Rapports d'études terminées et publiées

• Hyperbaric Oxygen Therapy and Piracetam decrease the early extension of deep partial thickness burns
• Hyperbaric oxygen therapy in the treatment of burns: evaluation of systemic lipid peroxidation and activation of oxygen-radical dependent inflammatory reactions

Etudes en cours

• -

HYPERBARIC OXYGEN THERAPY AND PIRACETAM
DECREASE THE EARLY EXTENSION OF DEEP PARTIAL THICKNESS BURNS

P.Germonpré, M.D.*, P.Reper, M.D.**, A.Vanderkelen, M.D.**
*Center for Hyperbaric Oxygen Therapy
**Burn Center, Military Hospital Queen Astrid
Brussels, Belgium

Abstract

During the first 24 hours, a progression of the burn wound in histological depth or extension is often noted. This can only partially be prevented by the routinely used protocols of fluid resuscitation and burn wound dressing. In a rat model of 5% TBSA burn, Hyperbaric Oxygen Therapy (HBOT) and Piracetam were evaluated for their ability to further prevent this early deepening of the burn wound. After infliction of the burn wound, the animals were treated with an accepted basic burn wound treatment consisting of mafenide 10% solution humid dressings. They were then randomized into three groups: a control group (n=10), receiving no other treatment, a HBOT group (n=17), receiving 60 minutes of HBOT (2 atmospheres absolute) twice daily, and a Piracetam group (n=19), receiving Piracetam (200mg/kg IM) twice daily. On the third day of treatment, the entire burn wound was excised and examined histologically. We found that both HBOT and piracetam had statistically significant effects on the preservation of skin appendages (p=0.003 and 0.007, respectively), epidermal basal membrane (p=0.001 and 0.002, respectively) and on the degree of subepidermal inflammation, as measured by leucocyte infiltration (p=0.001 and 0.038, respectively). The HBOT group showed furthermore significantly less leucocyte infiltration than the Piracetam group (p=0.001). We conclude that, although the clinical importance of the small effect on skin appendage and basal membrane preservation may be questionable, the effect on subepidermal leucocyte infiltration is striking and warrants further investigation to the anti-inflammatory effects of HBOT and possibly Piracetam.

Introduction

Since Jackson (1953) (1), it is generally accepted that a cutaneous burn wound consists of three distinct zones, each with its own histophysiological caracteristics: a central zone of coagulation necrosis, surrounded by a zone of established edema and capillary stasis, which is in turn surrounded by a zone of active edema formation. Another related concept, that of a progressive capillary stasis, reversible if dehydration of the burn wound is prevented, was described by Zawacki in 1974 (2). It is, however, a very common observation in serious burn patients, that areas that seemed to be "partial thickness" burns have to be regraded the next day as "full thickness", despite optimal fluid resuscitation and burn wound coverage (3). This progressive necrosis of tissue cells is closely linked to the degree of edema formation. Edema precedes capillary stasis, which in turn provokes capillary sludge by "rouleaux" formation of the red blood cells and finally capillary thrombosis. The end-point is tissular hypoxia, ischemia and cell death. Several therapeutic measures have been proposed to prevent this cascade. Some are widely accepted, such as early wound coverage and optimal osmotic vascular filling. Other drugs and physical measures have not gained wide acceptance.

We wanted to investigate if Hyperbaric Oxygen Therapy (HBOT) or Piracetam, associated to a classically accepted burn wound treatment, are able to limit the extent of this ongoing tissue damage during the early phase of burn injury.

HBOT is an effective means of augmenting the oxygen content of arterial blood. During HBOT administration, paO2 values as high as 1800 mmHg can be obtained, resulting in the physical dissolution of considerable quantities of oxygen in the plasma (Henry’s Law). This hyperoxia induces a generalized arteriolar vasoconstriction, without impairing the oxygen delivery to the tissues. Transcutaneous and intratissular measurements of pO2 in the limbs of patients undergoing HBOT, reveals that the peripheric oxygen delivery can be more than tenfold as high as in normobaric 100% oxygen breathing (4). Piracetam is a pharmacological substance, widely used in the treatment of cerebral vascular insufficiency, but also, because of its rheological properties, in the early phase after free tissue grafts with vascular micro-anastomosis, and in the treatment of frostbite.

