采用医学图像的三维建模技术及有限元技术，比较单独使用外侧锁定钢板（lateral locking plate，LLP）及单独使用内侧锁定钢板（medial locking plate,MLP）对肱骨近端骨折内侧柱支撑的生物力学稳定性差异，从而为此类骨折的临床处理提出相关建议。

通过Mimics 21.0、Geomagic Wrap 2017医学影像软件对获取的肱骨近端CT图像进行加工处理，利用Solidworks 2017造模软件设计肱骨近端的骨折缺损模型，并建立含有锁定钢板和螺钉的内固定模型。随后根据AO骨折治疗原则将内植物与已设计好的肱骨近端骨折模型进行装配，并分组：A组为LLP治疗组；B组为MLP治疗组，将以上两组模型以STEP格式导入ANSYS Workbench 2019有限元分析软件中进行网格划分后赋值计算。定义肱骨远端为完全固定，分别对A组和B组进行轴向压缩力、剪切力以及扭转力三种不同方式的载荷实验。比较两种骨折内固定模型在三种不同载荷条件下肱骨及内植物应力大小、应力分布、位移等情况。

在轴向压缩、剪切、扭转三种不同载荷加载情况下，肱骨的最大应力、最大位移、内植物的最大应力、最大位移以及骨折端的相对位移B组结果均小于A组。

在外侧皮质完整而内侧柱缺失的肱骨近端骨折中，MLP固定在轴向、剪切和扭转载荷下提供了优于LLP的结构稳定性，是治疗肱骨近端骨折内侧柱不稳定的有效方法，本项研究为这种新型内侧支撑技术的生物力学优势提供了证据从而为内侧柱缺失的肱骨近端骨折的手术治疗提供了更多的临床方案。

Proximal humerus fractures account for about 4%-5% of systemic fractures in clinical practice, ranking third in human incidence after hip and distal radial fractures. The classification and treatment of proximal humeral fractures are varied, and the optimal treatment is still widely debated. In recent years, internal fixators have been developed due to the continuous improvement of surgical technology. Various internal fixation methods, such as percutaneous internal fixation technology, locking plates, intramedullary nails, and shoulder joint replacement, are available. Studies have confirmed that using a lateral locking plate (LLP) for proximal humeral fractures has good biomechanical properties and become a commonly used fixation method in clinical practice for proximal humeral fractures with significant efficacy. It has the advantages of a stable structure and mature technology. However, LLP has shown a high reoperation rate with the increasing use of locking plates in proximal humeral fractures. Many associated complications have begun to trouble most orthopedic doctors, such as postoperative fracture varus deformity, malunion, screw release, bone nonunion, subacromial impingement, humeral head ischemic necrosis, infection, etc. Among them, the incidence of postoperative fracture varus deformity is high. Among the factors leading to this series of postoperative complications, inadequate medial column support is widespread in proximal humerus fractures. The lack of adequate medial column support for treating proximal humeral fractures with a LLP alone has not been identified. Some scholars have proposed adding humerus talus screws to the lateral plate. In addition, they also used the bone cement filling, the combined application of anterolateral and lateral plates, and the combined application of posterior and lateral plates to address the absence of the proximal humeral medial column. However, none of these surgical methods directly support the medial humerus to improve biomechanical stability. Yu He et al. used both the LLP and the medial locking plate (MLP) to provide strong support for the medial column to enhance the stability of the fracture site and reduce varus deformity. However, this kind of double-incision surgery will inevitably cause excessive damage to the soft tissue of the shoulder joint. In addition, due to the limited space at the proximal end of the humerus, the screws inevitably interact with each other. Therefore, to reduce the damage to the soft tissue of the shoulder joint and avoid the interaction between screws, and provide direct support to the medial column, we wondered if a MLP could be used alone to treat proximal humeral fractures.

To compare the biomechanical stability of medial column support for proximal humeral fractures using LLP and MLP with three-dimensional modeling technology and finite element analysis to put forward relevant suggestions for the clinical treatment of such fractures.

The CT DICOM data of the proximal humerus were obtained and processed by Mimics 21.0 and Geomagic Wrap 2017 software. The proximal humerus fracture defect model was designed by SolidWorks 2017 software, and the internal fixation model containing the locking plate and screw was established. Subsequently, according to the principles of AO fracture treatment, the implants were assembled with the designed proximal humeral fracture model and divided into group A (LLP group) and group B (MLP group). The above two groups of models were imported into the ANSYS Workbench 2019 finite element analysis software in STEP format for mesh assignment and calculation. The distal humerus was defined as complete fixation. The axial compression, shear, and torsional forces were tested in groups A and B, respectively. The humerus and implants' stress magnitude, distribution, and displacement were compared between the two internal fixation models under three different loading conditions.

Under axial compression, shear, and torsion, the maximum stress and displacement of the humerus, the maximum stress and displacement of the implant, and the relative displacement of the fracture end in group B were less than in group A.

In proximal humeral fractures with intact lateral cortex but absent medial column, the MLP fixation provides better structural stability than that of the LLP under axial, shear, and torsional loads, which is an effective method for the treatment of medial column instability in proximal humeral fractures. This study provides evidence for the biomechanical advantages of this new medial support technique. It thus offers additional clinical options for the surgical treatment of proximal humeral fractures with medial column loss.