Tattooing has been used to decorate the skin for different purposes since thousands of years ago. It is a very common practice in many cultures, and its popularity has increased in western countries. The number of reports of associated complications has increased in parallel, and recent publications indicate that the incidence of adverse reactions to permanent tattoos is approximately 2%.1–3 The reactions include infections, acute and chronic inflammatory reactions, and flare-ups of skin complaints due to the isomorphic response. Benign and malignant tumors have also been reported on tattoos, although no causal relationship between the 2 phenomena has been established.4,5 Current Spanish legislation adopts the resolution of the Council of Europe ResAP (2008) on permanent tattoos, which establishes regulations stating that the packaging must show the conditions of use, date of minimum durability, batch number, and the list of ingredients (the Chemical Abstract Service or Color Index number, according to international nomenclature) and guarantee that the content is sterile. However, since most inks used in Spain are imported from Asia and the United States, they are not subject to this regulation.6

The objective of the present study was to describe the clinical characteristics of patients seen at a skin allergy clinic with a tattoo-related complication between January 2002 and December 2016. We also propose an algorithm for managing patients with a suspected allergic reaction to ink and discuss the yield of patch tests and the usefulness of patch testing with the ink itself.

We performed a descriptive, retrospective study of all patients who were seen at the Skin Allergy Clinic of Hospital General Universitario de Alicante, Alicante, Spain with a tattoo-related complication. We excluded patients whose reaction was to a temporary tattoo. We recorded epidemiological and clinical variables, as well as the results of additional tests. We divided the cohort into 2 groups: patients who experienced early reactions (within the first month after the tattoo) (Table 1) and patients who experienced late reactions (more than 1 month after the tattoo) (Table 2). In addition, patients were grouped according to the clinical pattern observed based on the classification by Serup et al.5 Patients with a suspected allergic reaction to ink underwent patch testing, which was performed with the standard series of the Grupo Español de Investigación en Dermatitis de Contacto y Alergia Cutánea (Spanish Contact Dermatitis and Skin Allergy Research Group) or a specific textile series (Chemotechnique Diagnostics); in some cases, a specific metal series (Martí Tor) and the ink supplied by the patient were also used (Table 3).

A total of 23 patients consulted for a tattoo-related skin problem from January 2002 to December 2016. Men accounted for 43.5% (n=10) and women for 56.5% (n=13). The mean age was 36 years (range, 19-58 years). Of the 7 patients who had previous tattoos (28%), only 1 had simultaneous complications on the oldest tattoo (Fig. 1).

Figure 1.

A, Reaction in the form of an eczematous plaque on a red tattoo placed 1.5 months earlier. B) Simultaneous reaction on a red tattoo placed 2 years earlier.

(0.12MB).

Onset of symptoms was within the first month of having the tattoo in 9 patients (39%), in whom the initial suspicion was infection. Between November 2004 and April 2005, we saw 8 patients who had been tattooed with black ink during the month before the visit by the same tattooist. Subsequent culture of samples from 7 of these patients revealed Pseudomonas aeruginosa in 3 cases. We therefore performed an exhaustive epidemiological study, which revealed a cutaneous inflammatory reaction to the tattoo owing to a batch of black ink that was contaminated by P aeruginosa. One of the patients tattooed with the contaminated ink was coinfected with Mycobacterium chelonae, which was isolated in culture. The remaining patient was diagnosed with pyodermatitis caused by Staphylococcus aureus, with isolation of the microorganism in culture and compatible symptoms. In our study, early reactions on tattoos were caused by infection. All 8 patients underwent skin biopsy, which revealed the most common histologic pattern to be foreign body granuloma.

We included 14 patients with late reactions. Red ink was the problem color in 6 cases (42.85%), black ink in 5 (35.7%), and blue ink in 2 (14.2%). Only 1 patient experienced a reaction affecting 3 colors simultaneously (yellow, green, and red). The 3 inks shared the pigment CI11740 (pigment yellow 65: 4-methoxy-2-nitrophenyl-N-2-methoxyphenyl-3-oxo-butyramide), which could be the culprit substance (Fig. 2).

Figure 2.

Eczematous reaction in the green, yellow, and red areas. The 3 inks shared the pigment CI11740: Pigment yellow 65. The color in this figure can only be fully appreciated in the electronic version of the article.

(0.09MB).

