Light

Light is a fundamental form of energy with numerous medical applications. At the quantum level, light is composed of packets of energy, known as photons. Each photon carries a discrete amount of energy. Light is also an electromagnetic wave. The electromagnetic spectrum extends from low frequency radio waves to ultra-high-energy gamma rays. The energy carried by each photon is determined by its wavelength, which for visible light (400–700 nm) corresponds to its color. Laser is an acronym for light amplification by the stimulated emission of radiation. Stimulated emission is a quantum process by which one photon can stimulate the creation of another photon, by interacting with an excited atom or molecule. Lasers work by pumping many atoms into the excited state, from which a very large amount of stimulated emission can occur. Laser light is typically monochromatic, meaning that the output is composed of a single wavelength of light. A second characteristic of lasers is coherence, meaning that all waves of light travel in phase spatially and temporally. Laser light is also highly collimated, which allows the laser beam to travel long distances without divergence, and to be focused to a spot about equal to its own wavelength. These properties of lasers allow for unique forms of in vivo imaging, such as confocal microscopy and optical coherence tomography.

Lasers are also capable of producing extremely intense, short pulses of light. In dermatology and ophthalmology, pulsed lasers have become mainstream tools for precise surgery and target-selective treatments. Prior to 1983, lasers in dermatology were used primarily for non-specific tissue destruction. With the description of the theory of selective photothermolysis (SP) by Anderson & Parrish in 1983, applications of lasers in dermatology have evolved to a host of devices for more precise, targeted thermal damage, while minimizing non-specific tissue destruction. Non-laser flashlamp sources called intense pulsed light (IPL) have also been developed for some of the applications of SP that use millisecond pulses of light. Understanding the theory of SP is vital for making sense of the large number of laser and IPL devices and applications. An understanding of the optical properties of skin is also needed, since the whole endeavor of laser treatment starts with the absorption of light energy, inside the skin.

Lasers that vaporize a thin layer or column of tissue have also been developed. The concept of fractional photothermolysis (FP), reported by Manstein and colleagues in 2004, recently launched another era of lasers in dermatology, in which patterns of very small non-selective thermal damage zones are used to stimulate skin remodeling without scarring. Laser-stimulated remodeling is a complex process that mimics large wound healing in some aspects, with epidermal regeneration, induction of metalloproteinases, and formation of new dermal matrix including elastin fibrils and collagen types I and III. Compared with gross wound healing, there is minimal inflammation and no scarring. A ‘cookbook’ approach should be avoided when choosing among these devices for various applications. When treating a particular patient with a particular device, a combination of fundamental understanding, careful observation of the appropriate clinical end points, dexterity, and clinical experience is far better than a set of instructions (Box 1.1).

Light interactions with skin

Photons can be absorbed (giving up their energy to matter) or scattered (changing their direction of travel). Light that is scattered back from skin is called reflectance. For a given skin layer, light that passes through it is called transmittance. Scattering is inversely wavelength dependent, such that shorter wavelengths are scattered more and longer wavelengths (such as infrared) are scattered less. We are all familiar with these events – black objects become hot when placed in sunlight due to absorption and water droplets (clouds) or crystals (snow) appear bright white because they strongly scatter light, with little or no absorption. Similarly, light is both absorbed and scattered within the skin. Thus, skin layers are cloudy and colored depending on the mix of scattering and absorption. Penetration of light into (and beyond) skin is limited by both absorption and scattering. All effects of light on the skin begin with photon absorption, and the molecules that absorb light are called chromophores. Ablative lasers are those that vaporize tissue by rapidly boiling water inside the tissue. It should come as no surprise therefore, that the lasers intended for skin ablation are at wavelengths strongly absorbed by water. Non-ablative lasers do not vaporize tissue. There are many non-ablative lasers in dermatology, some of which are at wavelengths absorbed by water and some of which are absorbed by other chromophores such as melanin and / or hemoglobins.

