The edition · Orthopaedics
Fracture detection: 73% to 87% with a second reader that never tires
A meta-analysis of 17 studies puts numbers on what AI adds to fracture reading — and on who gains, which is mostly the junior in the emergency department at 3am. Plus a network comparison of distal tibial fixation, what psychiatric comorbidity does to hip arthroscopy at ten years, and a spacer finding that should change your bearing choice.
The edition in brief
Today's orthopaedics edition opens with a PRISMA-DTA systematic review and meta-analysis of AI-assisted versus unassisted fracture interpretation. Across 17 studies suitable for pooling, sensitivity rose from 73% (95% CI 69-78) to 87% (84-89) with AI assistance, while specificity held at 95% (92-97) against 94% (89-96). The summary ROC area under the curve rose from 0.849 to 0.929. Junior clinicians gained most, with a 21% absolute sensitivity increase, and meta-regression identified plain two-dimensional radiography and junior status as independent limits on the assisted ceiling — the authors keep a low threshold for cross-sectional imaging. A network meta-analysis of 29 studies and 2,145 patients compared five fixation strategies for distal tibial fractures. No strategy was consistently superior. Delayed union and nonunion did not differ. Minimally invasive plate osteosynthesis carried higher malunion risk than external fixation with limited open reduction (RR 3.26, 95% CI 1.08-9.80) and than suprapatellar intramedullary nailing (RR 4.03, 1.30-12.48), while open reduction and internal fixation had lower malunion risk than MIPO (RR 0.30, 0.11-0.82). Infection risk appeared higher with ORIF and MIPO. A 10-year cohort of 243 hip arthroscopies for femoroacetabular impingement found failure in 23.3% of those with psychiatric comorbidity versus 5.3% without. And in 302 two-stage revision hips, spacer dislocation raised the risk of dislocation after reimplantation more than sixfold.
What AI adds to fracture reading, and to whom
Use AI fracture assistance as a safety net for junior readers, and keep the same low threshold for CT when the film is negative and the clinical picture is not.
No fixation strategy wins outright in the distal tibia — but malunion separates them
Choose distal tibial fixation on the malunion-versus-infection trade this analysis quantifies, not on an expectation that one technique unites faster.
Psychiatric comorbidity quadrupled failure after hip arthroscopy at ten years
Record psychiatric comorbidity before hip arthroscopy and set expectations against a 23% ten-year failure rate rather than the 5% you quote otherwise.
Write down what the spacer did before you plan the reimplantation
Treat the spacer interval as diagnostic information about stability, and record what happened where the second-stage surgeon will see it.
A spacer that dislocated predicts a hip that will dislocate
If the spacer dislocated during the interval, reimplant with a higher-stability bearing — that history carries a sixfold risk of dislocation afterwards.
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 one 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