Abstract
Tattooed patients undergoing magnetic resonance imaging (MRI) can develop cutaneous complications during the procedure. Our aim was to review all published case reports on MRI-induced tattoo complications to identify a possible pattern. So far, 17 cases have been reported. Five (29%) of the cases were in cosmetic tattoos.
Symptoms are abrupt and painful with fast onset during MRI, sometimes requiring termination of the procedure. Clinical signs are absent or manifested as inflammation sensed as burning. No thermal skin burns have been recognized. Full recovery is fast, with no sequelae.
MRI-induced tattoo complications are uncommon. Patients with cosmetic and traditional tattoos can undergo routine MRI.
Introduction
Traditional tattoos and permanent make-up tattoos (PMU) have become mainstream through the past couple of decades. PMU is presently growing fast in popularity in many countries. It is estimated that some 15%–20% of Europeans have one or more tattoos, with the highest prevalence in the age group of 18–29 years (1). The tattoo trend is paralleled by an increasing number of tattoo complications needing medical treatment (2).
Magnetic resonance imaging (MRI) is often used in modern diagnostics, particularly in evaluating soft-tissue anatomic structures and a range of pathologies. MRI technology is evolving continuously with more powerful static magnetic fields, concomitant higher radio frequency transmission, and faster and stronger gradient magnetic fields. Awareness of foreign objects and particles in patients has always been in focus during MRI (3). Further to this, a relevant concern is tattoo adverse reactions and hazards following MRI since pigments used in commercial tattoo inks are foreign body particles in the nano- and micrometer range. Pigments can be organic or metallic, and tattoo inks often contain metallic contaminants. Thus, it is no surprise if tattoo pigments permanently deposited in the skin might be influenced by MRI.
Since the introduction of MRI in clinical routine occasional case reports of MRI-induced tattoo complications have been described. However, the literature is scarce and a systematic overview is not available. It is unanswered whether special precautions ought to be taken.
This is to our knowledge the first comprehensive review of the literature collecting all reported MRI-induced tattoo adverse events and hazards.
Material and Methods
References for this review were identified using the following index words in the PUBMED database: “MRI AND (tattoo*),” “MRI AND (permanent make-up),” and “MRI AND (permanent cosmetics).”
We included any published paper in any language reporting any cutaneous reactions or sensation in tattoos or PMU associated with MRI.
One investigator (Alsing et al.) independently assessed all publications. In the primary screening titles and abstracts were inspected for any reports of tattoo cutaneous sensations or reactions following MRI. Thereafter in the secondary screening, full-text versions of relevant articles were studied for final inclusion. Initially in the tertiary screening, reference lists of all enrolled articles were reviewed, and articles of interest studied.
Results
A search of the literature identified 55 citations, with inclusion of 15 papers reporting a total of 17 cases with sensory or/and cutaneous reactions after MRI scanning. All papers were included using the search index “MRI AND (tattoo*).” None of the two other search indices added any cases. No other relevant published literature was detected using Google Scholar.
More than half of MRI-induced reactions were reported in the period 2000–2010. Nine cases were women and eight were men (mean age = 31.9 years, age range = 18–62 years). The mean age of tattoos was 2.4 years (range = 4–10 years).
The reported tattoos with MRI-related events were mainly in dark tattoo colors, e.g. either dark blue or black pigment (9/15, 60%). Five (29%) of the reported cases were cosmetic tattoos, primarily permanent eyeliner. The remaining cases were traditional decorative tattoos located on the abdomen, back, upper arms, or lower extremities.
The tattoo shape and designs were predominantly thin lines either in curved shapes, i.e. lettering and symbols, or as straighter lines, e.g. eyeliners and lipliners. Only three reported designs had several curves and loops (4–6).
In most cases, it is not specified whether patients have undergone MRI before the current examination. But some cases had completed up to four identical MRI scans without symptoms before the reported event (7).
