* Laboratoire de Recherches chirurgicales, Unité de Recherche sur les substituts cutanés et Hôpital de la Conception, Centre des Brûlés, Marseille, France
** Laboratoire Biochimie C., Hôpital Edouard Herriot et Centre des Brûlés, Lyon, France (Echinard C.) Laboratoire de Recherches Chirurgicales, Faculté de médecine, Marseille, France

SUMMARY. Since 1987 we have been developing a cultured epidermis, using H. Green's technique. This epithelium was intended to cover skin losses such as extensive burns or excisions of giant naevi. Skin was harvested from the patient when he was admitted; 1.5 X 106 keratinocytes were cultured with 2 x 106 3T3 fibroblasts. After 10 days, a subculture was performed. On day 21, the sheets of cells were grafted onto patients or animals. Results with this single technique were not very satisfactory: the quality of the skin was poor and contractures often appeared within 2 months. A fibrotic scar was formed under the epithelium. For this reason we also developed an artificial dermis made of human collagen, chitosan and glycosaminoglycans. The epidermal ization of this dermis was then undertaken. This was first performed in vivo in rats and nude mice. These works are still in progress. In vitro epithelialization was also done using Green's or Boyee's technique on artificial dermis prior to grafting. Recent data about these experimental studies will be reported.

Since 1987, we have been performing epidermal cell cultures according to Green's technique (6). This allows us to obtain, within 21 days, an epidermal sheet grown from burned patient keratinocytes. Harvesting of the cells is done 24 hours after admission of the patient to hospital: a 3 CM2 full thickness skin piece is taken under sterile conditions.
Epidermal cells (EQ are separated from dermis (i.e. fibroblasts) by trypsin EDTA.
EC are then grown on a layer of irradiated 3T3 mice fibroblasts that are used as feeder cells: 2 x 106 3T3 together with 2 x 106 keratinocytes.
The culture medium, containing DMEM, HAM, FCS, hydrocortisone, insulin, cholera-toxin, glutamine, streptomycin and penicillin, is supplemented on day 3 with epidermal growth factor.
After 10 to 12 days cells are confluent. They are dissociated by trypsinization and regrown on another 3T3 feeder layer in order to perform a larger subculture. Within the next 10 days, EC are forming a nice sheet (Fig. 1) which can be detached from the bottom of the flask after incubation with dispase, and without dissociation of the cultured cells. EC sheets are then attached and clipped on gauzes and can be transferred to patients. Such an epithelium can be used for covering extensive burn patients or children with giant naevi.
However, though this technique is biologically excellent, it seems to give unsatisfactory clinical results. The quality of the reconstructed skin is poor, and shrinkage and contraction often occur. This is most frequently due to the granulation tissue of the underlying wound bed being infected and containing retracting myofibroblasts.

For this reason we have been simultaneously working on the fabrication of an artificial dermis, partly according to Burke's (2,7) theory, in order to obtain a hybrid artificial full thickness skin, comparable to normal skin with its two layers: epidermis and dermis.
This artificial dermis is constituted of an extracellular matrix made of human collagen (I, III, IV), chitosan and glycosaminoglycans. These last two components form an ionic network and give the matrix its specific three-dimensional structure. Our previous works, reported last year, showed no toxicity towards cells, good elasticity, and good tolerance to collagenase (Mediterranean Burns Club Ist meeting 1987).
In vivo biocompatibility was done in the Sprague Dawley rat, showing that the grafted artificial dermis is progressively recolonized by cells and vessels (3).
On day 2, the dermis is still separated from the receiving bed by oedema and exudate.
On day 20, the dermis has become a real neodermis with nicely orientated fibroblasts, very different from granulation tissue with its anarchic colonization and vertical vascuiarization (4,5).
We then endeavoured to epidermalize this artificial dermis. First studies were carried out in vivo in Sprague Dawley male rats and nude mice. When the sheets of artificial dermis and epidermis (grown according to Green's technique) were applied together at the same time on a full thickness wound of the back; results were not very satisfactory. Usually the epidermis became necrotic before the dermis could be completely colonized and revascularized. When the epidermis was grafted a week after the dermis was put on the back of the animal, we could obtain a living full thickness skin by day 20. However, due to the problems of dressings in animals, results were not as good as expected.
The next step was the graft of artificial dermis in burn patients.
A 30-year-old lady sustained a 45% full thickness flame burn. She was very deeply burned on the chest and abdomen. An early excision down to the fascia was performed on day 3 post-burn and artificial dermis was grafted at the same time (Fig. 2), a piece of healthy skin was harvested and an epidermal cell culture was begun. The aspect of the wound on day 5 after excision and grafting of the dermis was quite good.
The histological study carried out at that time showed a good recolonization of the artificial dermis. This phenomenon seems to happen faster in human beings than in animals. On day 21, the cultured epidermis was grafted on revascularized dermis: nearly the whole abdomen was grafted this way. A mesh graft was put on the breast.
Results were again not very satisfactory. 75% of the cultured epidermis was not adhering to the revascularized artificial dermis. However, in some places where the epidermal graft took, the histological study showed a beautiful aspect of reconstructed full thickness skin. On day 5, the epidermis was not perfectly attached to the dermis; on day 20, the dermal epidermal junction was complete.
These results led us to think that we could also fabricate a full thickness skin in vitro. Two techniques of cell cultures were used: Green's technique, as previously described, and Boyce's technique, using a definite medium without feeder cell layer (1). With Green's technique 5 to 6 layers of keratinocytes can be seen on day 12 on the top of a matrix in which there are no cells but a nice bundle of collagen. With Boyce's method, the same kind of aspect can be seen on day 10 (Fig. 3). Works in progress tend now to study the behaviour of such a full thickness skin prepared in vitro once it is grafted to animals of Daniere and Coll.
When this type of composite graft (Artificial Dermis + non confluent cultured keratinocytes) is put on the back of a nude mouse, after an 8-day culture, the whole graft becomes recolonized by mesenchymal cells and revascularized. On day 15 post-graft we can get a normal full skin (Fig. 4). This is the point we have reached. This kind of artificial full thickness skin, if it works on man, is still consistent with the early excision and immediate grafting concept, which can be done at the end of the first week post-burn.

