Regarding Tattoos: Title and subTitle BreakIs That Sunlight, or an Oncoming Train at the End of the Tunnel?

TATTOOS ARE increasingly popular among young adults around the world. This is not just a fad—almost every culture uses tattoos and the oldest human mummy sports several of them drawn more than 6000 years ago. Prevalence in the United States is probably about 3% to 5%, somewhat higher in young adults and in the South. However, good demographic data are unavailable. Many who get a tattoo will eventually want to get rid of it for a variety of good reasons.¹ Unfortunately, tattoos are far more difficult to remove than they are to acquire. The common misconception that lasers easily remove tattoos is driven in part by the media, advertisements, and wishful thinking. For example, in a television episode of "The Simpsons," Bart gets a tattoo that is quickly removed by a nerdy cartoon laser dermatologist with dollar signs in his eyeballs. Bart should have been depicted going to his dermatologist 6 to 10 times over the course of 1 year, wearing dressings to cover his painful, partially removed tattoo after each treatment, and being worried about the almost even chance that some of his tattoo will remain, after all. Whether Bart's dermatologist was properly depicted, I will leave to the reader to determine. Like it or not, we are charged with caring for the nation's skin problems, including self-inflicted ones. Unless something changes, we are going to disappoint the millions of persons getting tattoos, who eventually show up at a dermatologist's practice to get them removed.

The article by Timko et al² in this issue of the ARCHIVES is a laudable but small step in the right direction. If the most stubborn-to-remove tattoo inks can be identified, we might predict which patients will do poorly, and perhaps these inks can be taken off the market. Unfortunately, the element analysis technique used in this study does not identify chemical compounds and does not detect low-mass elements. Mass spectrometry combined with element analysis would have been more revealing. Furthermore, it is unlikely that anyone in clinical practice will perform chemical analysis prior to tattoo treatment. These techniques require an invasive sample and are not so simple to acquire. As suggested by Timko et al,² it would be nice if tattooists kept records of the specific inks used, which could be available to their clients years later. This and other issues suggest that regulatory standards for good practice should be applied to tattooists.

But, what else can be done? This editorial will address the following topics: (1) improving the clearance of tattoo ink particles after laser treatment, (2) optimizing the laser–tattoo ink interaction, (3) eliminating difficult-to-remove, antigenic and/or toxic tattoo inks from the market, and (4) designing new tattoo inks.

The first nonscarring removal of tattoos was demonstrated in Scotland by Reid et al,³ who used a Q-switched ruby laser. In my laboratory, we somewhat optimized this and other Q-switched lasers for tattoo removal about 10 years ago.^{4 - 5} In general, different laser wavelengths are needed for absorption by different tattoo ink colors. Tattoos consist of phagocytosed submicrometer ink particles trapped in the lysosomes of phagocytic dermal cells, mostly fibroblasts, macrophages, and mast cells. When extremely intense (100 million W/cm²), brief (billionths of a second) light pulses are absorbed by these intracellular ink particles, they reach extreme temperatures (at least 300°C). The particles fracture, undergo chemical changes, violently boil water in the cell cytoplasm, rupture the cells, and release laser-altered ink into the dermis. Some of this free ink is eliminated by lymphatic and transepidermal transport, but most of it is rephagocytosed by somatic dermal cells within a few days.⁶ This rephagocytosis accounts for the "residual" tattoo after each laser treatment; we found that essentially all of the residual tattoo ink particles were ultrastructurally altered by a single, previous laser treatment. Lymphatic transport seems to account for most of the ink removed, although ink is sometimes shed in a scale-crust after each treatment.

Therefore, tattoo removal might be enhanced by (1) increasing transepidermal elimination, (2) increasing lymphatic transport, (3) blocking rephagocytosis of ink by somatic dermal cells after laser treatment, or (4) increasing phagocytosis of ink by cells that "traffic" through the dermis just after laser treatment. Some of these ideas have been briefly examined. In a small controlled study, Ort et al⁷ removed the epidermis before laser treatment in an attempt to increase the elimination of ink. Unfortunately, there was no evidence of improved efficacy compared with tattoo treatment in adjacent intact skin. Massage is known to greatly increase lymphatic transport. Thus far, I have resisted the temptation of trying laser-plus-massage therapy for tattoos. In a pilot study Dierickx et al⁸ created a delayed-type hypersensitivity reaction to Candida antigen before laser tattoo treatment, with the intent of increasing phagocytosis by mobile inflammatory cells. Again unfortunately, this did not improve efficacy. However, corticosteroids, specific cytokines, and drugs affecting phagocytosis have not been examined in the context of tattoo removal.

