Lasers in Dermatology R. Steiner R. Kaufmann M. Landthaler O. Braun-Falco (Eds.) Lasers in Dermatology Proceedings of the International Symposium, Ulm, 26 September 1989 With 59 Figures Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Hong Kong Barcelona Budapest Professor Dr. Rudolf Steiner Institut fOr Lasertechnologien in der Medizin an der Universitat Ulm, Postfach 4066, Helmholtzstrasse 12, W-7900 Ulm, Fed. Rep. of Germany Priv.-Doz. Dr. med. Roland Kaufmann Dermatologische Klinik der Universitat Ulm, Oberer Eselsberg 40, W-7900 Ulm, Fed. Rep. of Germany Prof. Dr. med. Michael Landthaler Prof. Dr. med. Dr. h. c. mult. Otto Braun-Falco Dermatologische Klinik und Poliklinik der Ludwig-Maximilians-Universitat MOnchen, Frauenlobstrasse 9-11, W-8000 MOnchen 2, Fed. Rep. of Germany ISBN-13:978-3-642-75203-2 e-ISBN-13:978-3-642-75201-8 DO I: 10.1007/978-3-642-75201-8 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of 'translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only perrnitted under the provisions of the German Copyright Law of September 9,1965, in its current version, and a copyright fee must always be paid. Violations fall under the prosecution act of the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1991 Soft cover reprint of the hardcover 1st edititon 1991 The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. 56/3140-543210 - Printed on acid-free paper Preface Therapeutic laser applications in dermatology have become increasingly impor tant during the last ten years. There are indications such as portwine stain lesions where only the laser (argon or dye) can produce satisfying results. The other "classic" types of laser (C02 and Nd: YAG) are widely used for outpatient treat ment. Tissue removal or tissue coagulation are the two forms of eiLlJ.er thermal or ablative laser-tissue interaction. Fundamental research work has led to a com prehensive understanding of light distribution in skin and other biological tissues. Understanding the optical properties of a tissue, such as reflectance and transmit tance, is the first step towards predicting the best therapy. In addition, the laser parameters wavelength, energy and pulse duration modulate the tissue reactions, thus influencing the therapeutic result. This book is a notable summary of the state of the art of lasers in dermatol ogy. Starting with the basic theory of laser-tissue interaction and continuing with routine laser applications, this book also includes reports on new types of laser and their possible therapeutic potential. Future aspects of lasers in dermatology will concentrate mainly on pulsed laser techniques from the ultraviolet part of the spectrum to the infrared. The advantage of frequency doubling of the laser light and its effect on tissue reactions is also discussed. Physicians, and especially dermatologists, working in the field of laser applica tions will profit from this book, gaining a deeper understanding of the process of laser interaction with tissue and being therefore able to improve their techniques and methods of laser application. For newcomers this book is also the ideal intro duction to all the possible and established uses of the different types of laser for the maximum benefit of patients. UIm, R. Steiner December 1990 R. Kaufmann M. Landthaler O. Braun-Falco v Contents The Role of Skin Optics in Diagnostic and Therapeutic Uses of Lasers By S.L. Jacques (With 10 Figures) .......................... 1 Photodynamic Therapy in the Treatment of Diseases of the Skin By J.A.S. Carruth and S.R. Williams ......................... 22 The CO2 Laser in Dermatotherapy By M. Landthaler and U. Hohenleutner (With 1 Figure) ............ 26 The Argon Laser in Dermatotherapy By M. Landthaler, U. Hohenleutner, G. Donhauser, and 0. Braun-Falco (With 5 Figures) ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Dye Laser for Benign Cutaneous Vascular Lesions: Clinical and Technical Development By O.T. Tan (With 5 Figures) ............................. 60 The Neodymium YAG Laser in Dermatology By F.A. Bahmer (With 7 Figures) ........................... 73 Treatment of Deep Located Haemangiomas with the Nd:YAG Laser (1064nm) By C. Philipp, H.P. Berlien, and J. Waldschmidt (With 2 Figures) 85 Argon Laser Treatment of Port-Wine Stains and Quantitative Evaluation by Reflected Subject Color Analysis By R.A. Neumann, R.M. Knobler, and A. Lindmaier (With 3 Figures) 91 Fundamentals of Pulsed UV and Mid-infrared Laser Skin Ablation By R. Hibst and R. Kaufmann (With 9 Figures) ................. 102 The ArF Excimer Laser in Dermatology By S.L. Jacques (With 6 Figures) ........................... 116 Pulsed UV and Mid-infrared Laser Skin Ablation: Experimental and First Clinical Results By R. Kaufmann and R. Hibst (With 7 Figures) ................. 130 The Short Pulse Dye Laser in the Treatment of Port-Wille Stains By H. Strempel (With 4 Figures) ........................... 