Modification and validation of Lenke3 type adult idiopathic scoliosis finite element model / 中国组织工程研究

ABSTRACT

BACKGROUND: A Lenke3 type adult idiopathic scoliosis finite element model was established successfully using Mimics software. However, whether the model fits the actual conditions of individualized patients still requires a further revision and validation.OBJECTIVE: To modify and validate the Lenke3 type adult idiopathic scoliosis finite element model by finite element analysis software.METHODS: Based on the characteristics of Lenke3 adult idiopathic scoliosis model, the three-factor and three-level orthogonal experiment was used to optimize the finite element model, making it more close to the actual one. The vertebrae at T1-T4, T5-T8 and L6-S1 levels (sacral lumbarization) were loaded to simulate left and right lateral flexion,as well as extension and flexion, and the range of motion when left and right rotation were compared with Busscher and Yamamoto experiments in vitro.RESULTS AND CONCLUSION: (1) According to the orthogonal experiment, the mean difference and range of each factor and each level were calculated, and finally A1B2C3 combination was the optimal one that can make the model largely consistent with the real situation. The difference in Cobb angles between the clinical lateral flexion test and the parameter pre-modified model simulation was 54.44°, which was decreased to 2.11° after modification. Moreover, the maximum difference in each scoliosis Cobb angle of the modified model was 4.29°. (2) The simulation results of the modified model when compared with the X-ray images when left and right lateral flexion, the two data obeyed normal distribution, so the paired t test was used left lateral flexion, P =0.082 (P > 0.05); right lateral flexion, P=0.421 (P > 0.05);supine position, P=0.160 (P > 0.05). (3) The range of motion at T1-T4 segments was as followings left flexion, 3.25°;right flexion, 3.32°; anteflexion 2.52°; extension, 2.89°; left rotation, 3.73°; right rotation 3.76°; the range of motion at T5-T8 segments left flexion, 1.39°; right flexion, 1.43°; anteflexion 1.35°; extension, 1.34°; left rotation 2.09°; right rotation 2.11°; the range of motion at L6/S1 left flexion 5.17°; right flexion 5.19°; anteflexion 8.92°; extension 7.35°; left rotation 1.41°; right rotation 1.42°. The results were almost consistent with Busscher and Yamamoto experimental results. (4) To conclude, the model is in good agreement with the patient's actual properties after modification. The modified model has good reliability and validity, and provides valid data platform for simulating clinical operation in the future.