Detecting Avascular Necrosis of the Lunate from Radiographs Using a Deep-Learning Model.

Deep-learning (DL) algorithms have the potential to change medical image classification and diagnostics in the coming decade. Delayed diagnosis and treatment of avascular necrosis (AVN) of the lunate may have a detrimental effect on patient hand function. The aim of this study was to use a segmentation-based DL model to diagnose AVN of the lunate from wrist postero-anterior radiographs. A total of 319 radiographs of the diseased lunate and 1228 control radiographs were gathered from Helsinki University Central Hospital database. Of these, 10% were separated to form a test set for model validation. MRI confirmed the absence of disease. In cases of AVN of the lunate, a hand surgeon at Helsinki University Hospital validated the accurate diagnosis using either MRI or radiography. For detection of AVN, the model had a sensitivity of 93.33% (95% confidence interval (CI) 77.93-99.18%), specificity of 93.28% (95% CI 87.18-97.05%), and accuracy of 93.28% (95% CI 87.99-96.73%). The area under the receiver operating characteristic curve was 0.94 (95% CI 0.88-0.99). Compared to three clinical experts, the DL model had better AUC than one clinical expert and only one expert had higher accuracy than the DL model. The results were otherwise similar between the model and clinical experts. Our DL model performed well and may be a future beneficial tool for screening of AVN of the lunate.

In situ fixation of slipped capital femoral epiphysis carries a greater than 40% risk of later total hip arthroplasty during a long-term follow-up.

Aims: We aimed to assess the cumulative risk of total hip arthroplasty (THA) from in situ fixation for slipped capital femoral epiphysis (SCFE) after a follow-up of almost 50 years. Methods: In this study, 138 patients with 172 affected hips treated with in situ fixation were evaluated retrospectively. A total of 97 patients (70%) were male and the mean age was 13.6 years (SD 2.1); 35 patients (25%) had a bilateral disease. The median follow-up time was 49 years (interquartile range 43 to 55). Basic demographic, stability, and surgical details were obtained from patient records. Preoperative radiographs (slip angle; SA) were measured, and data on THA was gathered from the Finnish National Arthroplasty Register. Results: The preoperative SA was a mean of 39° (SD 19°). At follow-up, 56 of the patients had undergone THA for a hip previously fixed in situ for SCFE (41%) and 64 of all affected hips had been replaced (37%). Kaplan-Meier analysis gave a median prosthesis-free postoperative survival of 55 years (95% confidence interval (CI) 45 to 64) for the affected hips. In a multivariate analysis, female patients had a two-fold risk for THA (hazard ratio (HR) 2.42 (95% CI 1.16 to 5.07)) and a greater preoperative SA increased the risk of THA (HR 1.03 for every increment of 1° (95% CI 1.01 to 1.05)), while patient age at surgery, slip laterality, stability of slip, or diagnostic delay did not have a statistically significant effect on the risk of THA. Conclusion: SCFE treated primarily with in situ fixation may lead to THA in more than 40% of affected hips at a near 50-year follow-up. This risk is approximately 15-times the reported lifetime risk in the Finnish general population. Female sex and increasing preoperative SA significantly predicted higher risk of THA.

Long-Term Clinical Outcome After Titanium Lunate Arthroplasty for Kienböck Disease.

PURPOSE: Titanium lunate arthroplasty (TLA) for Kienböck disease was introduced in 1984 to address the silicone-wear particle problem common to silicone lunate implants. We sought to study the outcome of TLA. METHODS: We identified 11 patients from our hospital database who had undergone TLA between 2001 and 2010. We evaluated pain, range of motion (ROM), function, and radiological outcome at a mean 11 years after surgery. We compared preoperative ROM and radiological findings with final follow-up in the ipsilateral wrist and also made comparisons with the contralateral wrists. RESULTS: No implants were removed, and no wrist joints were fused. Pain on the visual analog scale averaged 0.5 at rest, 0.3 at night, and 2.7 during heavy exertion. Seven patients had no pain at rest and 9 had no pain at night. Range of motion reached 70% of that of the contralateral wrist, and strength reached 81%. The Disabilities of the Arm, Shoulder, and Hand (DASH) score averaged 9.6, optional DASH 9.7, and Mayo wrist score 67.7. Radiologically, only Ståhl and arthrosis indexes differed significantly between affected and unaffected wrists. Two patients had a dorsally dislocated implant, meaning that around 20% of our cases probably meet the criteria for failure. CONCLUSIONS: The longer-term results of TLA for stage III Kienböck disease are promising. TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.

Conservative treatment, plate fixation, or prosthesis for proximal humeral fracture. A prospective randomized study.

BACKGROUND: Proximal humerus fracture is the third most common fracture type after hip and distal radius fracture in elderly patients. A comprehensive study by Palvanen et al. demonstrated an increase in the annual fracture rate of 13.7% per year over the past 33 years. Should this trend continue, the fracture rate would triple over the next three decades. The increasing incidence of low-energy fractures raises questions about the optimal treatment in terms of functional outcome, pain, and rehabilitation time, as well as the economical impact. Despite the high incidence and costs of proximal humerus fractures, there is currently no valid scientific evidence for the best treatment method. Several publications, including a Cochrane review outline the need for high-quality, well-designed randomized controlled trials. METHODS/DESIGN: The study is a prospective, randomized, national multi-center trial. The hypothesis of the trial is that surgical treatment of displaced proximal humerus fractures achieves better functional outcome, pain relief, and patient satisfaction compared to conservative treatment. The trial is designed to compare conservative and surgical treatment of proximal humerus fractures in patients 60 years and older. The trial includes two strata. Stratum I compares surgical treatment with locking plates to conservative treatment for two-part fractures. Stratum II compares multi-fragmented fractures, including three- and four-part fractures. The aim of Stratum II is to compare conservative treatment, surgical treatment with the Philos locking plate, and hemiarthroplasty with an Epoca prosthesis. The primary outcome measure will be the Disabilities of the Arm, Shoulder and Hand (DASH) score and the secondary outcome measures will be the EuroQol-5D (EQ-5D) value, OSS, Constant-Murley Score, VAS, and 15D.Recruiting time will be 3 years. The results will be analyzed after the 2-year follow-up period. DISCUSSION: This publication presents a prospective, randomized, national multi-center trial. It gives details of patient flow, randomization, aftercare and also ways of analysis of the material and ways to present and publish the results. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT01246167.