Computer-assisted surgery (CAS) for total knee arthroplasty (TKA) has become popular largely based on claims that the technology improves accuracy of component positioning and alignment. Theoretically, that leads to superior patient-reported outcomes. However, the use of CSA has not reliably yielded improvements in implant survival or clinical outcomes. A large registry study by Roberts et al. in the April 1, 2020 issue of The Journal sheds additional light on this perplexing question.
An earlier study by the same author group used data from the same New Zealand Joint Registry and showed no difference in functional outcomes or implant survival between TKAs performed with and without CAS.1 However, that study did not account for the potential bias introduced by surgeons who use CAS only for complex cases.
In this study, Roberts et al. analyzed data from 2 carefully selected groups of surgeons: those who used CAS in 90% of their TKAs (“routine CAS” surgeons) and those who used CAS in <10% of their TKAs (“routine conventional” surgeons). Further limiting their analysis to surgeons with >50 TKAs recorded in the registry between 2006 and 2018, Roberts et al. identified 25 “routine CAS” surgeons and 22 “routine conventional” surgeons. This allowed a comparison between 9,501 TKAs performed by routine CAS surgeons and 7,672 TKAs performed by routine conventional surgeons. While analyzing revision rates and Oxford Knee Scores (OKS) at 6 months, 5 years, and 10 years, the authors also controlled for confounding variables such as age, body-mass index, and implant type.
With a mean follow-up of 4.5 years, the authors found a revision rate per 100 component-years of 0.437 in the group operated on by routine CAS surgeons, compared to a mean 4.9-year revision rate of 0.440 in the group operated on by routine conventional surgeons (p=0.724). When stratifying outcomes of patients <65 years old, the authors again found no statistical difference in revision rates. They also found no between-group differences in OKS within the full and <65 cohorts at 6 months, 5 years, or 10 years.
The findings prompt the authors (and I) to wonder whether continually improving design and durability of modern implants make it difficult to discern any advantage from computer assistance in implant positioning.
Matthew R. Schmitz, MD
JBJS Deputy Editor for Social Media
- Roberts TD, Clatworthy MG, Frampton CM, Young SW. Does computer assisted navigation improve functional outcomes and implant survivability after total knee arthroplasty? J Arthroplasty. 2015 Sep; 30(9)Suppl: 59-63.