Design of an artificial skin. II. Control of chemical composition

Influence of glycosaminoglycans on the collagen sponge component of a bilayer artificial skin

A bilayer artificial skin composed of a silicone membrane and a collagen sponge layer containing glycosaminoglycans (GAGs) was first developed by Yannas and Burke. They reported that GAGs contained in the collagen sponge layer contributed to the function of the artificial skin. In an attempt to assess the effect of GAGs in the collagen sponge layer, the electron microscopic structure, mechanical strength of collagen sponges, and cell proliferation were examined in vitro, using four kinds of collagen sponges containing: no GAG, chondroitin 6-sulphate (C6S), dermatan sulphate (DER), and hyaluronic acid (HYA). The results indicated that: (1) addition of GAGs scarcely affected the mechanical structure of collagen sponges; (2) addition of C6S and DER reinforced mechanical strength, while addition of HYA did not; (3) addition of C6S and DER significantly decreased cell proliferation.

Experimental study of a newly developed bilayer artificial skin

A bilayer artificial skin composed of an outer layer of silicone polymer and an inner sponge layer of collagen containing chondroitin 6-sulphate was developed by modifying the technique proposed by Yannas et al. The artificial skin was placed on the skin defects on the backs of rats. Histological observation indicated that fibroblasts and capillaries infiltrated into the pores and filled in lattice spaces, resulting in synthesis of the connective tissue matrix and absorption of the original network of collagen and chondroitin 6-sulphate. Epidermal cells migrated from the edge of the wound between the two layers. Post-operative contracture in the wound with the artificial skin was significantly less than in the control.

Two-ply biodegradable nerve guide: basic aspects of design, construction and biological performance

A synthetic biodegradable nerve guide was constructed of two polymeric layers: an inner microporous layer prepared from a copolymer of L-lactide and epsilon-caprolactone (pore size range 0.5-1 micron) and an outer microporous layer prepared from a polyurethane/poly(L-lactide) mixture (pore size range 30-70 microns). This nerve guide was used to bridge a 7 mm gap in the right sciatic nerve of rats. It enabled the sciatic nerve to regenerate across the gap, forming a new, well-defined nerve that effectively re-established the contact between the proximial and distal nerve end, as effective as an autograft.

The attachment and growth of an established cell line on collagen, chemically modified collagen, and collagen composite surfaces

The attachment and growth of an established cell line derived from mouse fibroblasts on collagen, chemically modified collagen, and collagen composite surfaces were compared. Tissue culture polystyrene dishes provided a suitable control. The substrates included native bovine dermal collagen, succinylated, acetylated and methylated collagen, and a series of composite materials formed from collagen and the glycosaminoglycans hyaluronic acid, chondroitin 4-sulphate and chondroitin 6-sulphate and the glycoprotein fibronectin. Attachment and growth of cells on each of these substrates were assessed by visual inspection under optical microscopy, by detachment of the cells using trypsinization and subsequent counting in a Coulter counter; and by 3H-thymidine incorporation studies. A very good correlation between the results was obtained by the three methods employed which showed that collagen, in comparison to polystyrene, is a relatively poor substrate for cellular attachment, growth and proliferation, but it may be improved by chemical modification and by incorporation of either fibronectin, chondroitin sulphate (5 and 10%), or low levels (less than 5%) of hyaluronic acid into the collagen matrix. Concentrations in excess of 5% hyaluronic acid into the collagen matrix, however, appeared to inhibit cellular attachment and growth and such materials provided a poorer substrate than native collagen.

In vivo evaluation and comparison of collagen, acetylated collagen and collagen/glycosaminoglycan composite films and sponges as candidate biomaterials

Native collagen, acetylated collagen, collagen/10% chondroitin sulphate, collagen/2.5% hyaluronic acid and collagen/20% hyaluronic acid were implanted both as film and as sponge into rat lumbar muscle for 7 and 14 d. After 7 d implantation, all materials elicited an acute inflammatory cell response characterized by numerous polymorphs and histocytes. The cell population after 14 d was principally mononuclear, i.e. leucocytes, neutrophils, macrophages, lymphocytes and fibroblasts. Both films and sponges followed a similar pattern. Native collagen elicited a subacute inflammatory response after 7 d. However, 14 d after implantation, a marked infiltration by neutrophils was apparent with subsequent degradation of existing collagen material. Acetylated collagen film evoked a much greater inflammatory cell response than native collagen. Both collagen/hyaluronic acid composites elicited a similar response. The collagen/10% chondroitin sulphate composite elicited the least inflammatory cell response at 7 d, whereas infiltration by host fibroblasts after 14 d implantation was clearly seen.

