Overinterpretation of joint replacement registry data may result in misleading conclusions
ORLANDO, Fla. — Data from joint replacement registries can be misleading when comparing implant types or designs because the information contained within the registries do not include factors such as indications for surgery or technique used, according to a presenter at the Current Concepts in Joint Replacement Winter Meeting, here.
"I think registers cannot reliably compare implant types, they cannot compare implant designs and they cannot reliably identify poor implants," David W. Murray, MD, FRCS, said. "I think such data is frequently overinterpreted, resulting in misleading conclusions."
David W. Murray
Murray pointed to data from the New Zealand Joint Registry where the thresholds for revision in unicompartmental knee arthroplasty (UKA) and total knee arthroplasty (TKA) differ. A poor TKA outcome could be lower than a poor UKA outcome despite both outcomes being within accepted thresholds.
"The difference in revision rate is a manifestation of the different threshold for revision. So, one has to conclude, therefore, that registry data cannot reliably compare implant types because of different thresholds for revision, and I would go further to say that these comparisons are misleading because all conservative procedures will have a higher revision rate even though they have better results," Murray said.
Factors such as surgical technique and easily revised implants also affect registry data and could lead to misinterpretations, he added. Registry data should instead be used as a starting point to test hypothesis in other ways, according to the abstract. – by Jeff Craven
Murray DW. Paper #25. Presented at: Current Concepts in Joint Replacement Winter Meeting; Dec. 12-14, 2013; Orlando, Fla.Disclosure: Murray receives royalties from Biomet.