches with 100000 human gingival fibroblasts previously labelled with a fluorescent, lipophilic dye, incubating for 24 hours at 37 C., removing non-adherent cells by washing the membranes in PBS, lysing adherent cells and quantifying them by measuring fluorescence at 485 nm. Fluorescence was normalized to a standard curve established with cell-seeded membrane punches that were not washed prior to lysis.

(77) The results obtained for the form stable resorbable membrane are represented in FIG. 5 which is a column diagram representing horizontally the % of cells capable to adhere on different types of dental membranes in percentage, the resorbable crosslinked form stable membrane of the invention and the Cystoplast PTFE membrane (Keystone Dental).

(78) FIG. 5 shows that adhesion to the resorbable crosslinked form stable membrane of the invention is about 10.5%, a value much closer to that of the Geistlich Bio-Gide membrane of about 13% than to that of the Cystoplast PTFE membrane of about 4%. The Geistlich Bio-Gide membrane is well known for its good healing properties with a low rate of dehiscence (Zitzmann et al. 1997 Resorbable versus non-resorbable membranes in combination with Bio-Oss for guided bone regeneration in Int J Oral Maxillofac Implants and Tal et al. 2008 Long-term bio-degradation of cross-linked and non-cross-linked collagen barriers in human guided bone regeneration in Clin Oral Implants Res) or no excessive inflammation (Jung et al. 2013 Long-term outcome of implants placed with guided bone regeneration (GBR) using resorbable and non-resorbable membranes after 12-14 years in Clin Oral Implants Res) This measured value of adhesion of human gingival fibroblasts to the resorbable crosslinked form stable membrane of the invention is predictive for soft tissue healing without adverse advents such as excessive inflammation or dehiscence.

(79) (5) Measurement of the Elongation of the Elastic Pretensed Collagen Material Layers

(80) To determine the amount of tensioning of the collagen layers, the dry collagen layer is mounted to a tensioning ring (FIG. 2, part a) using the not yet tensioned springs (FIG. 2, part b). In the centre of the membrane at least 4 points, which are several centimetres apart from each other, are marked using a pencil or pen. The distance between each point is measured using a ruler. The measured distances are defined as the initial lengths between each point. The collagen layer is submerged in water and tensioned to the desired force. The collagen layer is incubated in water for 30 minutes. Due to the viscoelastic nature of most collagen layers, the tension reduces. Therefore, the collagen layers need to be tensioned again. After 30-40 minutes of incubation the distance between each point is measured with a ruler. The percentage of strain is determined by subtracting the initial length from the length after tensioning, divided by the initial length multiplied by 100.

(81) Typical results such as to be in the linear region of the stress-strain curve are between 40 and 100% strain (elongation, stretching) for unsterile Geistlich Bio-Gide. Strain values measured by this method are not directly comparable to strain values obtained in a uniaxial elongation test.

(82) (6) Measurement of the Thickness of Collagen Hydroxyapatite Plate and Final Prototype

(83) The thickness of the final prototypes or the collagen/hydroxyapatite plate E can be measured as described above or by using a sliding calliper.

(84) (7) Analysis of the Mechanical Properties of Different Collagen Layers (FIG. 5)

(85) To compare different sources of collagen layers and estimate their mechanical properties, standard uniaxial tensioning of wet samples was used. A general setup for such an analytical method is described in ASTM D882-09 Standard Test Method for Tensile Properties of Thin Plastic Sheeting. Due to the high costs of the collagen membranes used, several parameters of the testing were adapted. Samples were cut into rectangular sheets of e.g. 21 cm, prewetted in isotonic phosphate buffered saline and mounted to a tensile testing machine with a distance of 1 cm between each sample holder. The samples were tensioned at a constant speed of 33% of initial length per minute. The prepressure, at which 100% initial length is recorded, was typically set to 50 kPa. The elongation of the sample was calculated using the distance between the two sample holders.

(86) The stress-strain curves of FIG. 5 were thus obtained.

EXAMPLE 5 PREPARATION OF PROTOTYPES OF THE RESORBABLE CROSSLINKED FORM STABLE MEMBRANE FOR USE OUTSIDE THE HUMAN CAVITY

(87) A resorbable crosslinked form stable nasal arch-shaped membrane was prepared as described below by assembling and gluing two layers of elastic pretensed collagen material on the two opposite faces of the hydroxyapatite/collagen plates (E)

(88) The following description will be better understood by referring to FIGS. 2 and 3.

(89) The assembly of a nasal arch-shaped membrane requires the use of bendable frames enabling the tensioning of the layers of collagen material.

