A fifty-nine-year-old woman with a trimalleolar ankle fracture with rupture of the syndesmosis (Fig. 1) underwent open reduction and internal fixation of the fractures with a titanium plate (catalogue number 441.370; Synthes, STRATEC Medical, Oberdorf, Switzerland) and screws (catalogue numbers 404.812, 404.810, 404.824, 407.040, 406.022, and 404.834; Synthes) from a standard small-fragment instrument set of pure titanium (Synthes). Follow-up visits were done at two to three-week intervals. The fracture was healed radiographically four months after the injury. Inflammatory skin reactions, however, began within a few days after open reduction and internal fixation. After removal of the sutures, there was delayed wound-healing with moist dermatitis but no sign of infection. The wound was finally healed by six months after surgery, although scaling dermatitis overlying the lateral malleolus persisted (Fig. 2-A). At the medial malleolus, where ligament repair had been done at the time of open reduction and internal fixation, the surgical incision healed well (Fig. 2-B). The titanium plate and screws were completely removed nine months after the initial procedure. At that time, tissue surrounding the implants was evaluated histologically and the patient was referred to an allergy specialist for the evaluation of suspected hypersensitivity to the titanium implants.
Anteroposterior radiograph of the left ankle, made after open reduction and internal fixation with use of titanium implants.
Fig. 2-A Photograph showing eczema-like skin changes adjacent to the surgical incision at the left lateral malleolus. Fig. 2-B Photograph showing the left medial malleolus with normal wound-healing.
There was no history of seasonal rhinoconjunctivitis or asthma, but the patient reported itching and erythema on contact with earrings or metal buttons. The patient was otherwise healthy. Corticosteroid ointment was used to treat the dermatitis, and patch testing was initiated. In addition, as described by Thomas et al.4, a lymphocyte transformation test with nickel, chromium, cobalt, and titanium was performed by stimulating peripheral blood mononuclear cells in quadruplicate over six days as follows. As control stimuli, we used the T-cell mitogen phytohemagglutinin (PHA; Biochrom, Berlin, Germany) at 2.4 µg/mL, the "recall antigen" tetanus toxoid (TT; Chiron Behring, Berlin, Germany) at 5 µg/mL, and culture medium alone. As metal stimuli, we used nickel sulphate (NiSO4), cobalt chloride (CoCl2), chromium chloride (CrCl3) (all three at 10-4M, 10-5M, and 10-6M), and titanium oxide (TiO2) (as particles with anatase structure, 10-4M and 10-5M) (Sigma, Steinheim, Germany for all). After five days, cells were pulsed with radioactively marked thymidine (e.g., 3H-thymidine) overnight and proliferation was assessed by incorporated radioactivity. The proliferative response was expressed as a stimulation index, which was calculated as the ratio of mean counts per minute (cpm) of stimulated to unstimulated cultures. A stimulation index of >3 was regarded as positive.
Patch testing was performed according to the guidelines of the German Contact Dermatitis Research Group on the upper back. We tested the standard series (including nickel, chromium, and cobalt), vehicles, disinfectants, and additional metals, including TiO2. The reactions were evaluated after two, three, and seven days.
For the histological analysis, a biopsy specimen from the tissue adjacent to the titanium miniplate was taken at the time of implant removal. Hematoxylin-eosin staining was performed and immunohistochemistry was performed with antibodies to visualize T-cell (anti-CD3), B-cell (anti-CD20), and macrophage (anti-CD68) infiltrates.
Atomic absorption spectrometry was performed to evaluate the removed implant materials. Eluates with 10 mL of distilled water each were prepared from the miniplate and from the screws together with shims. Determination of nickel in the eluates was performed by means of graphite furnace atomic absorption spectrometry with a Zeeman-effect background correction system (SIMAA 6100, THGA Graphite Furnace, autosampler AS 800; PerkinElmer, Rodgau, Germany) after microwave digestion of the samples. The metal content of the eluates is given as µg/L.
Patch testing gave a local eczematous contact allergic reaction (grade ++) to nickel and a local eczematous reaction (grade +) to cobalt. No reactions to the other tested metals were seen. A strong lymphocyte transformation test reactivity of the patient's peripheral blood mononuclear cells was found to nickel (maximal stimulation index = 14.95), but not to chromium, cobalt, or titanium.
Histologically, a dense lymphohistiocytic infiltrate was noted to form inconspicuous reactive lymph follicles with CD20+ B-lymphocytes and CD68+ tingible body macrophages in the centers (Fig. 3) and a CD3+ T-lymphocytic infiltrate surrounding these reactive follicles.
Photomicrograph of a biopsy specimen, showing a reactive germinal center (black arrows) consisting of densely packed lymphocytes, lymphoblasts, and dendritic cells (hematoxylin and eosin, ×40).
