- Email: [email protected]
Dermal cellular inflammation in burns. an insight into the function of dermal microvascular anatomy
Burns 27 (2001) 433– 438 www.elsevier.com/locate/burns
Dermal cellular inflammation in burns. an insight into the function of dermal microvascular anatomy Michael P.H. Tyler a,*, Andrew M.I. Watts a, Marta E. Perry b, Anthony H.N. Roberts a, D. Angus McGrouther a a
The Stoke Mande6ille Burns and Reconstructi6e Surgery Research Trust, Plastic Surgery Department, Stoke Mande6ille NHS Trust, Aylesbury HP21 8AL, UK b Di6ision of Anatomy and Cell Biology, UMDS Guy’s Hospital, London Bridge, London SE1 9RT, UK Accepted 4 December 2000
Abstract The damage caused by thermal trauma is augmented by the subsequent inflammatory response in a similar fashion to reperfusion injury. Animal studies have demonstrated a significant role for neutrophils in this delayed damage, but little is known about the numbers of neutrophils or other leucocytes that enter human skin following burns. We have longitudinally examined profiles of leucocyte migration into five cases of human partial thickness burns in relation to continued dermal microvascular destruction during the acute post-burn period. All burn wounds had a rapid influx of neutrophils that was followed by a delayed influx of macrophages. Compared to the controls, the two superficial burns also had rapid and sustained influx of CD4 and CD8 lymphocytes via patent post capillary venules in the dermal superficial vascular plexus, whilst in the three deeper burns, in which this superficial vascular plexus was occluded, the number of lymphocytes decreased. These results suggest that the patterns of leucocyte extravasation were dependent on the initial level of vascular occlusion, indicating that the dermal microvascular anatomy plays a pivotal role in determining the composition of the extravascular inflammatory cell infiltrates. The potential importance of this finding is highlighted by the differences in wound behaviour associated with the different leucocyte profiles. © 2001 Elsevier Science Ltd and ISBI. All rights reserved. Keywords: Macrophage; Neutrophil; T-lymphocyte; Vascular patency
1. Introduction One of the primary characteristics of the skin is its role as an immunocompetent organ [1]. The complex interaction of structure and function within the dermis allows the skin to act as an innate barrier and also as an integral part of the adaptive immune responses mediated by lymphocytes. Non-specific inflammatory reactions within the skin are characterised by the activation of humoral inflammatory cascades, such as complement and the prostaglandins, followed by neutrophil migration [2]. The acquired immune system is centred around antigen recognition and a lymphocyte-mediated response to that antigen [3]. * Corresponding author. Tel.: + 44-1296-315117; fax: + 44-1296315183. E-mail address: [email protected] (M.P.H. Tyler).
The vasculature of the skin is usually described as containing two plexuses: the superficial vascular plexus and the deep vascular plexus. The superficial plexus lies just beneath the dermal –epidermal junction and supplies capillary loops within the rete ridges. At this level, the post-capillary venules are the commonest type of vessel [4]. The deep vascular plexus lies at the base of the dermis and houses larger vessels, giving branches to the adnexal structures and anastomosing with the superficial plexus above [5,6]. Peripheral blood lymphocytes circulate through the skin, performing an immunosurveillance role. A simple laceration stimulates an acute neutrophil response at the edge of the cut dermis within 2 h, followed by the arrival of monocytes [7]. Lymphocytes are not usually involved in this acute form of inflammatory response, although their presence is recognised in delayed type hypersensitivity reactions such as acute and chronic
0305-4179/01/$20.00 © 2001 Elsevier Science Ltd and ISBI. All rights reserved. PII: S 0 3 0 5 - 4 1 7 9 ( 0 0 ) 0 0 1 5 4 - 6
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
434
dermatoses, which include psoriasis [7,8]. In acute dermatoses, the cellular infiltration within the dermis is initially centred around the superficial vascular plexus, which contains the many specialised post-capillary venules required for lymphocyte trafficking. Chronicity is associated with the presence of cellular infiltrates deeper in the dermis [8]. Burn wounds have long been thought to result in non-specific inflammatory responses, similar to those associated with minor trauma [9,10], although no longitudinal studies in humans have been carried out to confirm this. Research on animal models investigating the composition of inflammatory cell types in the infiltrates in acute burn wounds has shown that neutrophils and macrophages predominate [11]. Delayed microvascular damage occurring hours after the initial burn has been observed in humans [12] and measured in many experimental models [13– 15]. The role of neutrophils in this phenomenon has been illustrated by experiments both depleting and blocking neutrophil extravasation in burn wounds [15– 18]. We have designed a temporal study to assess the status of the dermal vasculature and the composition of developing cellular infiltrates early in human burn wounds.
