Chemical characterization of inks in skin reactions to tattoo
aSorbonne Université, Hôpital Rothschild, Service de Gériatrie-Plaies et Cicatrisation, Paris, France, bSorbonne Université, UPMC Paris 06, Institut National de la Santé et de la Recherche Médicale, Unité Mixte de Recherche S 1155, F-75020 Paris, France, cInstitut de Chimie Physique, Université Paris-Saclay, Orsay, France, dDiffAbs Beamline, Synchrotron SOLEIL, L'Orme des Merisiers, Départementale 128, 91190 Saint-Aubin, France, eSorbonne Université, CNRS, Laboratoire Lise UMR 8235, Paris, France, fSorbonne Université, Hôpital Tenon, Anatomie et Cytologie Pathologiques, Paris, France, gPrivate Practice, Paris, France, hSorbonne Université, Hôpital Tenon, Service de Dermatologie-Allergologie, Paris, France, iUniversité de Paris, Hôpital Cochin, Service de Dermatologie, Paris, France, jSorbonne Université, UMR 8233, Paris, France, kUniversité de Paris, Hôpital Bichat, Service d'Anatomie et Cytologie Pathologique, Paris, France, lUniversité de Paris, Hôpital Bichat, Service de Dermatologie, Paris, France, and mDepartment of Dermatology, Allergology and Venereology, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
*Correspondence e-mail: email@example.com
Skin reactions are well described complications of tattooing, usually provoked by red inks. Chemical characterizations of these inks are usually based on limited subjects and techniques. This study aimed to determine the organic and inorganic composition of inks using X-ray fluorescence spectroscopy (XRF), X-ray absorption spectroscopy (XANES) and Raman spectroscopy, in a cohort of patients with cutaneous hypersensitivity reactions to tattoo. A retrospective multicenter study was performed, including 15 patients diagnosed with skin reactions to tattoos. Almost half of these patients developed skin reactions on black inks. XRF identified known allergenic metals – titanium, chromium, manganese, nickel and copper – in almost all cases. XANES spectroscopy distinguished zinc and iron present in ink from these elements in endogenous biomolecules. Raman spectroscopy showed the presence of both reported (azo pigments, quinacridone) and unreported (carbon black, phtalocyanine) putative organic sensitizer compounds, and also defined the phase in which Ti was engaged. To the best of the authors' knowledge, this paper reports the largest cohort of skin hypersensitivity reactions analyzed by multiple complementary techniques. With almost half the patients presenting skin reaction on black tattoo, the study suggests that black modern inks should also be considered to provoke skin reactions, probably because of the common association of carbon black with potential allergenic metals within these inks. Analysis of more skin reactions to tattoos is needed to identify the relevant chemical compounds and help render tattoo ink composition safer.
Tattooing refers to introducing pigmented material into the dermis by puncturing the skin, to obtain a permanent symbolic or decorative design (Wohlrab et al., 2007). This practice dates back thousands of years, and probably started around the Neolithic period (Oeggl et al., 2007). In Europe, the number of tattooed people is estimated to be about 100 million (Kluger et al., 2019). Although the tattoo fashion has thus been around for centuries, the question of its health impact has really arisen lately (Bäumler, 2016). In fact, tattoos are associated with various complications, with allergic reactions dominating (Islam et al., 2016; Serup et al., 2015), which can strongly influence quality of life (Hutton Carlsen & Serup, 2015).
Tattoo inks are composed of pigments combined with additives such as dispersants and preservatives (Liszewski & Warshaw, 2019; Timko et al., 2001). Once the tattoo is performed, skin tissue is permanently exposed to these substances for life. Nowadays, natural pigments are very rarely used and most tattoo inks are synthetic, allowing cheaper mass production and a large variety of shades.
Tattoo ink composition is poorly regulated worldwide. In Europe, only a few recent guidelines exist. In 2008, the European Council passed a resolution regarding the composition of tattoo inks and permanent make-up (Piccinini et al., 2008). However, Hauri et al. reported that up to one third of tattoo inks contain prohibited pigments, illustrating the fact that these recommendations are unfortunately poorly observed (Hauri & Hohl, 2015; Desmedt et al., 2022).
Compounds in tattoo inks may be classified into two types: inorganic substances, such as iron, zinc and titanium and their oxides (Tighe et al., 2017), and organic substances, including polycyclic aromatic hydrocarbons (PAHs) mainly found in black inks (Regensburger et al., 2010), organic azo-colorants and quinacridone found in colored inks (Laux et al., 2016). Once in the skin, the original pigments can undergo metabolic changes, leading to the formation of other compounds, such as primary aromatic amines (PAA), cleavage products of azo-colorants (Laux et al., 2016). These metabolic changes are suspected to be the consequences of various phenomena, including enzymatic activity, thermal decomposition, or photodegradation occurring after UV light exposure or laser tattoo removal (Cui et al., 2005; Engel et al., 2007; Vasold et al., 2004). Some of these inorganic and organic compounds, as well as cleavage products of the latter, are known to be allergenic, carcinogenic, mutagenic and reproduction-toxic substances (Laux et al., 2016; Manso et al., 2019). Studies analyzing the relationship between color and allergic reactions show that colored tattoos, especially red, cause more allergic reactions than black (Serup et al., 2020; Kluger, 2016).