Of the clinical patterns observed in reactions to a tattoo proposed by Serup et al.,5 the most prevalent in our series were eczematous plaque (n=6, 42.8%) and a lichenoid-papular pattern (n=4, 28.5%), which were finally associated with a probable diagnosis of allergic contact dermatitis to the ink. An ulcerative-necrotic pattern was observed in 2 patients (14.28%), who had been diagnosed with allergic contact dermatitis to ink in one case and cutaneous sarcoidosis in the other. One patient presented a papulonodular pattern that affected a specific area of the tattoo but spared the remainder of the tattooed skin and was diagnosed as cutaneous sarcoidosis (Fig. 3). Another patient complained of pruritus and dysesthesia on a tattoo in black ink that had been placed 10 years previously; we interpret this finding as a neuropathic pattern, since the biopsy revealed no pathologic findings. Skin biopsies were performed in all cases. In the 10 patients with a probable allergic reaction to the ink color, histopathology revealed lichenoid interface dermatitis in 8 cases (80%). As for the remaining 20%, we observed a pseudolymphomatous pattern in one case and foreign body granuloma in the other. Biopsy confirmed the diagnosis of cutaneous sarcoidosis in 2 of the late-onset cases.

Figure 3.

Papulonodular pattern limited to an area of the tattoo. Histology compatible with cutaneous sarcoidosis.

(0.11MB).

Patch testing was performed in 9 patients whose clinical and histopathologic data were compatible with allergic contact dermatitis to tattoo ink. The standard series was applied, and in some cases the specific textile series and the specific metal series were also applied. Nickel sulphate was the most frequently positive substance (n=4, 44.4%), and its relevance was past in all cases. One patient tested positive for cobalt chloride (++), palladium chloride (++), and beryllium chloride (++). Given that the culprit ink was blue, relevance was thought to be probable for the cobalt chloride patch and unknown for the remaining positive patches. The metal series was used in 4 cases (44.4%), and the results were negative with unknown relevance. The specific textile series was applied in 8 patients (88%), and the results were negative in all cases. In 4 patients (44.4%), patch testing with the ink brought by the patient yielded a positive result in only 1 case.

Tattooing often exposes the skin to foreign bodies that can potentially induce a response.7 We can classify reactions as early (occurring within the first month after the tattoo) and late (occurring more than 1 month after the tattoo). We propose a diagnostic algorithm based on this classification and on our clinical experience (Fig. 4).

Figure 4.

Diagnostic algorithm. Although topical or intralesional corticosteroids are considered the first line of treatment, other valid options include Q-switched laser, carbon dioxide and erbium ablative lasers, and surgery.

(0.26MB).

The most frequent early complications are transient inflammatory reactions caused by the injury resulting from injection of ink into the skin. These usually occur during the first weeks after the tattoo is placed and do not generally require medical or dermatological attention, since tattooists themselves manage them. However, a potential infectious origin should be borne in mind in the case of an early reaction. Infection by microorganisms can occur through contact with nonsterile material, performance of the procedure under unsanitary conditions, and the use of contaminated ink or water. A recent study showed that 10% of marketed inks are contaminated with pathogenic bacteria and that there are no regulations to guarantee the sterility of ink in Spain.8 Clinical manifestations indicative of infection include pain, increased local temperature, suppuration, and ulceration. In our study, 6 of the patients who were infected through contaminated ink had atypical manifestations, with eczematous plaques (Fig. 5). In many cases, the histopathology findings were compatible with foreign body granuloma; abscess or abundant inflammatory infiltrate of polynuclear cells was observed in 3 cases, despite being an expected finding. We obtained negative cultures in 3 of the patients who had been tattooed with ink from the contaminated batch. This finding can be explained by the fact that people who have a tattoo are generally young and in good health, with sufficient immunity to protect them against the microorganisms involved.9,10

Figure 5.

“Atypical” skin reaction in 2 patients tattooed with ink from a batch contaminated with Pseudomonas aeruginosa.

(0.19MB).

Therefore, in any patient who presents with an early reaction in the area of the tattoo, we believe it is essential to culture the purulent secretion and start empiric antibiotic therapy if the reaction is clinically indicative of infection. If the patient's progress is poor, a biopsy will be necessary for the histopathology and microbiology workup (culture for bacteria and mycobacteria). It would be interesting to perform culture to rule out contamination if a sample of the ink used is available. If the patient presents an early reaction, with no signs of infection, we recommend starting treatment with topical corticosteroids. If the reaction is refractory to treatment, we recommended performing a biopsy.

Late complications are those that develop months or even years after the tattoo. A biopsy is essential in all such cases to rule out infections with late manifestations or even systemic infections. Furthermore, it would help us to make a diagnosis based on histologic patterns.

