Early surgery for periprosthetic hip fracture did not lower short-term mortality

Design

Target trial emulation with clone-censor-weight analysis, administrative database

Population

4,564 patients aged 45 and over admitted with periprosthetic hip fracture in Japan

Primary outcome

30- and 90-day mortality, early (days 0-2) vs delayed (days 3-10) surgery

Effect

90-day mortality 2.35% vs 2.31%, risk ratio 1.01 (0.59-1.68)

A target trial emulation (Clinical Orthopaedics and Related Research, 1 October 2026) used a Japanese administrative database from 2014 to 2024 covering 4,564 patients admitted with periprosthetic hip fracture (mean age 84, 80% women). Each patient was cloned into an early strategy (surgery within two days of admission) and a delayed strategy (days 3-10), an approach designed to remove the immortal time bias that affected earlier studies.

At 30 days, mortality was 1.72% with early and 0.83% with delayed surgery (risk difference 0.89%, 95% CI -0.32% to 1.72%). At 90 days, it was 2.35% versus 2.31% (risk ratio 1.01, 0.59-1.68). There was no consistent effect by age or sex.

For native hip fractures, surgery within 36-48 hours is the standard. Periprosthetic fractures are different operations: many need revision, specialist implants and a surgeon with revision experience. These data suggest that taking an extra day or two to assemble those does not, on average, cost lives in the short term.

The authors are clear that this does not justify avoidable delay, because they could not assess complications, pressure injury, delirium or function, and residual confounding is possible. The finding applies to mortality only.

• Prioritise having the right surgeon and implants over meeting a 48-hour clock for periprosthetic hip fracture
• Use any wait for medical optimisation and revision planning, not passive delay
• Keep delirium, pressure care and analgesia protocols running from admission
• Do not let this study justify delays for lack of theatre time or staffing

Why it matters

It challenges the reflex of applying native hip fracture timing targets to a very different operation.

Don't overread it

Mortality was the only outcome - complications, delirium and function were not assessed.

The statistics, in plain English

A target trial emulation imitates a randomised trial using observational data, but it can only adjust for what was recorded. The 30-day risk ratio of 2.06 looks large but its interval (0.64-3.73) includes no effect, and the absolute risks are small. At 90 days the two strategies are almost identical.

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.