- Design
- 12-week randomised, double-blind, placebo-controlled trial
- Population
- 120 patients over 40 with symptomatic knee osteoarthritis and VAS 4-7
- Primary outcome
- Change in WOMAC total score at 12 weeks
- Effect
- WOMAC fell 23.5 points (37.8%) with the combination and 13.2 points (21.6%) with celecoxib alone, reported as within-group changes
Celecoxib works in knee osteoarthritis and its ceiling is set by tolerability rather than efficacy, so an adjunct that lets you use less of it is worth looking at. This 12-week double-blind trial randomised 120 patients aged over 40 with a visual analogue pain score of 4 to 7 to celecoxib (200 mg/day tapered to 100 mg/day) plus either Jintiange capsules 3.6 g/day — a synthetic tiger-bone preparation used in traditional Chinese medicine — or placebo.
The headline numbers favour the combination: WOMAC total score fell 23.5 points (37.8%) in the combination arm against 13.2 points (21.6%) on celecoxib alone, with the pain subscale improving 51.6% against 33.3%, and a greater VAS reduction reported at P less than 0.001.
The problem is what is not reported. Both WOMAC figures are within-group changes from baseline, and a within-group change tells you the patients got better, not that one treatment beat the other. The number that answers the clinical question is the between-group difference with its confidence interval, and the abstract does not give it. Nor does it give the adverse event counts it says it collected, which matters when the stated rationale is reducing NSAID exposure. Read this as a signal worth a properly reported replication, not as a result you can act on. In India the preparation itself is not available, and the wider point — that a locally used adjunct plus an NSAID beats the NSAID alone — is exactly the claim that needs between-group numbers before it is believed.
- Keep NSAID choice and duration driven by cardiovascular, renal and gastrointestinal risk.
- Do not substitute an unstudied local herbal preparation on the strength of a trial of a different one.
- Ask about concurrent traditional remedies at every osteoarthritis review — many contain undeclared NSAIDs or steroids.
- Continue to lead with weight, quadriceps strengthening and activity modification; the drug is adjunctive.
- Record WOMAC or a simple pain and function measure so your own patients' response is visible.
Why it matters
It is the commonest shape of adjunct trial in osteoarthritis, and the commonest way one overstates itself.
Don't overread it
Within-group improvement in both arms is not evidence that one treatment is better than the other.
The statistics, in plain English
A 23.5-point fall in one arm and a 13.2-point fall in the other looks like a 10.3-point advantage, but that subtraction is not a valid estimate of treatment effect and carries no confidence interval. Both arms improved substantially, which is expected in a 12-week osteoarthritis trial where regression to the mean and placebo response are large. With 60 patients per arm, a between-group difference of 10 WOMAC points might well be statistically significant — but the trial has not shown us, and the minimum clinically important difference on WOMAC total is itself debated at around 10 to 20 points.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for orthopaedics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free