Systemic metal ion monitoring (Co;Cr) has proven to be a useful screening tool for implant performance to detect failure at an early stage in metal-on-metal hip arthroplasty. Several clinical studies have reported elevated metal ion levels after total knee arthroplasty (TKA), with fairly high levels associated with rotating hinge knees (RHK) and megaprostheses¹. In a knee simulator study, Kretzer², demonstrated volumetric wear and corrosion of metallic surfaces. However, prospective in vivo data are scarce, resulting in a lack of knowledge of how levels evolve over time. The goal of this study was to measure serum Co and Cr levels in several types TKA patients prospectively, evaluate the evolution in time and investigate whether elevated levels could be used as an indicator for implant failure.

The study was conducted at Ghent University hospital. 130 patients undergoing knee arthroplasty were included in the study, 35 patients were lost due to logistic problems. 95 patients with 124 knee prostheses had received either a TKA (primary or revision) (69 in 55 patients), a unicompartimental knee arthroplasty (7 UKA), a RHK (revision −7 in 6 patients) or a megaprosthesis (malignant bone tumours − 28 in 27 patients). The TKA, UKA and RHK groups were followed prospectively, with serum Co and Cr ions measured preoperatively, at 3,6 and 12 months postoperatively. In patients with a megaprosthesis, metal ions were measured at follow-up (cross-sectional study design).

In primary knees, we did not observe an increase in serum metal ion levels at 3, 6 or 12 months. Two patients with a hip arthroplasty had elevated preTKA Co and Cr levels. There was no difference between unilateral and bilateral knee prostheses. In the revision group, elevated pre-revision levels were found in 2 failures for implant loosening. In both cases, ion levels decreased postoperatively. In revisions with a standard TKA, there was no significant increase in metal ions compared to primary knee arthroplasty. RHK were associated with a significant increase in Co levels even at short-term (3–12 months). The megaprosthesis group had the highest metal ion levels and showed a significant increase in Co and Cr with time in patients followed prospectively. With the current data, we could not demonstrate a correlation between metal ion levels, size of the implant or length of time in situ.

In primary knee arthroplasty with a standard TKA or UKA, metal ion levels were not elevated till one year postoperatively. This suggests a different mechanism of metal ion release in comparison to metal-on-metal hip arthroplasties. In two cases of revision for implant loosening, pre-revision levels were elevated, possibly associated with component wear, and decreased after revision. With RHK, slightly elevated ion levels were found prospectively. Megaprostheses had significantly elevated Co and Cr levels, due to corrosion of large metallic surfaces and/or wear of components which were not perfectly aligned during difficult reconstruction after tumour resection. Further research is needed to assess the clinical relevance of metal ion levels in knee arthroplasty.