Design: Multicentre randomised trial, patient- and assessor-blinded, intention to treat
Population: 122 adults with a displaced patellar fracture
Primary outcome: KOOS sub-scales at 12 months
Effect: Symptoms +10.4 (4.4-16.4), sport +14.1 (3.4-24.7), QoL +10.9 (2.7-19.0) favouring plates

In this patient- and assessor-blinded multicentre randomised trial, 122 adults with a displaced patellar fracture were allocated to locking plate fixation (63) or tension-band wiring (59). Superiority was defined in advance as a significant difference exceeding the minimal clinically important difference in at least three of five KOOS sub-scales at 12 months.

The threshold was met. Plates improved symptoms (mean difference 10.4 points, 95% CI 4.4-16.4), sport and recreation (14.1, 3.4-24.7) and quality of life (10.9, 2.7-19.0). Hardware removal, fixation failure and reoperation were more frequent after tension-band wiring. Follow-up at 12 months was 90.5% and 88.1%.

Tension-band wiring has been the standard for decades despite prominent, irritating hardware and frequent removal. This is the first blinded randomised evidence that a plate gives patients a better knee, not just a stronger construct.

Where low-profile patellar plates are available and affordable, they should be the default for displaced fractures. Where cost limits access, as in many Indian public hospitals, the higher reoperation and removal burden of wiring is part of the true cost comparison.

Consider locking plate fixation as first choice for displaced patellar fractures.
Counsel patients having tension-band wiring about the higher likelihood of hardware removal.
Include reoperation and removal costs when comparing implant costs.
The trial reports 12-month outcomes; longer-term arthritis and function remain to be seen.

Why it matters

It overturns a decades-old default with blinded trial evidence that patients, not just constructs, do better with plates.

The statistics, in plain English

Each of the three significant differences was larger than the pre-set minimal clinically important difference, which is what makes this a clinically, not only statistically, significant result. The confidence intervals are wide because the trial was modest in size, so the lower bounds are closer to the threshold of what patients would notice.