Previous skin biopsy studies of in situ tattoo ink reactions identified multiple compounds: quinacridones (Serup et al., 2020; Gaudron et al., 2015), azo-pigments (Serup et al., 2020; Gaudron et al., 2015; Schreiver et al., 2017; Hutton Carlsen et al., 2016), phthalocyanine (Serup et al., 2020) and metal oxides, usually used as shading agents (Serup et al., 2020; Schreiver et al., 2017, 2019). Using sector field inductively coupled plasma mass spectrometry, Forte et al. (2009a) identified and quantified Cr (highest concentration, 315–4720 ng g−1), Ni (37.5–2318 ng g−1) and Cd (6.67–1150 ng g−1) in 13 tattoo inks, with Cr and Ni being well known allergenic metals (Thyssen et al., 2011). Involvement of these metals in skin reactions to tattoos has also been shown in vivo: De Cuyper et al. (2017), for example, have reported a patient with papulo-nodular infiltration of his tattoo that disappeared within a week after surgical removal of metal osteosynthesis implants. Nickel, chromium and titanium were identified both within the skin biopsy and the implant. However, these studies were either limited to a few techniques, or included a limited number of patients.
Wide-ranging chemical characterizations have been performed directly on tattoo inks (Arl et al., 2019; Poon et al., 2008; Forte et al., 2009a), pigments (Yakes et al., 2016) or skin samples from tattooed animals (Poon et al., 2008). Unfortunately, these studies are not directly relevant to human skin reactions; because of the specific metabolic changes tattoo inks undergo once introduced into human skin, it is clearly more pertinent to perform analyses on human skin samples themselves.
The present study aims to determine both the organic and inorganic chemical composition of inks in skin samples using multi-scale and multi-technique approaches on a cohort of patients with cutaneous reactions to tattoos. A specific approach has already been developed in related works and applied to different kinds of abnormal deposits identified in different parts of the human body (Bazin et al., 2012, 2016). We have also applied this methodology to skin (Colboc et al., 2016, 2019a,b, 2020, 2022a,b) as described below. Elemental composition and distribution of chemical elements can be determined using X-ray fluorescence (XRF) (Rouzière et al., 2016; Reguer et al., 2016), while chemical analyses are performed using a combination of Raman (Lucas et al., 2022) and X-ray absorption spectroscopy (mainly XANES, X-ray absorption near-edge structure) (Bazin et al., 2022a).
This study complied with Good Clinical Practices and the Declaration of Helsinki, and French law. We conducted a retrospective multicenter study in Tenon and Bichat Hospitals (Paris, France) and the Skin and Allergies Hospital (Helsinki, Finland) from January 2001 to December 2017. Patients diagnosed with cutaneous reactions to tattoo were included and their clinical and histopathological data collected.
Skin biopsy samples from these patients were all centralized in the Department of Pathology of Tenon Hospital. For each sample, 1.5 µm-thick sections from paraffin-embedded skin biopsies were deposited on both glass slides for hematoxylin–eosin–saffron (HES) staining and low-e microscope slides (MirrIR, Kevley Technologies, Tienta Sciences, Indianapolis, IN, USA) for Raman spectroscopy (Dessombz et al., 2011; Bazin & Daudon, 2015). XRF and XANES measurements were performed directly on the whole thickness of the paraffin-embedded tissue blocks. The entire block was analyzed except for three larger samples where a selected part of the sample was studied.
All X-ray based synchrotron experiments were performed at room temperature and atmospheric pressure on the DiffAbs beamline at the SOLEIL synchrotron facility (France), which offers multi-technique and multi-scale characterization of various materials (Reguer et al., 2016). This beamline provides a monochromatic X-ray beam, tunable in the 3–23 keV energy range. The Si(111) double-crystal monochromator is positioned between two cylindrical vertically focusing mirrors in order to monochromatize and focus the beam to about 300 µm diameter. A secondary optical system consisting of two trapezoidal orthogonally positioned curved mirrors under grazing incidence allows the X-ray beam to be focused down to a 5–10 µm-diameter beam spot with a flux close to 1010 photons s−1 at the sample position. In the present study, all samples were analyzed using a beam spot of 8 µm × 4 µm (horizontal × vertical, full width at half-maximum). XRF and XANES combined measurements were applied to highlight the localization and distribution of the chemical elements as well as the local environment of probed atoms (Dessombz et al., 2013; Bazin et al., 2022a,b).
Points of interest were selected on XRF maps, to measure Zn and Fe K-edge XANES spectra. The energy range for the spectral acquisition at the Zn K-edge was selected between 9.63 and 10 keV, with an energy step of 1 eV in the pre-edge and post-edge domain but 0.5 eV near the edge and a 1 s acquisition time. This K-edge results in a Zn signal more prominent than that of other elements, such as Ca or Ti, but does not indicate higher Zn content. Various correction procedures have to account for the self-absorbing matrix and the fact that measurements have been performed in air, the nature of the matrix, absorption by air, incident beam energy, and the ionization and X-ray emission cross sections associated with each element (Bazin et al., 2021).
The energy range for the spectral acquisition at the Fe K-edge was selected between 7.050 and 7.255 keV with an energy step of 1 eV in the pre-edge and post-edge region but 0.5 eV close to the edge and a 1 s acquisition time. XRF and XANES spectra were acquired using a four-element silicon drift detector (Vortex-ME4, Hitachi). Particular attention was paid to the data analysis which takes into account the shape and position of the edge, and the presence of pre-edge and post-edge first oscillation. These features indicate the effective charge of the atoms as well as the spatial distribution of atoms around Fe and Zn (Bazin et al., 2022a).
Raman spectroscopy was performed in order to characterize the organic as well as inorganic substances present within the tattoo inks (Yakes et al., 2016). It also allowed us to determine the chemical phase in which some of the metals are engaged, such as Ti cations which can be present as anatase or rutile (Frank et al., 2012). Spectra were collected using a micro-Raman system (LabRam HR-800 Evolution, Horiba, Japan) with a 785 nm laser excitation wavelength, 100× objective (Olympus, numerical aperture 0.9) and 300 grooves per mm grating (Lucas et al., 2022; Tamosaityte et al., 2021; Carden et al., 2003).
