- Email: [email protected]
Thermographic assessment of burns using a nonpermeable membrane as wound covering
Bums (1991) 17, (2), 117-122
Printed in Great Britain
117
Thermographic assessment of burns using a nonpermeable membrane as wound covering R. P. Cole, P. G. Shakespeare, H. G. Chissell and S. G. Jones” Wessex Centre for Plastic and Maxillofacial
Surgery, Odstock
Infrared fhewnography is a useful technique for the investigation of disorders which a&f skin blood flow. The damage to skin blood vessels caused by thermal injury is a major determinant of the capacity of fhe wound to heal. 7hewnographic assessment of this damage has been found to correlate with the healing time of bum wounds. However, fhe applicafion of fhewnography to the assessemenf of burns for early surgery has been limited because of the dificulfies involved in correctingfor cooling arfefacfs resulting from the effect of evaporative wafer loss (EWLJ at the wound surface. A wafer impermeable polyvinylchloride Flm (sold in fhe USA as Saran Wrap, in Ausfralia as Glad Wrap and in the UK as Clingfilm (CF),,was investigated as a wound coveting fo avoid this cooling effect. If was found that the CFabolished the coolingeflecfofEWL and did not significantly interfere with the measurement of surface temperature. This material provides a solution to the problems of fhewnographic examination of wounds such as burns where damage to the skin surface allows exudation or EWZ to occur.
Introduction Infrared thermography has been widely used for the investigation of rheumatoid diseases, breast cancer, vasospastic disorders and metabolic diseases. In all these conditions, the disease process leads to abnormalities in local circulation. The major factor controlling skin surface temperature is blood flow, so that measurement of skin temperature permits some investigation of the effects of underlying disease processes. Thermography has also been shown to have potential for the investigation of bum depth (Mladick et al., 1966; Watson and Vasilescu 1972; Hackett, 1974). However, the surface of the wound cools due to evaporative water loss (EWL). This cooling has been recognized as one of the main problems with early investigation of burn injuries (Anselmo and Zawacki, 1977). The EWL of a partial thickness burn is approximately 20 times that of normal skin (Lamke et al., 1977). This increased water loss may occur because of alterations in the lipid content of burned skin (Jelenko et al., 1968). The EWL from burned skin of any depth is further increased if the blisters are removed (Anselmo and Zawacki, X977), a common practice in the treatment of wounds. ‘Present address: Medical Laser Unit, Heriot-Watt carton, Edinburgh, UK. 0 1991 Butterworth-Heinemann 0305-4179/91/020117-06
Ltd
University, Ric-
Hospital, Salisbury, UK
The problem of EWL has been dealt with in previous studies by allowing a dry eschar to form on the bum surface, a process which takes up to 1 week. With the present trend towards early surgery of bums, however, it is necessary to assess the wound within 48 h of injury so that surgery can be performed within 3-5 days after injury. Another approach to the problem of effects of EWL on surface temperature has been to spray the wounds with water before examination in order to ‘standardize’ the surface conditions (Newman et al., 1981). Although this is an interesting way of addressing the problem it appeared to the present investigators that a more appropriate approach to the problem of examining bum wounds at an early stage after injury would be to use a method which would abolish the effect of EWL to produce cooling of the surface. The most effective way to attain this goal would be to use a covering over the wound which abolished evaporation from the wound surface. For infrared thermography to be of use in planning early surgery for bums, a wound covering is required which is impermeable to water but which does not affect the measured surface temperature in any other way. A PVC food wrapping film (Saran Wrap, Glad Wrap, Clingfilm, (CF)), which is widely used as a temporary wound covering, is a candidate for this use. In this study the properties of CF relevant to this potential use are investigated, in particular its apparent transparency, its effect on surface emissivity and its insulating effects on the surface. The properties were investigated both in vivo and in vitro and the effects of the material on the measured surface temperature of bum injuries were also investigated.
Materials and methods Physical effects of CF in vitro The effect of CF on measured surface temperature was investigated in a series of experiments. Surface temperature measurements were made using an Agema Systems 782 Thermal Scanner linked to a TC-SO0 thermal computer. A standard heat source consisting of a polished metal waterbath (an electric kettle) with three different surfaces was used. These surfaces were, a matt black stripe, a thin shaving of normal skin and a thin shaving of burned skin. Measurements were made of the temperature of all three
Bums (1991) Vol. 17/No. 2
118
surfaces with and without CF applied. First, the effect of CF upon surface temperature of the standard heat source was investigated. Measurements were made of the temperature of the matt black surface of the water-bath at internal temperatures between 30°C and 90°C. Images were recorded of this surface with and without CF applied. In addition, images of the black stripe were recorded with the CF over the camera lens at each temperature. The camera was kept at a distance of 0.5 m from the water-bath and viewed the surface at an angle approximating closely to 90”. Secondly, the effect of CF on the measured temperature of the surface of normal and burned skin was investigated. The thin shavings of both burned and normal skin were harvested from a child undergoing surgery for burns sustained 3 weeks previously. These shavings were applied to the polished surface of the water-bath alongside the matt black area. Previous calculations had shown the polished surface to have a low emissivity (approximately 0.25-0.3) assuming the matt black stripe to have an emissivity of 1.0. Thus the skin slices were thermographically visible as areas of high temperature against the apparent low temperature background of the low emissivity polished metal surface. Images of the skin slices uncovered and covered with CF were recorded. Images were recorded up to 23 min after the application of CF to the surface of the skin shavings. The temperature of the water-bath was maintained at 4045°C throughout. The temperature of the skin surface was recorded relative to the matt black surface.
