We found that the use of metal-on-metal prostheses and the prevalence of revisions and its causes following THA was similar in patch-tested dermatitis patients and “ordinary” patients from the DHAR (). The Department of Dermatology at Gentofte Hospital is a tertiary referral center where patch-testing of patients with moderate-to-severe allergic and eczematous disease is performed. Referred patients generally have increased delayed-type hypersensitivity immune responses upon cutaneous exposure to contact allergens compared to healthy individuals.
Accordingly, the prevalence of metal allergy was markedly higher in cases (and controls) than in subjects from the general population in Denmark (Nielsen and Menne 1992
, Thyssen et al. 2009b
). Metal allergy typically develops early in life following cutaneous exposure, and is therefore likely to precede THA in most patients (Thyssen et al. 2007a
). If subjects with metal allergy have an increased risk of complications following THA, e.g. aseptic loosening or reoperations, one would expect the prevalence to be higher in dermatitis patients (who had a high prevalence of metal allergy) than in “ordinary” patients from the DHAR (who were suspected of having a prevalence of metal allergy that was comparable to the prevalence in the general population). However, our study results indicate that dermatitis patients generally do not have an increased risk of complications following THA. The patch-test follow-up period following THA in this study was limited to 1–12 years. Thus, a longer period would possibly have revealed further complications. However, as the year of first operation and the year of patch-testing were equally distributed over the study years among cases, the follow-up time was reasonably long for most patients. Reed et al. published their clinical experience from 22 patients who underwent patch-testing before metal device implantation due to a history of contact allergy (Reed et al. 2008
). The authors concluded that patch-testing was helpful in guiding the choice of device selected.
A weakness of our study was an insufficient registration regarding various combinations of femoral heads and acetabular liner materials (i.e. ”metal-on-metal”, ”metal-on-polyethylene”, and ”ceramic-on-ceramic” or ”ceramic-on-polyethylene” prostheses) (). Thus, the study cannot confirm or exclude an association between metal allergy and second generation metal-on-metal prostheses in the entire study material, but only in a sub-sample. It is therefore possible that in patients with missing data (about 60%), an association between second-generation metal-on-metal prostheses and metal allergy, revision, or aseptic loosening might be found. The use of metal femoral stems was significantly higher in cases than in patients from the DHAR (82% vs. 76%). This finding may reflect the fact that ceramic bearings were mainly used in the 1990s and thus that few were present in both databases. Also, it could simply be a result of random error. However, it cannot be completely ruled out that an association exists because of greater metal exposure in patients receiving a metal head than in patients receiving a ceramic head. Our main finding, that metal allergy was not associated with revision or aseptic loosening, was supported by a sub-investigation among cases, which showed that the prevalence of metal allergy was similar in patients who had had no revision performed and in patients who had had one or more revisions (). However, one should be aware that the study size was small and that a higher number of patients could increase the validity of this finding. Finally, the overall use of various commercial femoral heads and acetabular components was similar in patients from the DHAR and in cases. The prevalence of metal allergy and the proportion of revisions were independent of the commercial subtype of femoral heads and acetabular components.
The prevalence of nickel allergy, cobalt allergy, and chromium allergy was similar in 356 patch-tested dermatitis patients with THA and in 712 patch-tested dermatitis patients without known THA (). The number of cases was relatively small, which could hide even small differences in the prevalence of metal allergy between cases and controls. However, previous linkage studies with fewer cases have demonstrated inverse associations between contact allergy and prevalent disorders such as diabetes (Engkilde et al. 2006
) and inflammatory bowel disease (Engkilde et al. 2007
). Furthermore, the prevalence of metal allergy in cases and in controls was similar, and was also the same as that of metal allergies registered in patients from private dermatology practice (Thyssen et al. 2009c
). Thus, these findings indicate that THA does not lead to higher prevalences of metal allergy. It is interesting that the prevalence of metal allergy (caused by low prevalences of nickel and cobalt allergy) was statistically significantly lower in 64 patients who were patch-tested after total hip arthroplasty than in 292 patients patch-tested before operation. This finding may be the result of random error. Also, the study sample was very small, which reduces validity. However, it is known that tolerance rather than hypersensitivity may develop in some individuals following systemic exposure to an allergen from e.g. dental braces or drinking water (Van et al. 1991
, Smith-Sivertsen et al. 1999
, Mortz et al. 2002
). Whether or not tolerance could explain the difference between the two groups is unknown. We had no knowledge about the indication for patch-testing; however, the vast majority of patients were patch-tested due to dermatitis and not as a result of surgical complications.
We investigated the overall association between contact allergy to selected metals (nickel, cobalt, and chromium) and total hip arthroplasty but were not able to specifically investigate an association with second-generation metal-on-metal prostheses. We found that the risk of surgical revision was not increased in patch-tested dermatitis patients with metal allergies in comparison to “ordinary” THA patients who were not registered in the patch-test database. Also, we found that the prevalence of metal allergy was not increased in patch-tested dermatitis patients who underwent THA in comparison to patch-tested dermatitis patients who were not operated. When interpreting these results, one should bear in mind that the study method had several limitations, such as small sample size and a case-control study design. Furthermore, although the overall prevalence appeared to be similar between the study groups, delayed-type hypersensitivity reactions would undoubtedly develop following exposure to metal implants in a few selected individuals (Davies et al. 2005
, Willert et al. 2005
). Furthermore, two prospective studies have found an increased incidence of metal allergy in patients with failed implants (Hallab et al. 2001
, Thomas et al. 2009
) whereas 2 prospective studies with unselected groups did not find any increase in allergy (Duchna et al. 1998
, Schuh et al. 2008
). Despite some important limitations of our study design, our findings add to the evidence that the risk of complications in metal allergic patients appears limited.