2. Materials and methods Five consecutive patients admitted to the Stoke Mandeville Hospital NHS Trust Burns unit were entered into this study following the specifications of the local ethical committee protocols (Table 1). Superficial areas of burns were examined clinically and by laser Doppler (Periflux PF 4000 with a dual channel recording on a 408 model probe from Perimed, Sweden) to confirm that the selected areas were of uniform thickness. These were then biopsied on admission, at 24 and 48 h post-burn. A single biopsy of normal skin was taken from a contralateral unburned site. On admission, two patients presented with superficial partial thickness burns and three with deep partial thickness burns. The material from each biopsy was divided into two parts: one was formaldehyde fixed, and the other was snap-frozen in liquid nitrogen. Routine histology was
performed to assess the microstructure of the tissues. The histological burn depth was calculated from the mean depth of patent and blocked dermal microvasculature of five sections, with three areas examined in each section. Individual inflammatory cell types were identified with immunocytochemical stains using antibodies to CD4 (CD4 lymphocytes), CD8 (CD8 lymphocytes), CD68 (monocytes and macrophages) and NP57 (neutrophil elastase), all obtained from Dako, Bucks, UK. Cell counts were made by a single observer examining 50 random high power fields (× 40 objective) per biopsy. The cell counts and histological depth assessments were repeated blindly by a second observer in 10 randomly selected sections to assess reproducibility of our histological techniques. The repeated cell counts had a bivarate Spearman rank correlation coefficient of 0.82. The differences between the two sets of cell counts were normally distributed with a mean of − 1.8 and a standard deviation of 4.98. The ‘‘limits of agreement’’ [19] between the two sets of measurements [5 and 95% confidence intervals (CI) for the mean difference] ranged between − 8.4 and 4.9 cells per high-power field. The repeated depth assessments had a bivarate Spearman rank correlation coefficient of 0.95. The differences between the two sets of cell counts were not normally distributed. The median of the differences was − 1.5%, and the Wilcoxon signed rank test was used to calculate that the ‘‘limits of agreement’’ [19] between the two sets of measurements (5 and 95% confidence intervals on the median difference) ranged between −8 and 5%.
3. Results The age, sex, total burn surface area and position of the biopsy sites are shown in Table 1. Clinically, the depths of the burns varied, two burns were superficial partial thickness, and three were deep partial thickness. The laser Doppler penetrated the dermis to a depth of between 1 and 1.5 mm. The superficial vascular plexus lay within this range, but the deep vascular
Table 1 Clinical details of the five patients studied Age (years)
Sex
Type of burn
Clinical estimation of burn depth
Site of biopsies
Percentage TBSA burned
52 28 51 25 44
F M M M F
Hot fat Flame Electrical flash Flame Flame
Superficial partial thickness Deep partial thickness Superficial partial thickness Deep partial thickness Deep partial thickness
Dorsum foot Ventral forearm Ventral forearm Anterior shoulder Ventral forearm
15 22 3 37 5
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
435
Fig. 1. The profiles of inflammatory cell infiltrates into burned dermis in the first 48 h post-burn: (a) neutrophils, (b) monocytes/macrophages, (c) CD4 lymphocytes, (d) CD8 lymphocytes.
plexus lay deeper than this in all our specimens. Initially, all laser Doppler measurements were greater than the control sites, but by 24 and 48 h, the two superficial partial thickness burns had very high doppler flux measurements indicative of patent, hyperaemic superficial vessels. However, the deeper burns at 24 and 48 h had low Doppler measurements, indicating a blocked superficial vasculature. Leucocyte influx into these burns was not uniform. All the burns presented had a rapid influx of neutrophils (Fig. 1a); the superficial burns, however, showed a reduction of this influx by 48 h, in contrast to the deeper burns whose neutrophil influx showed no signs of ending by 48 h. All burns had a delayed influx of monocytes/macrophages (Fig. 1b), but again, there were differences between the two depths of burn
with the speed and size of the monocyte/macrophage influx being delayed and reduced in the deeper burns. Finally, there were marked differences between the superficial and deep burns in the patterns of lymphocyte behaviour: the burns with a patent superficial vascular plexus had an influx of lymphocytes that was as rapid as the neutrophil influx and was sustained at 48 h. In contrast, the deeper burns with blocked superficial vascular plexus vessels had no noticeable extravasation of lymphocytes (Fig. 1c and d). The two groups of wounds behaved differently, as was observed histologically. Those with patent superficial vascular plexus and inflammatory responses, which included lymphocytes, did not deepen with time, whereas the deeper burns showed a measurable progression of dermal microvascular damage (Fig. 2).
436
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
4. Discussion The altered structure of burned skin acts as a highly potent immunogenic stimulus [20]. This observational study of progressive vascular occlusion in human partial thickness burns provides an opportunity for examining the patterns of cellular inflammation that occur when elements of the dermal vascular architecture have been destroyed. Human skin is unique, so histological studies such as this are not reproducible in animals. In this study, the differences in wound behaviour and inflammatory cell profiles did not correlate with the size of the burn, anatomical site, age or sex of the patient (Table 1), but there was a direct relationship with the depth of the burn. Little is known about the profiles of leucocyte extravasation following cutaneous burn injuries in humans, and no significant lymphocyte response has been reported previously. The stimulus for this migration
probably comes from the abnormal proteins that have been detected in burn wounds, some of which do not appear to be present in any other dermal inflammatory situations [21]. Polyacrylamide gel electrophoresis of the dermal products from burnt mice has revealed the presence of a highly toxic compound, lipid protein complex [22], though this specific ‘‘burns toxin’’ has been confused with the presence of other normal biochemical agents [23] and lipid peroxides [24]. The toxicity of lipid protein complex results from the thermally induced polymerisation of separate non-toxic precursor dermal proteins [25], and the lipid–protein complex has now been been detected intravenously in humans following burn injuries [26]. In-vitro experiments have also shown that the lipid–protein complex acts as an antigen, inducing interleukin-2 secretion and T-lymphocyte activation [27], and it can also activate monocytes to secrete interleukin-1[28]. The presence of the lipid– protein complex within burnt dermis may be contribu-
Fig. 1. (Continued)
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
437
Fig. 2. Measurement of continued dermal microvascular destruction in the first 48 h post-burn: the mean depth of blocked vessels is expressed as a percentage of the total depth of the dermis.