Clinical evaluation of a new bilayer "artificial skin" composed of collagen sponge and silicone layer

A bilayer "artificial skin" composed of an outer layer of silicone and an inner sponge layer of collagen and chondroitin sulphate has been developed by modifying the technique proposed by Yannas et al. (1980). Following experimental successes, the "artificial skin" was applied clinically. It was placed on the skin defects of 10 patients. Three weeks after application the outer layer of silicone sheet was peeled off and thin split thickness skin was grafted onto the newly synthesised dermis-like tissue. Secondary skin grafts took perfectly in all cases and postoperative appearance was satisfactory.

Collagen banded fibril structure and the collagen-platelet reaction

Bovine hide collagen dispersions were swollen in the pH range 1.6-7.0, treated with glutaraldehyde, and dialyzed to neutral pH. The intensity with which these collagens reacted with human platelets in plasma was studied by aggregometry and scanning electron microscopy. Collagen swollen at a pH below 4.25 +/- 0.30 and treated with glutaraldehyde exhibited greatly reduced platelet aggregating ability after restoration of neutral pH. In addition, the state of supramolecular order in these collagens was investigated by transmission electron microscopy and infrared spectroscopy. Native, insoluble collagen fibrils were found to lose their banded structure, as observed by transmission electron microscopy, reversibly when exposed to low ionic strength aqueous solutions below pH 4.25 +/- 0.30. During the disorder transition, which occurred by time dependent swelling of fibrils, but without their disaggregation, the packing order in the fibrils was largely abolished while the triple helical structure of individual collagen molecules was retained. Chemical modification of collagen by glutaraldehyde treatment was found to prevent recrystallization of collagen during dialysis to neutral pH but did not otherwise affect the collagen-platelet reaction. The results of altering collagen mass dose (concentration) demonstrated the critical importance of traces of banded fibrils which resisted disordering below pH 4.25. The data suggest that collagen preparations which are free of significant traces of banded fibrils, but which are made up of collagen molecules possessing triple helical structure do not induce platelet aggregation, irrespective of dose.

Structure of a collagen-GAG dermal skin substitute optimized for cultured human epidermal keratinocytes

Collagen and glycosaminoglycan (GAG) dermal skin substitutes (membranes) were studied as substrates for cultured human epidermal keratinocytes. Structure of dermal substitutes was optimized for pore size to promote ingrowth of fibrovascular tissue from the wound bed and for culture of human keratinocytes of the membrane's surface. Pore size of the freeze-dried material was regulated by control of the temperature of freezing between -50 degrees C and -20 degrees C and by concentration of starting materials between 0.17% and 1.62% wt/vol. A nonporous surface of collagen-GAG was laminated to the membranes to provide a planar substrate for cultured epidermal keratinocytes. Thickness of dermal substitutes was regulated by control of the volume and concentration of starting materials. Biotin was conjugated to solubilized collagen for binding with avidin of specific quantities of biologically active molecules. The optimized membranes are suitable substrates for the culture of human epidermal keratinocytes, and together with the cells yield a composite material that is histologically similar to skin.

Porous collagen sponge wound dressings: in vivo and in vitro studies

Collagen-based materials can be formed into a three-dimensional sponge for use as a wound dressing and as a support for cell cultured skin components. Factors such as biocompatibility, morphological structure and addition of non-collagenous molecules to collagen are analyzed and discussed. Large pores or channels, interchannel communications and combinations of macromolecules of the connective tissue enhance wound tissue infiltration in vivo as well as cell growth in vitro into collagen sponges. The presence of such factors can be useful in patients with excised burn wounds and pressure skin ulcers.

