Content area
Abstract
Adolescent idiopathic scoliosis (AIS) is a three-dimensional (3D) local and global deformation of the spine, which may require spinal instrumentation and fusion for severe deformities.
The objective of this study is threefold: the first aims to document and analyze 3D surgical correction goals for AIS as determined by a sample of experienced spine surgeons. Afterward, the second part aims to develop an optimization method, personalized to the specific correction objectives of a given surgeon, and to determine the most favourable surgical instrumentation strategy using a computer model implemented in a spine surgery simulator (S3). The third part aims to examine the effects of correction objectives on the optimal instrumentation strategy.
In the first part of the project, fifty surgeons from the Spinal Deformity Study Group were surveyed and asked to rank 20 parameters of scoliosis correction and to provide weights for correction in the coronal, sagittal, and transverse planes and for mobility (number of unfused vertebrae) according to their importance for an optimal 3-D correction. Responders were also asked to complete a more detailed survey where the correction objectives were assessed for each of the 6 Lenke curve types. In the second part, an optimization method using six instrumentation design parameters (e.g. limits of the instrumented segment, number, type and location of implants and rod shape) that were manipulated in a uniform experimental design framework was linked to a patient-specific biomechanical model. 702 surgical configurations iteratively were simulated using S3. Each configuration was assessed using an objective function that represented different correction objectives (to maximize) in the three anatomic planes and was oriented to minimize the number of instrumented levels. The relative weights of the objective function were defined by a spine surgeon according to his objectives for correction of scoliosis. An interpolation technique was used to build an approximation model from the simulation results and to locate instrumentation parameters minimizing the objective function. In the third part, eleven surgeons from the Spinal Deformity Study Group independently provided their respective correction objectives for the same patient (Lenke 2B curve type). For each surgeon, the optimization approach was used and the optimal strategy was found for the same patient. The influence of the eleven different correction objectives on the optimal surgical strategy was then evaluated.
For the assessment study of the scoliosis correction objective in AIS, twenty-five surgeons completed the first questionnaire. There was an overall agreement that sagittal and coronal balance were the most important parameters for an optimal correction. Apical vertebral rotation was the least important. All other parameters were highly variable. Results for individual parameters were in agreement with the weights given for an optimal 3D correction in the coronal (36%) and sagittal (34%) planes. A sub group of ten surgeons completed the second survey. Mobility was more important for the Lenke curve types 3–6 than for types 1–2 (p<0.032). The coronal plane was more important for curve types 2 and 4 than for the other types (p<0.032).
The optimization study of instrumentation strategies showed that small or no differences in the correction between the simulated optimal strategy and the real post-operative results of the instrumented segments were observed in the three planes. But the same overall correction was obtained by using fewer implants (only screws) and less instrumented levels.
The results of the third part showed that the fusion levels (T4-L2, T2-L4, T5-L1, etc), shape of the rod and the location of implants significantly were influenced by the correction objective strategies (p<0.05). The optimal number of implants was different (ranging from 8 to 13) and statistically significant (p< 0.05). The resulting Cobb angles, the kyphosis and the lordosis angles as well as the orientation of the plane of maximum deformity varied for the 11 simulated optimal strategies.
In conclusion, there is a large variability in scoliosis correction objectives. The variability is surgeon and curve-type dependent. This study demonstrated the potential and feasibility of using a spine surgery simulator to optimize the planning of surgical instrumentation in AIS for a specific correction objectives surgeon. Different surgeon-specified correction objectives produced different optimal instrumentation strategies for the same patient. A standardized decision-making protocol is required to minimize the inherent variability in defining AIS surgical correction objectives. (Abstract shortened by UMI.)