Important aspects of the physical and chemical processes initiated by laser pulses remain unknown. The classic concept of selective photothermolysis⁹ uses pulses short enough to provide confinement of heat in the "target," in this case an ink particle only 0.1 to 1 µm in diameter. However, it is also possible to confine pressure in the target by inertial confinement. Extreme stresses in the target can be created this way. For example, this approach was taken to initiate nuclear fusion reactions by laser pumping of small fuel pellets. In theory, inertial confinement should more thoroughly fracture the tattoo ink particles, perhaps into very small fragments that easily leave the skin. The present generation of Q-switched clinical lasers produce pulse durations from about 10 to 100 nanoseconds (billionths of a second). Q-switching is the equivalent of suddenly opening a window, letting out the light from a laser cavity. Lasers are usually a few feet long and, therefore, Q-switching yields a pulse that is about as long as it takes for light to travel several feet, ie, nanoseconds. While adequate for thermal confinement in a tattoo ink particle, these pulses are too long for inertial confinement. To achieve inertial confinement, the pulse duration needs to be less than the particle diameter divided by the speed of sound. For a 1-µm tattoo ink particle, this is a pulse duration less than 1 nanosecond.

Inertially confined laser pulses have been tested for tattoo removal and seem to be successful! Pilot studies in humans¹⁰ and animals¹¹ suggest that tattoo removal using picosecond (trillionths of a second) laser pulses is more efficient than with conventional nanosecond laser pulses. Unfortunately, standard technology for generating high-energy picosecond pulses is costly and, to my knowledge, there is not yet a commercial attempt to provide picosecond lasers for dermatology. Fortunately, new technology is emerging, notably microchip Q-switched lasers operating in the picosecond domain and wavelength-tunable ultrafast femtosecond (10^-15 seconds) lasers. A femtosecond is an almost inconceivably small fraction of a second, about the same as 1 minute compared with the age of our solar system. Femtosecond pulses have not yet been tested for efficacy in tattoo removal. Microchip Q-switched lasers work by the same principle as their larger ancestors, but emit picosecond pulses because their cavities are only a few millimeters in length. Perhaps we will see amplified versions of these in dermatology after a few years.

Although it is clear that chemical reactions occur during laser tattoo treatment, we know almost nothing about them. For carbon tattoos, the chemistry may be "good." Homemade tattoos made of carbon (india ink, graphite, ash) are easy to remove. Perhaps the reason for this is simply that carbon can burn. In a brief (R.R.A., unpublished data, 1993) laboratory experiment, I measured the emission spectrum from india ink particles in an agar plate during Q-switched Nd:YAG laser exposure. Spectral lines from extremely hot, unstable species such as dicarbon radical anion were present along with a continuum spectrum that came from temperatures of at least 2500 K. This strongly suggests that a plasma (hot, ionized matter) and/or combustion, occurs when carbon is the ink. For other tattoos, the laser-induced chemistry may be "bad." Cosmetic tattoos containing red iron oxide (Fe₂O₃) and titanium dioxide (TiO₂) turn black on Q-switched laser treatment, a phenomenon first discovered to my horror in a patient, but easily reproduced in vitro.¹² Ink darkening is probably a combination of chemical reduction and changes in particle size. The inks that darken could easily be identified and potentially removed from the market. Patients with a history of gold treatment for rheumatoid arthritis or other diseases may also have their skin turn black on Q-switched laser exposure, a phenomenon dubbed "laser-induced chrysiasis."¹³

This is a good idea. Many, if not most tattooists are professionals who genuinely care about the well-being and long-term satisfaction of their clients. These tattooists would gladly stop using problematic inks. The Food and Drug Administration, which does not approve any tattoo inks, should be willing to remove some from commerce if there were good data and a formal request to justify such action. For example, cinnabar (mercuric sulfide), an antigenic red ink that contains mercury, was removed from the US market long ago. Immune response to new red inks can be just as devastating, however.¹⁴ We need to clearly identify which inks are difficult or impossible to remove and which are most likely to darken on laser treatment. This could be done in animal models, with verification in humans by analysis of laser-resistant tattoo inks. The article by Timko et al² is a step in that direction.

Because tattooing extends into prehistory and will extend far into the future, it simply makes sense for us to figure out how to "do it right." In view of the care we take for testing and prescribing drugs, how should we feel about the fact that 1 in 20 Americans will be injected with unknown impure substances by people with little or no medical training? In theory, tattoo inks could be made that are safe, sterile, nontoxic, and designed for removal. Given good data, the Food and Drug Administration might even welcome clearing safe, removable tattoo inks as an alternative to the present situation.

There are several schemes that could be used to design easily removable tattoos. All tattoo inks consist of insoluble, indigestible solid particles that are small enough to be phagocytosed. Indeed, any material fitting this description appears to form a tattoo when introduced into the dermis. Colored particles that undergo controlled digestion, similar in concept to absorbable sutures, might be used to make time-limited tattoos lasting a predictable number of months or years. Particles designed to become soluble, to burn, to rupture, to bleach, or to be transported after activation by laser light or other external energy, might be made, tested, and approved by the Food and Drug Administration.

Years ago, I naively failed to anticipate that laser tattoo removal would inevitably lead to—more tattooing. This is sad, because I have never met a tattoo more beautiful than the skin onto which it was placed. With equal naivete perhaps, I suggest that we should continue to work on making tattoos safer and more removable than ever. Otherwise, what looks like sunlight at the end of this tunnel is surely the headlight of an oncoming train filled with unhappy, tattooed passengers.