Raman spectra of skin biopsy may display strong auto-fluorescence (as do many biological samples with protein matrices). In such cases, Raman signatures can be dominated or even totally masked by the fluorescence background, making their detection and analysis impossible. This experimental complication dictates a need to pay special attention to the wavelength of the incident laser.
Fifteen patients, mean age 32 [20–62], 9 men and 6 women, were included. All had developed papular erythematous eruption [Figs. 1(a), 1(c) and 1(d)] or lichenoid eruption [Fig. 1(b)] associated with itching and swelling 0.5 to 60 months following the tattoo procedure. Clinical and histopathological data are summarized in Table 1. Red ink was involved in 7 cases, black in 7, and blue in 1. These reactions were confined to one specific color when the tattoo contained multiple colors [Fig. 1(c)], suggesting the presence of a specific sensitizer compound in one of the inks used. Optical microscopy always identified the presence of intra-dermal colored pigments, associated with the histological reaction pattern which presented as lichenoid (6), lympho-histiocytic (5) or granulomatous (4).
In a critical review, Thum et al. showed that inflammatory complications related to tattooing are associated with a wide spectrum of histological reaction patterns. Consistent with what they described, and probably because of the variety of chemical substances present in modern tattoo inks, we could not directly associate a type of histopathological reaction to a specific color (Thum & Biswas, 2015).
One of the major strengths of µXRF is its ability to simultaneously determine and locate a large number of major and trace elements in a biopsy (Estève et al., 2016a,b, 2017). Furthermore, because we analyzed the full thickness of the paraffin-embedded skin biopsies, we were able to gather information on the entire sample (in contrast to Raman analysis performed on sections only 1.5 µm thick).
Various parameters allowed a correlation between optical microscopy results and XRF spectra. Calcium was used as a marker of the epidermis within the skin biopsy, as one of its layers, the stratum granulosum, is known to have a high concentration of calcium (Lee & Lee, 2018). Correlation was also made between skin vessels observed in HES staining and the linear Fe signal in XRF maps.
XRF maps performed at the micrometric scale clearly revealed the distribution of major elements such as Ca, Fe and Zn in most of the samples (Fig. 2). In addition, various minor elements were identified for the 15 patients: Ti in 11, Ni in 8, Cr in 8, Mn in 7, Cu in 7 and As in 1 (Figs. 3 and 4). These metals can have various origins in tattoo inks. For example, recent studies showed that Ti and Cu (among others) are used as colorants and shading, whereas As, Cr and Ni (again among others) tend to be contaminants (Forte et al., 2009a,b).
These results can be compared with the literature. In an investigation performed on 226 inks using X-ray emission spectroscopy, 15 metals and one halogen were present above the limit of quantification (Tighe et al., 2017). In order of abundance, these were Ti, Fe, Cr, Cu, Zr, Mn, Br, Ni, Nb, Sr, Zn, Ba, Mo, Pb, V and W (Tighe et al., 2017). The medical literature indicates quite clearly that several of these elements, namely Ti, Cr, Mn, Ni and Cu, all of which are commonly found in tattoo inks, are associated with allergic reactions (Thyssen et al., 2011). There is a case of a patient with a local allergic reaction to tattoos associated with systemic symptoms disappearing a week after removal of a metallic implant. XRF analysis showed that both the skin biopsy and the implant contained high levels of Ni, Cr and Ti, suggesting the role of these metals in both the local and the systemic clinical manifestations (Cuyper et al., 2017).
Some of the metals identified, such as Zn, Fe, Ca, Cu and Cr, are naturally present in human skin, and their detection can therefore reflect endogenous or exogenous origin, while others, such as Ti, Mn, Ni and As, are exclusively exogenous. The micrometre size of the incident monochromatic beam reveals the spatial distribution of these different elements, allowing precise determination of their origin.
Together with correlation to the location of pigment and inflammation observed by optical microscopy, this gives a clue to the exogenous or endogenous origin of Zn, Fe, Ca, Cu and Cr. Indeed, diffuse and homogeneous distribution of these elements probably reflects endogenous origin while a few intense spots co-located with the pigments probably indicates exogenous origin. This can be illustrated by the distribution, from XRF, of the two elements Zn and Ti. Let us discuss the biological significance of these two elements.
Two different Zn distribution patterns are observed. It may be diffusely distributed in the skin tissue and co-located with the inflammatory infiltrate – suggesting an endogenous origin [Fig. 5(a)] – or diffuse within the skin tissue but with a few intense areas of high Zn concentration, co-localized to the tattoo pigment – suggesting a combination of endogenous and exogenous origin within the same sample [Fig. 5(b)].
Zinc is an essential micronutrient (around 2–3 g for a 72 kg body) involved in the regulation of the innate and adaptive immune responses (Gammoh & Rink, 2017). The spatial correlation of Zn in our samples with the dermal inflammatory infiltrate highlights the role of this element in inflammatory responses.
The distribution of Ti, a metal not normally present in the human body, is also interesting. In the samples studied, Ti is usually co-located to the pigment as clearly shown in the XRF maps of patients 5, 6 and 11 [Figs. 6(a), 6(b) and 6(c)], but for one patient (patient 2) Ti was remarkably located in one area of the deeper dermis [Fig. 6(d)]. As a hair follicular orifice was visible directly above this by optical microscopy (Fig. 3, arrow head), Ti appeared aggregated around a hair follicle bulb in this case.