Of the 17 cases, 13 (76%) reported a burning sensation or heat in and around their tattoo. Other subjective sensations were described as “tingling,” “tightening,” or “stinging pain.” The severity of pain, if reported, was relatively strong, (visual analog scale [VAS] = 6–10. The pain seems to appear suddenly but also disappears as soon as the scanning terminates, or if the patient leaves the MRI suite.
Symptoms were primarily subjective with sudden onset of pain, sometimes fiery and necessitating immediate interruption of the examination. In 10 cases, this was accompanied by erythema and swelling in the tattoo and surrounding skin indicating a wheal and flare type of response. If any visual cutaneous reaction appeared, it was transient and remitted within 12–48 h (4,8–10).
Two papers rate the events as first-degree skin burns (4,11) and another as a second-degree burn (5). There has been no report of full thickness third-degree “burn” of a tattoo with necrosis and ulceration on exposure to MRI. There are no case reports of later distortion of the tattoo design, fade in tattoo color, ulceration, or scarring.
Discussion
The incidence of MRI-induced reactions in tattoos appears low albeit the events may have been neglected in the literature. The review identified five MRI-related events in PMU, 11 in a decorative tattoo, and one was not specified.
To investigate the incidence of MRI-related reaction in permanent cosmetics, Tope and Shellock (12) performed a survey study among 1032 patients who underwent MRI and identified 135 with PMU. Only 2/135 (1.5%) patients reported burning sensations and tingling during MRI (12). A more recent survey study by Callaghan et al. (13) estimated, using the Clopper–Pearson test, that the probability of an adverse reaction was 0.17%. These studies indicate that MRI-related complications in cosmetic tattoos are uncommon. However, MRI adverse events may be underreported since these patients have a medical problem of first priority at the forefront of their initial attention. Furthermore, the number of cases may rise in the future, due to the higher prevalence of tattoos among young people, who first need MRI examinations later in life.
It appears that PMU cause a reaction immediately when entering the MRI examination room or shortly after entering the static magnetic field (10–12) in distinction to most traditional tattoos reacting several minutes after initiation of the scan.
It is commonly believed in the medical literature that tattoo inks containing iron oxides are more likely to cause MRI-related cutaneous reactions due to their supposed special magnetic and conductive abilities. Iron oxide pigments were dominating in old school tattooing until the introduction of organic pigments a few decades ago. The iron oxide pigments have duller color tones, and therefore remain popular in inks made especially for PMU tattooing. However, iron oxide pigments are mineral pigments originating from natural sources and therefore invariably contaminated with other metals and minerals including nickel, chromium, and copper (15). Such contaminants also might respond to MRI in their pure form—Fe and Ni being ferromagnetic, Cu paramagnetic, and Cr antiferromagnetic—whereas their oxides may have very different magnetic properties.
Experimental studies by Alsing et al. and Kreidstein et al. both visualized how selected iron oxide pigments can be physically drawn under the influence of magnetic fields (9,14). However, in the study of Alsing et al. such “draw effect” was exceptional in the examined iron oxide-based inks and not related to measured concentrations of atomic iron or other metallic contaminants (measured say by mass spectrometry).
Taking into account the very small scope of events, the review supports the opinion that PMU tattoos are more prone to MRI although reactions may occur in any type of tattoo and at any anatomical site.
Nearly all cases were examined in scanners generating 1.5-T magnetic fields. However, Vahlensieck (4) reported that even as low as 0.5-T magnetic field strength caused a skin reaction, which he graded as a first-degree burn. Noureddine et al. (16) studied 7-T MRI in 108 volunteers with one or more tattoos and found no reaction in the tattoos. Noteworthy, in some cases, patients tolerated MRI in the past without tattoo complications; thus, the reaction appears to be acquired (7,10,17,18). It is known that tattoo pigments in the skin undergo chemical breakdown and physical change over time. Black carbon pigment has a special tendency to agglomerate and form larger pigment clusters, which can induce foreign body reactions with granuloma formation, autoimmunity, and induction of sarcoidosis (19). The review showed that MRI-related reactions especially occurred in dark or black tattoos, and there was no single report of such events in entirely red tattoos, despite the high popularity of this color. The apparent development of sensitivity could also point to a partial ordering of magnetic materials in the tattoo pigment induced by the exposure to high magnetic fields.
