Illustrative AI review sample — orthopedics

Key Points

1Single-centre Italian retrospective cohort of 64 Vancouver B periprosthetic femoral fractures comparing osteosynthesis versus revision THA; clinically relevant Vancouver classification-stratified results, but very small sample limits statistical power and no survival analysis is performed for the re-operation endpoint.

Major Issues

Methods / Design: Very small sample (n = 64 total; n = 32 per group) provides very low statistical power for detecting the observed re-operation rate difference (14.3% vs. 8.3%); post-hoc power calculation would reveal the study is underpowered to exclude a clinically meaningful difference between strategies.

Results: Re-operation as the primary endpoint is analysed as a proportion with chi-squared test; Kaplan–Meier time-to-re-operation curve is absent, preventing assessment of whether failure timing differs between osteosynthesis and revision groups.

Methods: Surgical strategy (osteosynthesis vs. revision) was allocated by surgeon discretion, not a pre-specified protocol; this introduces indication bias — surgeons may preferentially select osteosynthesis for less challenging B2 cases, confounding the outcome comparison.

Methods / Results: Harris Hip Score has well-documented ceiling effects in THA populations; a substantial proportion of patients may score in the ≥90 range regardless of treatment, making it insensitive to clinically relevant functional differences between strategies.

Periprosthetic femoral fractures in Total Hip Arthroplasty (THA): a comparison between osteosynthesis and revision in a retrospective cohort study

Key Points

Major Issues