Material and methods

The experimental model developed by Kaufman et al. (5), has been used. It has been shown to reproducibly create a partial thickness burn wound of approx. 5% TBSA, progressing to a full thickness wound if desiccation of the wound is not prevented. It also has a very low intrinsic mortality rate. A few adaptations have been apported, resulting in the following experimental protocol:

Experimental animal: Female Wistar rats of 200 grs (±20 grs) are used. The animals are housed in standard cages, submitted to a 12 hrs light/dark cycle and fed with standard rat chow and water ad libitum. 24 hrs before the test, the animals are shaved circumferentially on the abdomen and back, and carefully depilated with anti-allergic thioglycolate cream.

Burn wound: Three aluminium cylinders, of exact dimensions (diameter 3.76 cm, height 3.78 cm; total weight 500 grs) are placed in a hot water bath (75°C) for at least 1 hour before the beginning of the tests. The animal is anaesthetized with Hypnorm(R) (Fentanyl Citrate 0.315mg/ml + Fluanisole 10mg/ml) 0.15ml/100mg body weight I.M. This dosis provides for a good anaesthesia during approx. 3.5 hrs. The rat is then taken in the left hand, with the right flank exposed. One of the cylinders is taken out of the water bath, and is placed immediately on the exposed skin for 10 seconds. No supplemental pressure is applied. The application time is measured with a chronometre. However, neither extra pressure nor a prolongation of the application time would have had a significant influence on the thermal energy transfer to the skin (5). Surface temperature measurements are performed at the surface of the cylinder (Therm 2285-2, Enginel, Brussels), before and after each application, and show temperatures of 74°±1°C before, and 69°±2°C after the application. The three cylinders are used alternatively, so that each cylinder is allowed a rewarming time of about 10 minutes. In a series of preliminary experiments, biopsies have shown that this method results in a uniform deep partial thickness burn wound, progressing, over a 24-36 Hr period, to full-thickness even when humid dressings are applied.

Burn wound care and resuscitation: Immediately after burn injury, the wound is covered with a dry sterile gauze and the animal is left for four hours without any treatment. Then, two punch-biopsies (2mm²) are taken from standardized parts of the burn wound, and a wound dressing is applied, consisting of a sterile gauze, impregnated with 10% mafenide solution. The dressing is covered with an impermeable membrane (Opsite(R)), and with two layers of adhesive elastic circular bandage (Tensoplast(R)). No fluid resuscitation is given.

Ancillary Treatment: The burn wound dressing is changed daily, under light anesthesia (Hypnorm(R) 0.05ml/100g I.M.). The animals are randomized into three groups, by an independent person, not present at the time of burn wound infliction. The control group receives no ancillary treatment. The "Piracetam group" is given 200mg/kg Piracetam I.M. (UCB Pharma, Belgium), the first injection being given 4 hrs after the burn. The "HBOT" group is submitted to hyperbaric oxygen for 60 minutes, every 8 hours the first day, every 12 hours the following 2 days. The first treatment is given 4 hrs after the burn.

Hyperbaric Oxygen Therapy: HBOT is performed in a small experimental hyperbaric chamber of 60 litres. The animals are placed with their cage inside the chamber, which is preliminary flushed with 100% oxygen for 5 minutes. Then, over a 5 minutes time period, the chamber is pressurized with 100% oxygen to 2 atmospheres absolute (ATA). After a plateau phase of 1 hour at this pressure, ensuring a constant ventilation with 100% oxygen (to prevent CO2 build-up), the chamber is depressurized over a 5 minute period. This HBOT protocol is similar to the generally accepted protocols in the treatment of burns in the human patient.