Effect of CF on burned and normal skin temperature in vivo Fourteen burned patients and one normal subject were investigated, with and without CF applied to the surface of the skin. The normal control was studied at temperatures of 22.5” and 29°C. The hands and back of the normal subject were imaged both with and without CF applied to the surface of the skin. The CF was applied in intimate contact with the skin to minimize air pockets and surface irregularities. The 14 patients with burns had a total of 24 injured areas. The mean age of these patients was 33 years (range 1-74 years) and the mean body surface area burned was 12 per cent (range 2-41 per cent). Areas studied included hands, arms, legs and trunk. Bums were classified retrospectively as superficial if healing was complete within 2-3 weeks of injury, deep dermal if healing was incomplete within 3 weeks of injury, and full skin thickness if they had undergone excision and grafting within the first week after injury. The burned area was first imaged without CF in place and then 2 min after application of CF to the surface. Preliminary observations established that no further change in temperature occurred after the CF had been in place for longer than 2 min. For the normal subject, thermography was performed at environmental temperatures of 22.5”C and 29°C with negligible air movement. Humidity was not controlled but was in the region of 45 per cent relative humidity. The subject was clad in trousers, was bare to the waist and seated with elbows flexed to 90”. Conditions for the observation of burned patients were less well controlled. Cubicle temperatures varied between 20 and 25°C. Clothing for the subjects was variable but in general consisted of light cotton pyjamas or gowns. Where necessary the burns were imaged with the patient supine.
Table I. Effect of CF on the temperature of a high emissivity surface Surface temperature (W 305.7 319.4 325.4 329.9 335.8 353.3 361.2
+CFon surface
+ CF over scanner lens
(“K)
(OK)
305.1 318.2 323.6 328.3 333.8 350.5 358.7
304.8 318.0 323.3 327.8 333.2 349.4 357.7
Apparent transparency 0.991 0.987 0.981 0.981 0.977 0.967 0.971
Transparency is calculated as ‘photon transparency’, the ratio of the cube of the surface temperatures in the presence and absence of CF over the scanner lens. Temperatures are expressed as absolute temperatures on the Kelvin scale (OK).
Use of CF to facilitate thermographic prediction of healing time of burns A further 16 patients with a total of 29 burned areas were studied thermographically using a CF covering on both the wound and adjacent normal skin. The conditions for observation were the same as specified in the section above. The mean temperatures of the bum wound and of an area of adjacent skin were measured thermographically and the difference between the two was determined for each area injured. Bum depth was diagnosed retrospectively as described in the section above. Sixteen of the areas were classed as superficial, five as deep dermal and eight as full skin thickness in depth.
Results Physical properties of CF in vitro The measurements made using the standard heat source with three different surfaces are shown in Tables I-III. In the first experiment surface temperature measurements of the black stripe were made without CF, then with CF over the surface and finally with CF over the scanner lens remote from the black surface. The presence of CF over the scanner lens decreased the measured surface temperature of the black stripe owing to its incomplete transparency to infrared radiation. The reduction in temperature was used to estimate the transparency of the CF to the infrared radiation detected by the scanner. The intensity of the radiation emitted by a body is proportional to the fourth power of its absolute temperature*. The ratio of the fourth powers of the temperatures of the surface measured directly by the scanner and with CF interposed between the surface and the scanner lens provides an estimate of the transparency of the CF to infrared radiation detected by the scanner (3.6-5.5 pm). However, this method does not provide an accurate measurement of the absorption coefficient of the CF as the All objects at a temperature greater than absolute zero (- 273°C) emit radiation. The total amount of radiation emitted is described by the Stefan-Boltzmann equation and is proportional to the fourth power of the absolute temperature of the body. The wavelength of the radiation emitted is also determined by the object’s temperature. In the case of the human body the wavelength at which maximum intensity of radiation is emitted is approximately 9.5 pm, in the infrared. At this wavelength human skin behaves as a perfect radiator (it has a high emissivity and behaves as a ‘black body’). As a result skin temperature may be accurately and conveniently measured and mapped over extended areas using a thermographic camera.
119
Cole et al.: Thermographic assessment of burns Table II. Effect of CF on the surface temperature (“C) of normal skin in vitro 1 min after CF applied
Normal skin Black surface
5 min after CF applied
-CF (‘C)
+CF
Change (“C)
-CF (“Cl
+CF (“C)
Change
(“0
41.6 42.6
40.8 41.3
-0.8 -1.3
42.1 42.6
41.4 41.6
-0.7 -1.0
(“C)
CF was used to cover the surface of the skin and the black stripe.
Table III. Effect of CF on the surface temperature (“C) of burned skin in vitro 1 min after CF applied
Burned skin Black surface
5 min after CF applied
-CF
+CF
Change
-CF
+CF
Change
(“C)
07
(‘C)
(“C)
07
(“C)
36.2 39.9
39.1 39.0
+ 2.9 -0.9
38.7 39.8
39.1 39.3
+ 0.4 -0.5
CF was used to cover the surface of the burned skin and the black stripe,
