Evaluation and Real-World Performance Monitoring of Artificial Intelligence Models in Clinical Practice: Try It, Buy It, Check It.

The pace of regulatory clearance of artificial intelligence (AI) algorithms for radiology continues to accelerate, and numerous algorithms are becoming available for use in clinical practice. End users of AI in radiology should be aware that AI algorithms may not work as expected when used beyond the institutions in which they were trained, and model performance may degrade over time. In this article, we discuss why regulatory clearance alone may not be enough to ensure AI will be safe and effective in all radiological practices and review strategies available resources for evaluating before clinical use and monitoring performance of AI models to ensure efficacy and patient safety.

Hemiarthroplasty Versus Total Joint Arthroplasty for Hallux Rigidus: A Systematic Review and Meta-analysis.

BACKGROUND: When surgery is indicated for hallux rigidus, toe arthroplasty is an alternative procedure to arthrodesis for patients who wish to preserve toe range of motion. Our study investigated midterm outcomes of first metatarsophalangeal joint (MTPJ) arthroplasty in an effort to discern whether or not partial or total joint replacement confers benefit in these patients. METHODS: A systematic review of MTPJ arthroplasty was performed for the years 2000 to 2017. A Forest plot was created comparing preoperative and postoperative American Orthopedic Foot and Ankle Score (AOFAS), Visual Analogue Scale (VAS), and range of motion (ROM) results for both hemitoe and total-toe arthroplasty. Statistical analysis was performed. RESULTS: Mean postoperative AOFAS scores in patients undergoing hemiarthroplasty improved by 50.7 points (95% CI = 48.5, 52.8), whereas the mean AOFAS score improvement in total joint arthroplasty patients was 40.6 points (95% CI = 38.5, 42.8). VAS outcomes were comparable. Mean postoperative MTPJ ROM improved by 43.0° (95% CI = 39.3°, 46.6°) in hemitoe patients, which exceeded the mean ROM improvement of 32.5° (95% CI = 29.9°, 35.1°) found in total joint arthroplasty cases. A meta-analysis revealed no significant difference. CONCLUSION: Hemisurface implants in MTPJ arthroplasty may improve postoperative AOFAS and ROM results to a greater extent than total-toe devices. LEVEL OF EVIDENCE: Level IV: Systematic review.

Trends in surgical treatment of femoral neck fractures in the elderly.

OBJECTIVE: To analyze recent demographic and medical billing trends in treatment of femoral neck fracture of American elderly patients. METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was analyzed from 2006 to 2015, for patients aged 65 years and older, using the Current Procedural Terminology codes 27130, 27125, 27235, and 27236. Patient demographics, postoperative complications, and frequency of codes were compared and analyzed over time. Our sample had 17,122 elderly patients, in that, 70% were female, mean age of 80.1 years (standard deviation±6.6 years). RESULTS: The number of cases increased, but age, gender, body mass index, rates of diabetes and smoking did not change over time. Open reduction internal fixation was the most commonly billed code, with 9,169 patients (53.6%), followed by hemiarthroplasty with 5,861 (34.2%) patients. Combined estimated probability of morbidity was 9.8% (standard deviation±5.2%), and did not change significantly over time. Postoperative complication rates were similar between treatments. CONCLUSION: Demographics and morbidity rates in femoral neck fractures of elderly patients did not change significantly from 2006 to 2015. Open reduction internal fixation was the most common treatment followed by hemiarthroplasty.

Total hip arthroplasty for posttraumatic osteoarthritis following acetabular fracture: A systematic review of characteristics, outcomes, and complications.

PURPOSE: Posttraumatic arthritis (PTA) may develop years after acetabular fracture, hindering joint function and causing significant chronic musculoskeletal pain. Given the delayed onset of PTA, few studies have assessed outcomes of delayed total hip arthroplasty (THA) in acetabular fracture patients. This study systematically reviewed the literature for outcomes of THA in patients with PTA and prior acetabular fracture. METHODS: Pubmed, EMBASE, SCOPUS, and Cochrane library were searched for articles containing the keywords "acetabular", "fracture", "arthroplasty", and "post traumatic arthritis" published between 1995 and August 2017. Studies with less than 10 patients, less than 2 years of follow-up, conference abstracts, and non-English language articles were excluded. Data on patient demographics, surgical characteristics, and outcomes of delayed THA, including implant survival, complications, need for revision, and functional scores, was collected from eligible studies. RESULTS: With 1830 studies were screened and data from 10 studies with 448 patients were included in this review. The median patient age on date of THA was 51.5 years, ranging from 19 to 90 years. The median time from fracture to THA was 37 months, with a range of 27-74 months. Mean follow-up times ranged from 4 to 20 years. The mean Harris hip scores (HHS) improved from 41.5 pre-operatively, to 87.6 post-operatively. The most prevalent postoperative complications were heterotopic ossification (28%-63%), implant loosening (1%-24%), and infection (0%-16%). The minimum 5-year survival of implants ranged from 70% to 100%. Revision rates ranged from 2% to 32%. CONCLUSION: Despite the difficulties associated with performing THA in patients with PTA from previous acetabular fracture (including soft tissue scarring, existing hardware, and acetabular bone loss) and the relatively high complication rates, THA in patients with PTA following prior acetabular fracture leads to significant improvement in pain and function at 10-year follow-up. Further high quality randomized controlled studies are needed to confirm the outcomes after delayed THA in these patients.