British Journal of Dermatology (1976) 94, 487.
Vascular changes in human skin after ultraviolet irradiation C.A.RAMSAY* AND A.V.J.CHALLONERj St John's Hospital for Diseases of the Skin, Guy's Hospital, and flnstitute of Dermatology, London Accepted for publication 14 October 1975
SUMMARY
Blood-flow changes in human skin after ultraviolet irradiation at 250 and 300 nm have been measured by three separate methods. Those methods which measure overall blood-flow changes in the skin showed increased flow after irradiation at both wavelengths. A method which measured flow only in the superficial vessels showed a slight increase in flow after low doses of both wavelengths, but in contrast, after higher doses, this method showed a marked reduction in flow through the upper dermal vessels. This reduction in flow is probably due to stasis in these superficial vessels, perhaps secondary to vascular damage. Contrary to previous reports, blood-flow changes after irradiation at 250 and 300 nm are similar and may be mediated by identical mechanisms.
Although the mechanism responsible for the development of erythema following ultraviolet irradiation (UVR) is unknown, two main theories have been postulated. The first is that UVR has a direct dilating effect on dermal blood vessels (Finsen, 1899), the second that UVR acts by releasing a vasoactive material which diffuses to the dermal blood vessels producing dilatation (Lewis, 1927). It has been observed for many years that there are differences between the appearance of the erythema produced by different parts of the UV spectrum, and in the past the areas which have usually been chosen for study and comparison have been in the region of 250 and 300 nm, although 250 nm does not appear in terrestrial sunlight. The studies of van der Leun (1966) have provided a great deal of information on vascular changes in the skin after 250 and 300 nm UVR. The results of his studies led van der Leun to conclude that 300 nm erythema was associated with arteriolar dilatation and increased blood flow in the skin, whilst 250 nm erythema was not. Furthermore, he suggested that erythema from these two different parts of the UV spectrum is produced by two different mechanisms. Some earlier investigations (Ramsay & Cripps, 1970) indicated that vascular changes in the skin following UVR at 250 and 300 nm were identical, although these conclusions have been criticized (van der Leun, 1972). We wish to report fur±er studies on erythema induced by UVR at 250 and 300 nm using various methods of assessing changes in cutaneous blood flow. * Present address: Department of Medicine, Toronto Western Hospital, Toronto, Ontario, Canada. 487
488
C.A.Ramsay and A. V.J.Challoner MATERIALS AND METHODS
Irradiation sources We would have preferred to have used a monochromator for irradiating the skin at both wavelengths but the irradiated area obtainable from the prism monochromator available (Magnus et al., 1959) is small and would have made vascular studies difficult. Repeated exposures would have been necessary to increase the size of the irradiated area leading to possible errors because of overlapping zones. Radiation predominantly at about 250 nm was therefore provided by a low pressure or 'germicidal' mercury lamp (Philips TUV 30W) and at around 300 nm by a Westinghouse 20W fluorescent 'Sunlamp'. The lamps were aged for 100 h before use and the spectral emission was measured on a Beckman DBG recording spectrophotometer. These measurements showed that more than 98% of the emission from the 'germicidal' lamp was at the 2537 nm 'line' whilst the 'sunlamp' had an emission between 285 and 360 nm with a peak at approximately 310 nm. Absolute intensity of the lamps was measured by chemical actinometry (Hatchard & Parker, 1956). Subjects The normal skin of the back of the upper trunk was used; sixteen subjects took part (10 males aged 33-70 years and 6 females aged 26-56 years). There was no history of photosensitivity in any of the subjects, who were either normal volunteers or patients suffering from localized dermatoses, e.g. hypostatic leg ulcers, well away from the irradiated area. The minimal erythema dose (MED) for each source was determined for each subject by irradiating separate areas of skin, each 2 cm in diameter, with increasing doses. Results were read at 24 h and the lowest dose in the series which just produced erythema was taken as the MED. Assessment of vascular changes With each method, readings from the irradiated site were always compared with readings from an adjacent unirradiated site, each subject thus acted as his own control. Skin thermometry Thermistors were used, temperature being recorded directly with a Light Laboratories electric thermometer on which temperature changes of o-O5°C can be easily measured. The measured AT is the difference between skin temperatures at irradiated and adjacent unirradiated sites in each subject. Photoelectric plethysmography The method has been described in detail elsewhere (Challoner & Ramsay, 1974). AF is the difference in pulse height at irradiated and unirradiated sites in each subject. Thermal conductance The instrument used was that described by Challoner (1975). AC is the difference in thermal conductance at irradiated and unirradiated sites. Experimental procedure Eight subjects were examined with the 250 nm source and eight wi± the 300 nm source. With each source, reactions following both 2 and 5 times the MED were examined with the thermal conductance method but reactions from only 5 times the MED with skin thermometry and the photoelectric plethysmograph.
