Ankle fracture classification using deep learning: automating detailed AO Foundation/Orthopedic Trauma Association (AO/OTA) 2018 malleolar fracture identification reaches a high degree of correct classification

Background and purpose - Classification of ankle fractures is crucial for guiding treatment but advanced classifications such as the AO Foundation/Orthopedic Trauma Association (AO/OTA) are often too complex for human observers to learn and use. We have therefore investigated whether an automated algorithm that uses deep learning can learn to classify radiographs according to the new AO/OTA 2018 standards.Method - We trained a neural network based on the ResNet architecture on 4,941 radiographic ankle examinations. All images were classified according to the AO/OTA 2018 classification. A senior orthopedic surgeon (MG) then re-evaluated all images with fractures. We evaluated the network against a test set of 400 patients reviewed by 2 expert observers (MG, AS) independently.Results - In the training dataset, about half of the examinations contained fractures. The majority of the fractures were malleolar, of which the type B injuries represented almost 60% of the cases. Average area under the area under the receiver operating characteristic curve (AUC) was 0.90 (95% CI 0.82-0.94) for correctly classifying AO/OTA class where the most common major fractures, the malleolar type B fractures, reached an AUC of 0.93 (CI 0.90-0.95). The poorest performing type was malleolar A fractures, which included avulsions of the fibular tip.Interpretation - We found that a neural network could attain the required performance to aid with a detailed ankle fracture classification. This approach could be scaled up to other body parts. As the type of fracture is an important part of orthopedic decision-making, this is an important step toward computer-assisted decision-making.

A higher association of medial collateral ligament injury of the knee in pronation injuries of the ankle

INTRODUCTION

To evaluate the prevalence of medial collateral ligament (MCL) injury of the knee among ankle-fracture patients and to determine the risk factors associated with MCL injury in this patient group.

MATERIALS AND METHODS

303 patients (303 affected ankles) who underwent surgical treatment for an ankle fracture were assessed. Supination versus pronation injury, Danis-Weber classification, age, sex, body mass index (BMI), limb dominance, and mechanism of injury were reviewed to identify factors related to MCL injury.

RESULTS

Prevalence of MCL injury of the knee among the total number of patients with an ankle fracture was 3.96% (12 out of 303 injuries). Multivariable logistic and linear regression analysis with adjustment of possible confounding factors confirmed that female sex and pronation injury were associated significantly (p < 0.05) with MCL injury.

CONCLUSIONS

The prevalence of MCL injury among females and the pronation type of ankle injury was 8.19% (10 out of 122 females) and 10.75% (10 out of 93 pronation injuries), respectively. More careful physical examination of the knee joint is strongly recommended in patients with ankle fractures, especially if the patient is female or the ankle-fracture pattern corresponds to the pronation type of injury.

Predictors of Adverse Events for Ankle Fractures: An Analysis of 6800 Patients

Ankle fractures are one of the most common injuries seen by orthopedic surgeons. It is therefore essential to understand the risks associated with their treatment. Using the American College of Surgeons National Surgical Quality Improvement Program(®) database from 2006 to 2013, the patient demographics, comorbidities, and 30-day complications were collected for 5 types of ankle fractures. A bivariate analysis was used to compare the patient demographics, comorbidities, and complications across all Common Procedural Terminology codes. A multivariable logistic regression model was then used to assess the odds of minor and major postoperative complications within 30 days after open treatment. A total of 6865 patients were included in the analysis. Of these patients, 2507 (36.5%) had bimalleolar ankle fractures. The overall rate of adverse events for ankle fractures was low. Bimalleolar fractures had the greatest rate of major (2.6%, n = 64), minor (3.8%, n = 94), and total (5.7%, n = 143) complications. When controlling for individual patient characteristics, bimalleolar fractures were associated with 4.92 times the odds (95% confidence interval 1.80 to 13.5; p = .002) of developing a complication compared with those with a medial malleolar fracture. The risk factors driving postoperative complications for all ankle fractures were age >65 years, obesity, diabetes, American Society of Anesthesiologists score >2, and functional status (p < .05). Although the overall rate of adverse events for ankle fractures was low, bimalleolar fractures were associated with 5 times the odds of developing a complication compared with medial malleolar fractures. Orthopedic surgeons must be aware of the risk factors that increase the rate of ankle fracture complications to improve patients' quality of care.

Absorbable implants versus metal implants for the treatment of ankle fractures: A meta-analysis

This meta-analysis was performed to evaluate the efficiency and the safety of absorbable implants. Five major electronic databases (PubMed, Embase, Cochrane Library, SinoMed and Wanfang Data) were systematically searched for randomized controlled trials (RCTs) from their establishment to November 2012. Studies on absorbable implants and metal implants for ankle fractures were selected. The meta-analysis was performed using RevMan 5.1. Ten studies with 762 patients were included and analyzed. Compared with metal implants, absorbable implants used for the internal fixation of ankle fractures produce similar radiographic and functional outcomes (P= 0.52). Normally, removal of the internal fixation is unnecessary (P<0.0001) and the incidence of palpable implants is lower (P=0.02) for absorbable implants. No statistically significant difference was observed between the two groups with regard to foreign body reactions (P=0.07), infection (P= 0.69), osteoarthritis (P= 0.39), pain (P= 0.06), refracture (P=0.67), skin necrosis (P=0.99), deep vein thrombosis (P=0.21) and nerve injury (P=0.94). Absorbable implants used in ankle fractures rarely require reoperation and result in similar functional outcomes and complications compared with metal implants. These characteristics make them efficient and reasonably safe for the treatment of ankle fractures.