Laser dosimetry is extremely important for safe and effective results. In order to remove tissue, ablative lasers must raise local tissue temperature beyond the boiling point of 100^oC, plus add much more energy needed for changing water into steam. The fundamental unit of energy is a joule (J). It takes 4.2 J to heat 1 cm³ of water by 1^oC. In order to vaporize the same 1 cm³ of water, more than 2000 J are required. An ablative laser must deliver about 2500 J of energy per cm³ of vaporized tissue. Not only is a lot of energy required to ablate skin tissue – the energy must be delivered quickly to remove the hot tissue before heat is conducted deeply into the skin, causing a burn. The standard ablative lasers in dermatology are erbium (2940 nm) and CO₂ (10 600 nm). The desired interaction of these ablative lasers is to precisely remove a thin layer for resurfacing or narrow column for fractional treatment of skin, leaving behind minimal residual thermal damage. A thin residual thermal damage layer, typically about 0.1 mm, is useful in practice for hemostasis. Minimum residual thermal injury is achieved with ablative lasers by a combination of wavelength, pulse duration, and power density (W/cm²) at the skin surface. A common mistake made by beginning laser users is to ‘turn down’ the power of a surgical CO₂ laser in a misguided attempt to exercise caution. Unfortunately, turning down the power can cause burns because the process turns from rapid, precise vaporization with minimal thermal damage to bulk heating of the skin from unwanted residual heat. Fortunately, many of the ablative lasers made specifically for dermatology are designed to stay within a range of dosimetry for rapid tissue ablation, making this scenario less likely. The safest erbium and CO₂ lasers are those emitting high power, high energy, and short (less than a few ms) pulses, designed specifically for dermatologic use with minimal residual thermal damage. Despite whatever safeguards an ablative laser may offer, the most reliable safeguard is an ability to recognize the desired and undesired immediate response end points. For example, immediate contraction of the skin is always a sign that substantial thermal injury of the dermis has occurred (Fig. 1.1).

Figure 1.1 Scheme of ablative laser vaporizing skin, leaving a residual thermal damage layer.

Fluence is defined as the energy delivered per unit area of skin, and its units are typically expressed in J/cm². One can think of fluence as the local ‘dose’ of laser energy applied to skin. Pulse duration (also called pulsewidth, or exposure duration) is simply the time for which laser energy is delivered, expressed in seconds. Power is defined as the rate of energy delivery. Power is measured in watts (W), a familiar unit because of common devices such as light bulbs. One W is defined by 1 W = 1 J/second. A common incandescent light bulb consumes 100 W of electrical power, but emits less than 10 W of light. In contrast, common lasers in dermatology produce 10 to 1 000 000 000 (a billion) W of light power. The Q-switched lasers, which we commonly use to remove tattoos and pigmented lesions, produce more power than a typical nuclear power plant! However, these lasers emit that impressive power for only 10–100 nanoseconds (ns, billionths of a second). Thus, the fluence for treatment of a child with a nevus of Ota using a 10 ns Q-switched laser, and of a child with port-wine stain using a 1 millisecond (ms) pulsed dye laser, can be similar – about 5–10 J/cm² – but the pulsed dye laser has 100 000 times less power than the Q-switched laser.

Skin optics

In skin, the most important chromophores are hemoglobin, melanin, exogenous pigments (e.g. tattoo ink, some drugs), water, and lipids. The intended target chromophore, depth of the target structures, and absorption of light by adjacent tissue influence the appropriate wavelength selection. Absorption spectra of various chromophores across the electromagnetic spectrum are summarized in Figure 1.2.

Figure 1.2 Absorption spectra of chromophores commonly used for selective photothermolysis laser surgery.

This figure was modified from Sakamoto et al, 2007.

