Validation of a patient-specific finite element analysis framework for identification of growing rod-failure regions in early onset scoliosis patients

PURPOSE

Growing rods are the gold-standard for treatment of early onset scoliosis (EOS). However, these implanted rods experience frequent fractures, requiring additional surgery. A recent study by the U.S. Food and Drug Administration (FDA) identified four common rod fracture locations. Leveraging this data, Agarwal et al. were able to correlate these fractures to high-stress regions using a novel finite element analysis (FEA) framework for one patient. The current study aims to further validate this framework through FEA modeling extended to multiple patients.

METHODS

Three patient-specific FEA models were developed to match the pre-operative patient data taken from both registry and biplanar radiographs. The surgical procedure was then simulated to match the post-operative deformity. Body weight and flexion bending (1 Nm) loads were then applied and the output stress data on the rods were analyzed.

RESULTS

Radiographic data showed fracture locations at the mid-construct, adjacent to the distal and tandem connector across the patients. Stress analysis from the FEA showed these failure locations matched local high-stress regions for all fractures observed. These results qualitatively validate the efficacy of the FEA framework by showing a decent correlation between localized high-stress regions and the actual fracture sites in the patients.

CONCLUSIONS

This patient-specific, in-silico framework has huge potential to be used as a surgical tool to predict sites prone to fracture in growing rod implants. This prospective information would therefore be vital for surgical planning, besides helping optimize implant design for reducing rod failures.

Reducing residual forces in spinal fusion using a custom-built rod bending machine

BACKGROUND AND OBJECTIVE

As part of spinal fusion surgery, shaping the rod implant to align with the anatomy is a tedious, error-prone, and time-consuming manual process. Inadequately contoured rod implants introduce stress on the screw-bone interface of the pedicle screws, potentially leading to screw loosening or even pull-out.

METHODS

We propose the first fully automated solution to the rod bending problem by leveraging the advantages of augmented reality and robotics. Augmented reality not only enables the surgeons to intraoperatively digitize the screw positions but also provides a human-computer interface to the wirelessly integrated custom-built rod bending machine. Furthermore, we introduce custom-built test rigs to quantify per screw absolute tensile/compressive residual forces on the screw-bone interface. Besides residual forces, we have evaluated the required bending times and reducer engagements, and compared our method to the freehand gold standard.

RESULTS

We achieved a significant reduction of the average absolute residual forces from for the freehand gold standard to (p=0.0015) using the bending machine. Moreover, our bending machine reduced the average time to instrumentation per screw from to . Reducer engagements per rod were significantly decreased from an average of 1.00±1.14 to 0.11±0.32 (p=0.0037).

CONCLUSION

The combination of augmented reality and robotics has the potential to improve surgical outcomes while minimizing the dependency on individual surgeon skill and dexterity.

Residual Stresses in Surgical Growing Rods

The treatment of early onset scoliosis using surgical growing rods suffers from high failure rate. Fatigue resistance can be improved by inducing compressive residual stresses within the near surface region. An in-depth investigation of the residual stresses profile evolution is performed through the sequence of material processing steps followed by surgeons handling operations, in connection to material properties. The final goal is to guide further improvements of growing rod lifetime. Residual stress evaluation was carried out on Ti-6Al-4V rods using digital image correlation applied to microbeam ring-core milling by focused ion beam. This provided experimental stress profiles in shot-peened rods before and after bending and demonstrated that compressive residual stresses are maintained at both concave and convex rod sides. A finite element model using different core and skin conditions was validated by comparison to experiments. The combination of an initial shot peening profile associated with a significant level of backstress was found to primarily control the generation of compressive stresses at the rod surface after bending. Guidelines to promote larger compressive stresses at the surface were formulated based on a parametric analysis. The analysis revealed the first order impact of the initial yield strength, kinematic hardening parameters and intensity of the shot peening operation, while the bending angle and the depth of shot peening stresses were found to be of minor importance. Materials exhibiting large kinematic hardening and low yield strength should be selected in order to induce compressive residual stresses at key fatigue initiation site.

Global stiffness and residual stresses in spinal fixator systems: A validated finite element study on the interconnection mechanism

Posterior spinal fixation systems are the gold standard to treat different column disorders using rods and screws. The proper connection between them is guaranteed by the Interconnection Mechanism (IM), consisting of different metallic subcomponents held together through the application of tightening torque. The response of the fixation system is defined by its overall stiffness, which in turn is governed by the local residual stress field arising during tightening. Although literature computational models for studying spinal fixation are becoming increasingly anatomically complex, most studies disregard completely the realistic modeling of the IM, namely choosing elastic-plastic material models and proper contact interactions. In this frame, the present study aims at increasing awareness in the field of spinal fixation modeling by investigating the mechanical response of the IM in terms of overall stiffness and local residual stresses. Once validated through dedicated experiments, the results of the proposed model have been compared with the current literature, highlighting the key role of the IM in the reliable modeling of spinal fixation.