Morphologic Assessment: Each animal is weighed daily, before treatment. The punch biopsies, taken on day one, are immediately fixed in Bouin's solution for one hour, then in a 15% formaldehyde solution. At the end of the study period, the animals are terminated by ether inhalation and the burn wound is excised entirely, to the fascia. A small (1cm²) flap of unburned skin is included in the excision at the proximal end of the wound, to serve as a intra-individual control. The excised wound is fixed in Bouin's solution for one hour, then in a 15% formaldehyde solution. After fixation, three histologic preparations are made. The first one is a horizontal cut through the middle of the entire specimen and the flap of unburned skin. The other two are horizontal cuts through each of the remaining parts.

Standard haematoxylin-eosin coloration is performed, and the specimens are assessed by an independent pathologist, experienced in the appreciation of burn wound specimens, in the following way: - punch biopsies: a global histologic appreciation is made, classifying the specimens in first, second or third degree burn, according to the standard histologic criteria of epidermal necrosis, subepidermal dehiscence and skin appendage necrosis. - excised burn wound: - reference skin flap: the number of skin appendages per microscopic field (100x) is counted and is considered an indicator of the general type of skin. - burn wound: four microscopic fields (100x) are analysed in the central horizontal preparation, and two more in the upper and lower part of the burn wound (six fields in total). In each microscopic field, the following parameters are evaluated: a- number of destroyed skin appendages, criteria for destruction being: presence of round or pyknotic cells, disappearance of hair follicle roots, abnormal coloration of the cytoplasm b- degree of destruction of the epidermal cover, with estimation (in n/4) of the integrity of the basal epidermal cell layer c- degree of inflammation: per microscopic field, attribution of a score: 0 (no dermal nor hypodermal inflammation), 1 (moderate inflammation, leucocyte presence concentrating around hair follicles), or 2 (severe inflammation, abundant leucocytes present in all skin layers).

Statistical Analysis: For each animal, a global score for all 6 microscopic fields is calculated for each of the three histologic parameters. The means per group are then calculated, and the groups are compared using a one-tailed Student's t-test, with the null hypothesis being the similarity of all groups. All statistical analyses are performed on an IBM PC, by means of SPSS-PC 4.0.

Results

50 rats have entered the study. Early mortality was 4/50 (8%). No exact cause of the deaths could be given; however, anafylaxis due to the Hypnorm(R) injection could not be excluded, since one of the animals died even before the epilation could be performed. The 46 remaining animals were randomized, after the burn wound, into three groups: Control group (n=10), Piracetam group (n=19) and HBOT group (n=17).

The evolution of body weight is given in Table 1. No statistical differences are observed between the 3 groups.

Analysis of the excised normal skin flaps does not show any significant difference between the three groups (Table 2)

All punch biopsies on day one have been classified as "superficial partial thickness wound", with edema formation, subepidermal dehiscence, preservation of appendages to the superficial third of the dermis.

Percentage of destroyed skin appendages on day 3: significantly less appendages are destroyed in both experimental groups (Table 3). The difference between the Piracetam group and the HBOT group (0.64 vs 0.61) does not attain statistical significance (p=0.17).

Destruction of epithelial basal layer: here again, significantly less destruction has taken place in both experimental groups (Table 4). Also, the difference between Piracetam (0.78) and HBOT (0.73) groups is significant (p=0.02).

Degree of dermal and subepidermal leucocyte infiltration (Table 5): here, the Piracetam (0.60) and HBOT (0.47) groups are significantly different from the Control group (p=0.038 and 0.001, respectively), but i also highly significant between the experimental groups themselves (p=0.001).

Discussion

a. Burn Depth

Depth and extension of the burn wound surface are two of the most important determinants of both mortality and morbidity of the burn injury. Although, in recent years, immediate mortality - due to "burn wound shock" - has decreased significantly, owing to the more efficiënt fluid resuscitation protocols and the more aggressive surgical approach (early tangential excision), a considerable portion of the burn-related deaths now occurs during the weeks after the insult. These deaths are mostly due to multi-organ failure, respiratory insufficiency and/or systemic infection (6). As for morbidity, the risk of delayed burn wound healing and hypertrophic scar formation is directly dependent of the depth of the burn wound and hence its chances of re-epithelialising spontaneously. The longer the duration of the healing process, the higher the risk of wound infection, further compromising proper burn wound healing (7, 8, 9).

