Purpura Without Vessel Structural Damage

Lasers Med Sci 1998, 13:299–303 © 1998 Springer-Verlag London Limited

SHORT REPORT Purpura Without Vessel Structural Damage M.A. Trelles1, L.O. Svaasand2, W. Verkruysse3, E. Mayayo1 and M. Ve´lez1 1 Instituto Me´dico Vilafortuny, Fundacion Antoni de Gimbernat, Tarragona, Spain; 2Laser Center, Academic Medical Hospital, Amsterdam, The Netherlands; 3University of Trondheim, Dept. of Electrical Engineering & Computer Science, Norwegian Institute of Technology, Trondheim, Norway

Abstract. Experiments on an animal model for vascular lesions, the chicken comb model, have demonstrated a bluish-grey discoloration phenomenon similar to that observed immediately after pulsed dye laser treatment of port wine stains. In the model, erythrocytes and cell nuclei are found in the extravascular matrix even where there is no sign of vessel wall rupture. It is believed that the vapour pressure of the boiling blood forces erythrocytes through passages in an elastically expanding vessel wall. Keywords: Laser treatment; Port wine stain; Pulsed dye lasers; Purpura

INTRODUCTION Treatment of port wine stains (PWS) with pulsed lasers at 585 nm wavelength with 300– 450
s pulses, introduces a characteristic bluish-grey discoloration in the vascular tissue. This temporary discoloration is, however, more bluish-grey than the usual purpura caused by haemorrhage within the skin, such as that found after trauma [1]. Experiments on a widely-used animal model for vascular lesions, the chicken comb model [2], have demonstrated a discoloration phenomenon similar to that observed after pulsed dye laser treatment of port wine stains. However, the vessel walls of the comb seem to remain intact with no sign of rupture.

MATERIALS AND METHODS The comb of the chicken (Rhode Island Strain 4 months old) was used as the model. Tests on various areas of the comb were carried out on three chickens, using single 585 nm pulses, 360 ns long with a dye-laser. The energy density per pulse was 6 J/cm2. No local anaesthesia was used. Macroscopically observed changes were noted, in order to analyse the characteristics of tissue discoloration or any Correspondence to: Professor M.A. Trelles, Instituto Me´dico Vilafortuny, Fundacion Antoni de Gimbernat, E-43850 Cambrils, Tarragona, Spain. Fax: +34 9 77 79 1024.

other significant changes. Biopsy samples of the target tissue were taken immediately after laser irradiation. Processing of tissue for optical microscopy was done using routine methods. Samples were stained with haematoxylin–eosin (H&E). The evaluation of tissue characteristics was entrusted to a pathologist unfamiliar with the experiment who selected samples randomly. Evaluation criteria included vessel morphology, structure, blood deposition (inside and outside the vessels) and any visible transformation occurring in the skin layers and its structure.

RESULTS Within seconds after exposure the irradiated regions developed a bluish-grey appearance which was very similar to that seen in conventional dye-laser treatment of PWS, with 6–10 J/ cm2 fluence at 585 nm wavelength. The pulse length used in the present experiments, 360 ns, is about a factor of 1000 shorter than the commonly used pulse length of 300–450
s in the clinical situation. Histological samples taken shortly after irradiation revealed that only a few erythrocytes were left within the vessel lumen (Figs 1–4). Red blood cells and cell fragments such as cell nuclei, were found in the extravascular matrix (Fig. 5). No apparent damage to the vessel wall was observed. The contour of

300

M.A. Trelles et al.

Fig. 1. Hen comb tissue. Histological sample before laser treatment. Observe the dilated vessels. Erythrocytes appear occupying the vessel lumen. H&E×250.

Fig. 2. Hen comb tissue. Magnification of a random sample before laser treatment. Characteristic dilated vessels similar to those of PWS structure can be seen. Blood is seen in the lumen of the vessels. The erythrocytes, unlike human ones, have a nucleus. H&E×400.

Purpura Without Vessel Structural Damage

301

Fig. 3. Hen comb tissue. Histology immediately after laser irradiation using 360 ns pulse, 6 J/cm2, 585 nm dye-laser. Vessels are empty of blood and without apparent changes in the wall characteristics. H&E×250.

Fig. 4. Hen comb tissue. Magnification of a random histology sample after laser treatment. Vessels are empty of blood without apparent changes. Note (arrows) that the wall structure of the vessels is intact. H&E×400.

302

M.A. Trelles et al.

Fig. 5. Hen comb tissue after laser treatment. Some red cells and cell nuclei of red cells are located outside the vessels, although there are also red cells inside the vessel lumen. H&E×250.

the vessel wall with an intact lining of endothelial cells was the same as in control samples taken from non-irradiated regions.

