RÉSUMÉ
BACKGROUND:Internal fixation and open reduction with locking plate is the main treatment for proximal humeral fractures with medial column instability.However,reduction failure is one of the main postoperative complications,and accurate risk factor assessment is beneficial for screening high-risk patients and clinical decision selection. OBJECTIVE:To construct four types of prediction models by different machine learning algorithms,compare the optimal model to analyze and sort the risk variables according to their weight scores on the impact of outcome,and explore their significance in guiding clinical diagnosis and treatment. METHODS:262 patients with proximal humeral fractures with medial column instability,aged(60.6±10.2)years,admitted to Foshan Hospital of Traditional Chinese Medicine between June 2012 and June 2022 were included.All patients underwent open reduction with locking plate surgery.According to the occurrence of reduction failure at 5-month follow-up,the patients were divided into a reduction failure group(n=64)and a reduction maintenance group(n=198).Clinical data of patients were collected,and model variables and their classification were determined.The data set was randomly divided into a training set and a test set according to a 7:3 ratio,and the optimal hyperparameters were obtained in the training set according to a 5-fold cross-over test.Four machine learning prediction models of logistic regression,random forest,support vector machine,and XGBoost were constructed,and the performance of different algorithms was observed in the test set using AUC,correctness,sensitivity,specificity,and F1 scores,so as to comprehensively evaluate the prediction performance of the models.The best-performing model was evaluated using SHAP to assess important risk variables and to evaluate its clinical guidance implications. RESULTS AND CONCLUSION:(1)There were significant differences between the two groups in deltoid tuberosity index,fracture type,fracture end with varus deformity before operation,fragment length of inferior metaphyseal of humerus,postoperative reduction,cortical support of medial column of proximal humerus,and insertion of calcar screw(P<0.05).(2)The best-combined performance of the four machine models was XGBoost.The AUC,accuracy,and F1 scores were 0.885,0.885,and 0.743,respectively;followed by random forest and support vector machine,with both models performing at approximately equal levels.Logistic regression had the worst combined performance.The SHAP interpretation tool was used in the optimal model and results showed that deltoid tuberosity index,medial humeral column cortical support,fracture type,fracture reduction quality,and the status of the calcar screw were important influencing fators for postoperative fracture reduction failure.(3)The accuracy of using machine learning to analyze clinical problems is superior to that of traditional logistic regression analysis methods.When dealing with high-dimensional data,the machine learning approach can solve multivariate interaction and covariance problems well.The SHAP interpretation tool can not only clarify the importance of individual variables but also obtain detailed information on the impact of dummy variables in each variable on the outcome.
RÉSUMÉ
BACKGROUND: Flatfoot is a commonly seen disease in foot and ankle surgery, and stage II adult acquired flatfoot is mostly seen in clinic, so this stage is a key to treatment. However, medial column instability occurs in stage II adult acquired flatfoot, which is an important cause for arch collapse. Medial column stabilization can correct the deformity to great extent, but there is a lack of biomechanical study to assess the effect of medial column stabilization on the whole foot. OBJECTIVE: To investigate the biomechanical effects of medial column stabilization on stage II adult acquired flatfoot. METHODS: A three-dimensional finite element model of stage IIa and IIb adult acquired flatfoot was established. Geomagic software, Solidwork software and Abaqus software were used to simulate medial column stabilization operation (naviculocuniform joint fusion, tarsometatarsal joint fusion, and both fusion). The maximum pressure of plantar soft tissue, medial column bone and medial ligaments was compared before and after simulated single-foot weight loading. Meanwhile, the related parameters were measured to carry out a comprehensive comparison. RESULTS AND CONCLUSION: (1) The maximum plantar stress was located under the first metatarsal head after the simulated medial column stabilization operation. The maximum plantar stress increased significantly after the medial column stabilization in stage IIa flatfoot model, but did not change significantly after the medial column stabilization in stage IIb model. (2) After medial column fusion, the stress of the corresponding joint was reduced, but increased for the other joints of the first metatarsal column. (3) The stress of medial ligament and plantar fascia was not alleviated after medial column fusion. (4) These results indicate that simple medial column stabilization surgery cannot reduce the pressure of medial column of flatfoot in stage II acquired flatfoot adults. It can only be used as a combined surgery to stabilize joints with excessive motion and correct the deformity of supination of forefoot.
RÉSUMÉ
BACKGROUND: The medial column fracture of the proximal humeral fracture is a type of fracture with high postoperative complications after internal fixation. Proximal humeral internal locking system is widely used in the proximal humeral fracture. The biomechanical stability of two different internal fixation models for the treatment of the poor medial column support proximal humeral fractures with proximal humeral internal locking system and proximal humeral intemal locking system with fibular allograft augmentation is still unclear. OBJECTIVE: To investigate the biomechanical stability of the proximal humeral internal locking system and proximal humeral internal locking system with fibular allograft augmentation in the poor medial column support proximal humeral fracture with finite element analysis, and to compare the difference in fixation modes in humeral calcar supporting screws under different fixing methods. METHODS: The proximal CT data of osteoporosis were obtained. According to the 5-mm osteotomy of the humerus, the posterior medial column of the humerus was established and divided into two groups. Group A was the proximal humeral intemal locking system plate group, in which the A1 group was placed in the proximal end of the plate, the A2 group was the E screw missing, the A3 group was the D screw missing; the B group received the proximal humeral intemal locking system plate combined with fibular allograft augmentation, and all screws was placed in the proximal end of the plate in the B1 group. In the B2 group, E screw was missing; in the B3 group, D screw was missing. The three-dimensional finite element method was used to analyze the biomechanical stability of different groups of models under indirect violence. RESULTS AND CONCLUSION: (1) In terms of structural stability, B group was significantly better than A group. The structural stability of A1 group and A3 group of A group was comparable and better than that in A2 group. In B group, the structural stability of B1 group and B2 group was comparable and superior to B3 group. (2) For the poor medial column support proximal humeral fracture, the mechanical stability of the proximal humeral intemal locking system plate combined with fibular allograft augmentation was better than that of the proximal humeral internal locking system plate alone. When the fibular support was combined, the screw placement in the proximal end of the plate is optimal, and D screw has an important stabilizing effect on the support of the poor medial column support proximal humeral fracture.