tory to the cytokine cascade that occurs locally during the acute post-burn period [29]. Skin ‘homing’ lymphocytes access the quiescent dermis in their role of immunosurveillance through specialised post-capillary venules of the superficial vascular plexus [2]. This function is enhanced during inflammatory states by the increased expression of intercellular adhesion molecule1 and E-selectin adhesion molecules on the activated vascular endothelium [30]. In the superficial burns in this study, the majority of the superficial vascular plexus remained intact, explaining the brisk extravasation of lymphocytes as a part of the immunological reaction to the altered dermal proteins. Interestingly, the acute lymphocyte and macrophage rich infiltrate, seen in the superficial partial thickness burns, closely resembles the cellular inflammatory composition of the delayed type hypersensitivity reactions as seen in some dermatoses [31]. A possible explanation for the differential migration of lymphocytes into superficial and deep dermal injuries may lie in the structural anatomy of human skin. There is no evidence, clinically or experimentally, that the nature of the inflammatory stimulus varies with the depth of the burn, so patient-management plans are made with reference to the area of burn [32,33] and not to the depth of burn. Assuming, therefore, that in all our patients, the immunogenic stimulus was similar, the deeper burns failed to produce a lymphocytic response because the specialised superficial vascular plexus was destroyed, and the remaining deep vascular plexus did not contain the specialised structures [5,6], which could support lymphocyte migration. However, the rate of neutrophil extravasation was not dependent on the patency of the superficial vascular plexus, suggesting a less specialised route for their transmigration. Burns, in common with reperfusion injuries [34], also cause a complex oxygen-derived free radical driven injury response. Free radicals are mainly produced
from local tissue hypoxia, which increases the activity of xanthine oxidases converted from xanthine dehydrogenase [35]. Normal mechanisms for free-radical scavenging are impaired by the reduction in vascular supply [36] allowing the free radical to cause endothelial and other cell damage, which results in further ischaemia and necrosis [37]. Free radicals have been implicated in local and systemic inflammatory events [38]. The local release of inflammatory mediators, including complement factors [39], as well as the chaotic production of many cytokines [40–42], plays an integral role in initiating the leucocyte response to burn injuries. The neutrophil response itself has been implicated as an important factor in the progressive endothelial damage and tissue necrosis seen in the first 48 h post-burn [15,17,18]. The novel finding in our study has shown that progression of vascular damage after a burn injury was only measurable in the three deeper burns, despite the significant neutrophil infiltration seen in all burns (Fig. 1a). Interestingly, it was the lack of lymphocyte infiltration and, to a lesser degree, the delay in monocyte/macrophage infiltration, which mirrored the observed progression in vascular damage, suggesting a role for these two cells types in neutrophil control and removal. The implications of these differential patterns of cellular inflammation on progressive microvascular occlusion in the acute post-burn period are the subject of ongoing research. However, this study does highlight the importance of the superficial vascular plexus in mediating dermal inflammation and provides an anatomical basis for differences in wound behaviour.
Acknowledgements We are most grateful for the time and patience of the five patients who entered into this study; for the help of
438
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
Mr Y. Mustafa, of UMDS, Guy’s Hospital and the Staff in the Pathology Department, Stoke Mandeville NHS Trust Hospital for all their help in the preparation and staining of the histological specimens; for the help provided by the Staff of the Burns Unit at Stoke Mandeville; and to Dr B Shine of Stoke Mandeville NHS Trust Hospital for his advice on the statistics.
References [1] Streilein JW. Skin associated lymphoid tissues (SALT): origins and functions. J Invest Dermatol 1983;80:12s – 6s. [2] Bos JD, Kaspenberg ML. The skin immune system (SIS): its cellular constituents and their interactions. Immunol Today 1986;7:235 – 40. [3] Bos JD, Kapsenberg ML. The skin immune system: progress in cutaneous biology. Immunol Today 1993;14:75 –8. [4] Yen A, Braverman IM. Ultrastructure of the human dermal microcirculation: The horizontal plexus of the papillary dermis. J Invest Dermatol 1976;66:131 –42. [5] Higgins JC, Eady RAJ. Human dermal microvasculature: I. Its segmental differentiation. Light and electron microscopic study. Br J Dermatol 1981;104:117 –29. [6] Braverman IM, Yen A. Ultrastructure of the human dermal microcirculation. III. The vessels in the mid and lower dermis and subcutaneous fat. J Invest Dermatol 1981;77:297 – 304. [7] Cotran RS, Kumar V, Robbins SL. Robbins Pathologic Basis of Disease. London: WB Saunders, 1994. [8] Ackerman BA. Histologic Diagnosis of Inflammatory Skin Diseases. London: Lea and Febiger, 1936. [9] Sevitt S. Burns Pathology and Therapeutic Applications. London: Butterworth, 1957. [10] Panke TW, McLeod CGJr. Classification of burn wound injury & mechanisms of repair. In: Anonymous, editor. Pathology of Thermal Injury: A Practical Approach. Orlando, FL: Grune & Stratton, 1986:8 – 25. [11] Winter GD. Histological aspects of burn wound healing. Burns 1973;1:191 – 6. [12] Jackson DM. The diagnosis of the depth of burning. Br J Surg 1953;40:588 – 96. [13] Robson MC, Kucan JO, Paik KI, Eriksson E. Prevention of dermal ischemia after thermal injury. Arch Surg 1978;113:621 – 5. [14] Zawacki BE. Reversal of capillary stasis and prevention of necrosis in burns. Ann Surg 1974;180:98 –102. [15] Mulligan MS, Till GO, Smith CW, et al. Role of leukocyte adhesion molecules in lung and dermal vascular injury after thermal trauma of skin. Am J Pathol 1994;144:1008 –15. [16] Mulligan MS, Yeh CG, Rudolph AR, Ward PA. Protective effects of soluble CR1 in complement- and neutrophil-mediated tissue injury. J Immunol 1992;148:1479 –85. [17] Choi M, Rabb MD, Arnaout MA, Ehrlich HP. Prevention of infiltration of leukocytes by monoclonal antibody blocks the development of progressive ischaemia in rat burns. Plast Reconst Surg 1995;96:1177 – 85. [18] Bucky LP, Vedder NB, Hong HZ, et al. Reduction of burn injury by inhibiting CD18-mediated leukocyte adherence in rabbits. Plast Reconst Surg 1994;93:1473 –80. [19] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307 – 10. [20] Allgower M, Schoenenberger GA, Sparkes BG. Burning the largest immune organ. Burns 1995;21:S7 – S47. .