Apart from the direct cellular death by heat-induced denaturation of proteins, a delayed and progressive necrosis is observed in the zones around the primary burn injury. A number of factors have been associated with this progression of cellular necrosis.

Zawacki (2) has described a progressive capillary stasis in second degree burn wounds, and was able to distinguish two phases:

• 0-4 hours: period of edema formation, progressive capillary stasis
• 4-24 hours: stabilisation of edema and capillary stasis.

By preventing dessiccation of the burn wound in this model, a reversal of capillary stasis could be observed after 24 hours, up to the epidermal layers. In more severe burn wound models, however, as often observed in the clinical situation, these preventive measures, even when associated with a "state of the art" vascular filling and hemodynamic resuscitation, do not succeed in reversing this capillary stasis sufficiently to permit the survival of dermal cells (10).

One of the reasons for this may be found in the sequence of events causing this capillary stasis. During the first hours, the slowing of the capillary blood stream is essentially due to a mechanical compression of the capillaries by edema formation at the tissular end. There, the direct thermal energy overload causes cell lysis and liberation of oncotic substances in the intercellular space, with attraction of fluids from the capillary vascular bed, and elevation of the capillary blood viscosity (11). This corresponds to the first phase observed by Zawacki. At a certain point, this will induce rouleaux formation of the red blood cells, progressive desaturation of their hemoglobin and progressive hypoxia in the capillaries adjacent to the initial burn injury. Consequences of this hypoxia, such as endothelial cell swelling and initiation/propagation of inflammatory reactions will increase the permeability of the capillary wall and augment the edema formation in these zones (12, 13). The causes of edema formation will thus be progressively shifting from initial extracapillary (increased oncotic pressure and hence increased afterload) to local structural defects (augmented capillary permeability and endothelial cell damage).

A key factor in both the endothelial cell damage and the initiation of the various inflammatory cascades (complement activation, activation of arachidonic acid cycle, coagulation cascade, activation of polymorphonuclear leucocytes) seems to be hypoxia-induced oxygen free radical (OFR) formation (14, 15, 16, 17).

In fact, the OFR mediated reactions taking place in the vicinity of the burn wound show striking similarities with those observed in most ischemia-reperfusion models (18). As in those, hypoxia induces OFR formation by means of at least two mechanisms: the conversion of Xanthine-Reductase (X-R) to Xanthine-Oxidase (X-O) (15, 19), and the increase of the natural Superoxide Radical spill by the reduction of the enzymes of the Electron Tranfer Chain in the mitochondria of the endothelial cell (20). The increased production of Superoxide Radical initiates an auto-aggravating process, with the leucocytes playing a key role in continuing OFR production (21, 22). This leads to increasing local tissue damage (by the deleterious action of the various OFR species on all cellular membranes), but may also induce distant pathologic changes related to (pulmonary, hepatic, renal) migration and subsequent translocation of neutrophils, or to increased production of humoral factors (TNFa, IL6...) (14, 16).

Any reduction of the hypoxic capillary and tissue damage would likely reduce not only the final extent of the local insult, but also the risks of subsequent systemic complications in a burned patient.

b. Hyperbaric Oxygen Therapy (HBOT) and Piracetam in the treatment of burns

Since 1965, the possible beneficial influence of HBOT on burn wound healing has been suggested (23). Because of a lack of randomised prospective clinical trials and financial and/or practical constraints, the therapeutic use of oxygen under pressure has not yet gained widespread acceptance, on the contrary.

There have been, however, a considerable number of experimental reports that document the effects of HBOT when used in the acute phase after the burn injury.

• a decrease of the amount of plasma extravasation (25% vs. 41%) in dogs, submitted to a 40% TBSA 3rd degree burn (24),
• an acceleration and more complete restoration of the capillary permeability (measured by the Chinese Ink infusion technique) in a rat model of 5% TBSA partial thickness burn (25),
• a preservation of the tissue ATP levels in zones adjacent to the burn wound (26),
• a decrease of edema formation and exsudation rate in and around a (5mm diameter) experimental partial thickness burn wound in a human model (27).