In this specific case, we can first hypothesize that Ti arises from the tattoo ink and has migrated through the dermis towards the hair follicle after tattooing. Indeed, a migration phenomenon from tattooed skin to lymph nodes has been shown previously for various organic and inorganic tattoo ink substances, including metals such as Ti (Schreiver et al., 2017; Hayakawa et al., 2018), Ni (Schreiver et al., 2019) and Hg (Hayakawa et al., 2018). Another possibility is that Ti arises from cosmetics applied to the tattoo by the patient, possibly sunscreen, often used to provide sun protection to the tattoo following the procedure and known to contain nanosized particles of Ti (Newman et al., 2009). Indeed, Ti penetration through the hair follicular orifice has been demonstrated (Lademann et al., 1999), which might therefore cause Ti accumulation around the hair follicle in patient 2. Furthermore, Ti is suspected of being associated with frontal fibrosing alopecia, an inflammatory skin disease (Kidambi et al., 2017). In this hypothesis, the association of the sensitizer compounds in the tattoo ink, and Ti from topical application on the tattoo, might have both contributed to the hypersensitivity phenomenon, and potentiated each other, showing the complexity and multifactorial processes underlying these reactions.
Although the distribution of the metal within the skin biopsy gives us clues regarding its origin as endogenous or exogenous, this should be confirmed by analysis of its precise chemical structure. Indeed, some of the chemical phases in which Zn and Fe are engaged – such as zincite – are not naturally present within the human body, proving that their presence must be due to tattoo inks. To confirm this, XANES measurements were performed on four skin biopsies (patients 2, 8, 13 and 15) at the Zn K-edge and on six (patients 2, 5, 10, 11, 13 and 14) at the Fe K-edge.
For example, four XANES spectra were measured at the Zn K-edge on points of interests (POI) in the XRF maps of patients 2 and 8 in order to identify the Zn environment (Fig. 7). Some spectra identify zincite (ZnO), such as three spectra of patient 8. Other spectra (one from patient 8, and those acquired on all other patients) show the maximum of the white line shifted to lower-energy and less structured oscillations, similar in appearance to those observed in human parkin protein (Konovalov et al., 2020), which contains multiple Zn binding sites, suggesting endogenous Zn associated with dermal proteins.
However, a difference is observed as the shoulder at 7150 eV is visible for patients 10 and 7, while it is less visible for patients 11 and 14. In addition, the first oscillation at 7180 eV, related to oxygen neighbors, shows a modified shape. The overall shape of the spectra is similar to that of iron oxy-hydroxides, such as goethite or ferrihydrite compounds, with Fe3+ in an octahedral environment. Exogenous Fe oxides, such as magnetite or hematite oxides that have very specific spectral profiles, might be present in these cases (Rueff et al., 2004; Wilke et al., 2001).
Raman spectroscopy identified six different pigments. Black tattoos were always composed of carbon black, whereas various pigments were identified in colored tattoos. Red tattoos were composed of PR170 (an azo pigment) (where PR is pigment red) in three cases (Scherrer et al., 2009), PR2 (azo pigment) in two cases (Fig. 9), PR202 (quinacridone) in one case and PR122 (quinacridone) in one case (Longoni & Bruni, 2021) (Fig. 10). The blue tattoo was composed of PB15 (PB is pigment blue) (phthalocyanine) (Fig. 11).
Black tattoos are the most common, and their main component is carbon black (Jacobsen & Clausen, 2015). Manufacturing of black inks usually involves thermal combustion of feedstock oil, a process also associated with the formation of PAHs (Regensburger et al., 2010). Although the carcinogenic potential of PAHs is well known, they are, unlike metals and colored pigments, usually not associated with hypersensitivity reactions to tattoos. In our patients, carbon black was identified in most cases, but, with one exception, it was always associated with a known allergenic metal. The occurrence of skin reactions in this one case (patient 9) is therefore quite unexpected and highlights the complexity of these phenomena. Further studies, analyzing the skin allergenic potential of PAHs, are needed, especially since data suggest that these substances stimulate inflammatory responses and enhance allergic reactions (Nel et al., 1998).
Azo pigments are the ones most commonly implicated in skin reactions to tattoos (Serup et al., 2020). Consistent with our findings, quinacridone has also been reported in hypersensitivity reactions (Greve et al., 2003; Hutton Carlsen et al., 2016). We also identified a phthalocyanine in one patient, which, to the best of our knowledge, has never been clearly identified in hypersensitivity reaction to tattoo. However, in all the cases except one, these colored pigments were associated with at least one metal with known allergenic properties (either Ti, Cr, Mn, Ni and/or Cu), making the exact identification of the causative compounds impossible without thorough allergenic exploration.
A list of carcinogenic chemicals that should be excluded from tattoo inks was published in a European resolution in 2008 (Council of Europe, 2008). Although we did not identify any prohibited pigments in our patients, this has not been the case in other studies. A study on 396 tattoos and 55 permanent make-up inks taken from the Swiss market between 2009 and 2017 identified 28 different pigments, four of them prohibited (Niederer et al., 2017). These authors also compared the pigments found with those declared on tattoo ink labels and convincingly showed that banned pigments are rarely declared, but rather masked by listing non-present legal pigments and label forging. These results highlight the importance of performing rigorous chemical analysis, like ours, of skin reactions to tattoos.
Ti was the metal the most often associated with colored pigments (Figs. 9, 10 and 11), as it was found in all cases except one. Ti is an efficient photocatalyst and can produce reactive oxygen species under UVA irradiation (Chen & Mao, 2007; Ancona et al., 2018; Wydra et al., 2015). In our patients, this metal might have potentiated photodegradation of these colored pigments and promoted metabolic changes, possibly contributing to the formation of sensitizing substances.
Because the photocatalytic properties of these metal oxide deposits depend on their size and form, it is important to provide precise chemical information, as we did using Raman spectroscopy. This allowed us to identify anatase (with a characteristic strong band at 145 cm−1 and weaker bands at 395, 517 and 641 cm−1) in four cases (Frank et al., 2012; Wypych et al., 2014). Recent investigations emphasize that anatase and amorphous forms of nano-TiO2 show greater cytotoxicity than the rutile form (Xue et al., 2010; Huang et al., 2017).
