Issue: July 2021
Perspective from Vasili Karas, MD, MS
Source: Shaw JH, et al. J Arthroplasty. 2021;doi:10.1016/j.arth.2021.02.038.
Disclosures: Shaw reports no relevant financial disclosures.
July 15, 2021
2 min read

Study detects differences in patient-reported outcomes for robotic vs manual TKA

Issue: July 2021
Perspective from Vasili Karas, MD, MS
Source: Shaw JH, et al. J Arthroplasty. 2021;doi:10.1016/j.arth.2021.02.038.
Disclosures: Shaw reports no relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Mixed differences in patient-reported outcomes were seen for patients after total knee arthroplasty performed with robotics vs. without robotics, according to published results.

Jonathan H. Shaw, MD, Jason J. Davis, MD, and colleagues reviewed 1,160 consecutive patients in the prospective Michigan Arthroplasty Registry Collaborative Quality Initiative database who underwent robotic (Mako, Stryker; n=260) or manual (n=900) TKA from December 2017 to October 2019. The goal of the study was to determine whether patients who underwent robotic-assisted TKA showed higher rates of early and intermediate postoperative improvement in minimal clinically important differences (MCIDs) for patient-reported outcomes compared with patients who underwent manual TKA.

“Exclusion criteria I want to emphasize is that any patient who did not have an early outcome was excluded from our database. Early outcomes were defined 2 to 6 weeks postoperative and intermediate outcomes were defined as 4 to 8 months postoperatively. We defined our MCID as one-half the standard deviation of the preoperative scores,” Shaw said at the American Association of Hip and Knee Surgeons Annual Meeting, where he presented the results.

Jonathan H. Shaw, MD
Jonathan H. Shaw

MCID achievement

Primary outcome measures used included the Knee Injury and Osteoarthritis Outcome Score for Joint Replacement and patient reported outcomes measurement information system – global health (PROMIS-GH) scales, which include PROMIS physical health (PH) and mental health (MH) scales.

“When it comes to PROMIS physical health, you do see the similar trending increases in postoperative scores like you do with KOOS JR scores,” Shaw told Orthopedics Today, but noted this trend did not hold true regarding the PROMIS-MH scale score.

“Multivariate analysis showed similar short-term and intermediate MCID achievement in these patient-reported outcome systems and there are some interesting findings in the low preoperative performers vs. high preoperative performers that should be included in future studies when looking at these patient-reported outcome systems,” he said at the meeting.

Investigators saw a statistically significant improvement in MCID achievement for the early KOOS JR scores that favored manual TKA. However, when it came to the multivariate analysis, this did not hold true.

“There was a significant improvement in mean scores for PROMIS mental health in robotic total knees compared with manual total knees, but the MCID achievement did not show any significance,” Shaw said.

The single-surgeon subanalysis demonstrated PROMIS-PH MCID achievement that favored robotic-assisted TKA in the early postoperative period, which, Shaw said, shows how confounders from the multi-surgeon database used may cause skewing of the data. Therefore, a large single-surgeon database may be necessary to control for these confounders, he said.

Controlling for variabilities

Shaw told Orthopedics Today results of this study reflect some of the variabilities that need to be controlled for in future research to provide a good conclusion. This is especially the case with early reported outcomes that may indicate a subtle difference between the manual and robotic TKA cohorts, he said.

“If you have these differences ... it is probably going to be subtle and, if that is the case, then any type of confounders that you can find that would skew the data, can skew it one way or the other. So, you may have to have a pure study that controls for that,” Shaw said.