These possible effects would be related to

• a generalized precapillary vasocontriction, hyperoxia-induced, diminishing the blood flow through the damaged capillaries (28),
• an increase of the quantity of oxygen transported per unit of blood, by physical dissolution of oxygen in plasma (up to 5ml/100ml plasma) (29),
• an increase of the intracapillary pressure of oxygen, resulting in an increase of the pericapillary diffusion distance of oxygen (30),
• a possible increase of the plasticity of the red blood cells, diminishing the capillary sludge (31).

A number of clinical reports of the use of HBOT exist, that seem to confirm these experimental findings. Notably, a reduction in resuscitation fluid requirements, in the needs for surgical interventions, in duration of hospital stay and in total hospitalisation costs, has been noted in patient groups, comparable in age, type of burn, and TBSA burned (32). These ongoing studies, of a more economical nature, seem to gain importance. They are however subject to criticism because of their retrospective nature and a lack of formal randomisation.

For Piracetam, no animal or human studies are available regarding its use in burns treatment, some being "en route". However, the rheological properties of this widely used drug, together with the complete absence of noticeable side effects, even at extremetly high dosages, warranted its formal evaluation for this field of application.

c. Experimental Setup

Although the influence of HBOT as a therapeutic measure "on its own" has been demonstrated in the experimental setting, few animal studies have been done to evaluate the supplemental benefit of HBOT to a classical burn wound treatment protocol, in preventing the extension of the burn injury.

Therefore, our protocol was designed to resemble a "realistic" burn patient treatment scenario:

• a delay of 3 to 4 hours before initiation of advanced burn wound management
• daily wound dressing changes, with application of an antimicrobial agent
• utilisation of a HBOT protocol that is currently accepted and employed in the ancillary treatment of burned patients (33)
• utilisation of a currently recommended high dosage of Piracetam (200mg/kg b.i.d. IM)

As antimicrobial agent, we have chosen mafenide hydrochloride. Although silver sulfadiazine 1% cream is more commonly used in burn centers throughout the world, its greasy component is incompatible with external high pressure oxygen exposure (risk of explosion). Mafenide, although not commercially available as an aqueous solution, can be obtained in powdery form, and is in our burn center commonly used in a 10% solution, as in the commercially available cream. This unusual pharmacological presentation does not, to our view, affect the validity of the results of our study. Severely burned patients need, in our opinion, to be treated in large multiplace hyperbaric chambers, compressed with air and equipped with advanced intensive care monitoring and life support apparatus. In this "intensive care" hyperbaric chambers, greasy wound dressings are not prohibited since the external environment of the patient consists solely of compressed air, the high pressure oxygen being breathed or administered via an isolated breathing circuit.

For this study, no hematological or serological parameters have been studied, for the burn wound inflicted was small (5% TBSA) and would not routinely need important intravenous fluid resuscitation. Also, the risk of distant complications is, in this type of injury, virtually non-existant. Lung tissue biopsies, taken from animals of preliminary groups after completion of the study period, showed no gross pathologic alterations (data not shown) in either group. Any serologic alterations would be likely to be either unmeasurable, or of no clinical significance in this burn wound model.

d. Results

HBOT and Piracetam were both able to decrease the amount of (epi)dermal cellular destruction, as well as the degree of inflammatory reaction (dermal leucocyte infiltration), when applied early after the burn infliction. This represents an added benefit when compared to a "classical" burn wound treatment only. Some reservations have to be made, however, as to the interpretation of these results.

• It is not known how big a fraction of skin appendages needs to be preserved to ensure a spontaneous burn wound healing. The difference in appendage destruction is significant but small. Whether this will result in faster healing of the burn wound, is impossible to appreciate from this study.
• The same remark can be made for the percentage of destruction of the surface epithelium. This destruction is, obviously, mainly induced by the direct thermal energy directed to these cells. The fact that HBOT or Piracetam, either via an increased availability of oxygen or by some other mechanism, can preserve a bigger portion of these cells, is interesting, but here again, the difference is very small, and the clinical importance is questionable.