[21] Feifel H, Bruchelt G, Schmidt K. Effect of constituents of burned skin and in vivo skin burning on the respiratory activity of rat liver mitochondria. Burns 1992;18:308 – 12. [22] Allgower M, Burri C, Gruber UL, Nagel G. Toxicity of burned mouse skin in relation to burn temperature. Surg Forum 1963;14:37 – 9. [23] Rocha M, Silva E, Rosenthal SR. Release of pharmacologically active substances from the rat’s skin in vivo following thermal injury. J Pharmacol Exp Ther 1961;132:110 – 6. [24] Hiramatsu M, Izawa Y, Hagahara M. Serum lipid peroxide levels of patients suffering from thermal injury. Burns 1984;11:111 – 6. [25] Schoenenberger GA, Burkhardt F, Kalberer F, Muller W, Stagtler K, Vogt P. Experimental evidence for a significant impairment of host defence for gram-negative organisms by a specific cutaneous toxin produced by severe burn injuries. Surg Gynecol Obstet 1975;141:555 – 61. [26] Sparkes BG, Monge G, Marshall S, Peters W, Allgower M, Schoenenberger GA. Plasma levels of cutaneous burn toxin and lipid peroxides in thermal injury. Burns 1990;16:118 – 22. [27] Sparkes BG. Influence of burn-induced lipid-protein complex on IL-2 secretion by PBMC in vitro. Burns 1991;17:129 – 35. [28] Schoenenberger GA. Burn toxins isolated from mouse and human skin. Monogr Allergy 1975;9:72. [29] Camp R, Jones RR, Brain S, Woollard P, Greaves M. Production of intraepidermal microabscesses by topical application of leucotriene B4. J Invest Dermatol 1984;82:202 – 4. [30] Nwariaku FE, Mileski WJ, Lightfoot E, Sikes PJ, Lipsky PE. Alterations in leukocyte adhesion molecule expression after burn injury. J Trauma 1995;39:285 – 8. [31] Waldorf HA, Walsh LJ, Schechter NM, Murphy GF. Early cellular events in evolving cutaneous delayed hypersensitivity in humans. Am J Pathol 1991;138:477 – 85. [32] Muir IKF, Barclay TL. Burns and their Treatment. London: Lloyd Luke, 1962. [33] Boswick JA. The Art and Science of Burn Care. Rockville: Aspen, 1987. [34] McCord JM. Oxygen-derived free radicals in post-ischaemic tissue injury. New Engl J Med 1985;312:159 – 63. [35] McKelvey TG, Ho¨ llwarth ME, Granger DN, Engerson TD, Landler U, Jones HP. Mechanisms of conversion of xanthine dehydrogenase to xanthine oxidase in ischaemic rat liver and kidney. Am J Physiol 1988;254:753 – 60. [36] Till GO, Guilds LS, Mahrougui M, Friedl HP, Trentz O, Ward PA. Role of xanthine oxidase in thermal injury of skin. Am J Pathol 1989;135:195 – 202. [37] Germonpre P, Reper P, Vanderkelen A. Hyperbaric oxygen therapy and piracetam decrease the early extension of deep partial-thickness burns. Burns 1996;22:468 – 73. [38] Till GO, Hatherill JR, Tourtellotte WW, Lutz MJ, Ward PA. Lipid peroxidation and acute lung injury after thermal trauma to skin — Evidence of a role for hydroxyl radical. Am J Pathol 1985;119:376 – 84. [39] Sharma VK, Agarwell DS, Satyanand SK. Profile of complement components in patients with severe burns. J Trauma 1980;20:976 – 8. [40] Cannon JG, Friedburg JS, Gelfand JA, Tompkins RG, Burke JF, Dinarello CA. Circulating interleukin-1b and tumour necrosis factor-a concentrations after burn injury in humans. Crit Care Med 1992;20:1414 – 9. [41] Endo S, Inada K, Yamada Y, et al. Plasma tumour necrosis factor-a (TNF-a) levels in patients with burns. Burns 1993;19:124 – 7. [42] Rodriguez JL, Miller CG, Garner WL, et al. Correlation of the local and systemic cytokine response with clinical outcome following trauma. J Trauma 1993;34:684 – 94.