147 VII Future Aspects of Lasers in Dermatology By R. Steiner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 154 Index of Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 159 VIII The Role of Skin Optics in Diagnostic and Therapeutic Uses of Lasers S.L. Jacques Laser Biology Research Laboratory, University of Texas, M.D. Anderson Cancer Center - 17, 1515 Holcombe Blvd. Houston, TX 77030, USA 1. Introduction Light must penetrate the skin before it can exert therapeutic effect or offer diagnostic information. Laser radiation will penetrate some tissues better than others due to variation in tissue optical properties. Skin is a particularly difficult tissue to penetrate because of the strong scattering properties of the dermis. Fortunately, skin is a relatively thin tissue, and the shallow penetration of light is sufficient for strong interaction with the epidermis and upper dermis. The availability of the skin surface for topical irradiation makes the use of lasers in dermatology an attractive modality. This paper will discuss the basics of skin optics that are pertinent to proper dosimetry of laser irradiation in dermatology. Examples of dosimetry for wavelengths pertinent to photodynamic therapy and for laser therapy of portwine stain lesions at 577-nm wavelength are presented. 2. Tissue optical properties 2.1 Clinical perspective The movement of photons through a tissue is determined by the intrinsic optical properties of the tissue: the absorption coefficient, Ila, the scattering coefficient, Ils, and the anisotropy of scattering, g. In contrast to intrinsic properties, observable parameters such as reflectance (R), transmittance (T), and the internal light distribution (<») will change with geometrical factors such as tissue thickness, surface boundary conditions (eg., air/tissue versus water/tissue), and angle of incident irradiation, despite constant instrinsic optical properties. Understanding the optical properties of a tissue is the first step toward comprehensive understanding of laser dosimetry for a variety of delivery conditions. For example, consider a freshly peeled potato. Slice a thin layer from the potato, and irradiate both the thin slice and the whole potato with light. The R, T, and <» for the slice and the whole potato will differ, although both have the same optical properties. The geometrical factor of potato thickness influences the observable parameters despite constant optical properties. We deduce the optical properties from observable parameters such as Rand T under carefully controlled geometric conditions (flat thin slab). Then we use the optical Lasers in Dermatology Editors: R. Steiner· R. Kaufmann· M. Landthaler· O. Braun-Falco © Springer-Verlag Berlin, Heidelberg 1991 properties to predict laser dosimetry in terms of light (<», W/cm2) or heat (Jla<», W/cm3) for a particular geometric situation (laser delivery, tissue size and shape, boundary conditions) that is pertinent to a clinical protocol. 2.2 Absorption coefficient, /-La The absorption coefficient (Ila) is defined as the incremental loss of radiant fluence rate (<\» that occurs over a unit of path length (L): Ila = -d<»/dL. In simple terms, Ila specifies the transmission (T) without absorption over a pathlength (L), expressed as: T = exp(-llaL). The mean free path, mfp, between absorption events is 1/1la, and if L equals the mfp then transmission equals 37%. For example, a typicailla value for bloodless dermis at the yellow 577-nm wavelength is 2 cm-1, and the mfp is 1/2 cm, or 5 mm. After an ensemble of photons has traveled 5 mm, only 37% of the photons remain unabsorbed and 63% of the photons have been absorbed. Note that the pathlength L need not be in a straight line. Many scattering events may deflect each photon many times. The pathlength L refers to the total path traveled by a photon regardless of directional changes due to scattering. 2.3 Scattering coefficient, /-Ls The scattering coefficient, Ils, specifies the transmission T without scattering over a path L, expressed as: T = exp(-llsL). The mfp between scattering events is 1I lls. For example, a typicaills value for bloodless dermis at the yellow 577-nm wavelength is 200 cm-1, and the mfp is 1/200 cm, or 50 11m. Compared to the mfp between absorption events, 5 mm, the occurence of scattering events is quite frequent in the dermis for yellow light. On the average at the yellow 577-nm wavelength, there are 100 scattering events before an absorption event finally occurs. 2.4 Anisotropy of scattering, g When a photon is scattered, the direction of photon deflection must be considered. e, The anisotropy, g, describes the average angle, of deflection when a scattering event occurs. The g is defined as the average value of the cosine of the angle of deflection: g = <cose>. The definition of g is discussed more fully in the next paragraph, but first let us consider the basic importance of anisotropy. If the deflection angle is totally random = (isotropic scatterin@, g 0) in a medium, then a single scattering event will cause a photon to lose any sense of its initial direction of delivery to the medium, and the photon will participate in a random walk within the media with mean free steps of 1/1ls between e scattering events. In tissues however, the deflection angle is usually