Routine radiographs after joint replacement rarely change care; newer AI model hallucinates less; FDA approves pimicotinib for TGCT

Listen to this post
read by AI voices · all episodesRoutine radiographs after uncomplicated elective hip and knee arthroplasty changed management in one of 5,555 imaging encounters, a meta-analysis in the Journal of Arthroplasty reported Nov. 10. A newer large language model extracted hip arthroplasty operative note data with fewer hallucinations than a 2024 model. The Food and Drug Administration approved pimicotinib for symptomatic tenosynovial giant cell tumor and cleared OsteoSight Spine for bone density assessment on routine X-rays.
- Research Nov 10, 2026A systematic review and meta-analysis in the Journal of Arthroplasty included 17 studies, 13 of which (10,853 imaging encounters) contributed to the primary analysis of routine plain radiographs after primary elective total hip or knee arthroplasty. The pooled management change rate was 0.02% (95% CI 0 to 0.34), and all seven observed events occurred after hip arthroplasty. The number needed to image was 5,555, at an estimated $1,623,813 per management change, and the authors said the findings do not support routine postoperative imaging in uncomplicated elective cases.Sources: Journal of Arthroplasty, PubMed
- AI Nov 12, 2026Researchers tested two locally hosted large language models, one released in 2024 and one in 2026, against manual abstraction of surgical approach, fixation and bearing surface from 1,000 primary total hip arthroplasty notes by 47 surgeons, according to the Journal of Arthroplasty study. The newer model was more accurate for approach (98% versus 96%), fixation (98% versus 96%) and bearing surface (93% versus 89%). In 100 reviewed reasoning outputs per model, hallucination rates were 5%, 3% and 11% for the newer model versus 49%, 20% and 49% for the older one; the authors said remaining hallucinations limit trustworthiness.Sources: Journal of Arthroplasty, PubMed
- FDA Nov 10, 2026The Food and Drug Administration (FDA) approved pimicotinib (Pimjetmi) on Nov. 10 for adults with symptomatic tenosynovial giant cell tumor in whom surgical resection could potentially worsen functional limitations, according to OncLive and CancerNetwork. In part 1 of the randomized phase 3 MANEUVER trial, 94 patients were assigned 2:1 to pimicotinib 50 mg once daily or placebo for 24 weeks, and the objective response rate at week 25 was 54.0% versus 3.2%, CancerNetwork reported. The most common adverse reactions included pruritus, rash, face edema, periorbital edema, fatigue, nausea, headache and peripheral edema.Sources: OncLive, CancerNetwork
- AI Nov 12, 2026Naitive Technologies received Food and Drug Administration (FDA) 510(k) clearance for OsteoSight Spine, software that analyzes routine lumbar spine, thoracic spine and chest X-rays taken for other clinical concerns in patients 50 and older to identify those at risk for low bone mineral density, the company said Nov. 12. It provides a report to aid the interpreting physician, according to the company, which said it develops AI-driven software. The clearance follows the FDA's September 2025 clearance of OsteoSight for hip and pelvis X-rays. Neither the company release nor OrthoWorld reported performance data.Sources: Naitive Technologies, OrthoWorld
The weekly innovation roundup for orthopedic surgery, published every Sunday for the week before; every item links to its sources. Latest orthopedic surgery news · All specialties