In a previous study, we identified the association of PR170 with metal oxides (either Ti, Fe or Zn) in three cases of keratoacanthoma, a type of skin tumor (Colboc et al., 2020). The fact that, in the present study, the same pigment with metal oxide was associated with hypersensitivity but not carcinogenic processes highlights the pathogenic complexity of skin reactions to tattoos.
Interestingly, Raman spectroscopy failed to identify Ti in seven other cases in which it had been identified by XRF. This may be due to strong (auto)fluorescence background, as mentioned earlier. Another explanation may be that, because Raman spectroscopy is performed on a 1.5 µm-thick section of the paraffin-embedded biopsy (and not on the whole biopsy, in contrast to XRF), it failed to identify some of the chemical substances that were indeed present. This highlights once again the importance of using various complementary techniques for precise chemical characterization of tattoo inks, XRF being highly sensitive for elemental detection, while Raman provides information on the nature (composition, structure) of the inorganic (metal oxide) and of the organic compounds.
Because we wanted our study to have a consistent approach from a clinical point of view, we focused on hypersensitivity reactions after the tattoo, thus excluding several cases of infection and skin cancer encountered during this period in the collaborating dermatological centers. This focus necessarily limited the number of samples.
The possibility of chemical modifications of the inks after sample fixation with formol and embedding in paraffin must also be considered. Unfortunately all routine human skin biopsies undergo this process, so such modifications cannot be excluded. However, it is worth pointing out that the color of the formol containing the sample is unchanged after fixation, excluding at least massive dilution of the tattoo inks. Besides, the pigments identified by our techniques faithfully correspond to the color of the tattoo, evidence that their chemistry is unaltered by fixation.
The samples themselves present some challenges. These include: poor regulation of tattoo ink composition, undocumented use of ink mixtures from different manufacturers, and possible photodegradation perhaps accelerated by metal oxide catalysis; these must be acknowledged as significant difficulties in such research.
We report Raman spectroscopic, XRF and XANES results from skin reactions to tattoos from, to the best of our knowledge, the largest patient cohort yet analyzed. Using these innovative and complementary techniques, we were able to provide wide-ranging information on the organic and inorganic compounds found in skin reactions to tattoos. In each of our patients, we identified various suspected allergens, highlighting the necessity of partnering such studies with allergenic investigations.
XRF identified the metals Fe, Zn, Ti, of Ni, Cr, Mn, Cu and As, and Raman spectroscopy the organic pigments carbon black, PR170, PR2, PR122, PR202 and PB15. Raman spectroscopy also identified the TiO2 polymorph, anatase. XANES spectroscopy distinguished exogenous Zn and Fe from these elements of endogenous origin.
Our study also highlights the complexity of tattoo-related reactions and physiological responses: these involve multiple different colored pigments, a variety of organic and inorganic compounds, and diverse histopathological and clinical presentations. Pathogenesis is likely to be multifactorial, including needle trauma during the tattoo procedure, inflammatory reactions, UV radiation, and individual and lifestyle background.
In contrast to previous studies, almost half our patients had reactions to black tattoos, possibly because carbon black was usually associated with allergenic metals such as Ti, Cr and Ni. These results show that modern black inks should therefore be considered as provocative of hypersensitivity reactions as colored inks, as they are unlikely to be pure carbon black.
Additional analyses of skin reactions to tattoos using the same techniques are needed to allow a more reliable definition of the allergenic factors and so help render tattoo ink composition safer.
We acknowledge SOLEIL for provision of synchrotron radiation facilities (proposal Nos. 20170374 and 20180362) and we thank the DiffAbs beamline team for assistance during experiments.
We thank the clusters of excellence MiChem and Matisse (ANR-11-IDEX-0004-02) led by Sorbonne Université.