Vascular changes after ultraviolet irradiation
489
Each subject was examined before and where possible at 3, 6, 24, 48, 72 and 144 h after irradiation. In addition, with skin thermometry and photoelectric plethysmography, some subjects were examined 96 h after irradiation. Each subject acted as his own control, readings from the irradiated site were always compared with readings from an adjacent unirradiated site. Thus it would be expected that if nei±er site was irradiated the result would be zero for skin thermometry and unity for photoelectric plethysmography and thermal conductance. In order to test this and to obtain a statistical analysis of the results, eight subjects were examined at exactly the same intervals as the irradiated group. Measurements were taken from one site on the back of the trunk and compared with an adjacent site. Thus the procedure in this group was exactly the same as in the irradiated group, except that the skin was not irradiated. RESULTS
The MED with the 250 nm source varied between 37 and 74 mj/cm^ and with the 300 nm source between 160 and 320 mj/cm^. In the unirradiated group of eight subjects it is seen that the results vary around zero or unity (Figs 1-5). The results of the vascular studies were analysed statistically using the Wilcoxon Rank Test to compare irradiated and unirradiated groups. 250 nm source
Skin thermometry at 5 MED. A T increased 3 h after irradiation (Fig. i), reached a maximum at 24 h and then slowly fell. The difference between the irradiated and control groups was statistically significant at all the time intervals with Photoelectric plethysmography at 5 MED. Pulse height increased 6 h after irradiation (Fig. 2) and AF reached a maximum between 24 and 48 h (P
Vascular changes in human skin after ultraviolet irradiation C.A.RAMSAY* AND A.V.J.CHALLONERj St John's Hospital for Diseases of the Skin, Guy's Hospital, and flnstitute of Dermatology, London Accepted for publication 14 October 1975
SUMMARY
Blood-flow changes in human skin after ultraviolet irradiation at 250 and 300 nm have been measured by three separate methods. Those methods which measure overall blood-flow changes in the skin showed increased flow after irradiation at both wavelengths. A method which measured flow only in the superficial vessels showed a slight increase in flow after low doses of both wavelengths, but in contrast, after higher doses, this method showed a marked reduction in flow through the upper dermal vessels. This reduction in flow is probably due to stasis in these superficial vessels, perhaps secondary to vascular damage. Contrary to previous reports, blood-flow changes after irradiation at 250 and 300 nm are similar and may be mediated by identical mechanisms.