Influence of double rods and interbody cages on range of motion and rod stress after spinopelvic instrumentation: a finite element study

PURPOSE

To compare instrumentation configurations consisting of bilateral single or double rods and additional interbody cages (IBCs) at different levels in terms of Range of Motion (ROM) and distribution of von Mises stress in rods.

METHODS

A previously validated L1-pelvis finite element model was used and instrumented with configurations consisting of single or double bilateral rods and IBCs at multiple levels. Pure moments of 7.5 N.m were applied to L1 in main directions in addition to a follower load of 280 N. Global, segmental ROM and distribution of von Mises stress in rods were studied.

RESULTS

All configurations reduced segmental and global ROM from 50 to 100% compared to the intact spine. Addition of IBCs slightly increased ROM at levels adjacent to the IBC placement. The simple rod configuration presented the highest von Mises stress (457 MPa) in principal rods at L5-S1 in flexion. Doubling rods and IBC placement reduced this value and shifted the location of maximum von Mises stress to other regions. Among studied configurations, double rods with IBCs at all levels (L2-S1) showed the lowest ROM. Maximal von Mises stresses in secondary rods were lower in comparison to main rods.

CONCLUSIONS

Double rods and IBCs reduced global and segmental ROM as well as von Mises stress in rods. The results suggest a possible benefit in using both strategies to minimize pseudarthrosis and instrumentation failure. However, increased ROM in adjacent levels and the shift of maximal von Mises stress to adjacent areas might cause complications elsewhere.

How do spine instrumentation parameters influence the 3D correction of thoracic adolescent idiopathic scoliosis? A patient-specific biomechanical study

BACKGROUND

Patient-specific models promises to support the surgical decision-making process, particularly in adolescent idiopathic scoliosis. The present computational biomechanical study investigates how specific instrumentation parameters impact 3D deformity correction in thoracic scoliosis.

METHODS

1080 instrumentation simulations of a representative patient were run. The independent instrumentation parameters were: screw pattern, upper and lower instrumented vertebrae, rod curvature and rod stiffness. ANOVA and correlation analyses analyzed how the instrumentation parameters influenced the 3D correction.

FINDINGS

Coronal plane correction was affected by the lower instrumented vertebra and rod stiffness (explaining 84% and 11%, respectively, of its overall variance). The sagittal profile was controlled by rod curvature and the upper vertebra (56% and 36%). The transverse plane vertebral rotation was influenced by lower, upper instrumented vertebra and screw pattern (35%, 32% and 19%). The Cobb angle correction was strongly correlated with the number of fused vertebrae, particularly when grouped by the upper instrumented vertebra (r = -0.91) and rod stiffness (r = -0.73). Thoracic kyphosis was strongly correlated with the number of fused vertebrae grouped by rod curvature (r = 0.84). Apical vertebral rotation was moderately correlated with the number of fused vertebrae grouped by upper/lower instrumented vertebra (r = 0.55/0.58), although variations were minimal.

INTERPRETATION

Instrumenting the last vertebra touching the central sacral vertical line improves 3D correction. A trade-off between a more cranial vs. caudal upper instrumented vertebra, respectively beneficial for coronal/sagittal vs. transverse plane correction, is required. High rod stiffness, differential rod contouring, and screw pattern were effective for coronal correction, thoracic kyphosis, and axial vertebral derotation, respectively.

Load-sharing biomechanics of lumbar fixation and fusion with pedicle subtraction osteotomy

Pedicle subtraction osteotomy (PSO) is an invasive surgical technique allowing the restoration of a well-balanced sagittal profile, however, the risks of pseudarthrosis and instrumentation breakage are still high. Literature studied primary stability and posterior instrumentation loads, neglecting the load shared by the anterior column, which is fundamental to promote fusion early after surgery. The study aimed at quantifying the load-sharing occurring after PSO procedure across the ventral spinal structures and the posterior instrumentation, as affected by simple bilateral fixation alone, with interbody cages adjacent to PSO level and supplementary accessory rods. Lumbar spine segments were loaded in vitro under flexion–extension, lateral bending, and torsion using an established spine tester. Digital image correlation (DIC) and strain-gauge (SG) analyses measured, respectively, the full-field strain distribution on the ventral surface of the spine and the local strain on posterior primary rods. Ventral strains considerably decreased following PSO and instrumentation, confirming the effectiveness of posterior load-sharing. Supplemental accessory rods considerably reduced the posterior rod strains only with interbody cages, but the ventral strains were unaffected: this indicates that the load transfer across the osteotomy could be promoted, thus explaining the higher fusion rate with decreased rod fracture risk reported in clinical literature.