The decrease in polymorphonuclear leucocyte infiltration is the most striking observation. Being admittedly only an imprecise indicator of the extent of the inflammatory reactions that are taking place, neutrophil adherence to the capillary wall, followed by rolling and translocation, is one of the earliest and most easily observable signs of their activation (17, 22, 34, 35). It seems that both HBOT and Piracetam are able to significantly reduce this leucocyte migration.

Conclusions

This study addressed the possible benefits of HBOT and Piracetam when added to the burn wound treatment - from a morphological, descriptive point of view. No information was obtained as to the final clinical consequences of these ancillary treatments, because animals were sacrified before wound healing. However, the effects of Piracetam, but more notably of HBOT on the degree of inflammatory leucocyte response are important, and warrant further investigation. A biochemical study, with a more important burn injury and subsequent serological measurements, as well of its effects on OFR production, will therefore be undertaken. In the mean time, this study somehow adds support to the claims of HBOT of having a place in the combined early management of the burn injury.

(Supported by a grant from the Brussels Capital Region Energy Department, Belgium)

Litterature

1. Jackson, D.M.: The diagnosis of the depth of burning. Br J Surg 1953, 40: 588
2. Zawacki, B.E.: The natural history of reversible burn injury. Surg Gynaecol Obstet 1974; 139: 867-872.
3. Boykin , J.V., Eriksson, E., Pittman, N.: In vivo microcirculation of a scald burn and the progression of postburn dermal ischemia. Plast Reconstr Surg 1980, 81: 741-754.
4. Mathieu, D., Wattel, F., Pellerin, Ph.: Oxygénothérapie hyperbare et chirurgie réparatrice. In: Wattel, F., Mathieu, D., Eds. Oxygénothérapie et réanimation. Paris: Masson, 1990, pp.192-198
5. Kaufman, T., Lusthaus, S.N., Sagher, U., Wexler, M.R.: Deep partial skin thickness burns: a reproducible animal model to study burn wound healing. Burns 1990; 16: 13-16.
6. Warden, G.D.: Burn shock resuscitation. World J Surg 1992; 16: 16-23.
7. Deitch, E.A., Wheelahan, T.M., Paige Rose, M., Clothier, J., Cotter, J.: Hypertrophic burn scars: analysis of variables. J Trauma 1983; 23 (10): 895-898.
8. McDonald, W.S., Deitch, E.A.: Hypertrophic skin grafts in burned patients: a prospective analysis of variables. J Trauma 1987; 27 (2): 147-150.
9. Robson, M.C., Barnett, R.A., Leitch, I.O.W., Hayward, P.G.: Prevention and treatment of postburn scars and contractures. World J Surg 1992; 16: 87-96.
10. Heimbach, D., Engrav, L., Grube, B., Marvin, J.: Burn depth: a review. World J Surg 1992; 16: 10-15.
11. Arturson, G., Mellander, S.: Acute changes in capillary filtration and diffusion in experimental burn injury. Acta Physiol Scand 1964; 62: 457.
12. Ganrot, K., Jacobson, S., Rothman, V.: Transcapillary passage of plasma proteins in experimental burns. Acta Physiol Scand 1974; 91: 297.
13. Lund, T., Onarheim, H., Reed, R.K.: Pathogenesis of edema formation in burn injuries. World J Surg 1992; 16: 2-9.
14. Till, G.O., Hatherill, J.R., Tourtelotte, W.W., Lutz, M.J., Ward, P.A.: Lipid peroxidation and acute lung injury after thermal trauma to skin. Evidence of a role for hydroxyl radical. Am J Pathol 1985; 119: 376-384.