Since the clinical and histological reaction indicated a hypersensitivity reaction, and in view of the marked nickel allergy of the patient, we questioned whether there could have been some nickel exposure caused by the implant material. Atomic absorption spectrometry analysis of the titanium miniplate eluate demonstrated no nickel. However, in the eluate prepared from the screws and shims that had been removed nine months following open reduction and internal fixation, significant amounts of released nickel were found (56.7 µg/L) (Table I). The control analysis of the pure eluent (e.g., the pure double-distilled water needed for elution) showed no nickel contamination. After implant removal, the eczema gradually resolved under concomitant dermatologic treatment.
Titanium and its alloys are widely used for medical purposes such as arthroplasty, osteosynthesis, pacemaker encasing, and orthodontic wires as well as in items in daily use, such as eyeglass frames. Potential delayed-type hypersensitivity to titanium was first reported as pacemaker dermatitis, but its existence has been questioned because of incomplete allergy workup and insufficient patch test preparations. In 1984, Peters et al.5 described a patient in whom cardiac pacemakers had been repeatedly implanted and removed because pruritus, redness, and swelling of the skin overlying the pacemakers had developed within weeks after implantation. As there was a positive patch test reaction to a thin square of metallic titanium applied with artificial sweat, a contact allergy to the encasing of the pacemaker was postulated. In addition, contact allergy to nickel was found. One year later, Verbov6 described a patient with impaired wound-healing and moist dermatitis after each of four consecutive insertions of a titanium-encased pacemaker. Histological examination of the tissue surrounding the implant showed a granulomatous reaction. Granulomatous dermatitis in association with a titanium-containing pacemaker was also observed by Brun and Hunziker7 and Viraben et al.8. With regard to orthopaedic implants, Lalor et al.9 described a series of patients with a failed titanium total hip replacement in whom tissue around the implants showed lymphohistiocytic inflammation. The authors unsuccessfully tried to detect titanium sensitivity by testing with titanium salicylate, titanium tannate, titanium dioxide, and titanium peroxide at different concentrations. However, some patients reacted to an ointment containing a mixture of those components. Thomas et al. reported on a patient who had local eczema and impaired healing following titanium-based implant fixation of a metacarpal fracture10. No metal allergy was found, but there was a marked lymphocyte transformation test reactivity to titanium, which disappeared along with the eczema after implant removal. To date, no standardized titanium preparations exist for patch testing and, because of its very limited solubility, titanium has been used in particulate form for the lymphocyte transformation test. Given that titanium hypersensitivity exists, the few previous case reports have all reflected its very rare occurrence. However, in the dental literature, cases of titanium sensitivity have been reported11.
The potential for other sources of reactions to titanium-based materials has been rarely addressed. Romaguera and Grimalt12 reported on a positive reaction (graded as +++) to epoxy resins in a patient who had received a pacemaker and subsequently developed local eczema. In another patient with eczema that was related to a titanium-based pacemaker, stainless steel screws were exposed to surrounding tissue, and contact allergy to nickel, chromium, and cobalt was found13. Furthermore, "hidden" allergy causes may be found in materials that are claimed to be made of titanium; for example, palladium and nickel have been found in "pure titanium" eyeglass frames14,15. It is well known that the exposure of nickel-containing titanium-based orthodontic appliances to an acidic environment and even to saliva will provoke a nickel release16. Small amounts of nickel impurities have been found even in "pure" titanium-based orthopaedic-surgical materials. With the restriction of very low levels, standard titanium alloys (TiAl6Nb7, TiAl6V4) and pure titanium discs supplied by five different titanium manufacturers were shown to contain up to 0.034 weight percent (wt%) nickel, with iodide-titanium having the lowest percentage (0.002 wt%)17.
In our patient, the one-third tubular plate, screws, and shims were made of pure titanium according to ISO (International Organization for Standardization) standard 5832-2, as indicated by the manufacturer. However, the local eczema and impaired wound-healing together with lymphocyte-dominated inflammation on histological examination were suggestive of a hypersensitivity reaction. Thus, patch testing was done (with the restriction that, in contrast to nickel, cobalt, or chromium, no valid titanium test preparation exists) and a lymphocyte transformation test was performed.
In allergy diagnostics in our patient, there was no hyperreactivity to titanium, but an allergy to cobalt and particularly to nickel was found, and a lymphocyte transformation test demonstrated a strong reactivity to nickel. To look for a potential local nickel exposure, eluates from the removed implants were prepared and nickel release was found from the screws and shims. Because the eczema resolved after removal of the titanium material, we concluded that the impaired wound-healing and eczema had been caused by nickel hypersensitivity. Thus, in cases of suspected titanium hyperreactivity, we recommend first looking for other potential causes, particularly a nickel allergy.