Ancona, A., Dumontel, B., Garino, N., Demarco, B., Chatzitheodoridou, D., Fazzini, W., Engelke, H. & Cauda, V. (2018). Nanomaterials, 8, 143. Web of Science CrossRef Google Scholar
Arl, M., Nogueira, D. J., Schveitzer Köerich, J., Mottim Justino, N., Schulz Vicentini, D. & Gerson Matias, W. (2019). J. Hazard. Mater. 364, 548–561. Web of Science CrossRef CAS PubMed Google Scholar
Bäumler, W. (2016). Dtsch Arztebl Int. 113, 663–664. Web of Science PubMed Google Scholar
Bazin, D. & Daudon, M. (2015). Ann. Biol. Clin. 73, 517–534. CAS Google Scholar
Bazin, D., Daudon, M., Combes, C. & Rey, C. (2012). Chem. Rev. 112, 5092–5120. Web of Science CrossRef CAS PubMed Google Scholar
Bazin, D., Foy, E., Reguer, S., Rouzière, S., Fayard, B., Colboc, H., Haymann, J. P., Daudon, M. & Mocuta, C. (2022b). C. R. Chim. 24(S1), 165–188. Google Scholar
Bazin, D., Jouanneau, C., Bertazzo, S., Sandt, C., Dessombz, A., RéFrégiers, M., Dumas, P., Frederick, J., Haymann, J.-P., Letavernier, E., Ronco, P. & Daudon, M. (2016). C. R. Chim. 19, 1439–1450. Web of Science CrossRef CAS Google Scholar
Bazin, D., Papoular, R. J., Elkaim, E., Weil, R., Thiaudière, D., Pisapia, C., Ménez, B., Hwang, N., Tielens, F., Livrozet, M., Bouderlique, E., Haymann, J.-P., Letavernier, E., Hennet, L., Frochot, V. & Daudon, M. (2021). C. R. Chim. 25(S1), 343–354. Google Scholar
Bazin, D., Reguer, S., Vantelon, D., Haymann, J. P., Letavernier, E., Frochot, V., Daudon, M., Esteve, E. & Colboc, H. (2022a). C. R. Chim. 25(S1), 189–208. Google Scholar
Carden, A., Rajachar, R. M., Morris, M. D. & Kohn, D. H. (2003). Calcif. Tissue Int. 72, 166–175. Web of Science CrossRef PubMed CAS Google Scholar
Chen, X. & Mao, S. S. (2007). Chem. Rev. 107, 2891–2959. Web of Science CrossRef PubMed CAS Google Scholar
Colboc, H., Bazin, D., Moguelet, P., Reguer, S., Amode, R., Jouanneau, C., Lucas, I., Deschamps, L., Descamps, V. & Kluger, N. (2020). J. Eur. Acad. Dermatol. Venereol. 34, e313–e315. Web of Science CrossRef CAS PubMed Google Scholar
Colboc, H., Bazin, D., Moguelet, Ph., Frochot, V., Weil, R., Letavernier, E., Jouanneau, Ch., Francès, C., Bachmeyer, C., Bernaudin, J.-F. & Daudon, M. (2016). C. R. Chim. 19, 1631–1641. Web of Science CrossRef CAS Google Scholar
Colboc, H., Moguelet, P., Bazin, D., Bachmeyer, C., Frochot, V., Weil, R., Letavernier, E., Jouanneau, C., Daudon, M. & Bernaudin, J. F. (2019a). J. Eur. Acad. Dermatol. Venereol. 33, 198–203. Web of Science CrossRef CAS PubMed Google Scholar
Colboc, H., Moguelet, P., Bazin, D., Carvalho, P., Dillies, A.-S., Chaby, G., Maillard, H., Kottler, D., Goujon, E., Jurus, Ch., Panaye, M., Frochot, V., Letavernier, E., Daudon, M., Lucas, I., Weil, R., Courville, Ph., Monfort, J. B., Chasset, F. & Senet, P. (2019b). JAMA Dermatol. 155, 789–796. Web of Science CrossRef PubMed Google Scholar
Colboc, H., Moguelet, Ph., Letavernier, E., Frochot, V., Bernaudin, J.-F. R., Weil, S., Rouzière, P., Seneth, C., Bachmeyeri, N., Laporte, I., Lucas, V., Descamps, R., Amodek, F., Brunet-Possentik, N., Kluger, L., Deschamps, A., Dubois, S., Reguer, A., Somogyi, K., Medjoubi, M., Refregiers, M., Daudon & Bazin, D. (2022b). C. R. Chim. 25(S1), 445–476. Google Scholar
Colboc, H. J., Fontaine, D., Bazin, V., Frochot, E., Letavernier, M., Daudon, N., Laporte, S., Rouzière, M., Reby, A., Galezowski, Ch., Forasassi, S. & Meaume, S. (2022a). J. Gerontol. A Biol. Sci. Med. Sci. 77, 27–32. Web of Science CrossRef PubMed Google Scholar
Council of Europe (2008). Resolution ResAP(2008)1: on requirements and criteria for the safety of tattoos and permanent make-up, https://rm.coe.int/16805d3dc4. Google Scholar
Cui, Y., Churchwell, M. I., Couch, L. H., Doerge, D. R. & Howard, P. C. (2005). Drug Metab. Dispos. 33, 1459–1465. Web of Science CrossRef PubMed CAS Google Scholar
Cuyper, C. de, Lodewick, E., Schreiver, I., Hesse, B., Seim, C., Castillo-Michel, H., Laux, P. & Luch, A. (2017). Contact Dermatitis, 77, 397–405. Web of Science PubMed Google Scholar
Desmedt, B., Vanhamme, M. & Deconinck, E. (2022). Regul. Toxicol. Pharmacol. 129, 105123. Web of Science CrossRef PubMed Google Scholar
Dessombz, A., Bazin, D., Dumas, P., Sandt, C., Sule-Suso, J. & Daudon, M. (2011). PLoS One, 6, e28007. Web of Science CrossRef PubMed Google Scholar
Dessombz, A., Nguyen, C., Ea, H. K., Rouzière, S., Foy, E., Hannouche, D., Réguer, S., Picca, F. E., Thiaudière, D., Lioté, F., Daudon, M. & Bazin, D. (2013). J. Trace Elem. Med. Biol. 27, 326–333. Web of Science CrossRef CAS PubMed Google Scholar