15. Woolliscroft, J.O., Prasad, J.K., Thomson, P., Till, G.O., Fox, I.H.: Metabolic alterations in burn patients: detection of adenosine triphosphate degradation products and lipid peroxides. Burns 1990; 16 (2): 92-96.
16. Youn, Y.K., Lalonde, C., Demling, R.: The role of mediators in the response to thermal injury. World J Surg 1992; 16: 30-36.
17. Ward, P.A., Till, G.O.: The autodestructive consequences of thermal injury. J Burn Care Rehab 1985; 6 (3): 251-255.
18. Traystman, R.J., Kirsch, J.R., Koehler, R.C.: Oxygen radical mechanisms of brain injury following ischemia and reperfusion. J Appl Physiol 1991; 71 (4): 1185-1195.
19. McCord, J.M.: Oxygen-derived free radicals in postischemic tissue injury. N Eng J Med 1985; 312 (3): 159-163.
20. Flaherty, J.T., Zweier, J.L.: Role of oxygen radicals in myocardial reperfusion injury: experimental and clinical evidence. Klin Wochenschr 1991; 69: 1061-1065.
21. Ward, P.A., Mulligan, M.S.: New insights into mechanisms of oxyradical and neutrophil mediated lung injury. Klin Wochenschr 1991; 69: 1009-1011.
22. D'Alesandro, M.M., Gruber, D.F.: Quantitative and functional alterations of peripheral blood neutrophils after 10% and 30% thermal injury. J Burn Care Rehab 1990; 11 (4): 295-300.
23. Wada, J., Ikeda, T., Kamata, K., Ebuoka, M.: Oxygen hyperbaric treatment for carbon monoxyde poisoning and severe burns in coal mine gas explosion. Igakunoayumi (Japan) 1965, 54: 68-701.
24. Wells, C.H., Hilton, J.G.: Effects of hyperbaric oxygen on post-burn plasma extravasation. In: Davis, J., Hunt, T.K., Eds. Hyperbaric Oxygen Therapy. Bethesda, Md.: UHMS 1977; 259-265.
25. Miller, T.A., Korn, H.N.: Epithelial burn injury and repair. Arch Surg 1977; 112: 732-737.
26. Stewart, R.J., Yamaguchi, K.T., Cianci, P.A., Knost, P.M., Samadani, S., Mason, S.W., Roshdieh, B.B.: Effects of hyperbaric oxygen on adenosine triphosphate in thermally injured skin. Surg Forum 1988, 39: 87-90.
27. Hammarlund, C., Svedman, C., Svedman, P.: Hyperbaric oxygen treatment of healthy volunteers with u.v.-irradiated blister wounds. Burns 1991; 17 (4): 296-301.
28. Eggers, G.W., Paley, H.W., Leonard, JJ., Warren, J.V.: Hemodynamic responses to oxygen breathing in man. J Appl Physiol 1962, 17: 75-79.
29. Bauer, P., Larcan, A., Broussole, B.: Effets cardio-respiratoires de l’oxygène hyperbare. In: Wattel, F, Mathieu, D., Eds: Oxygénothérapie hyperbare et réanimation. Masson: Paris, 1990, p.11.
30. Sheffield, P.J.: Tissue oxygen measurements. In: Davis, J.C., Hunt, T.K., Eds: Problem Wounds. The role of oxygen. Elsevier: London, 1988, p.37-44.
31. Mathieu, D., Coget, J., Saulnier, F., Durocher, A., Wattel, F.: Filtrabilité érythrocytaire et oxygénothérapie hyperbare. Méd Sub Hyp 1984; 3: 100-104.
32. Cianci, P., Lueders, H.W., Lee, H., Shapiro, R.L., Sexton, J., Williams, C., Sato, R.: Adjunctive hyperbaric oxygen therapy reduced length of hospitalisation in thermal burns. J Burn Care Rehab 1989; 19:432-435.
33. Hart, G.B., Grossman, A.R.: Thermal Burns. In: Kindwall, E.P., Goldmann, R.W., Eds. Hyperbaric Medicine Procedures. Milwaukee, St.Luke's Medical Center, 1988; 98-108.
34. Zamboni, W.A., Roth, A.C., Russell, R.C., Graham, B., Suchy, H., Kucan, J.O.: Morphological analysis of the microcirculation during reperfusion of ischaemic skeletal muscle and effect of hyperbaric oxygen. Plast Reconstr Surg 1993, 91: 1110-1123.
35. Arturson, G.: The pathophysiology of severe thermal injury. JCBR 1985, 6: 129-146