Dermal cellular inflammation in burns. an insight into the function of dermal microvascular anatomy Michael P.H. Tyler a,*, Andrew M.I. Watts a, Marta E. Perry b, Anthony H.N. Roberts a, D. Angus McGrouther a a
The Stoke Mande6ille Burns and Reconstructi6e Surgery Research Trust, Plastic Surgery Department, Stoke Mande6ille NHS Trust, Aylesbury HP21 8AL, UK b Di6ision of Anatomy and Cell Biology, UMDS Guy’s Hospital, London Bridge, London SE1 9RT, UK Accepted 4 December 2000
Abstract The damage caused by thermal trauma is augmented by the subsequent inflammatory response in a similar fashion to reperfusion injury. Animal studies have demonstrated a significant role for neutrophils in this delayed damage, but little is known about the numbers of neutrophils or other leucocytes that enter human skin following burns. We have longitudinally examined profiles of leucocyte migration into five cases of human partial thickness burns in relation to continued dermal microvascular destruction during the acute post-burn period. All burn wounds had a rapid influx of neutrophils that was followed by a delayed influx of macrophages. Compared to the controls, the two superficial burns also had rapid and sustained influx of CD4 and CD8 lymphocytes via patent post capillary venules in the dermal superficial vascular plexus, whilst in the three deeper burns, in which this superficial vascular plexus was occluded, the number of lymphocytes decreased. These results suggest that the patterns of leucocyte extravasation were dependent on the initial level of vascular occlusion, indicating that the dermal microvascular anatomy plays a pivotal role in determining the composition of the extravascular inflammatory cell infiltrates. The potential importance of this finding is highlighted by the differences in wound behaviour associated with the different leucocyte profiles. © 2001 Elsevier Science Ltd and ISBI. All rights reserved. Keywords: Macrophage; Neutrophil; T-lymphocyte; Vascular patency
1. Introduction One of the primary characteristics of the skin is its role as an immunocompetent organ [1]. The complex interaction of structure and function within the dermis allows the skin to act as an innate barrier and also as an integral part of the adaptive immune responses mediated by lymphocytes. Non-specific inflammatory reactions within the skin are characterised by the activation of humoral inflammatory cascades, such as complement and the prostaglandins, followed by neutrophil migration [2]. The acquired immune system is centred around antigen recognition and a lymphocyte-mediated response to that antigen [3]. * Corresponding author. Tel.: + 44-1296-315117; fax: + 44-1296315183. E-mail address: [email protected] (M.P.H. Tyler).
The vasculature of the skin is usually described as containing two plexuses: the superficial vascular plexus and the deep vascular plexus. The superficial plexus lies just beneath the dermal –epidermal junction and supplies capillary loops within the rete ridges. At this level, the post-capillary venules are the commonest type of vessel [4]. The deep vascular plexus lies at the base of the dermis and houses larger vessels, giving branches to the adnexal structures and anastomosing with the superficial plexus above [5,6]. Peripheral blood lymphocytes circulate through the skin, performing an immunosurveillance role. A simple laceration stimulates an acute neutrophil response at the edge of the cut dermis within 2 h, followed by the arrival of monocytes [7]. Lymphocytes are not usually involved in this acute form of inflammatory response, although their presence is recognised in delayed type hypersensitivity reactions such as acute and chronic
0305-4179/01/$20.00 © 2001 Elsevier Science Ltd and ISBI. All rights reserved. PII: S 0 3 0 5 - 4 1 7 9 ( 0 0 ) 0 0 1 5 4 - 6
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
434
dermatoses, which include psoriasis [7,8]. In acute dermatoses, the cellular infiltration within the dermis is initially centred around the superficial vascular plexus, which contains the many specialised post-capillary venules required for lymphocyte trafficking. Chronicity is associated with the presence of cellular infiltrates deeper in the dermis [8]. Burn wounds have long been thought to result in non-specific inflammatory responses, similar to those associated with minor trauma [9,10], although no longitudinal studies in humans have been carried out to confirm this. Research on animal models investigating the composition of inflammatory cell types in the infiltrates in acute burn wounds has shown that neutrophils and macrophages predominate [11]. Delayed microvascular damage occurring hours after the initial burn has been observed in humans [12] and measured in many experimental models [13– 15]. The role of neutrophils in this phenomenon has been illustrated by experiments both depleting and blocking neutrophil extravasation in burn wounds [15– 18]. We have designed a temporal study to assess the status of the dermal vasculature and the composition of developing cellular infiltrates early in human burn wounds.