Selective photothermolysis

SP relies on fundamental choices being made correctly – wavelength, pulse duration, fluence, exposure spot size, and use of skin cooling. First, a wavelength (or, with IPLs, a range of wavelengths) must be used that is preferentially absorbed by the intended ‘target’ structures such as hair follicles, microvessels, tattoo inks, or melanocytes. Thus far, all lasers utilizing SP operate in the visible and near-infrared (NIR) spectrum. Generally, in the visible light spectrum, a target chromophore is treated using wavelengths of light of a complimentary color. For example, red tattoo ink absorbs green light and can be effectively treated with a frequency doubled Q-switched Nd : YAG laser operating at the green wavelength of 532 nm. Similarly, green tattoo ink is best removed with a red Q-switched laser, such as the ruby laser at 694 nm. Preferential absorption implies the avoidance of competing chromophores, not simply strong absorption in the intended target. For example, when treating dermal targets such as blood vessels it is important to minimize unwanted damage to the epidermis. Since every photon that reaches a blood vessel must first travel through the overlying epidermis, the best wavelengths for port-wine stain treatment are not simply those with strong absorption by blood. The proper wavelength(s) must also penetrate deeply enough to reach the intended targets. Across the visible and near-infrared spectrum from 400 to 1200 nm, longer wavelengths penetrate deeper into tissue. These reasons account for the use of yellow light pulsed dye lasers rather than the very strongly absorbed blue wavelengths for treating superficial vascular lesions. Long-pulsed dye lasers are the first example of a laser designed specifically for a medical application: treatment of port-wine stains in children (see Case study 2). On the microscopic scale, microvessels are selectively heated and damaged, with minimal injury to the rest of the skin structures. However, for a hypertrophic or deep vascular lesion, such as many adult port-wine stains and venous malformations, much better efficacy is often obtained using the deeply penetrating 755 nm near-infrared alexandrite laser, as detailed by Izikson et al in 2009. (Looking at Fig. 1.2, it is easy to observe that hemoglobin absorbs yellow light much more strongly than at 755 nm, a wavelength that is also well absorbed by melanin.) When alexandrite lasers are used for vascular lesion treatment, it is therefore imperative to use excellent skin cooling for epidermal protection; see Chang & Nelson 1999 and Altschuler et al 2000.

Melanin absorbs across a wide spectrum of wavelengths. Eumelanin, the primary chromophore in the epidermis and darkly pigmented hair follicles, has a broad absorption spectrum spanning from ultraviolet light to the near-infrared region. Eumelanin is the chromophore targeted in lentigo simplex. It is also the target in laser hair removal with the secondary target being the follicular stem cells, as reported by Grossman and colleagues in 1996. In fair-skinned individuals with dark hair, wavelengths in the near-infrared range (810 nm diode; 755 nm alexandrite) are ideal for laser depilation. However, a common mistake is to use these popular devices for hair removal of red or blond hair, which is primarily composed of pheomelanin. These laser wavelengths are poorly absorbed by pheomelanin and are therefore ineffective for permanent removal of red or blond hair.

In general, water is not a useful target for selective photothermolyis because it is present at high concentration in almost every skin structure. Water absorption gradually increases starting in the near IR range and peaking within the mid-IR spectrum. When used in conjunction with appropriate epidermal cooling devices, lasers within this wavelength spectrum can function as non-ablative modalities for photorejuvenation by targeting water within the dermis, thereby generating heat and controlled thermal damage. This wounding of the dermis subsequently results in collagen remodeling, as well as neocollagenesis, contributing to the modest improvement in the appearance of rhytides.

More recently, near-infrared lasers have been used by Sakamoto and co-workers and by Anderson et al to target lipid-rich tissue. Unlike the targeting of traditional chromophores, which is based on electronic charge, lasers to target lipids are based on the vibrational modes of the molecules. Lipid molecules are selectively destroyed at 1210 nm and 1720 nm where their absorption is slightly higher than that of water. Although there are no commercial devices yet available, the application of these forthcoming devices offers an appealing, alternative, non-invasive methodology of targeting lipids.

