Anterior approach total hip arthroplasty (AA-THA) has a steep learning curve, with higher complication rates in initial cases. Proper surgical case selection during the learning curve can reduce early risk. This study aims to identify patient and radiographic factors associated with AA-THA difficulty using Machine Learning (ML). Consecutive primary AA-THA patients from two centres, operated by two expert surgeons, were enrolled (excluding patients with prior hip surgery and first 100 cases per surgeon). K- means prototype clustering – an unsupervised ML algorithm – was used with two variables - operative duration and surgical complications within 6 weeks - to cluster operations into difficult or standard groups. Radiographic measurements (neck shaft angle, offset, LCEA, inter-teardrop distance, Tonnis grade) were measured by two independent observers. These factors, alongside patient factors (BMI, age, sex, laterality) were employed in a multivariate logistic regression analysis and used for k-means clustering. Significant continuous variables were investigated for predictive accuracy using Receiver Operator Characteristics (ROC). Out of 328 THAs analyzed, 130 (40%) were classified as difficult and 198 (60%) as standard. Difficult group had a mean operative time of 106mins (range 99–116) with 2 complications, while standard group had a mean operative time of 77mins (range 69–86) with 0 complications. Decreasing inter-teardrop distance (odds ratio [OR] 0.97, 95% confidence interval [CI] 0.95–0.99, p = 0.03) and right-sided operations (OR 1.73, 95% CI 1.10–2.72, p = 0.02) were associated with operative difficulty. However, ROC analysis showed poor predictive accuracy for these factors alone, with area under the curve of 0.56. Inter-observer reliability was reported as excellent (ICC >0.7). Right-sided hips (for right-hand dominant surgeons) and decreasing inter-teardrop distance were associated with case difficulty in AA-THA. These data could guide case selection during the learning phase. A larger dataset with more complications may reveal further factors

Summary. The ankle X-ray has moderate diagnostic power to identify syndesmotic instability, showing large sensitivity ranges between observers. Classification systems and radiographic measurements showed moderate to high interobserver agreement, with extended classifications performing worse. Introduction. There is no consensus regarding the diagnosis and treatment of ankle fractures with respect to syndesmotic injury. The diagnosis of syndesmotic injury is currently based on intraoperative findings. Surgical indication is mainly made by ankle X-ray assessment, by several classification systems and radiographic measurements. Misdiagnosis of the injury results in suboptimal treatment, which may lead to chronic complaints, like instability and osteoarthritis. This study investigates the diagnostic power and interobserver agreement of three classification methods and radiographic measures, currently used to assess X-ankles and to identify syndesmotic injury. Patients and Methods. Twenty patients (43.2 ± 15.3yrs) with an ankle fracture, indicated for surgery, were prospectively included. All patients received a preoperative ankle X-ray, which was assessed by several observers: two orthopaedic surgeons, one trauma surgeon and two radiologists. The ankle X-ray was assessed on syndesmotic injury/stability and presence of fractures (fibula, medial/tertius malleolus). Three classification systems were used: Weber, AO-Müller (short-version n=3 options; extended-version n=27 options), Lauge-Hansen (short-version n=5 options; extended-version n=17 options) and two radiographic measurements were done: tibiofibular overlap (TFO) and ratio medial clearspace/superior clear space (MCS/SCS). All observers were instructed about the assessments before the measurements. During surgery, a proper intraoperative description of the syndesmosis was noted. Agreement (%), Intraclass Correlation Coefficients (ICC) and Kappa were calculated to determine interobserver agreement. Kappa statistic was interpreted according to Landis and Koch. To test the diagnostic power of ankle X-rays to identify syndesmotic instability, sensitivity and specificity were calculated with intraoperative findings serving as golden standard. Results. Six of 20 ankles showed syndesmotic instability intraoperatively. An overall sensitivity of 43% (specificity: 78) was found for X-rays in identifying syndesmotic instability, showing a wide range in sensitivity between observers (17–83%), with radiologists performing better (range 50–83%) than surgeons (range: 17–33%). Overall, substantial to perfect interobserver agreement (range 70–100%) was found for all short classification systems, showing an average kappa ≥0.60. The agreement reduced for more extended classification systems. E.g. observer agreement for the AO-Muller classification with 3, 9 and 27 options was respectively 85% (kappa 0.66), 68% (kappa 0.57) and 55% (kappa 0.51). One observer deviated slightly from others in all classification assessments. Removing this observer resulted in excellent agreement for all classification systems (>90%). Radiographic measurements showed moderate to high interobserver agreement, with TFO performing best (avg. ICC 0.88). Discussion/Conclusion. In ankle fractures, a preoperative X-ray has low sensitivity in detecting syndesmotic instability, showing large sensitivity ranges between observers. Further study is needed to investigate the contribution of classification systems in determining the best treatment method for syndesmotic injury. Ankle X-ray assessment using the three classification systems and radiographic measures was consistent among observers. Disagreement between observers can be attributed to intrinsic differences among the systems (e.g. stepwise classification vs. single assessment). No preference for one specific classification was found, as all showed comparable interobserver agreement. However classification systems with few options are recommended, as the observer agreement reduced with more extending classifications