2. Materials and methods Five consecutive patients admitted to the Stoke Mandeville Hospital NHS Trust Burns unit were entered into this study following the specifications of the local ethical committee protocols (Table 1). Superficial areas of burns were examined clinically and by laser Doppler (Periflux PF 4000 with a dual channel recording on a 408 model probe from Perimed, Sweden) to confirm that the selected areas were of uniform thickness. These were then biopsied on admission, at 24 and 48 h post-burn. A single biopsy of normal skin was taken from a contralateral unburned site. On admission, two patients presented with superficial partial thickness burns and three with deep partial thickness burns. The material from each biopsy was divided into two parts: one was formaldehyde fixed, and the other was snap-frozen in liquid nitrogen. Routine histology was
performed to assess the microstructure of the tissues. The histological burn depth was calculated from the mean depth of patent and blocked dermal microvasculature of five sections, with three areas examined in each section. Individual inflammatory cell types were identified with immunocytochemical stains using antibodies to CD4 (CD4 lymphocytes), CD8 (CD8 lymphocytes), CD68 (monocytes and macrophages) and NP57 (neutrophil elastase), all obtained from Dako, Bucks, UK. Cell counts were made by a single observer examining 50 random high power fields (× 40 objective) per biopsy. The cell counts and histological depth assessments were repeated blindly by a second observer in 10 randomly selected sections to assess reproducibility of our histological techniques. The repeated cell counts had a bivarate Spearman rank correlation coefficient of 0.82. The differences between the two sets of cell counts were normally distributed with a mean of − 1.8 and a standard deviation of 4.98. The ‘‘limits of agreement’’ [19] between the two sets of measurements [5 and 95% confidence intervals (CI) for the mean difference] ranged between − 8.4 and 4.9 cells per high-power field. The repeated depth assessments had a bivarate Spearman rank correlation coefficient of 0.95. The differences between the two sets of cell counts were not normally distributed. The median of the differences was − 1.5%, and the Wilcoxon signed rank test was used to calculate that the ‘‘limits of agreement’’ [19] between the two sets of measurements (5 and 95% confidence intervals on the median difference) ranged between −8 and 5%.
3. Results The age, sex, total burn surface area and position of the biopsy sites are shown in Table 1. Clinically, the depths of the burns varied, two burns were superficial partial thickness, and three were deep partial thickness. The laser Doppler penetrated the dermis to a depth of between 1 and 1.5 mm. The superficial vascular plexus lay within this range, but the deep vascular
Table 1 Clinical details of the five patients studied Age (years)
Sex
Type of burn
Clinical estimation of burn depth
Site of biopsies
Percentage TBSA burned
52 28 51 25 44
F M M M F
Hot fat Flame Electrical flash Flame Flame
Superficial partial thickness Deep partial thickness Superficial partial thickness Deep partial thickness Deep partial thickness
Dorsum foot Ventral forearm Ventral forearm Anterior shoulder Ventral forearm
15 22 3 37 5
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
435
Fig. 1. The profiles of inflammatory cell infiltrates into burned dermis in the first 48 h post-burn: (a) neutrophils, (b) monocytes/macrophages, (c) CD4 lymphocytes, (d) CD8 lymphocytes.
plexus lay deeper than this in all our specimens. Initially, all laser Doppler measurements were greater than the control sites, but by 24 and 48 h, the two superficial partial thickness burns had very high doppler flux measurements indicative of patent, hyperaemic superficial vessels. However, the deeper burns at 24 and 48 h had low Doppler measurements, indicating a blocked superficial vasculature. Leucocyte influx into these burns was not uniform. All the burns presented had a rapid influx of neutrophils (Fig. 1a); the superficial burns, however, showed a reduction of this influx by 48 h, in contrast to the deeper burns whose neutrophil influx showed no signs of ending by 48 h. All burns had a delayed influx of monocytes/macrophages (Fig. 1b), but again, there were differences between the two depths of burn
with the speed and size of the monocyte/macrophage influx being delayed and reduced in the deeper burns. Finally, there were marked differences between the superficial and deep burns in the patterns of lymphocyte behaviour: the burns with a patent superficial vascular plexus had an influx of lymphocytes that was as rapid as the neutrophil influx and was sustained at 48 h. In contrast, the deeper burns with blocked superficial vascular plexus vessels had no noticeable extravasation of lymphocytes (Fig. 1c and d). The two groups of wounds behaved differently, as was observed histologically. Those with patent superficial vascular plexus and inflammatory responses, which included lymphocytes, did not deepen with time, whereas the deeper burns showed a measurable progression of dermal microvascular damage (Fig. 2).
436
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
4. Discussion The altered structure of burned skin acts as a highly potent immunogenic stimulus [20]. This observational study of progressive vascular occlusion in human partial thickness burns provides an opportunity for examining the patterns of cellular inflammation that occur when elements of the dermal vascular architecture have been destroyed. Human skin is unique, so histological studies such as this are not reproducible in animals. In this study, the differences in wound behaviour and inflammatory cell profiles did not correlate with the size of the burn, anatomical site, age or sex of the patient (Table 1), but there was a direct relationship with the depth of the burn. Little is known about the profiles of leucocyte extravasation following cutaneous burn injuries in humans, and no significant lymphocyte response has been reported previously. The stimulus for this migration
probably comes from the abnormal proteins that have been detected in burn wounds, some of which do not appear to be present in any other dermal inflammatory situations [21]. Polyacrylamide gel electrophoresis of the dermal products from burnt mice has revealed the presence of a highly toxic compound, lipid protein complex [22], though this specific ‘‘burns toxin’’ has been confused with the presence of other normal biochemical agents [23] and lipid peroxides [24]. The toxicity of lipid protein complex results from the thermally induced polymerisation of separate non-toxic precursor dermal proteins [25], and the lipid–protein complex has now been been detected intravenously in humans following burn injuries [26]. In-vitro experiments have also shown that the lipid–protein complex acts as an antigen, inducing interleukin-2 secretion and T-lymphocyte activation [27], and it can also activate monocytes to secrete interleukin-1[28]. The presence of the lipid– protein complex within burnt dermis may be contribu-
Fig. 1. (Continued)
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
437
Fig. 2. Measurement of continued dermal microvascular destruction in the first 48 h post-burn: the mean depth of blocked vessels is expressed as a percentage of the total depth of the dermis.