(90) Forming of Nasal Arch-Shaped Membrane (Membrane for Rhinoplasty), Orbital Fracture Membrane or Posterolateral Spinal Fusion Membrane Prototypes

(91) FIG. 2 is a schematic view of equipment suitable for enabling the tensioning of the layers of collagen material prior to their assembling into a nasal arch-shaped membrane prototype of the invention.

(92) That equipment consists of a frame (a), which can be made of any suitable material, e.g. steel or aluminum. The main purpose for the frame is to anchor the springs (b), which tension the two wet collagen layers (c). The hydroxyapatite/collagen plate (E) was positioned in between the two collagen layers (c).

(93) For a nasal arch-shaped membrane prototype, a negative form (e) for bending the collagen plate (E) and frames with hinges (f) are used, thus leading to nasal arch-shaped membrane.

(94) Collagen material layers of unsterile Geistlich Bio-Gide Collagen layers were pretensed by elongating or stretching 40 to 100% of initial length through tensioning each spring by 2-3 N, such as to be in the linear region of the stress-curve of the collagen material. Within this linear region, the elastic modulus is highest and therefore the highest stiffness is achieved

(95) Due to the viscoelastic nature of collagenous tissues, wet and tensioned materials were kept for approximately 30 minutes in tensioned state. Due to the relaxing of the pretensed collagen membrane, the springs were tensioned again to 1-3 N, such as to be in the linear region of the stress-curve of the collagen material.

(96) Upon completion of this step, the hydroxyapatite/collagen plates (E) were wetted on both faces with the collagen fibre glue (C) and then, the hydroxyapatite/collagen plate was placed between the two elastic pretensed collagen layers. The central bar (e) as well as the hinges (f) are necessary to produce nasal arch-shaped membrane prototypes (see below).

(97) The Bio-Oss plate (E) obtained in Example 2 was shortly submerged in prewarmed collagen fibre glue (D) and placed between the two elastic pretensed collagen membranes.

(98) Polyamide nets, as well as sponges (of thickness 5 cm, density of approx. 20-25 mg/cm.sup.3, containing interconnected pores, made of polyurethane), were placed on both sides, compressed by 50-95% leading to compression pressures of up to 120 kPa.

(99) See FIG. 3, which represents the assembly of a flat form stable membrane, wherein (1) is a steel plate, (2) is a compressed polyurethane sponge, (3) is a polyamide net, (4) is a layer of elastic pretensed collagen and (5) is a crosslinked hydroxyapatite-collagen plate.

(100) Subsequently, the construct was dried in a vacuum oven at 40 C. with a steady decrease in air pressure down to 10 mbar for a total of 32 hours.

(101) The skilled person will readily adapt the apparatus of FIGS. 2 and 3 and the method described above to the forming of nasal arch-shaped membrane prototypes, the orbital fracture membrane prototype or the posterolateral spinal fusion membrane prototype, by bending the construct over an appropriate negative form and replacing one of the sponges by a thinner polyurethane sponge or a fibre free paper towel.

(102) Cross-Linking of the Nasal Arch-Shaped Membrane, Orbital Fracture Membrane or Posterolateral Spinal Fusion Membrane Prototypes

(103) Nasal arch-shaped membrane prototypes, the orbital fracture membrane prototype or the posterolateral spinal fusion membrane prototype (F) were cut into the desired dimensions using scissors or a small circular saw. The prototypes were then crosslinked chemically or by dehydrothermal treatment (DHT).

(104) Cross-linking was performed by dehydrothermal treatment (DHT) at 0.1-10 mbar at 80-160 C. for 1-4 days. In this case no subsequent drying method was necessary. Prototypes obtained by the above described methods are wetted in water, saline or PBS within an hour or two. Prototypes are cut to its final shape and are punctured with a needle bed (as described above). At this time the perforation of one or two sides with an appropriate needle bed is possible too.

(105) Drying of Crosslinked Nasal Arch-Shaped Membrane, Orbital Fracture Membrane or Posterolateral Spinal Fusion Membrane Prototypes

(106) The crosslinked prototypes were dehydrated by immersion in ethanol for 15 minutes a total of 5 times. They were then dried at 10 mbar and 40 C. in a vacuum oven. Another option is to freeze dry the with water wet device (freezing below 10 C. and drying by conventional lyophilisation treatment).

(107) The thickness of the crosslinked form stable membrane of the different prototypes in wet state was from 1.0 to 2.0 mm, for most of them from 1.2 to 1.8 mm. The dried prototypes were optionally sterilized by x-ray-irradiation at 27-33 kGy.

(108) While the invention has been illustrated and described in details in the drawings and forgoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive: the invention is not limited by the disclosed embodiments.

(109) Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the amended claims.

(110) In the claims, the word comprising does not exclude other elements; the definite article a or an does not exclude a plurality.