slight (eg., between 5° to 45°) and the photon continues its trajectory in the forward direction with only a small deflection. The photon will require approximately 1/(1-g) scattering events before it loses any sense of its initial direction of delivery to the tissue, and the photon will then propagate as a random walk with mean free steps of 1/(lls(1-g)). For example, yellow light at 577-nm wavelength has a g value of about 0.8 for dermis. Note that the cos(37°) equals 0.8, and 2 E isotropic scattering g g = 0, lis = 187 cm·1 ..c D.. ~ 2 3~----------~------------------~ Figure 1: Propagation of photons in human dermis. (Left) The trajectories of 25 photons at 633-nm wavelength are charted by Monte Carlo simulation, using the optical properties of bloodless dermis: Ila = 2.7 cm-1, Ils = 187 cm-1, and g = 0.82. (Right) The trajectories of 25 photons in an isotropically scattering media (g = 0) with the same Ila and Ils values. The figure illustrates how the forward-directed nature of scattering in dermis allows deep penetration of photons into dermis. The trajectories terminate when an absorption event occurs. the average equivalent e is -37°. The photon requires 1/(1-0.8) or 5 scattering events before achieving a random walk. Since the mfp between scattering events is 50 Ilm, the photon travels 250 Ilm before its propagation is truly a random walk. Each photon still has 95 (= 100 -5) scattering events, or 4.75 mm (= 5 mm -250 Ilm) of travel before there is a 63% probability of absorption. Therefore, most of the photon's life is spent in random walk. The size of the mean free steps in the photon's random walk is very important in determing the lateral spread and deep penetration of a laser beam into skin. The mean free steps for tissue equal 1/(lls(1-g», but if g equals zero the steps are only 1I lls. In our example of skin at 577-nm wavelength, the steps are 250 Ilm because g = 0.8, but if the scattering were isotropic (g = 0) then the steps would be only 50-llm steps. The forward directed nature of scattering in skin yields large equivalent steps in the photon's random walk. These larger steps allow photons to diffuse further in skin than in an isotropically scattering media of equal Ils before a given amount of photon absorption occurs. Figure 1 illustrates the propagation of 25 photons (633 nm, HeNe laser) in human dermis (g = 0.82), and for an equivalent isotropically scattering tissue (g = 0). In both cases, the absorption (Ila) and scattering (Ils) coefficients are 2.7 and 187 cm-1, respectively. Note how the isotropic scattering causes photons to remain near the point of entry, but the anisotropic scattering characteristic of dermis allows photon to propagate much deeper into tissue. For a moment, let us discuss more carefully the concept of anisotropy. When photons are scattered, the distribution of the deflection angle, e, is conventionally described by a probability distribution function, p(e), as observed in a single plane of 3 observation. The integration of p(8) over the three dimensions of spherical coordinates must equal unity to conserve energy: fIt p(8) 21tsin8d8 = 1 o (1 ) The integration of the product p(8)cos(8) yields the expectation value <cos(8», which is the definition of g: fIt g == <cos8> = p(8)cos8 21tsin8d8 o (2) In 1941, Henyey and Greenstein [1] suggested a convenient expression for p(8) which approximates the forward-directed Mie scattering which is typical for media such as galaxies or biological tissue [2] whose particles or scattering structures have a size that is similar to the photon wavelength: 2 312 (1 + gHG -29HGcos8) (3) If this PHG(8) is substituted for p(8) in eq. 2, the g value calculated by eq. 2 will equal the gHG of eq. 3. Henyey and Greenstein were rather clever. For example, COnsider the angular scattering of red light (633 nm, HeNe laser) in human dennis. Jacques et al [2] showed that on the average 90% of photons were = scattered as a Henyey-Greenstein function with gHG 0.91, and 10% were scattered = isotropically, gisotropic O. The overall g value is equal to the product gHG(90%), or 0.82 (illustrated in Fig. 1). The authors used alternative experiments (integrating sphere and collimated transmission experiments, see below) to specify an effective g value of 0.81, which agreed with the 0.82 value from angular scattering experiments. 3. Measurement of tissue optical properties 3.1 Clinical perspective Now that the optical properties are defined, let us consider how optical properties are measured. Generating such information is normally not the task of the laser clinician. Therefore, this chapter simply provides background on how optical properties are specified experimentally in the laboratory. Skin is particularly heterogeneous, and therefore optically complicated. The pigmented epidermis, the nonpigmented dermis, and the dermal blood vessels constitute a challenge to optical modeling. Heterogeneous skin optics will be discussed in section 4. At this time, we will discuss the optical properties of three component tissue types: (1) pigmented epidermis, (2) bloodless dermis, and (3) blood. 4