The second essential factor for SP is to use a pulse duration that allows heat to be confined during the laser pulse in or near the target structures. The moment that heat is formed in a target by preferential absorption of photons, the target begins to cool by conduction. Therefore, heating of the target is a balance between the rate of photon absorption and the rate of cooling. The concept of a particular target’s thermal relaxation time (TRT) is useful in clinical practice to pick the correct pulse duration. TRT is simply defined as the time required for substantial cooling of the target structure. TRT is strongly related to target size, and this variation accounts for the wide range of laser pulse durations needed for optimal dermatological lasers. A simple and useful approximation is that TRT ≈ d², when TRT is in units of seconds, and d is the target size in millimeters. For example, a 1 mm leg vein cools in about 1 second, while a 0.2 mm telangiectasia, typical for rosacea, cools in about 0.04 seconds (40 ms), and a 0.03 mm venule in a child’s port-wine stain cools in about 0.001 seconds (1 ms). The optimal laser or IPL pulse duration is typically about equal to the TRT. In this example, a very long exposure from a low-power KTP (532 nm) laser would be appropriate for treating the leg vein. A higher power KTP or pulsed dye (595 nm) laser operated at about 20–40 ms would be appropriate for the rosacea-associated telangiectasia, and a pulsed dye laser operated at about 1 ms would be appropriate for the pediatric port-wine stain. This extreme dependence of TRT on target size applies all the way down to the nano-scale of subcellular targets. Q-switched lasers are used in dermatology because their 10–100 nanosecond pulse durations are shorter than the TRT for targets such as tattoo ink particles, melanosomes, and drug pigmentation deposits (see Case study 1).

The matching of TRT and pulse duration is clinically important to achieve efficacy, avoid side effects, and even to define the targets that will respond. For example, consider a young man with both nevus of Ota and a dark beard on his face. Both the nevus and his beard hair contain high concentrations of the same chromophore, melanin. A Q-switched alexandrite laser (~755 nm wavelength) will be highly effective for fading his nevus of Ota, because the targets are small, isolated melanocytes scattered deeply throughout his dermis. The appropriate end point is immediate whitening, due to microscopic gas bubbles formed when the target melanocytes in his dermis are fractured. However, this Q-switched alexandrite laser will not permanently remove his hair, because its pulse duration is a million times shorter than the TRT for a terminal hair follicle. This laser merely vaporizes the hair shaft (which is already dead) before heat can flow to the hair follicle epithelium and the patient can be informed with confidence that his beard will not be accidentally removed. In contrast, a long-pulse (3–30 ms) alexandrite laser at the identical wavelength could permanently remove his beard without affecting his nevus of Ota. This long pulse duration is incapable of providing thermal confinement in something as small as an isolated melanocyte, but allows plenty of time for heating of the entire hair follicle without vaporizing its pigmented hair shaft.

A common professional liability issue related to pulse duration is the use of long-pulsed sources such as IPLs, broadly available for laser hair removal, for the treatment of tattoos. Long-pulsed lasers and IPLs emit millisecond domain pulses that heat the tattooed skin at large instead of the individual ink particles, because the pulse duration greatly exceeds the TRT of the ink particle. The surrounding dermis is therefore heated causing unselective thermal damage, blistering, dyschromia and scarring, as reported by Wenzel and co-workers in 2009. Unfortunately, similar mistakes are commonly observed due to the lack of a full understanding of SP and incorrect choice of treatment pulse width.

The third factor for optimal SP is sufficient fluence to affect the targets. In general, the fluence necessary is inversely related to absorption by the target structures – stronger absorption requires lower fluence, and vice versa. This is the reason, for example, that a typical alexandrite laser fluence for treatment of a port-wine stain (see Case study 2) is 40 J/cm², while that for pulsed dye laser treatment of the same lesion may be only 8 J/cm².

For pulsed sources, it is not uncommon to see frequency as one of the parameters. Frequency is the measurement of repetition rate of a laser pulse in a given period of time (seconds) and is measured in hertz (Hz), where 1 Hz is 1 pulse/second. It is useful to use higher repetition rates for treatments that require a large number of laser pulses (e.g. large tattoos). Although it is tedious to treat a large tattoo using 1 pulse/second, increasing the frequency of pulses makes the treatment less time consuming (and more challenging to distribute the pulses uniformly).