scanner measures over an appreciable bandwidth and there is a variation in the numbers of photons emitted by the surface at the different temperatures at which the measurements are made. A different measure of the transparency can be obtained by considering the ratio of the third powers of the measured temperatures. This ratio gives an estimate based on the change in photon numbers associated with the increases in temperature over the range of the measurements. These results are presented in Table 1. The results indicate that the CF is more than 90 per cent transparent to infrared radiation over the range of wavelengths detected by the scanner. This estimate of 90 per cent or greater transparency is supported by the infrared absorption spectrum for CF and that of the polythene bags used routinely for hand bum dressings, and in which hands are examined using the scanner (Cole et al., 1990). These spectra are shown in &we Ia and lb. The transmittance of both materials between 3.6 and 5.5 pm is approximately 90 per cent. It should be noted that at the peak emission wavelength for normal skin at normal surface temperatures (approx 9.5 pm), the CF shows considerable absorption of energy, while the polythene bag shows a low absorption of energy. This observation suggests that the imaging of burned hands inside the poiythene bag dressings will not cause degradation of the thermal image because of infrared transmittance problems (Cole et al., 1990). Temperature measurements were then made with CF on the surface of the black stripe instead of over the scanner lens. The decrease in temperature was smaller when the CF was applied directly to the surface. These results are shown in Table 1. The effect of transparency of the CF should be independent of the placement of the CF, whether on the surface or over the lens. It appears that the presence of CF on the surface has a small apparent warming effect. This is unlikely to be due to an increase in emissivity and is most probably caused by an insulating effect at the surface of the water-bath. This effect, however, is very small (Table I). These results indicate that placement of CF on a high emissivity surface is unlikely to produce significant changes in the measured temperature of that surface. In the second experiment, when shavings of normal and burned skin were placed alongside the matt black stripe, different effects of CF were seen. When normal skin was
examined it was found that the effect of the CF was similar to its effect on the surface temperature of the high emissivity black stripe (Table II). After 5 min under CF the surface temperatures of the normal skin and the black stripe were within 0.2”C. When burned skin was examined alongside the black stripe, without CF cover, the burned skin was found to have a much lower surface temperature (some 3.7”C lower). When CF was applied to the surface of the burned skin the temperature of its surface became almost identical to the temperature of the matt black stripe and the normal skin (O.l”C difference) (Table III). Serial readings of the temperatures of the surfaces were taken over a period of 23 min. During this time the CF-covered burned skin maintained its temperature very close to that of the black stripe. In contrast the uncovered burned skin had become dessicated after this time and its apparent temperature had risen by 2.3”C, approaching that of the covered burned skin. When the CF was removed from the previously covered burned skin after 23 min, its temperature fell by l.S”C immediately. These results demonstrate that the cooling effect of EWL from the surface of the burned skin is abolished by the application of CF to its surface. Effect of CF on normal and burned skin in vivo The effect of CF on the temperature of normal hands and back skin in vivo was investigated at environmental temperatures of 22.5”C and 29°C (Table IV). The CF caused a slight decrease in the surface temperature of the skin. This decrease was similar at both environmental temperatures. After 8 min the nett effect of CF was negligible, particularly at an environmental temperature of 22.5”C. This suggests that the transparency effect and the insulation effect which had been demonstrated in vitro tended to cancel each other out in vivo. When CF was applied to the injured skin of burned patients, it increased the surface temperature of the covered area of injured skin compared to the same area of uncovered injured skin (Table V). This increase occurred in injured skin with all depths of burn. The increase in temperature caused by the presence of CF on the surface gives a measure of evaporative loss from the bum wound. Full skin thickness bums showed a smaller evaporative water loss than did
Bums (1991) Vol. 17/No. 2
120
s
40,
2 2 :: 5 $ E ir
20
0 2.5
a
4.5
a.5
6.5 Wavelength
of i.r
10.5 radiation
12.5
14.5
12.5
b
(urn)
Wavelength
of i.r.
radiation
14.5
(vm)
Figure 1. The transparency of films used to cover wounds for thermographic examination. a, The transparency of CF; b, the transparency
of polythene bags used in thermography spectrophotometer.
of burned hands. In both cases the spectrum was obtained using a Perkin-Elmer 1700 i.r.
Table IV. Surface temperature measurements (“C) of normal skin in vivo Environmental
Hand CF CF Back CF CF
temp. 22.5X
Environmental
temp. 29°C
-CF
+CF
+CF
(“C)
Change (‘C)
-CF
(‘C)
(“C)
(“C)
Change (‘C)
(dorsal surface) appt. 2 min appl. 8 min
31 .I 32.1
30.7 32.0
-0.4 -0.1
33.9 33.6
33.6 33.4
-0.3 -0.2
appl. 2 min appl. 8 min
32.1 31.9
31.6 31.7
-0.5 -0.2
33.9 33.9
33.3 33.5
-0.6 -0.4
CF was applied to the surface of the skin.
Table V. Effect of covering burned skin in vivo with CF Burn depth
Areas
Table VI. Mean temperature difference between burned and normal skin
Temp. change Areas
Temp. difference (“C)
16 5 8
+1.19*0.97 -1.40f1.17 - 2.21 f 1 .16
07 Burn depth Superficial Deep dermal Full thickness
15 5 4
+1.77fl.03 +1.96&0.77 +0.95+0.72
Burns were examined within 48 h of injury. The figures represent the change in average temperature (“C) when the burn was covered with CF. Maximum change was observed after 3 min covering with CF. Subsequently the average temperature did not increase.
superficial and deep partial skin thickness bums. There was substantial overlap between the values observed between these two latter depths of bum. Use of CF to facilitate themographic prediction of healing time The differences between burned skin covered with CF and the adjacent unburned skin is shown in Table VI Superficial bums were found to be warmer than the adjacent unburned skin. Deep dermal and full skin thickness bums were colder than adjacent unburned skin. These observations suggest that measurement of surface temperature of bumed skin covered with CF will permit those bums which will not heal spontaneously within 3 weeks to be identified as a separate group.
Superficial Deep dermal Full thickness
Burns were examined within 48 h after injury. The figures represent the difference (“C) between the burned area covered in CF and adjacent unburned skin in the same area of the body.