Although the mechanism responsible for the development of erythema following ultraviolet irradiation (UVR) is unknown, two main theories have been postulated. The first is that UVR has a direct dilating effect on dermal blood vessels (Finsen, 1899), the second that UVR acts by releasing a vasoactive material which diffuses to the dermal blood vessels producing dilatation (Lewis, 1927). It has been observed for many years that there are differences between the appearance of the erythema produced by different parts of the UV spectrum, and in the past the areas which have usually been chosen for study and comparison have been in the region of 250 and 300 nm, although 250 nm does not appear in terrestrial sunlight. The studies of van der Leun (1966) have provided a great deal of information on vascular changes in the skin after 250 and 300 nm UVR. The results of his studies led van der Leun to conclude that 300 nm erythema was associated with arteriolar dilatation and increased blood flow in the skin, whilst 250 nm erythema was not. Furthermore, he suggested that erythema from these two different parts of the UV spectrum is produced by two different mechanisms. Some earlier investigations (Ramsay & Cripps, 1970) indicated that vascular changes in the skin following UVR at 250 and 300 nm were identical, although these conclusions have been criticized (van der Leun, 1972). We wish to report fur±er studies on erythema induced by UVR at 250 and 300 nm using various methods of assessing changes in cutaneous blood flow. * Present address: Department of Medicine, Toronto Western Hospital, Toronto, Ontario, Canada. 487
488
C.A.Ramsay and A. V.J.Challoner MATERIALS AND METHODS
Irradiation sources We would have preferred to have used a monochromator for irradiating the skin at both wavelengths but the irradiated area obtainable from the prism monochromator available (Magnus et al., 1959) is small and would have made vascular studies difficult. Repeated exposures would have been necessary to increase the size of the irradiated area leading to possible errors because of overlapping zones. Radiation predominantly at about 250 nm was therefore provided by a low pressure or 'germicidal' mercury lamp (Philips TUV 30W) and at around 300 nm by a Westinghouse 20W fluorescent 'Sunlamp'. The lamps were aged for 100 h before use and the spectral emission was measured on a Beckman DBG recording spectrophotometer. These measurements showed that more than 98% of the emission from the 'germicidal' lamp was at the 2537 nm 'line' whilst the 'sunlamp' had an emission between 285 and 360 nm with a peak at approximately 310 nm. Absolute intensity of the lamps was measured by chemical actinometry (Hatchard & Parker, 1956). Subjects The normal skin of the back of the upper trunk was used; sixteen subjects took part (10 males aged 33-70 years and 6 females aged 26-56 years). There was no history of photosensitivity in any of the subjects, who were either normal volunteers or patients suffering from localized dermatoses, e.g. hypostatic leg ulcers, well away from the irradiated area. The minimal erythema dose (MED) for each source was determined for each subject by irradiating separate areas of skin, each 2 cm in diameter, with increasing doses. Results were read at 24 h and the lowest dose in the series which just produced erythema was taken as the MED. Assessment of vascular changes With each method, readings from the irradiated site were always compared with readings from an adjacent unirradiated site, each subject thus acted as his own control. Skin thermometry Thermistors were used, temperature being recorded directly with a Light Laboratories electric thermometer on which temperature changes of o-O5°C can be easily measured. The measured AT is the difference between skin temperatures at irradiated and adjacent unirradiated sites in each subject. Photoelectric plethysmography The method has been described in detail elsewhere (Challoner & Ramsay, 1974). AF is the difference in pulse height at irradiated and unirradiated sites in each subject. Thermal conductance The instrument used was that described by Challoner (1975). AC is the difference in thermal conductance at irradiated and unirradiated sites. Experimental procedure Eight subjects were examined with the 250 nm source and eight wi± the 300 nm source. With each source, reactions following both 2 and 5 times the MED were examined with the thermal conductance method but reactions from only 5 times the MED with skin thermometry and the photoelectric plethysmograph.
Vascular changes after ultraviolet irradiation
489
Each subject was examined before and where possible at 3, 6, 24, 48, 72 and 144 h after irradiation. In addition, with skin thermometry and photoelectric plethysmography, some subjects were examined 96 h after irradiation. Each subject acted as his own control, readings from the irradiated site were always compared with readings from an adjacent unirradiated site. Thus it would be expected that if nei±er site was irradiated the result would be zero for skin thermometry and unity for photoelectric plethysmography and thermal conductance. In order to test this and to obtain a statistical analysis of the results, eight subjects were examined at exactly the same intervals as the irradiated group. Measurements were taken from one site on the back of the trunk and compared with an adjacent site. Thus the procedure in this group was exactly the same as in the irradiated group, except that the skin was not irradiated. RESULTS
The MED with the 250 nm source varied between 37 and 74 mj/cm^ and with the 300 nm source between 160 and 320 mj/cm^. In the unirradiated group of eight subjects it is seen that the results vary around zero or unity (Figs 1-5). The results of the vascular studies were analysed statistically using the Wilcoxon Rank Test to compare irradiated and unirradiated groups. 250 nm source
Skin thermometry at 5 MED. A T increased 3 h after irradiation (Fig. i), reached a maximum at 24 h and then slowly fell. The difference between the irradiated and control groups was statistically significant at all the time intervals with Photoelectric plethysmography at 5 MED. Pulse height increased 6 h after irradiation (Fig. 2) and AF reached a maximum between 24 and 48 h (P