Numerous factors could theoretically contribute to the interference between MRI and tattoos, such as the age of the tattoo, pigment density, tattoo design, field strength, conductive abilities of the pigment, the strength of radio frequency coils, etc.
The age of the tattoo—and therefore the density of tattoo pigment particles—may affect the risk of MRI interaction. A newly applicated tattoo would allow only a little time for macrophage pigment engulfing, dispersion, and fibroblasts growth to separate and bind pigment particles. With a denser mass of pigments particles, the ability to interact with static magnetic fields or radio frequency energy may increase.
The pathophysiology behind the reported sensations and cutaneous reactions still remains unsubstantiated; however, since hypothesized by Kreidstein et al. (9), it has been a popular opinion that MRI causes thermal heating. The proposed mechanism is radio frequency-induced induction of heating and eventually skin burns (5,7,12). The risk of MRI thermal injuries is well-known in association with foreign objects primarily with elongated shapes such as monitoring cables, electrodes, guidewires, and sensor leads. These injuries are believed to arise from induced currents within conductive material by the MRI coils (3). Several authors address the theory that if tattoos are drawn in loops or spiral shapes, whether conductive material will result in hazardous voltage enough to raise the intercellular water temperature and cause a burn (5–7). The most egregious case imaginable would be an unbroken loop of highly conducting material with a circumference one-quarter of the incoming wavelength, which in the 3-T case, would correspond to a radius of approximately 9.3 cm, hence a rather large circle (20). However, in our comprehensive review, only three cases with reported skin burn sensations had tattoo designs with loops or spirals (4,6,7); contrarily, most designs were either straight and curved lines (eyeliners or lettering) or bigger tattooed skin patches (18) (Table 1).
Characteristics of case reports.
F, female; M, male; MRI, magnetic resonance imaging; NA, not applicable; T, Tesla; VAS, visual analog scale (0–10).
In addition, a recent ex vivo study by Alsing et al. investigated the thermal effects and magnetic behavior of tattoo pigments, and could not confirm any clinically relevant temperature increase of the tattoo pigment after MRI (14). So far, none of the published case reports have been able to document any thermal heating or conclusive clinical signs of skin burns like bullae or ulcerations. The majority of authors (76%) describe cases with burning pain in the tattooed skin area followed in some cases by erythema with mild edema and rate it as a skin burn – even though this does not exclusively validate a skin burn. Without verifying signs of a skin burn, the “sensation of burn” has become misleading in the literature, since similar neural sensations can be triggered by, for example, hypothermia, etching, reactive oxygen species [ROS] reactions, torque, or physical nerve damage, etc.
In our opinion, it is too early to draw any firm conclusions. Several contributing factors can cause observed skin reactions, which makes the provoking mechanism hard to envisage. At the same time, these events are expected to be underreported due to the relatively rare and spontaneous occurrence, lack of academic interest and knowledge among physicians, and the relatively mild transient nature of skin reactions that are probably treated locally. In our opinion, there is no clear evidence to suspect thermodynamic energy or MRI induction currents in tattoo spirals or loops to cause tattoo skin burn during MRI.
The immediate sensations of burn and the mild irritations of the tattooed skin, without any permanent skin damage, make it more plausible that torque and attractive forces of magnetic and paramagnetic ink particles are causing the reported events.
The possible risk of tattooed patients undergoing MRI is in no way near the diagnostic value of the scanning which can be crucial for the future wellbeing and prognosis of patients (3).
For future patients experiencing MR-related sensations, it is advisable to apply cold pressure dressings and ice packs of the affected tattooed skin as a precautionary measure, to reduce possible pain and edema. Additional topical analgesics such as lidocaine gel or sprays could be beneficial to reduce induced pain.
In conclusion, because of the presumably rare occurrence of incidents besides the short-term and reversible reactions that may develop, patients with tattoos should be allowed to undergo MRI.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