Discussion The trend in recent years for early surgery of bums demands an accurate preoperative diagnosis of the depth of bum. The degree of bum injury depends on two variables. First, the immediate thermal damage to the epidermis and the epidermal elements within the dermis. Secondly, the progressive ischaemia resulting from damage to the dermal and subdermal blood vessels. Damage to the skin matrix may also be involved. It has been shown that a full skin thickness bum can be successfully used as a donor site for a split thickness skin graft if this is harvested within a few hours of the bum injury (Brauer and Spira, 1966). After this, vascular damage leads to loss of the epidermal elements within the dermis that had survived the original thermal insult. The
Cole et al.: Thermographic
121
assessment of burns
condition of the skin blood vessels is therefore of importance in determining the severity of injury, the depth of the bum and the likelihood of spontaneous healing. One result of damage to the blood vessels in the skin will be a decrease or cessation in superficial skin blood flow, which will result in a decreased surface temperature. The most appropriate method of assessing temperature changes over a large area of skin is by thermography. The use of thermography in the assessment of skin surface temperature after bum injury is complicated by the cooling effect of evaporative water loss (EWL) (Anselmo and Zawacki, 1977). It has been shown that the EWL in g/mz/h from open granulating wounds is 214f 9, from partial thickness bums 178 f 6, and from full skin thickness bum wounds 143 f 5 (Lamke et al., 1977). This compares with a value of 8.5 f 0.5 for normal skin. Previous studies have shown thermography to be helpful in the assessment of bum depth (Mladick et al., 1966; Watson and Vasilescu, 1972; Hackett, 1974) and these studies have addressed the problem of EWL by waiting for up to I week for a ‘dry’ eschar to form on the wound surface. The present study has shown that EWL may interfere with surface temperature measurements for up to 3 weeks from the time of injury. A similar demonstration of the increased loss of water by evaporation from the burned surface in vim for a period of up to 3 weeks after injury has been provided by Davies et al. (1974). In addition to the persistence of the EWL problem, and because early surgery is usually performed between 3 and 5 days following injury, a delay of 1 week defeats the objective of the early intervention. The requirement is for a wound covering which abolishes the cooling effect of EWL without otherwise significantly changing the physical characteristics of the surface. This study has demonstrated that covering with CF causes a small decrease in the measured surface temperature of normal skin. This probably occurs because CF is not completely transparent to infrared radiation, although its transparency is estimated at greater than 90 per cent. When CF is applied to the skin the surface appears slightly warmer than when the skin is viewed with the CF over the scanner lens remote from the surface of the skin. This effect is probably due to the CF acting as an insulator on the skin surface and decreasing the rate of heat loss to the surrounding environment. These two effects tend to cancel each other out in practice. These observations suggest that the physical characteristics of CF are such that it does not make a significant difference to the measured surface temperature of skin and is a suitable material for covering a bum or other wound in order to obtain a measurement of its surface temperature. The main effect of CF on the skin surface temperature has been shown to be a reduction of the EWL from skin damaged by bum injury. It was shown that cooling due to EWL from normal skin was negligible, but that EWL from burned skin may produce surface cooling of over 3°C. The clinical requirements from thermography are, first, for it to be possible to examine the bums within 48 h of injury and, secondly, for it to help with the identification of deep partial skin thickness wounds which will not heal spontaneously within 3 weeks. These bums would benefit from early surgery in order to reduce hospital stay and, probably, to improve long-term functional outcome. Identification of deep dermal bums is a difficult clinical problem,
whereas superficial and full skin thickness bums are often clearly identifiable on clinical examination. Although Mladick et al. (1966) and Watson and Vasilescu
(1972) found thermography to be helpful in examining bums, they did not dea! with the identification of critical deep partial skin thickness wounds. Hackett (1974) successfully addressed this problem but his technique did not allow thermography to be used within 48 h after injury. It should also be noted that, although the s-day or so wait for a ‘dry’ surface to form on the bum wound will relieve the problem of cooling due to EWL, it will also add an extra unknown factor to the problem of interpreting the measurements. A variable thickness of crust will form on the surface of the wound. This will change the thermal conductivity characteristics of the surface and so introduce an uncertainty as to whether the measured temperature genuinely reflects the temperature of the bum wound surface. This uncertainty is not present in the method reported in the present study which clearly demonstrates the effectiveness of using CF as a covering for the wounds in order to produce ‘standard conditions for temperature measurement either by thermography or any other method. The study further demonstrates that superficial wounds are warmer than the surrounding unburned skin. It is possible that if further more detailed studies confirm the preliminary results reported here, any bum colder than the surrounding skin should be considered as a candidate for early surgery.
Acknowledgements We acknowledge the interest and encouragement in relation to this project from Mr L. F. A. Rossi, Director of the Wessex Regional Bums Unit. The assistance of the Bums Unit Staff with the collection of data and images is also acknowledged. We are grateful to the Chemistry Division of the Chemical Defence establishment, Porton Down who performed the i.r. transparency spectra. The advice and assistance of Dr Francis Ring, Clinical Measurement Department, Royal National Hospital for Rheumatic Diseases, Bath, in connection with the design and presentation of the work is gratefully acknowledged.
References Anselmo V. J. and Zawacki B. E. (1977) Effect of evaporative water loss on thermographic assessment of bum depth. Radiology
123,331. Brauer R. 0. and Spira M. (1966) Full thickness bums as a source of donor graft in the pig. Phsf. Recomtr. Surg. 3 7, 21. Cole R. P., Jones S. G. and Shakespeare P. G. (1990) Thermographic assessment of hand burns. Berms 16,60. Davies J. W. L., Lamke L-O. and Liljedahl S-O. (1974) A guide to the rate of non-renal water loss from patients with burns. Br. J, Pht. Surg. 27,325. Jelenko C. (III)., Smuylan W. I. and Wheeler M. L. (1968) The role of lipids in the transmissivity of membranes. Ann. Surg. 167,
531. Hackett M. E. J. (1974) The use of thermography
in the assessment of depth of bum and blood supply of flaps with preliminary reports on its use in Dupuytren’s contracture and treatment of varicose ulcers. Br. 1. Plast. Swg. 2 7,311. Lamke L-O., Nilsson G. E. and Reithner H. L. (1977) The evaporative water loss from burns and the water vapour permeability of grafts and artificial membranes used in the treatment of bums. BLOTS3, 159.
122
Bums (1991)Vol. 17iNo.2
Mladick R., Georgiade N. and Theme F. (1966)A clinical evaluation of the use of thermography in determining the depth of bum injury. Pkzst.Reumsfr. Surg. 38,512. Newman P., Pollock M., Reid W. H. et al. (1981) A practical technique for the thermographic examination of bum depth: a preliminary report. Burns 8,59. Watson A. C. H. and Vasilescu C. T. (1972) Thermography in plastic surgery. 1. R. Co/l. Sttrg. Minb. 17,‘247.
Paper accepted 15 October
1990.
C~~~ondence should be addressed fo: Dr P. G. Shakespeare, Lting Bum Research Laboratory, Odstock Hospital, Salisbury, Wilt&e, SP2 8BJ, UK.
1st European Conference on Advances in Wound Management Conference on Advances in Wound Management will take place on 4-6 September 1991 at the University of Cardii, Wales - a multidisciplinary conference aimed at all healthcare professionals concerned with Wound Management.
The 1st European
Call for Papers Papers are invited for concurrent
sessions - any subject is welcome but priority will be
given to the following provisional topics: 0 0 0 0
Experimental Wound Healing Prevalence Surveys of Wounds Research Methodology Nutritional Aspects ??Infection 0 Specific Dressings for Specific Wounds 0 Management of Complex Wounds 0 Leg Ulcers 0 Pressure Sores 0 Acute Trauma
Deadline for papers is May 31 1991. Delegate Registrations Registrations are now being taken for the conference and an agenda is available on request. For further details and a registration form call the Conference Office on 071 836 6633 or write to the First European Conference on Advances in Wound Management, Macmillan
Conferences, 4 Little Essex Street, London WC2R 3LF.
