Comparison among load-, ROM-, and displacement-controlled methods used in the lumbosacral nonlinear finite-element...

For lumbosacral nonlinear analysis, the characteristics and differences between the load- and range-of-motion (ROM)-controlled methods (LCM and RCM) were compared using the numerical approach.

This study aimed to discuss the LCM and RCM problems inherent in the method assumption and calculation procedure. A displacement-controlled method (DCM) based on the nodal movement at the lumbosacral top was proposed to offer a more efficient and equivalent comparison between the evaluated models.

Both LCM and RCM have been extensively used to evaluate the biomechanical performance of lumbosacral implants. The LCM models were subject to the same loads as the intact model. The ROMs of the RCM models were controlled in the same way by iteratively adjusting some of the applied loads. However, the different strategies for adjusting lumbar loads might affect the predicted results and the execution might be inefficient. To the best of the authors' knowledge, the kinematic, mechanical, and computational comparisons between the 2 methods have still not been extensively investigated.

An intact lumbosacral model was developed and validated with the cadaveric and numerical data from the literature studies. The intact model was then modified as a degenerative model, in which the moderately dehydrated L4-L5 segment was instrumented with transpedicular fixation. Lumbosacral flexion was simulated by ligament interconnection, muscular contraction, and weight compression. One LCM, 3 RCM, and 1 DCM models were developed to evaluate their effects on biomechanical results and the computational efficiency of the lumbosacral nonlinear analysis.

Both solution feasibility and calculation time were closely related to the loading sequence that was defined as the time curves of the load-incremental control. The calculation of the RCM models was the most time-consuming. The calculation time of the DCM model was about 17 times faster than that of the RCM counterparts. Apart from the LCM model, the total ROM of the other models could be consistently controlled with the same value as that of the intact model. The intersegmental ROMs of all models were quite comparable. However, the LCM model predicted the least value of the screw stress and averaged 15.6% and 19.9% less than the RCM and DCM models. In general, the computational efficiency between the models was the most different, followed by the mechanical stress; the kinematic results were the most comparable.

The superiority of the computational efficiency of the DCM compared with its counterparts makes it the improved strategy for executing lumbosacral nonlinear analysis.