DISCUSSION The thermal di#usion length, ]0.2
m, is much shorter than the thickness of the endothelial layer. (The di#usion length is approximately ]√, where  is the duration of the pulse and  is the thermal di#usivity, i.e. 1.2–1.410
7 m2/s [3,5]. Di#usion of heat during the pulse is therefore quite negligible, and only the primary site of absorption of optical energy, i.e. the blood, will be heated during the pulse. The required optical fluence for heating blood in situ from 35C to 100C and then evaporate it completely is about 9 J/cm2. The fluence is TC+H/
a, where T is the temperature rise (65C), C is the specific heat (3.6106 J/m3, 85% water), H is the latent heat of evaporation (1.9109 J/m3, 85% water) and
a is the absorption coe$cient of whole blood at 585 nm (25103 m
1) [5]. The corresponding irradiant dose from a broad beam is typically 2.5 times less than in the in situ fluence in

the upper dermis, i.e. in the range of 3–4 J/cm2 [5], because of the large backscattering of light in the dermis and epidermis. The entire blood volume in superficial vessels with diameters comparable to or less than the optical penetration depth in blood, i.e. 1/
a =40
m, will therefore undergo volatile vaporisation during the laser pulse. In larger vessels the same kind of vaporisation is expected to occur in blood within about 40
m of the endothelial layer. The resulting vapour pressure will expand the vessel walls and allow the so-called pores, the 5
m diameter channels in the vessel wall, to open up. Intact erythrocytes and cell nuclei can, therefore, be transported to the perivascular matrix without physical rupture of the vessel wall. Actual rupture of the wall will not only depend on the vapour pressure, but also on the mechanical strength and elasticity of the wall. Garden et al. [6] have demonstrated that the threshold fluence for production of the bluegrey discoloration rises with increasing pulse length in the range from 1
s to 1 ms. This observation is in accordance with expectations. A 1-
s pulse will only heat the blood itself, since the thermal di#usion length is less

Purpura Without Vessel Structural Damage

than 1
m, i.e. =0.4
m. A 1-ms pulse, which corresponds to a di#usion length larger than 10
m, i.e. =11
m, will also heat the vessel wall and perivascular tissues. Thus, the required optical dose for evaporating the blood will increase with a lengthening of the pulse. The detailed mechanism of the bluish-grey appearance is not fully understood. However, it is expected to be partly related to the Tyndall e#ect. This e#ect, which explains the bluish colour of subcutaneous veins and blue naevi, arises from the reflection of the shorter (blue) wavelengths of incident light by finely dispersed particles situated above absorbing layers of pigments such as melanin deposits or haemoglobin, in subcutaneous vessels. This blue-enhanced scattering is caused by objects smaller than the wavelength of the incident light, such as globules of protein or lipid or very small vesicles of vapour and gas. The duration of the pulses is, as discussed previously, much shorter than the thermal relaxation time of the vessel wall. The very volatile heating of blood might introduce small vapour vesicles into the extravascular matrix, and thereby enhance the reflection of blue light.

CONCLUSIONS The present findings indicate that the bluishgrey discoloration seen after pulsed laser irradiation of port wine stains is not necessarily caused by haemorrhage from ruptured vessels.

303

This discoloration is also observed in cases with no sign of rupturing of the vessel wall.

ACKNOWLEDGEMENTS Part of the histological material used in this short report has been taken from research performed by one of the authors (MAT) and K. Siomos of the Institute for Material Structure and Laser Physics of Chania (Crete), Greece, to whom we express our gratitude. The conclusions shown in this paper are part of the material of the Preparatory phase of the project ‘Comportamiento Vascular y Accio´ n Laser’ of the Antoni de Gimbernat Foundation, 1996 Cambrils, Instituto Me´ dico Vilafortuny, Spain.

REFERENCES 1. Verkruysse W, Trelles MA, Pikering JW. Defining purpura. J Am Acad Dermatol 1993; 4, 28:666 2. Trelles MA, Pons JM, Mayayo E et al. A model for investigation of haemangioma elimination. Lasers Med Sci 1987; 2:243–7 3. Duck FA. Physical Properties of Tissues. London, Academic Press, 1990 4. Svaasand LO, Fiskerstrand EJ, Kopstad G et al. Therapeutic response during pulsed laser treatment of portwine stains: dependence on vessel diameter and depth in dermis. Lasers Med Sci 1995; 10:253–43 5. Svaasand LO, Norvang LT, Fiskerstrand EJ et al. Tissue parameters determining the visual appearance of normal skin and port-wine stains. Lasers Med Sci 1995; 10:55–65 6. Garden JM, Tan OT, Kerschmann R et al. E#ect of dye laser pulse duration on selective cutaneous vascular injury. J Invest Dermatol 1986; 87:653–7

Received for publication 23 June 1997; accepted following revision 17 March 1998.