Printed in Great Britain
117
Thermographic assessment of burns using a nonpermeable membrane as wound covering R. P. Cole, P. G. Shakespeare, H. G. Chissell and S. G. Jones” Wessex Centre for Plastic and Maxillofacial
Surgery, Odstock
Infrared fhewnography is a useful technique for the investigation of disorders which a&f skin blood flow. The damage to skin blood vessels caused by thermal injury is a major determinant of the capacity of fhe wound to heal. 7hewnographic assessment of this damage has been found to correlate with the healing time of bum wounds. However, fhe applicafion of fhewnography to the assessemenf of burns for early surgery has been limited because of the dificulfies involved in correctingfor cooling arfefacfs resulting from the effect of evaporative wafer loss (EWLJ at the wound surface. A wafer impermeable polyvinylchloride Flm (sold in fhe USA as Saran Wrap, in Ausfralia as Glad Wrap and in the UK as Clingfilm (CF),,was investigated as a wound coveting fo avoid this cooling effect. If was found that the CFabolished the coolingeflecfofEWL and did not significantly interfere with the measurement of surface temperature. This material provides a solution to the problems of fhewnographic examination of wounds such as burns where damage to the skin surface allows exudation or EWZ to occur.
Introduction Infrared thermography has been widely used for the investigation of rheumatoid diseases, breast cancer, vasospastic disorders and metabolic diseases. In all these conditions, the disease process leads to abnormalities in local circulation. The major factor controlling skin surface temperature is blood flow, so that measurement of skin temperature permits some investigation of the effects of underlying disease processes. Thermography has also been shown to have potential for the investigation of bum depth (Mladick et al., 1966; Watson and Vasilescu 1972; Hackett, 1974). However, the surface of the wound cools due to evaporative water loss (EWL). This cooling has been recognized as one of the main problems with early investigation of burn injuries (Anselmo and Zawacki, 1977). The EWL of a partial thickness burn is approximately 20 times that of normal skin (Lamke et al., 1977). This increased water loss may occur because of alterations in the lipid content of burned skin (Jelenko et al., 1968). The EWL from burned skin of any depth is further increased if the blisters are removed (Anselmo and Zawacki, X977), a common practice in the treatment of wounds. ‘Present address: Medical Laser Unit, Heriot-Watt carton, Edinburgh, UK. 0 1991 Butterworth-Heinemann 0305-4179/91/020117-06
Ltd
University, Ric-
Hospital, Salisbury, UK
The problem of EWL has been dealt with in previous studies by allowing a dry eschar to form on the bum surface, a process which takes up to 1 week. With the present trend towards early surgery of bums, however, it is necessary to assess the wound within 48 h of injury so that surgery can be performed within 3-5 days after injury. Another approach to the problem of effects of EWL on surface temperature has been to spray the wounds with water before examination in order to ‘standardize’ the surface conditions (Newman et al., 1981). Although this is an interesting way of addressing the problem it appeared to the present investigators that a more appropriate approach to the problem of examining bum wounds at an early stage after injury would be to use a method which would abolish the effect of EWL to produce cooling of the surface. The most effective way to attain this goal would be to use a covering over the wound which abolished evaporation from the wound surface. For infrared thermography to be of use in planning early surgery for bums, a wound covering is required which is impermeable to water but which does not affect the measured surface temperature in any other way. A PVC food wrapping film (Saran Wrap, Glad Wrap, Clingfilm, (CF)), which is widely used as a temporary wound covering, is a candidate for this use. In this study the properties of CF relevant to this potential use are investigated, in particular its apparent transparency, its effect on surface emissivity and its insulating effects on the surface. The properties were investigated both in vivo and in vitro and the effects of the material on the measured surface temperature of bum injuries were also investigated.
Materials and methods Physical effects of CF in vitro The effect of CF on measured surface temperature was investigated in a series of experiments. Surface temperature measurements were made using an Agema Systems 782 Thermal Scanner linked to a TC-SO0 thermal computer. A standard heat source consisting of a polished metal waterbath (an electric kettle) with three different surfaces was used. These surfaces were, a matt black stripe, a thin shaving of normal skin and a thin shaving of burned skin. Measurements were made of the temperature of all three
Bums (1991) Vol. 17/No. 2
118
surfaces with and without CF applied. First, the effect of CF upon surface temperature of the standard heat source was investigated. Measurements were made of the temperature of the matt black surface of the water-bath at internal temperatures between 30°C and 90°C. Images were recorded of this surface with and without CF applied. In addition, images of the black stripe were recorded with the CF over the camera lens at each temperature. The camera was kept at a distance of 0.5 m from the water-bath and viewed the surface at an angle approximating closely to 90”. Secondly, the effect of CF on the measured temperature of the surface of normal and burned skin was investigated. The thin shavings of both burned and normal skin were harvested from a child undergoing surgery for burns sustained 3 weeks previously. These shavings were applied to the polished surface of the water-bath alongside the matt black area. Previous calculations had shown the polished surface to have a low emissivity (approximately 0.25-0.3) assuming the matt black stripe to have an emissivity of 1.0. Thus the skin slices were thermographically visible as areas of high temperature against the apparent low temperature background of the low emissivity polished metal surface. Images of the skin slices uncovered and covered with CF were recorded. Images were recorded up to 23 min after the application of CF to the surface of the skin shavings. The temperature of the water-bath was maintained at 4045°C throughout. The temperature of the skin surface was recorded relative to the matt black surface.
