A majority of knee replacement constructs were worse than contemporary benchmark constructs, according to a study published in BMJ Open.
Researchers conducted a non-inferiority analysis of 449 different combinations of total (405) and partial (44) knee replacement constructs from April 1, 2003, to Dec. 31, 2016. They compared construct performance to an internally identified benchmark construct. Researchers considered the first linked surgical revision to be the point of product failure.
What they found:
1. At three years post implant, the researchers noted the NexGen cemented, unconstrained TKR with a fixed bearing was the most effective implant with 34,558 procedures at risk with a 1.10 percent failure rate.
2. Seventy-three constructs had more than 500 procedures at risk three years post implant compared to the benchmark.
2. Thirty-nine constructs had at least a 20 percent higher risk of failure than the benchmark.
3. Nine of the 73 inferior constructs had at least 100 percent higher relative failure risks.
4. At five years post implant, researchers noted the Profix uncemented, unconstrained TKR with a fixed bearing construct was the most effective implant with 1,910 procedures at risk and a 1.54 percent failure rate.
5. Sixty-five constructs had more than 500 procedures at risk five-years post implant compared to the benchmark.
5. Thirty-six constructs were inferior to the benchmark with at least a 20 percent higher risk of failure.
6. Twelve of the 36 constructs had at least 100 percent higher relative failure risks.
7. Concerning partial knee replacement constructs, all eight constructs with more than 500 procedures at risk had at least 100 percent higher relative failure risks.
Researchers concluded: "Many of those implanted in sufficient number have here been demonstrated to be inferior in terms of construct survivorship, while very few TKRs have been demonstrated to be non-inferior to a contemporary benchmark."
Researchers noted the vast majority of constructs didn't have enough cases to allow for meaningful analysis.