Observations on the development and clinical use of artificial skin--an attempt to employ regeneration rather than scar formation in wound healing

Artificial skin, a bilaminar membrane, is grafted on an excised wound immediately following injury. This bilayer membrane, made of a dermal and epidermal portion, is populated in place on the wound bed by the patient's own fibroblasts and epidermal cells producing a permanent skin replacement with an anatomically functioning dermis and epidermis. The dermal portion is a porous collagen-chondroitin 6-sulfate fibrous matrix arranged in a three dimensional pattern closely resembling the fiber pattern of normal dermis. A thin silastic covering serves as a temporary epidermis immediately after grafting until the patient's epidermal cells, seeded on the "neodermis", grow into a confluent epidermal replacement. The autogenous "neodermis" is produced as fibroblasts and vessels migrate from the wound bed into the artificial dermal template and, using the artificial fibers as a scaffolding, synthesize new connective tissue in the collagen fiber pattern of normal dermis rather than the pattern of scar while slowly biodegrading the artificial fibers. This replacement dermis functions as normal dermis and not as scar tissue. The patient's epidermal cells seeded on the "neodermis" grow into a confluent normal appearing epidermis and with the neodermis produce a permanent skin composed of normal functioning dermal and epidermal components produced in situ by the patient's own cells. Artificial skin has been successfully used to permanently replace skin destroyed by burn injuries ranging from 10 to over 95% BSA. The long-term functional results in these patients have been excellent and the long term cosmetic results in preliminary studies tend to be superior to autograft. Artificial skin appears to provide a successful physiologic and cosmetic skin replacement in severe burn injury.

In vitro and in vivo characterization of synthetic polymer/biopolymer composites

Collagen, extracted from rat tail tendons using dilute acetic acid, was fabricated into films for subsequent characterization and biocompatibility testing. The reconstituted collagen was characterized with infrared spectroscopy, solution viscosity, contact angle, and tensile testing techniques and was found to be pure with molecular and physical properties consistent with findings of previous researchers. Composites composed of collagen coated on urethane and Silastic Rubber films were fabricated to give improved tear resistance. The biocompatibility of the composites and individual polymers was evaluated by discs implanted in the paravertebral muscle of rabbits. After four weeks none of the materials induced any gross changes in the muscle. Histopathological evaluation revealed a fibrous capsule around all of the materials. Collagen and collagen composites exhibited a stronger reaction as evidenced by a larger fibroblast layer and a variety of inflammatory cells, lymphocytes, eosinophils, and macrophages. The urethane was rated with a response index of 1.5 versus 3.25 for the urethane/collagen composite; Silastic Rubber rated a response index of 1.67 versus 3.12 for the Silastic Rubber/collagen composite; collagen rated a response index of 3.3. The polyester sutures also induced a reaction with a larger fibrous capsule but fewer inflammatory cells as compared to collagen and collagen composites.

Regeneration of epidermis by cells grown in tissue culture

Pig epidermal cells were grown in vitro for 21 to 180 days, forming multilayered sheets of epidermis varying from thirteen to forty layers. For an evaluation of their use in wound coverage, they were transplanted onto surgically prepared full-thickness wound beds on domestic swine. Autologous epidermal cells were studied on twenty-five of the animals, and allogeneic epidermal cells were studied on thirteen. Autologous epidermal cells grown in vitro provided functional wound coverage 10 to 15 days post transplantation and inhibited scar formation. Allogeneic epidermal cells "melted" 15 days post transplantation. There was a 7-day prolongation of allograft survival with cultured, as compared with noncultured, epidermal allografts. Results obtained are related to other advances made in this field.