tory to the cytokine cascade that occurs locally during the acute post-burn period [29]. Skin ‘homing’ lymphocytes access the quiescent dermis in their role of immunosurveillance through specialised post-capillary venules of the superficial vascular plexus [2]. This function is enhanced during inflammatory states by the increased expression of intercellular adhesion molecule1 and E-selectin adhesion molecules on the activated vascular endothelium [30]. In the superficial burns in this study, the majority of the superficial vascular plexus remained intact, explaining the brisk extravasation of lymphocytes as a part of the immunological reaction to the altered dermal proteins. Interestingly, the acute lymphocyte and macrophage rich infiltrate, seen in the superficial partial thickness burns, closely resembles the cellular inflammatory composition of the delayed type hypersensitivity reactions as seen in some dermatoses [31]. A possible explanation for the differential migration of lymphocytes into superficial and deep dermal injuries may lie in the structural anatomy of human skin. There is no evidence, clinically or experimentally, that the nature of the inflammatory stimulus varies with the depth of the burn, so patient-management plans are made with reference to the area of burn [32,33] and not to the depth of burn. Assuming, therefore, that in all our patients, the immunogenic stimulus was similar, the deeper burns failed to produce a lymphocytic response because the specialised superficial vascular plexus was destroyed, and the remaining deep vascular plexus did not contain the specialised structures [5,6], which could support lymphocyte migration. However, the rate of neutrophil extravasation was not dependent on the patency of the superficial vascular plexus, suggesting a less specialised route for their transmigration. Burns, in common with reperfusion injuries [34], also cause a complex oxygen-derived free radical driven injury response. Free radicals are mainly produced
from local tissue hypoxia, which increases the activity of xanthine oxidases converted from xanthine dehydrogenase [35]. Normal mechanisms for free-radical scavenging are impaired by the reduction in vascular supply [36] allowing the free radical to cause endothelial and other cell damage, which results in further ischaemia and necrosis [37]. Free radicals have been implicated in local and systemic inflammatory events [38]. The local release of inflammatory mediators, including complement factors [39], as well as the chaotic production of many cytokines [40–42], plays an integral role in initiating the leucocyte response to burn injuries. The neutrophil response itself has been implicated as an important factor in the progressive endothelial damage and tissue necrosis seen in the first 48 h post-burn [15,17,18]. The novel finding in our study has shown that progression of vascular damage after a burn injury was only measurable in the three deeper burns, despite the significant neutrophil infiltration seen in all burns (Fig. 1a). Interestingly, it was the lack of lymphocyte infiltration and, to a lesser degree, the delay in monocyte/macrophage infiltration, which mirrored the observed progression in vascular damage, suggesting a role for these two cells types in neutrophil control and removal. The implications of these differential patterns of cellular inflammation on progressive microvascular occlusion in the acute post-burn period are the subject of ongoing research. However, this study does highlight the importance of the superficial vascular plexus in mediating dermal inflammation and provides an anatomical basis for differences in wound behaviour.
Acknowledgements We are most grateful for the time and patience of the five patients who entered into this study; for the help of
438
M.P.H. Tyler et al. / Burns 27 (2001) 433–438
Mr Y. Mustafa, of UMDS, Guy’s Hospital and the Staff in the Pathology Department, Stoke Mandeville NHS Trust Hospital for all their help in the preparation and staining of the histological specimens; for the help provided by the Staff of the Burns Unit at Stoke Mandeville; and to Dr B Shine of Stoke Mandeville NHS Trust Hospital for his advice on the statistics.
References [1] Streilein JW. Skin associated lymphoid tissues (SALT): origins and functions. J Invest Dermatol 1983;80:12s – 6s. [2] Bos JD, Kaspenberg ML. The skin immune system (SIS): its cellular constituents and their interactions. Immunol Today 1986;7:235 – 40. [3] Bos JD, Kapsenberg ML. The skin immune system: progress in cutaneous biology. Immunol Today 1993;14:75 –8. [4] Yen A, Braverman IM. Ultrastructure of the human dermal microcirculation: The horizontal plexus of the papillary dermis. J Invest Dermatol 1976;66:131 –42. [5] Higgins JC, Eady RAJ. Human dermal microvasculature: I. Its segmental differentiation. Light and electron microscopic study. Br J Dermatol 1981;104:117 –29. [6] Braverman IM, Yen A. Ultrastructure of the human dermal microcirculation. III. The vessels in the mid and lower dermis and subcutaneous fat. J Invest Dermatol 1981;77:297 – 304. [7] Cotran RS, Kumar V, Robbins SL. Robbins Pathologic Basis of Disease. London: WB Saunders, 1994. [8] Ackerman BA. Histologic Diagnosis of Inflammatory Skin Diseases. London: Lea and Febiger, 1936. [9] Sevitt S. Burns Pathology and Therapeutic Applications. London: Butterworth, 1957. [10] Panke TW, McLeod CGJr. Classification of burn wound injury & mechanisms of repair. In: Anonymous, editor. Pathology of Thermal Injury: A Practical Approach. Orlando, FL: Grune & Stratton, 1986:8 – 25. [11] Winter GD. Histological aspects of burn wound healing. Burns 1973;1:191 – 6. [12] Jackson DM. The diagnosis of the depth of burning. Br J Surg 1953;40:588 – 96. [13] Robson MC, Kucan JO, Paik KI, Eriksson E. Prevention of dermal ischemia after thermal injury. Arch Surg 1978;113:621 – 5. [14] Zawacki BE. Reversal of capillary stasis and prevention of necrosis in burns. Ann Surg 1974;180:98 –102. [15] Mulligan MS, Till GO, Smith CW, et al. Role of leukocyte adhesion molecules in lung and dermal vascular injury after thermal trauma of skin. Am J Pathol 1994;144:1008 –15. [16] Mulligan MS, Yeh CG, Rudolph AR, Ward PA. Protective effects of soluble CR1 in complement- and neutrophil-mediated tissue injury. J Immunol 1992;148:1479 –85. [17] Choi M, Rabb MD, Arnaout MA, Ehrlich HP. Prevention of infiltration of leukocytes by monoclonal antibody blocks the development of progressive ischaemia in rat burns. Plast Reconst Surg 1995;96:1177 – 85. [18] Bucky LP, Vedder NB, Hong HZ, et al. Reduction of burn injury by inhibiting CD18-mediated leukocyte adherence in rabbits. Plast Reconst Surg 1994;93:1473 –80. [19] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307 – 10. [20] Allgower M, Schoenenberger GA, Sparkes BG. Burning the largest immune organ. Burns 1995;21:S7 – S47. .