Effect of CF on burned and normal skin temperature in vivo Fourteen burned patients and one normal subject were investigated, with and without CF applied to the surface of the skin. The normal control was studied at temperatures of 22.5” and 29°C. The hands and back of the normal subject were imaged both with and without CF applied to the surface of the skin. The CF was applied in intimate contact with the skin to minimize air pockets and surface irregularities. The 14 patients with burns had a total of 24 injured areas. The mean age of these patients was 33 years (range 1-74 years) and the mean body surface area burned was 12 per cent (range 2-41 per cent). Areas studied included hands, arms, legs and trunk. Bums were classified retrospectively as superficial if healing was complete within 2-3 weeks of injury, deep dermal if healing was incomplete within 3 weeks of injury, and full skin thickness if they had undergone excision and grafting within the first week after injury. The burned area was first imaged without CF in place and then 2 min after application of CF to the surface. Preliminary observations established that no further change in temperature occurred after the CF had been in place for longer than 2 min. For the normal subject, thermography was performed at environmental temperatures of 22.5”C and 29°C with negligible air movement. Humidity was not controlled but was in the region of 45 per cent relative humidity. The subject was clad in trousers, was bare to the waist and seated with elbows flexed to 90”. Conditions for the observation of burned patients were less well controlled. Cubicle temperatures varied between 20 and 25°C. Clothing for the subjects was variable but in general consisted of light cotton pyjamas or gowns. Where necessary the burns were imaged with the patient supine.
Table I. Effect of CF on the temperature of a high emissivity surface Surface temperature (W 305.7 319.4 325.4 329.9 335.8 353.3 361.2
+CFon surface
+ CF over scanner lens
(“K)
(OK)
305.1 318.2 323.6 328.3 333.8 350.5 358.7
304.8 318.0 323.3 327.8 333.2 349.4 357.7
Apparent transparency 0.991 0.987 0.981 0.981 0.977 0.967 0.971
Transparency is calculated as ‘photon transparency’, the ratio of the cube of the surface temperatures in the presence and absence of CF over the scanner lens. Temperatures are expressed as absolute temperatures on the Kelvin scale (OK).
Use of CF to facilitate thermographic prediction of healing time of burns A further 16 patients with a total of 29 burned areas were studied thermographically using a CF covering on both the wound and adjacent normal skin. The conditions for observation were the same as specified in the section above. The mean temperatures of the bum wound and of an area of adjacent skin were measured thermographically and the difference between the two was determined for each area injured. Bum depth was diagnosed retrospectively as described in the section above. Sixteen of the areas were classed as superficial, five as deep dermal and eight as full skin thickness in depth.
Results Physical properties of CF in vitro The measurements made using the standard heat source with three different surfaces are shown in Tables I-III. In the first experiment surface temperature measurements of the black stripe were made without CF, then with CF over the surface and finally with CF over the scanner lens remote from the black surface. The presence of CF over the scanner lens decreased the measured surface temperature of the black stripe owing to its incomplete transparency to infrared radiation. The reduction in temperature was used to estimate the transparency of the CF to the infrared radiation detected by the scanner. The intensity of the radiation emitted by a body is proportional to the fourth power of its absolute temperature*. The ratio of the fourth powers of the temperatures of the surface measured directly by the scanner and with CF interposed between the surface and the scanner lens provides an estimate of the transparency of the CF to infrared radiation detected by the scanner (3.6-5.5 pm). However, this method does not provide an accurate measurement of the absorption coefficient of the CF as the All objects at a temperature greater than absolute zero (- 273°C) emit radiation. The total amount of radiation emitted is described by the Stefan-Boltzmann equation and is proportional to the fourth power of the absolute temperature of the body. The wavelength of the radiation emitted is also determined by the object’s temperature. In the case of the human body the wavelength at which maximum intensity of radiation is emitted is approximately 9.5 pm, in the infrared. At this wavelength human skin behaves as a perfect radiator (it has a high emissivity and behaves as a ‘black body’). As a result skin temperature may be accurately and conveniently measured and mapped over extended areas using a thermographic camera.
119
Cole et al.: Thermographic assessment of burns Table II. Effect of CF on the surface temperature (“C) of normal skin in vitro 1 min after CF applied
Normal skin Black surface
5 min after CF applied
-CF (‘C)
+CF
Change (“C)
-CF (“Cl
+CF (“C)
Change
(“0
41.6 42.6
40.8 41.3
-0.8 -1.3
42.1 42.6
41.4 41.6
-0.7 -1.0
(“C)
CF was used to cover the surface of the skin and the black stripe.
Table III. Effect of CF on the surface temperature (“C) of burned skin in vitro 1 min after CF applied
Burned skin Black surface
5 min after CF applied
-CF
+CF
Change
-CF
+CF
Change
(“C)
07
(‘C)
(“C)
07
(“C)
36.2 39.9
39.1 39.0
+ 2.9 -0.9
38.7 39.8
39.1 39.3
+ 0.4 -0.5
CF was used to cover the surface of the burned skin and the black stripe,
scanner measures over an appreciable bandwidth and there is a variation in the numbers of photons emitted by the surface at the different temperatures at which the measurements are made. A different measure of the transparency can be obtained by considering the ratio of the third powers of the measured temperatures. This ratio gives an estimate based on the change in photon numbers associated with the increases in temperature over the range of the measurements. These results are presented in Table 1. The results indicate that the CF is more than 90 per cent transparent to infrared radiation over the range of wavelengths detected by the scanner. This estimate of 90 per cent or greater transparency is supported by the infrared absorption spectrum for CF and that of the polythene bags used routinely for hand bum dressings, and in which hands are examined using the scanner (Cole et al., 1990). These spectra are shown in &we Ia and lb. The transmittance of both materials between 3.6 and 5.5 pm is approximately 90 per cent. It should be noted that at the peak emission wavelength for normal skin at normal surface temperatures (approx 9.5 pm), the CF shows considerable absorption of energy, while the polythene bag shows a low absorption of energy. This observation suggests that the imaging of burned hands inside the poiythene bag dressings will not cause degradation of the thermal image because of infrared transmittance problems (Cole et al., 1990). Temperature measurements were then made with CF on the surface of the black stripe instead of over the scanner lens. The decrease in temperature was smaller when the CF was applied directly to the surface. These results are shown in Table 1. The effect of transparency of the CF should be independent of the placement of the CF, whether on the surface or over the lens. It appears that the presence of CF on the surface has a small apparent warming effect. This is unlikely to be due to an increase in emissivity and is most probably caused by an insulating effect at the surface of the water-bath. This effect, however, is very small (Table I). These results indicate that placement of CF on a high emissivity surface is unlikely to produce significant changes in the measured temperature of that surface. In the second experiment, when shavings of normal and burned skin were placed alongside the matt black stripe, different effects of CF were seen. When normal skin was
