One injection or a catheter after knee replacement: the trade is not in the pain score

Design

Open-label randomised controlled non-inferiority trial

Population

100 patients undergoing unilateral total knee arthroplasty

Primary outcome

Cumulative NRS area under the curve from PACU discharge to 72 hours, at rest and on activity

Effect

Non-inferiority met (upper 95% CI 18.03 rest, 19.82 activity, margin 20.7); catheter group used less rescue opioid and mobilised better

One hundred patients having total knee arthroplasty were randomised to a continuous parasacral ischial plane block or a single shot of liposomal bupivacaine in the same plane, with the cumulative pain score area under the curve to 72 hours as the primary outcome and a pre-specified non-inferiority margin of 20.7.

The single shot met non-inferiority in both per-protocol and intention-to-treat analyses, at rest and on activity. But the catheter group used less rescue opioid, took longer to need the first dose, and did better on range of motion, two-minute walking distance and the WOMAC index in the early period. They also had less nausea and insomnia.

Against that, the catheter failed in ways a single injection cannot: catheterisation failure 4.1 per cent, leak 2.2 per cent, dislodgement 4.3 per cent, blockage 6.7 per cent. So the decision turns on what the unit can support and what the patient needs — a pain score that is no worse, or the opioid sparing and early mobilisation that came with the catheter.

• If the outcome you care about is early rehabilitation, the catheter did better in this trial.
• If catheter care is unreliable on your wards, the single shot delivers comparable pain scores.
• Count catheter failures in your own service before assuming the trial's rates apply.
• Liposomal bupivacaine costs substantially more in India than plain bupivacaine; factor that in.
• The trial was open-label, so subjective outcomes carry expectation effects both ways.

Why it matters

It reframes the choice from which block works to what your ward can look after.

Don't overread it

Open-label design, 100 patients, and a non-inferiority margin set by the investigators.

The statistics, in plain English

Non-inferiority was declared because the upper limits of the confidence intervals (18.03 and 19.82 per protocol) sat below the pre-specified margin of 20.7 — a margin chosen by the investigators, and the whole conclusion rests on that choice being clinically reasonable. Note that the secondary outcomes went the other way: non-inferior on the primary endpoint and worse on opioid use and rehabilitation is a common and honest pattern, and the secondary results were not adjusted for multiple testing.

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.