[21] Feifel H, Bruchelt G, Schmidt K. Effect of constituents of burned skin and in vivo skin burning on the respiratory activity of rat liver mitochondria. Burns 1992;18:308 – 12. [22] Allgower M, Burri C, Gruber UL, Nagel G. Toxicity of burned mouse skin in relation to burn temperature. Surg Forum 1963;14:37 – 9. [23] Rocha M, Silva E, Rosenthal SR. Release of pharmacologically active substances from the rat’s skin in vivo following thermal injury. J Pharmacol Exp Ther 1961;132:110 – 6. [24] Hiramatsu M, Izawa Y, Hagahara M. Serum lipid peroxide levels of patients suffering from thermal injury. Burns 1984;11:111 – 6. [25] Schoenenberger GA, Burkhardt F, Kalberer F, Muller W, Stagtler K, Vogt P. Experimental evidence for a significant impairment of host defence for gram-negative organisms by a specific cutaneous toxin produced by severe burn injuries. Surg Gynecol Obstet 1975;141:555 – 61. [26] Sparkes BG, Monge G, Marshall S, Peters W, Allgower M, Schoenenberger GA. Plasma levels of cutaneous burn toxin and lipid peroxides in thermal injury. Burns 1990;16:118 – 22. [27] Sparkes BG. Influence of burn-induced lipid-protein complex on IL-2 secretion by PBMC in vitro. Burns 1991;17:129 – 35. [28] Schoenenberger GA. Burn toxins isolated from mouse and human skin. Monogr Allergy 1975;9:72. [29] Camp R, Jones RR, Brain S, Woollard P, Greaves M. Production of intraepidermal microabscesses by topical application of leucotriene B4. J Invest Dermatol 1984;82:202 – 4. [30] Nwariaku FE, Mileski WJ, Lightfoot E, Sikes PJ, Lipsky PE. Alterations in leukocyte adhesion molecule expression after burn injury. J Trauma 1995;39:285 – 8. [31] Waldorf HA, Walsh LJ, Schechter NM, Murphy GF. Early cellular events in evolving cutaneous delayed hypersensitivity in humans. Am J Pathol 1991;138:477 – 85. [32] Muir IKF, Barclay TL. Burns and their Treatment. London: Lloyd Luke, 1962. [33] Boswick JA. The Art and Science of Burn Care. Rockville: Aspen, 1987. [34] McCord JM. Oxygen-derived free radicals in post-ischaemic tissue injury. New Engl J Med 1985;312:159 – 63. [35] McKelvey TG, Ho¨ llwarth ME, Granger DN, Engerson TD, Landler U, Jones HP. Mechanisms of conversion of xanthine dehydrogenase to xanthine oxidase in ischaemic rat liver and kidney. Am J Physiol 1988;254:753 – 60. [36] Till GO, Guilds LS, Mahrougui M, Friedl HP, Trentz O, Ward PA. Role of xanthine oxidase in thermal injury of skin. Am J Pathol 1989;135:195 – 202. [37] Germonpre P, Reper P, Vanderkelen A. Hyperbaric oxygen therapy and piracetam decrease the early extension of deep partial-thickness burns. Burns 1996;22:468 – 73. [38] Till GO, Hatherill JR, Tourtellotte WW, Lutz MJ, Ward PA. Lipid peroxidation and acute lung injury after thermal trauma to skin — Evidence of a role for hydroxyl radical. Am J Pathol 1985;119:376 – 84. [39] Sharma VK, Agarwell DS, Satyanand SK. Profile of complement components in patients with severe burns. J Trauma 1980;20:976 – 8. [40] Cannon JG, Friedburg JS, Gelfand JA, Tompkins RG, Burke JF, Dinarello CA. Circulating interleukin-1b and tumour necrosis factor-a concentrations after burn injury in humans. Crit Care Med 1992;20:1414 – 9. [41] Endo S, Inada K, Yamada Y, et al. Plasma tumour necrosis factor-a (TNF-a) levels in patients with burns. Burns 1993;19:124 – 7. [42] Rodriguez JL, Miller CG, Garner WL, et al. Correlation of the local and systemic cytokine response with clinical outcome following trauma. J Trauma 1993;34:684 – 94.