According to the literature, the presence of noninfectious granulomas is uncommon; however, in our series, this pattern was observed in 2 patients, probably because the particles of pigment that are deposited in the dermis are identified as microscopic foreign bodies.11–13 Such a reaction should lead us to rule out systemic granulomatous disease. Patch testing should be ordered in cases where the symptoms point to an allergic reaction.

The pseudolymphomatous pattern is characterized by a dense lymphocytic infiltrate, which was shown to be polyclonal in the immunohistochemical evaluation. The cause of this type of immune response is unknown, although chronic antigenic stimulation is thought to be responsible.13,14 We propose starting treatment with topical corticosteroids for such reactions. If the lesion persists and clinical manifestations compatible with an allergic reaction appear, we should consider using patch tests.

Lichenoid interface dermatitis was the most frequent histologic pattern in the late reactions and was associated with allergic reactions. This inflammatory infiltrate of the papillary dermis and basal layer of the epidermis is composed of mononuclear cells and macrophages loaded with intracellular pigment. Eosinophils are not a constant finding.14 In our experience, this histologic pattern was the most frequent in cases of probable allergic reaction. Allergic reactions to ink are not common. Their onset is usually late, since they normally result from primary sensitization. Once initiated, the reaction is permanent and does not usually respond to topical corticosteroids. Three clinical patterns that point to allergic reaction to ink were recently reported, although they are not specific. The 3 patterns are plaque-like, hyperkeratotic, and ulcerative-necrotic.5 In the sample we report, we were able to show a different pattern that also points to allergic contact dermatitis, namely, a lichenoid-papular pattern involving nonconfluent pruriginous papules arranged homogeneously on specific areas of color in the tattoo but that do not affect the other colors (Fig. 6). In the present series, the eczematous plaque pattern was the most common, although the lichenoid-papular pattern was also observed in 3 patients, probably because the pigment is injected into the dermis and does not come into direct contact with the epidermis; therefore, the clinical and histologic reaction may differ from the typical manifestations of contact dermatitis. We propose that both clinical patterns point to an allergic reaction. Furthermore, the lesions are usually monomorphic and uniform, affecting all the areas tattooed with the problem color. Patch tests are sometimes ordered for patients with a late reaction that does not affect the whole area tattooed with the problem color; therefore, by definition, the reaction is not an allergic reaction to ink and the yield of the tests is zero. The appearance of identical skin manifestations in areas of the same color on previous tattoos supports an allergic reaction. We observed this phenomenon in a patient with allergic contact dermatitis to red ink, which, according to data from the literature, is the color that most frequently causes allergic reactions. In our experience, there are no differences between reactions to red ink and reactions to black ink, probably because black tattoos are more frequent.

Figure 6.

Papular-lichenoid pattern in a patients with a probable allergic reaction to red tattoo ink. The color in this figure can only be fully appreciated in the electronic version of the article.

(0.11MB).

When a patient with a suspected allergic reaction to a tattoo is assessed, most skin allergy units apply their national standard series, a specific metal series, and a textile colorant series, as well as the problem ink. The standard metal series is used because inks can contain pigments manufactured from inorganic metal salts. However, metal salts are being replaced by organic molecules, such as azo pigments and quinacridones.15–17 The specific battery for textile colorants includes several pigments, although we cannot be certain that it contains those used in the tattooed patient. The usefulness of patch tests in this context is not completely clear. In most cases the result is negative or inconsistent, and cases with positive results and an appropriate clinical correlation are exceptional.18 This lack of sensitivity and specificity of patch tests generates several hypotheses. Some authors propose that negative results could be attributed to poor penetration of the skin by the ink used in the tests. Others state that the real allergen is not present in its native form in inks and believe that these are formed in the dermis after a haptenization process. Sunlight could play a role in the process, thus explaining photosensitivity in reactions to tattoos.18–21

Despite the data reported above, we believe that patch testing should be used in all patients with a late reaction to a tattoo, compatible symptoms, and biopsy-proven lichenoid interface dermatitis. In such cases, only the ink and the standard series should be used, since neither the metal series nor the textile series has proven useful.18 Conceptually, intradermal tests would be more specific for the assessment of these patients, although they are not without risks and their use is not ethical.22–24 We hope that in the future, more information on the composition of the inks will enable us to perform patch testing to detect allergens with greater accuracy and thus demonstrate the compound responsible for allergic contact dermatitis.

The authors declare that they have no conflicts of interest.