Cement or no cement in a robotic knee: still unanswered

Design

systematic review and random-effects meta-analysis of ten observational comparative studies, ROBINS-I and GRADE, PROSPERO registered

Population

patients undergoing primary robotic-assisted total knee arthroplasty with cementless or cemented fixation

Primary outcome

mid-term all-cause revision, manipulation under anaesthesia, operative time and length of stay

Effect

revision RR 0.63 (95% CI 0.09–4.53); manipulation 1.03 (0.09–11.74); operative time −12.91 min (−36.38 to 10.55, I² 94.4%); certainty very low throughout

Cementless fixation is often argued for on the basis that robotic bone preparation is accurate enough to make press-fit reliable. A systematic review tested whether the argument shows up in outcomes, pooling ten comparative studies (11 reports) of cementless against cemented primary robotic-assisted total knee arthroplasty.

It does not, in either direction. Mid-term all-cause revision gave a pooled risk ratio of 0.63 with a confidence interval from 0.09 to 4.53 — an interval so wide it is compatible with cementless being far better or far worse. Manipulation under anaesthesia was 1.03 (0.09 to 11.74). Operative time appeared shorter with cementless by a mean 12.91 minutes, but the interval crossed zero (−36.38 to 10.55) with heterogeneity of 94.4%, meaning the studies disagreed almost entirely. Length of stay differed by 0.24 days with an interval spanning more than three days either way. Certainty was rated very low for every pooled outcome by GRADE, and the included studies were observational, assessed with ROBINS-I.

The practical reading is that this question has not been answered and should not be argued as though it had. Choose fixation on the patient in front of you — bone quality, age, expected demand, and what you and your unit do reliably — and treat any claim that robotic assistance has settled the cementless case as unsupported. Cost matters here too: cementless implants generally cost more, and there is no outcome advantage in this evidence to set against that.

• Do not cite robotic accuracy as evidence for cementless fixation; the comparative data do not support it.
• Choose fixation on bone quality, age and demand, and on what your unit does consistently well.
• Weigh implant cost explicitly — no outcome benefit was demonstrated to offset a price difference.
• Treat the operative-time saving as unproven; heterogeneity of 94.4% means it varied by centre, not by technique.
• Watch for randomised evidence; ten observational studies rated very low certainty will not settle this.

Why it matters

The case for cementless fixation in robotic knees is being made on a mechanism argument that the outcome data have not backed.

Don't overread it

These are observational comparisons with very low certainty — 'no difference found' here means the studies could not detect one, not that none exists.

The statistics, in plain English

A risk ratio of 0.63 looks like a 37% reduction until you read the interval: 0.09 to 4.53 spans a tenfold reduction and a fourfold increase, which means the three studies contributing to it had almost no revisions between them. Very wide intervals like this are a count problem, not a measurement problem. An I² of 94.4% on operative time says the studies were measuring genuinely different things — different theatres, different teams, different case mixes — so pooling them produces a number that describes no real setting. GRADE certainty of very low across every outcome is the summary the authors intend you to act on.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.