Engel, E., Spannberger, A., Vasold, R., König, B., Landthaler, M. & Bäumler, W. (2007). JDDG, 5, 583–589. CrossRef PubMed Google Scholar
Estève, E., Bazin, D., Jouanneau, C., Rouzière, S., Bataille, A., Kellum, A., Provost, K., Mocuta, C., Reguer, S., Jorissen, K., Rehr, J. J., Hertig, A., Rondeau, E., Letavernier, E., Haymann, J. P., Daudon, M. & Ronco, P. (2016a). C. R. Chim. 19, 1586–1589. Google Scholar
Estève, E., Bazin, D., Jouanneau, C., Rouzière, S., Bataille, A., Kellum, A., Provost, K., Mocuta, C., Reguer, S., Thiaudière, D., Jorissen, K., Rehr, J. J., Hertig, A., Rondeau, E., Letavernier, E., Daudon, M. & Ronco, P. (2016b). C. R. Chim. 19, 1580–1585. Google Scholar
Esteve, E., Reguer, S., Boissiere, C., Chanéac, C., Lugo, G., Jouanneau, C., Mocuta, C., Thiaudière, D., Leclercq, N., Leyh, B., Greisch, J.-F., Berthault, J., Daudon, M., Ronco, P. & Bazin, D. (2017). J. Synchrotron Rad. 24, 991–999. Web of Science CrossRef CAS IUCr Journals Google Scholar
Forte, G., Petrucci, F., Cristaudo, A. & Bocca, B. (2009a). Open Chem. Biomed. Methods J. 2, 42–47. CrossRef CAS Google Scholar
Forte, G., Petrucci, F., Cristaudo, A. & Bocca, B. (2009b). Sci. Total Environ. 407, 5997–6002. Web of Science CrossRef PubMed CAS Google Scholar
Frank, O., Zukalova, M., Laskova, B., Kürti, J., Koltai, J. & Kavan, L. (2012). Phys. Chem. Chem. Phys. 14, 14567–14572. Web of Science CrossRef CAS PubMed Google Scholar
Gammoh, N. Z. & Rink, L. (2017). Nutrients, 9, 624. Web of Science CrossRef Google Scholar
Gaudron, S., Ferrier-Le Bouëdec, M. C., Franck, F. & D'Incan, M. (2015). Contact Dermatitis, 72, 97–105. Web of Science CrossRef CAS PubMed Google Scholar
Greve, B., Chytry, R. & Raulin, C. (2003). Contact Dermatitis, 49, 65–266. Web of Science CrossRef Google Scholar
Hauri, U. & Hohl, C. (2015). Curr. Probl. Dermatol. 48, 164–169. CrossRef PubMed Google Scholar
Hayakawa, A., Sano, R., Takei, H., Takahashi, Y., Kubo, R., Tokue, H., Hirasawa, S., Shimada, T., Awata, S., Yuasa, M., Uetake, S., Akuzawa, H. & Kominato, Y. (2018). Leg. Med. 35, 9–11. Web of Science CrossRef CAS Google Scholar
Huang, Y.-W., Cambre, M. & Lee, H.-J. (2017). Int. J. Mol. Sci. 18, 2702. Web of Science CrossRef Google Scholar
Hutton Carlsen, K., Køcks, M., Sepehri, M. & Serup, J. (2016). Skin Res. Technol. 22, 460–469. Web of Science CrossRef CAS PubMed Google Scholar
Hutton Carlsen, K. & Serup, J. (2015). Skin Res. Technol. 21, 101–107. Web of Science CrossRef CAS PubMed Google Scholar
Islam, P. S., Chang, Ch., Selmi, C., Generali, E., Huntley, A., Teuber, S. S. & Gershwin, M. E. (2016). Clin. Rev. Allerg. Immunol. 50, 273–286. Web of Science CrossRef Google Scholar
Jacobsen, N. R. & Clausen, P. A. (2015). Curr. Probl. Dermatol. 48, 170–175. CrossRef PubMed Google Scholar
Kidambi, A. D., Dobson, K., Holmes, S., Carauna, D., Del Marmol, V., Vujovic, A., Kaur, M., Takwale, A., Farrant, P., Champagne, C., Harries, M. & Messenger, A. (2017). Br. J. Dermatol. 175, 46. Google Scholar
Kluger, N. (2016). Int. J. Dermatol. 55, 764–768. Web of Science CrossRef PubMed Google Scholar
Kluger, N., Misery, L., Seité, S. & Taieb, C. (2019). J. Am. Acad. Dermatol. 81, 607–610. Web of Science CrossRef PubMed Google Scholar
Konovalov, O. V., Novikova, N. N., Kovalchuk, M. V., Yalovega, G. E., Topunov, A. F., Kosmachevskaya, O. V., Yurieva, E. A., Rogachev, A. V., Trigub, A. L., Kremennaya, M. A., Borshchevskiy, V. I., Vakhrameev, D. D. & Yakunin, S. N. (2020). Materials, 13, 4635. Web of Science CrossRef Google Scholar
Lademann, J., Weigmann, H., Rickmeyer, C., Barthelmes, H., Schaefer, H., Mueller, G. & Sterry, W. (1999). Skin Pharmacol. Physiol. 12, 247–256. Web of Science CrossRef CAS Google Scholar
Laux, P., Tralau, T., Tentschert, J., Blume, A., Dahouk, S. A., Bäumler, W., Bernstein, E., Bocca, B., Alimonti, A., Colebrook, H., de Cuyper, Ch., Dähne, L., Hauri, U., Howard, P. C., Janssen, P., Katz, L., Klitzman, B., Kluger, N., Krutak, L., Platzek, Th., Scott-Lang, V., Serup, J., Teubner, W., Schreiver, I., Wilkniß, E. & Luch, A. (2016). Lancet, 387, 395–402. Web of Science CrossRef PubMed Google Scholar
Lee, S. E. & Lee, S. H. (2018). Ann. Dermatol. 30, 265–275. Web of Science CrossRef CAS PubMed Google Scholar
Liszewski, W. & Warshaw, E. M. (2019). J. Am. Acad. Dermatol. 81, 379–385. Web of Science CrossRef CAS PubMed Google Scholar
Longoni, M. & Bruni, S. (2021). J. Chem. Educ. 98, 966–972. Web of Science CrossRef CAS Google Scholar
Lucas, I. T., Bazin, D. & Daudon, M. (2022). C. R. Chim. 25(S1), 83–103. Google Scholar