Skin cooling: limiting thermal damage to the intended targets

Sparing of the epidermis and superficial dermis is important for selective destruction of deeper structures and can be improved by the use of appropriate skin cooling. Cooling can be applied before (pre-cooling), during (parallel cooling), and after the laser pulse (post-cooling). Similarly to laser-induced tissue heating, cooling should be applied keeping in mind the histological target. According to Zenzie et al, the greater the depth of an anatomical structure, the longer the cooling should be applied. For epidermal protection, Sakamoto and co-workers reported that 20–50 ms is enough, while for epidermal and dermal protection (e.g. for targeting subcutaneous fat) cooling should be applied for 5–10 seconds. Cooling can be applied using direct solid contact cooling (e.g. cold sapphire window), automated cryogen spray (DCD™, direct cooling devices) or by blowing direct cold air. Cold-air cooling has the advantage of bulk skin cooling, which limits pain, edema, and the risk of burns from residual heat.

For the choice of proper dosimetry, it is crucial to be familiar with the particular device being used, and to carefully observe skin response to treatment. The combination of laser wavelength, pulse duration, spot size, skin cooling, and dosimetry can suggest initial treatment parameters, but only careful observation of immediate clinical end points will ensure efficacy (Figs 1.3–1.5), helping to avoid side effects. Common clinical end points are summarized in Table 1.1.

Figure 1.3 Expected clinical end point after laser treatment of pigmented lesions using a Q-switched 755 nm alexandrite laser for solar lentigines. Immediate response observed with epidermal whitening due to steam bubbles.

Figure 1.4 Expected clinical end point after laser treatment of a vascular lesion: (A) Child’s chest with port-wine stain previously treated with laser, before PDL session, with immediate response after a single 585 nm pulsed-dye laser pulse, 3 ms pulse width, 6 mm spot size, 9 J/cm² with dynamic cooling (arrow). (B) Note immediate round-shaped purpura where laser is applied, indicating vessel coagulation.

Figure 1.5 Expected clinical end point after laser treatment of a vascular lesion: (A) Wrist with port-wine stain. (B) Immediate response after a single (arrow) 755 nm alexandrite laser pulse, 1.5 ms pulse width, 6 mm spot size, 80 J/cm² with dynamic cooling. Note immediate bluish color, indicating vessel coagulation.

Fractional photothermolysis

Fractional photothermolysis (FP) uses microbeams of laser to target the tissue, inducing microthermal zones (MTZ) of injury, as reported by Manstein and colleagues in 2004. Each MTZ is typically 100–300 µm in diameter. The depth and density (number per unit area) of the microlaser beams applied to the tissue can be adjusted depending on the clinical indication. The advantage of this technique is that it spares untreated skin surrounding each MTZ, allowing fast healing and reducing the risk of side effects. A typical FP treatment session provides laser exposure to about 10–50% of the skin.

Soon after its introduction in 2004, the concept of FP has been widely embraced in dermatology. A number of new devices, laser wavelengths, and clinical indications have been developed with success. In principle, FP can be applied with a wide variety of energy sources capable of producing an array of small zones of skin damage. These include various non-ablative NIR lasers (1320–1550 nm; 1927 nm thulium) and ablative lasers (2940 nm erbium; 10 600 nm CO₂). Currently, even visible light and other technologies such as ultrasound and radiofrequency devices have been using fractionated applicators. Photoaging and pigmentary alterations, scar treatment, melasma, striae, and xanthelasma are examples of the variety of clinical indications that can be treated with FP (see respectively the studies by Manstein et al, Alster et al, Tannous & Astner, Kim et al and Katz et al).

Interestingly, in addition to local thermal destruction and stimulation, fractionated devices may also play an important role for drug delivery into the tissue and for extruding material out of the skin, as in the studies by Haedersdal et al. This has also been recently reported by Ibrahimi et al using an ablative fractionated erbium : YAG laser to treat an allergic tattoo reaction with success. Whereas conventional treatment of allergic tattoo reactions with a Q-switched laser alone could likely increase immunogenicity of the tattoo pigment post-treatment and the risk of a systemic allergic response, the ablative fractional laser has shown the ability to remove allergic tattoo pigment as an alternative method without inducing a systemic allergic reaction.

Conclusion

It is likely that fractional photothermolysis lasers and similar fractional treatment technology will continue to evolve into unforeseen applications. Many interesting, fundamental, and clinically important questions remain to be answered about lasers in dermatology.