Ions of chromium and cobalt typically are elevated in patients with metal-on-metal hip implants, and authorities in all countries agree that patients with these implants should be followed, said Alister Hart, MD, of University College London.
The U.K. Medicines and Healthcare Products Regulatory Agency (MHRA) has proposed that these ions be used in surveillance for failing hips, but the FDA has been less sure. Their actual diagnostic ability is unknown, Hart said at the annual meeting here of the American Academy of Orthopaedic Surgeons.
"We've revised more than 300 metal-on-metal hips, and have collected more than 2,000 implant failures and we now run a multidisciplinary clinical advisory service to guide surgeons on who, when, and why to revise," he said.
Their experience has shown that there are three groups of patients. "Those with either very severe or very few symptoms are straightforward. The difficult group includes patients who have moderate hip pain or abnormal metal ion tests," he said.
Hart's group previously demonstrated that the metal ion tests, which are simple and cheap, have good ability to identify poorly functioning implants, but there has not yet been a large cohort study examining the usefulness of these tests for predicting negative outcomes in patients with moderate pain.
They conducted a prospective case-control study that included 597 patients who had undergone either total hip replacement or resurfacing and who had been followed for at least 1 year after the surgery.
Patients with poorly functioning hips were classified as those who had already undergone revision, those with pain and awaiting revision, and those with continued inadequate function as defined by scores below 30 on the Oxford Hip Scoring System.
Comparing blood metal ion levels overall found significant elevations in the poor function group (P<0.01), and similar elevated levels in patients who had undergone total hip replacement as well as in those treated with hip resurfacing.
The ability of the metal ion levels to discriminate between well functioning and poorly functioning hips was shown by the finding of areas under the curve of 0.72 to 0.84, with cobalt being more useful than chromium, Hart said.
When they looked at different cutoff levels of the ions, they found that a maximum level of 7 µg/L, which is the recommended MHRA cutoff, had a specificity of 87.7% and a sensitivity of 53.2% for identifying a failed hip implant.
In contrast, using a threshold level of 25 parts per billion give almost perfect specificity, at 99%, but only 24.5% sensitivity for hip failure.
"No cutoff for levels of chromium and cobalt ions had ideal test characteristics. This is a poor screening tool, which would pick up many who don't need revision," he said.
Therefore, he declined to recommend any threshold as the optimal cutoff for revision of total hip arthroplasty.
"We continue to use conventional assessment supplemented by metal ions, but none of the tests in isolation provide the key," he concluded.
Hart A, et al "Surveillance of metal-on-metal hip arthroplasties: is blood metal ion measurement useful?" AAOS 2013; Abstract 66.