Fig. 2: Excision of a deep burn on the chest, and immediate coverage by artificial dermis. Fig. 3: In vitro, culture of human keratinocytes on the extra-cellular matrix of dermis (day 10).

Early surgical treatment of extensive burns is related to two major conditions: rapidity of the procedure and quality of the result.
Rapidity means early excision: this makes it possible to avoid metabolic disorders and immunosuppression which lead to infection.
The quality of the result is due to immediate coverage of the wound in order to avoid granulation tissue responsible for hypertrophic scars and contractures.
The isolated techniques of Green, Bell or Prunieras are excellent biological techniques, but they need 3 weeks to be achieved and therefore do not allow early excisions. On the other hand, the use of epidermis alone is not satisfactory.
Reconstruction of full thickness skin is like building a house: a good foundation is early excision of the burn, the body of the house is the dermis. The roof is the epidermis. Between them the beams are constituted by the dermal/epidermal junction.
This full thickness skin is a hybrid artificial skin: the biological part is the epidermis made of living keratinocytes. The artificial part is the extra-cellular matrix that becomes also a real living tissue once it is colonized by the patient's own fibroblasts.

RÉSUMÉ. Un derme artificiel français a été développé ces dernières années pour couvrir les grands brûlés. Il est constitué de collagène humain, de glycosaminoglycannes et de chitosan. Ce dernier constituant permet une rêticulation stable et non chimique entre les fibres de cette matrice extracellulaire. Il a, en outre, une action sur l'immunité locale, sur l'hémostase et peut être utilisé comme support de cultures cellulaires. Le derme artificiel mis en place est complètemente réhabité et revascularisé en quelques jours et devient un véritable tissu vivant dont les qualités sont identiques à un derme normal. Nous présentons ici les tentatives d'épidermisation de cette matrice extracellulaire, in vivo et in vitro, en un temps ou en deux temps de manière à obtenir une peau bioartificielle totale utilisable en chirurgie plastique et chez les grands brûlés.

BIBLIOGRAPHY

1. Boyee C., Ham R.G.: Ca-regulated differentiation of normal human epidermal keratinocytes in chemically defined clonal culture and free serial culture. J. Invest. Dermatol., 9-2: 83-93, 1983.
2. Burke J.F., Yannas I., Quiriby C., Bondoc C., Jung W.K.: Successful use of a physiologically acceptable artificial skin in treatment of extensive burn injury. Annals of Surgery, 194: 413-428, 1981.
3. Echinard G, Dantzer E., Damour 0., Poinsignon F., Chabert B., Collombel C.: Use of artificial dermis for skin loss repair. 11 Trattamento delle ferite, 85-87, Monduzzi Editor, 1987.
4. Echinard C., Dantzer E., Poinsignon F. et al.: Mise au point d'un derme acellulaire, un pas vers la peau artificielle totale. Ann. Chir. Plast. Esth. 1988. In press.
5. Echinard C., Damour 0., David M.F., Vescovali C. et al.: In vivo and in vitro studies of a hybrid artificial skin. In: "Surgical Updating 1988". International College of Surgeons, Montorsi Publ. Vol. 111, Monduzzi Editore. Bologna, Italia.
6. Green H., Kehinde 0., Thomas J.: Growth of cultured human epidermal cells into multiple epithelia suitable for grafting. Proceedings of the National Academy of Sciences, 76: 5665-5666, 1979.
7. Yannas I., Burke L, Orgill D., Skrabut E.: Worend tissue can utilise a polymeric template to synthesize a functional extension of skin. Science, 215: 174-176, 1982.