Evaluation of collagen crosslinking techniques

The properties of collagen films crosslinked by physical and chemical techniques were compared to the properties of films crosslinked with glutaraldehyde (GTA). Physical techniques studied include exposure to short wave (254 nm) u.v. irradiation and severe dehydration. Chemical techniques studied include immersion of collagen films in aqueous solutions of cyanamide or GTA. Collagen films exposed to combinations of aqueous solutions of cyanamide and severe dehydration had moduli of elasticity, swelling ratios and resistance to bacterial collagenase similar to films crosslinked with GTA. Theoretical calculations based on amino acid composition indicate that approximately seven times as many amino acid residues are capable of forming crosslinks using cyanamide or severe dehydration procedures as compared to GTA crosslinking. In addition, using severe dehydration or cyanamide forms crosslinks involving both amino and carboxyl residues which may allow these procedures to act synergistically. Based on our studies this two-step procedure effectively crosslinks collagen-based biomaterials while the only by-product of this reaction is water-soluble urea. Preliminary biocompatibility studies suggest that this crosslinking procedure may allow for pronounced tissue ingrowth.

Considerations on manufacturing principles of a synthetic burn dressing: a review

This review presents various considerations on the construction of a synthetic burn dressing, based mainly on collagen protein. Membranous wound covers are compared with sponge-felt types, monocomponental with composite. The importance of collagen crosslinking agent and the nonextractibility of any component from the dressing material are discussed. According to the type of the burn the dressing should be used dry or wet, plain or medicated, and changed often to reduce substantially the presence of necrotic tissue, inflammatory cell of the granulation tissue, and bacterial contamination.

Wound tissue can utilize a polymeric template to synthesize a functional extension of skin

Prompt and long-term closure of full-thickness skin wounds is guinea pigs and humans is achieved by applying a bilayer polymeric membrane. The membrane comprises a top layer of a silicone elastomer and a bottom layer of a porous cross-linked network of collagen and glycosaminoglycan. The bottom layer can be seeded with a small number of autologous basal cells before grafting. No immunosuppression is used and infection, exudation, and rejection are absent. Host tissue utilizes the sterile membrane as a culture medium to synthesize neoepidermal and neodermal tissue. A functional extension of skin over the entire wound area is formed in about 4 weeks.

Successful use of a physiologically acceptable artificial skin in the treatment of extensive burn injury

A bilayer artificial skin composed of a temporary Silastic epidermis and a porous collagen-chondroitn 6-sulfate fibrillar dermis, which is not removed, has been used to physiologically close up to 60% of the body surface following prompt excision of burn wounds in ten patients whose total burn size covered 50--95% body surface area (BSA). Following grafting, the dermal portion is populated with fibroblasts and vessels from the wound bed. The anatomic structure of the artificial dermis resembles normal dermis and serves as a template for the synthesis of new connective tissue and the formation of a "neodermis," while it is slowly biodegraded. This artificial skin has physiologically closed excised burn wounds for periods of time up to 46 days before the Silastic epidermis was removed. At the time of election when donor sites are ready for reharvesting, the Silastic epidermis is removed from the vascularized artificial dermis and replaced with 0.004 autoepidermal graft in sheet or meshed form. Clinical and histologic experience in a relatively short follow-up period (2--16 months) indicates that "neodermis" retains some of the anatomic characteristics and behavior of normal dermis, thus promising improvement in the functional and cosmetic results, as well as providing physiologic function as a skin substitute. The artificial skin is easily sterilized and stored at room temperature, capable of large scale production, and immediately available for grafting, indicating its potential for easy and relatively economic use in the burn patient.

Design of an artificial skin. Part III. Control of pore structure

Several methods are compared for preparing collagen-glycosaminoglycan (GAG) membranes of high or low porosity. Collagen-GAG membranes have been used to cover satisfactorily large experimental full-thickness skin wounds in guinea pigs over the past few years. Methods studied as means for controlling pore size are confined to purely physical processes which do not require use of additives or chemical reagents to form the porous membrane. We find that membranes, initially swollen in distilled water or saline, shrink linearly to no less than 94% of original dimension after freeze drying; to 75% after critical point drying (from CO2, following water-ethanol exchange); and to 41% of original dimension following air drying from the swollen state. Scanning electron microscopic study of the pore structure resulting from eah drying procedure confirms our major conclusion: A carefully designed freeze drying process, two variants of which are described in detail, yields membranes with the highest mean pore size, as measured by quantitative stereological procedures. Critical point drying gave significantly more shrinkage and a lower mean pore size than either one of the two freeze drying procedures used.