examined it was found that the effect of the CF was similar to its effect on the surface temperature of the high emissivity black stripe (Table II). After 5 min under CF the surface temperatures of the normal skin and the black stripe were within 0.2”C. When burned skin was examined alongside the black stripe, without CF cover, the burned skin was found to have a much lower surface temperature (some 3.7”C lower). When CF was applied to the surface of the burned skin the temperature of its surface became almost identical to the temperature of the matt black stripe and the normal skin (O.l”C difference) (Table III). Serial readings of the temperatures of the surfaces were taken over a period of 23 min. During this time the CF-covered burned skin maintained its temperature very close to that of the black stripe. In contrast the uncovered burned skin had become dessicated after this time and its apparent temperature had risen by 2.3”C, approaching that of the covered burned skin. When the CF was removed from the previously covered burned skin after 23 min, its temperature fell by l.S”C immediately. These results demonstrate that the cooling effect of EWL from the surface of the burned skin is abolished by the application of CF to its surface. Effect of CF on normal and burned skin in vivo The effect of CF on the temperature of normal hands and back skin in vivo was investigated at environmental temperatures of 22.5”C and 29°C (Table IV). The CF caused a slight decrease in the surface temperature of the skin. This decrease was similar at both environmental temperatures. After 8 min the nett effect of CF was negligible, particularly at an environmental temperature of 22.5”C. This suggests that the transparency effect and the insulation effect which had been demonstrated in vitro tended to cancel each other out in vivo. When CF was applied to the injured skin of burned patients, it increased the surface temperature of the covered area of injured skin compared to the same area of uncovered injured skin (Table V). This increase occurred in injured skin with all depths of burn. The increase in temperature caused by the presence of CF on the surface gives a measure of evaporative loss from the bum wound. Full skin thickness bums showed a smaller evaporative water loss than did
Bums (1991) Vol. 17/No. 2
120
s
40,
2 2 :: 5 $ E ir
20
0 2.5
a
4.5
a.5
6.5 Wavelength
of i.r
10.5 radiation
12.5
14.5
12.5
b
(urn)
Wavelength
of i.r.
radiation
14.5
(vm)
Figure 1. The transparency of films used to cover wounds for thermographic examination. a, The transparency of CF; b, the transparency
of polythene bags used in thermography spectrophotometer.
of burned hands. In both cases the spectrum was obtained using a Perkin-Elmer 1700 i.r.
Table IV. Surface temperature measurements (“C) of normal skin in vivo Environmental
Hand CF CF Back CF CF
temp. 22.5X
Environmental
temp. 29°C
-CF
+CF
+CF
(“C)
Change (‘C)
-CF
(‘C)
(“C)
(“C)
Change (‘C)
(dorsal surface) appt. 2 min appl. 8 min
31 .I 32.1
30.7 32.0
-0.4 -0.1
33.9 33.6
33.6 33.4
-0.3 -0.2
appl. 2 min appl. 8 min
32.1 31.9
31.6 31.7
-0.5 -0.2
33.9 33.9
33.3 33.5
-0.6 -0.4
CF was applied to the surface of the skin.
Table V. Effect of covering burned skin in vivo with CF Burn depth
Areas
Table VI. Mean temperature difference between burned and normal skin
Temp. change Areas
Temp. difference (“C)
16 5 8
+1.19*0.97 -1.40f1.17 - 2.21 f 1 .16
07 Burn depth Superficial Deep dermal Full thickness
15 5 4
+1.77fl.03 +1.96&0.77 +0.95+0.72
Burns were examined within 48 h of injury. The figures represent the change in average temperature (“C) when the burn was covered with CF. Maximum change was observed after 3 min covering with CF. Subsequently the average temperature did not increase.
superficial and deep partial skin thickness bums. There was substantial overlap between the values observed between these two latter depths of bum. Use of CF to facilitate themographic prediction of healing time The differences between burned skin covered with CF and the adjacent unburned skin is shown in Table VI Superficial bums were found to be warmer than the adjacent unburned skin. Deep dermal and full skin thickness bums were colder than adjacent unburned skin. These observations suggest that measurement of surface temperature of bumed skin covered with CF will permit those bums which will not heal spontaneously within 3 weeks to be identified as a separate group.
Superficial Deep dermal Full thickness
Burns were examined within 48 h after injury. The figures represent the difference (“C) between the burned area covered in CF and adjacent unburned skin in the same area of the body.
Discussion The trend in recent years for early surgery of bums demands an accurate preoperative diagnosis of the depth of bum. The degree of bum injury depends on two variables. First, the immediate thermal damage to the epidermis and the epidermal elements within the dermis. Secondly, the progressive ischaemia resulting from damage to the dermal and subdermal blood vessels. Damage to the skin matrix may also be involved. It has been shown that a full skin thickness bum can be successfully used as a donor site for a split thickness skin graft if this is harvested within a few hours of the bum injury (Brauer and Spira, 1966). After this, vascular damage leads to loss of the epidermal elements within the dermis that had survived the original thermal insult. The
Cole et al.: Thermographic
121
assessment of burns