Manso, M., Pessanha, S., Guerra, M., Reinholz, U., Afonso, C., Radtke, M., Lourenço, H., Carvalho, M. L. & Buzanich, A. G. (2019). Biol. Trace Elem. Res. 187, 596–601. Web of Science CrossRef CAS PubMed Google Scholar
Nel, A. E., Diaz-Sanchez, D., Ng, D., Hiura, T. & Saxon, A. (1998). J. Allergy Clin. Immunol. 102, 539–554. Web of Science CrossRef CAS PubMed Google Scholar
Newman, M. D., Stotland, M. & Ellis, J. I. (2009). J. Am. Acad. Dermatol. 61, 685–692. Web of Science CrossRef PubMed CAS Google Scholar
Niederer, M., Hauri, U., Kroll, L. & Hohl, C. (2017). F1000Res, 6, 2034. Google Scholar
Oeggl, K., Kofler, W., Schmidl, A., Dickson, J. H., Egarter-Vigl, E. & Gaber, O. (2007). Quaternary Sci. Rev. 26, 853–861. Web of Science CrossRef Google Scholar
Piccinini, P., Pakalin, S., Contor, L. & Bianchi, I. (2008). Safety of tattoos and permanent make-up. Adverse health effects and experience with the Council of Europe Resolution. Publications Office of the European Union. Google Scholar
Poon, K. W. C., Dadour, I. R. & McKinley, A. J. (2008). J. Raman Spectrosc. 39, 1227–1237. Web of Science CrossRef CAS Google Scholar
Regensburger, J., Lehner, K., Maisch, T., Vasold, R., Santarelli, F., Engel, E., Gollmer, A., König, B., Landthaler, M. & Bäumler, W. (2010). Exp. Dermatol. 19, e275–e281. Web of Science CrossRef PubMed Google Scholar
Reguer, S., Mocuta, C., Thiaudière, D., Daudon, M. & Bazin, D. (2016). C. R. Chim. 19, 1424–1431. Web of Science CrossRef CAS Google Scholar
Rouzière, S., Bazin, D. & Daudon, M. (2016). C. R. Chim. 19, 1404–1415. Google Scholar
Rueff, J.-P., Journel, L., Petit, P.-E. & Farges, F. (2004). Phys. Rev. B, 69, 235107. Web of Science CrossRef Google Scholar
Scherrer, N. C., Stefan, Z., Francoise, D., Annette, F. & Renate, K. (2009). Spectrochim. Acta A Mol. Biomol. Spectrosc. 73, 505–524. Web of Science CrossRef PubMed Google Scholar
Schreiver, I., Hesse, B., Seim, Ch., Castillo-Michel, H., Anklamm, L., Villanova, J., Dreiack, N., Lagrange, A., Penning, R., De Cuyper, Ch., Tucoulou, R., Bäumler, W., Cotte, M. & Luch, A. (2019). Part. Fibre Toxicol. 16, 33. Web of Science CrossRef PubMed Google Scholar
Schreiver, I., Hesse, B., Seim, Ch., Castillo-Michel, H., Villanova, J., Laux, P., Dreiack, N., Penning, R., Tucoulou, R., Cotte, M. & Luch, A. (2017). Sci. Rep. 7, 11395. Web of Science CrossRef PubMed Google Scholar
Serup, J., Carlsen, K. H. & Sepehri, M. (2015). Tattooed Skin Health, 48, 48–60. CrossRef Google Scholar
Serup, J., Hutton Carlsen, K., Dommershausen, N., Sepehri, M., Hesse, B., Seim, C., Luch, A. & Schreiver, I. (2020). Contact Dermatitis, 82, 73–82. Web of Science CrossRef CAS PubMed Google Scholar
Tamosaityte, S., Pucetaite, M., Zelvys, A., Varvuolyte, S., Hendrixson, V. & Sablinskas, V. (2021). C. R. Chim. 25(S1), 73–82. Google Scholar
Thum, C. K. & Biswas, A. (2015). Am. J. Dermatopathol. 37, 54–66. Web of Science CrossRef PubMed Google Scholar
Thyssen, J. P., Menné, T., Schalock, P. C., Taylor, J. S. & Maibach, H. I. (2011). Br. J. Dermatol. 164, 473–478. Web of Science CAS PubMed Google Scholar
Tighe, M. E., Libby, D. K., Dorn, S. K., Hosmer, J. R. & Peaslee, G. F. (2017). J. Environ. Prot. 08, 1243–1253. CrossRef Google Scholar
Timko, A. L., Miller, C. H., Johnson, F. B. & Ross, V. (2001). Arch. Dermatol. 137, 143–147. Web of Science PubMed CAS Google Scholar
Vasold, R., Naarmann, N., Ulrich, H., Fischer, D., König, B., Landthaler, M. & Bäumler, W. (2004). Photochem. Photobiol. 80, 185–190. Web of Science CrossRef PubMed CAS Google Scholar
Wilke, M., Farges, F., Petit, P.-E., Brown, G. E. Jr & Martin, F. (2001). Am. Mineral. 86, 714–730. CrossRef CAS Google Scholar
Wohlrab, S., Stahl, J. & Kappeler, P. M. (2007). Body Image, 4, 87–95. Web of Science CrossRef PubMed Google Scholar
Wydra, R. J., Oliver, C. E., Anderson, K. W., Dziubla, T. D. & Hilt, J. Z. (2015). RSC Adv. 5, 18888–18893. Web of Science CrossRef CAS PubMed Google Scholar
Wypych, A., Bobowska, I., Tracz, M., Opasinska, A., Kadlubowski, S., Krzywania-Kaliszewska, A., Grobelny, J. & Wojciechowski, P. (2014). J. Nanomater. 2014, 124814. Web of Science CrossRef Google Scholar
Xue, C., Wu, J., Lan, F., Liu, W., Yang, X., Zeng, F. & Xu, H. (2010). J. Nanosci. Nanotechnol. 10, 8500–8507. Web of Science CrossRef CAS PubMed Google Scholar
Yakes, B. J., Michael, T. J., Perez–Gonzalez, M. & Harp, B. P. (2016). J. Raman Spectrosc. 48, 736–743. Web of Science CrossRef Google Scholar
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.