Other writing
Precise at What?
Published here

A robotic arm helped surgeons place knees closer to the planned alignment. What's the argument?
RACER-Knee. The Lancet. 339 patients. 10 hospitals across the NHS. Mako robotic arm versus conventional total knee. Patients didn't know which one they got. Neither did the people scoring them. Forgotten Joint Score at one year.
The robot hit the planned alignment better than the surgeon did by hand. Took about ten minutes longer. Cost around £950 more per knee.
At one year the trial found no clinically meaningful patient benefit. Pain and function were comparable. So were the serious complications.
Well done. Strong data.
Precise at what, though. The robot hit the plan. That doesn't settle which plan is best. Mechanical. Kinematic. Functional. Restricted kinematic. Every camp has a target and a paper to back it. What does hitting the plan more precisely prove if we still disagree about the target?
Then the commentary. Press. Posts. Perceptions. Opinions.
The conventional guys feel validated. They've been saying this for years and now they have a Lancet paper.
The robot guys have a list of reasons the study doesn't say what it says. Too early. Wrong score. Only one system. And more.
Fine. Which score would have counted. What year of follow up. How many patients. And whose robot.
Then the questions I actually care about.
How many surgeons change anything? Does a robot user put the arm back in the corner. And the surgeon who was about to switch, does he hold off. My guess is very few on either side.
We've had the data on arthroscopic meniscectomy in arthritic knees for years. They still get done. And the surgeons strongly justify why they need to get done.
And the manufacturers. A robot has never been just a robot. With many systems, buy the robot and you buy the implants that go with it. A robot can grow an implant business, too. Watch how the companies respond. The implant agnostic systems will be first in line to say they weren't in the trial.
When a study agrees with us we post it the same day. When it disagrees we go straight to the methods section looking for the flaw.
I'll be waiting for the ten year follow up. Until then, let's not bury our heads in the sand.