condition of the skin blood vessels is therefore of importance in determining the severity of injury, the depth of the bum and the likelihood of spontaneous healing. One result of damage to the blood vessels in the skin will be a decrease or cessation in superficial skin blood flow, which will result in a decreased surface temperature. The most appropriate method of assessing temperature changes over a large area of skin is by thermography. The use of thermography in the assessment of skin surface temperature after bum injury is complicated by the cooling effect of evaporative water loss (EWL) (Anselmo and Zawacki, 1977). It has been shown that the EWL in g/mz/h from open granulating wounds is 214f 9, from partial thickness bums 178 f 6, and from full skin thickness bum wounds 143 f 5 (Lamke et al., 1977). This compares with a value of 8.5 f 0.5 for normal skin. Previous studies have shown thermography to be helpful in the assessment of bum depth (Mladick et al., 1966; Watson and Vasilescu, 1972; Hackett, 1974) and these studies have addressed the problem of EWL by waiting for up to I week for a ‘dry’ eschar to form on the wound surface. The present study has shown that EWL may interfere with surface temperature measurements for up to 3 weeks from the time of injury. A similar demonstration of the increased loss of water by evaporation from the burned surface in vim for a period of up to 3 weeks after injury has been provided by Davies et al. (1974). In addition to the persistence of the EWL problem, and because early surgery is usually performed between 3 and 5 days following injury, a delay of 1 week defeats the objective of the early intervention. The requirement is for a wound covering which abolishes the cooling effect of EWL without otherwise significantly changing the physical characteristics of the surface. This study has demonstrated that covering with CF causes a small decrease in the measured surface temperature of normal skin. This probably occurs because CF is not completely transparent to infrared radiation, although its transparency is estimated at greater than 90 per cent. When CF is applied to the skin the surface appears slightly warmer than when the skin is viewed with the CF over the scanner lens remote from the surface of the skin. This effect is probably due to the CF acting as an insulator on the skin surface and decreasing the rate of heat loss to the surrounding environment. These two effects tend to cancel each other out in practice. These observations suggest that the physical characteristics of CF are such that it does not make a significant difference to the measured surface temperature of skin and is a suitable material for covering a bum or other wound in order to obtain a measurement of its surface temperature. The main effect of CF on the skin surface temperature has been shown to be a reduction of the EWL from skin damaged by bum injury. It was shown that cooling due to EWL from normal skin was negligible, but that EWL from burned skin may produce surface cooling of over 3°C. The clinical requirements from thermography are, first, for it to be possible to examine the bums within 48 h of injury and, secondly, for it to help with the identification of deep partial skin thickness wounds which will not heal spontaneously within 3 weeks. These bums would benefit from early surgery in order to reduce hospital stay and, probably, to improve long-term functional outcome. Identification of deep dermal bums is a difficult clinical problem,
whereas superficial and full skin thickness bums are often clearly identifiable on clinical examination. Although Mladick et al. (1966) and Watson and Vasilescu
(1972) found thermography to be helpful in examining bums, they did not dea! with the identification of critical deep partial skin thickness wounds. Hackett (1974) successfully addressed this problem but his technique did not allow thermography to be used within 48 h after injury. It should also be noted that, although the s-day or so wait for a ‘dry’ surface to form on the bum wound will relieve the problem of cooling due to EWL, it will also add an extra unknown factor to the problem of interpreting the measurements. A variable thickness of crust will form on the surface of the wound. This will change the thermal conductivity characteristics of the surface and so introduce an uncertainty as to whether the measured temperature genuinely reflects the temperature of the bum wound surface. This uncertainty is not present in the method reported in the present study which clearly demonstrates the effectiveness of using CF as a covering for the wounds in order to produce ‘standard conditions for temperature measurement either by thermography or any other method. The study further demonstrates that superficial wounds are warmer than the surrounding unburned skin. It is possible that if further more detailed studies confirm the preliminary results reported here, any bum colder than the surrounding skin should be considered as a candidate for early surgery.
Acknowledgements We acknowledge the interest and encouragement in relation to this project from Mr L. F. A. Rossi, Director of the Wessex Regional Bums Unit. The assistance of the Bums Unit Staff with the collection of data and images is also acknowledged. We are grateful to the Chemistry Division of the Chemical Defence establishment, Porton Down who performed the i.r. transparency spectra. The advice and assistance of Dr Francis Ring, Clinical Measurement Department, Royal National Hospital for Rheumatic Diseases, Bath, in connection with the design and presentation of the work is gratefully acknowledged.
References Anselmo V. J. and Zawacki B. E. (1977) Effect of evaporative water loss on thermographic assessment of bum depth. Radiology
123,331. Brauer R. 0. and Spira M. (1966) Full thickness bums as a source of donor graft in the pig. Phsf. Recomtr. Surg. 3 7, 21. Cole R. P., Jones S. G. and Shakespeare P. G. (1990) Thermographic assessment of hand burns. Berms 16,60. Davies J. W. L., Lamke L-O. and Liljedahl S-O. (1974) A guide to the rate of non-renal water loss from patients with burns. Br. J, Pht. Surg. 27,325. Jelenko C. (III)., Smuylan W. I. and Wheeler M. L. (1968) The role of lipids in the transmissivity of membranes. Ann. Surg. 167,
531. Hackett M. E. J. (1974) The use of thermography
in the assessment of depth of bum and blood supply of flaps with preliminary reports on its use in Dupuytren’s contracture and treatment of varicose ulcers. Br. 1. Plast. Swg. 2 7,311. Lamke L-O., Nilsson G. E. and Reithner H. L. (1977) The evaporative water loss from burns and the water vapour permeability of grafts and artificial membranes used in the treatment of bums. BLOTS3, 159.
122
Bums (1991)Vol. 17iNo.2
Mladick R., Georgiade N. and Theme F. (1966)A clinical evaluation of the use of thermography in determining the depth of bum injury. Pkzst.Reumsfr. Surg. 38,512. Newman P., Pollock M., Reid W. H. et al. (1981) A practical technique for the thermographic examination of bum depth: a preliminary report. Burns 8,59. Watson A. C. H. and Vasilescu C. T. (1972) Thermography in plastic surgery. 1. R. Co/l. Sttrg. Minb. 17,‘247.
Paper accepted 15 October
1990.
C~~~ondence should be addressed fo: Dr P. G. Shakespeare, Lting Bum Research Laboratory, Odstock Hospital, Salisbury, Wilt&e, SP2 8BJ, UK.
1st European Conference on Advances in Wound Management Conference on Advances in Wound Management will take place on 4-6 September 1991 at the University of Cardii, Wales - a multidisciplinary conference aimed at all healthcare professionals concerned with Wound Management.
The 1st European
Call for Papers Papers are invited for concurrent
sessions - any subject is welcome but priority will be
given to the following provisional topics: 0 0 0 0
Experimental Wound Healing Prevalence Surveys of Wounds Research Methodology Nutritional Aspects ??Infection 0 Specific Dressings for Specific Wounds 0 Management of Complex Wounds 0 Leg Ulcers 0 Pressure Sores 0 Acute Trauma
Deadline for papers is May 31 1991. Delegate Registrations Registrations are now being taken for the conference and an agenda is available on request. For further details and a registration form call the Conference Office on 071 836 6633 or write to the First European Conference on Advances in Wound Management, Macmillan
Conferences, 4 Little Essex Street, London WC2R 3LF.