In silico patient-specific optimization of correction strategies for thoracic adolescent idiopathic scoliosis

BACKGROUND

With modelling and simulation (or in silico) techniques, patient-specific optimization algorithms represent promising tools to support the surgical decision-making process, particularly in 3D correction of adolescent idiopathic scoliosis, where the best intraoperative instrumentation strategy and the correction goals are debated.

METHODS

1080 biomechanical intraoperative simulations of a representative pediatric thoracic curve were run according to a full-factorial design approach. Widely accepted instrumentation configurations (5 screw patterns, 4 upper and 3 lower instrumented vertebrae, 6 rod curvatures and 3 rod stiffnesses) were analyzed, assuming concave rod rotation and en bloc derotation as main correction maneuvers. Results in terms of 3D correction and mobility were rated using an objective function for thoracic scoliosis also including surgeon-dependent correction objectives. An extensive sensitivity analysis on correction objectives was performed.

FINDINGS

Multiple optimal strategies were identified, depending on the selected correction objective. They provided significantly better coronal (67% vs. 55%) correction, using comparable instrumented levels (9.9 ± 1.6 vs. 10.7 ± 2.1), screw patterns and significantly higher implant density (1.6 ± 0.3 vs. 1.4 ± 0.2 screws/vertebra) compared to worst ones. Optimal strategies typically included the neutral and the last touching vertebrae in the construct and high stiffness (CoCr, 6 mm) differentially/highly contoured rods.

INTERPRETATION

The computerized algorithm determined the best instrumentation parameters to achieve optimal correction for the considered thoracic case. Multiple clinically equivalent strategies may be used, as supported by the variety of considered correction objectives. The current approach could be translated to any scoliotic curves, including surgeon preferences in terms of instrumentation parameters, intraoperative correction maneuvers and correction objectives.

Correction objectives have higher impact than screw pattern and density on the optimal 3D correction of thoracic AIS: a biomechanical study

STUDY DESIGN

Assessment of screw pattern, implant density (ID), and optimization of 3D correction through computer-based biomechanical models.

OBJECTIVE

To investigate how screw pattern and ID affect intraoperative 3D correction of thoracic curves in adolescent idiopathic scoliosis, and how different correction objectives impact the optimal screw pattern. Screw pattern, ID, correction objectives and surgical strategies for posterior fusion of AIS are highly variable among experienced surgeons. The "optimal" instrumentation remains not well defined.

METHODS

10 patient-specific multibody models of representative adolescent idiopathic scoliosis Lenke 1A cases were built and used to compare alternative virtual correction surgeries. Five screw patterns and IDs (average: 1.6 screws/instrumented level, range: 1.2-2) were simulated, considering concave rod rotation, en bloc derotation, and compression/distraction as primary correction maneuvers. 3D correction descriptors were quantified in the coronal, sagittal and transverse planes. An objective function weighting the contribution of intraoperative 3D correction and mobility allowed rating of the outcomes of the virtual surgeries. Based on surgeon-dependent correction objectives, the optimal result among the simulated constructs was identified.

RESULTS

Low-density (ID ≤ 1.4) constructs provided equivalent 3D correction compared to higher (ID ≥ 1.8) densities (average differences ranging between 2° and 3°). The optimal screw pattern varied from case to case, falling within the low-density screw category in 14% of considered scenarios, 73% in the mid-density (1.4 < ID < 1.8) and 13% in the high-density. The optimal screw pattern was unique in five cases; multiple optima were found in other cases depending on the considered correction objectives.

CONCLUSIONS

Low-density screw patterns provided equivalent intraoperative 3D correction to higher-density patterns. Simulated surgeon's choice of correction objectives had the greatest impact on the selection of the optimal construct for 3D correction, while screw density and ID had a limited impact.

LEVEL OF EVIDENCE

N/A.

Biomechanical in vitro comparison between anterior column realignment and pedicle subtraction osteotomy for severe sagittal imbalance correction

PURPOSE

To investigate the biomechanical effects of anterior column realignment (ACR) and pedicle subtraction osteotomy (PSO) on local lordosis correction, primary stability and rod strains.

METHODS

Seven cadaveric spine segments (T12-S1) underwent ACR at L1-L2. A stand-alone hyperlordotic cage was initially tested and then supplemented with posterior bilateral fixation. The same specimens already underwent a PSO at L4 stabilized by two rods, a supplemental central rod (three rods) and accessory rods (four rods) with and without adjacent interbody cages (La Barbera in Eur Spine J 27(9):2357-2366, 2018). In vitro flexibility tests were performed under pure moments in flexion/extension (FE), lateral bending (LB) and axial rotation (AR) to determine the range of motion (RoM), while measuring the rod strains with strain gauge rosettes.

RESULTS

Local lordosis correction with ACR (24.7° ± 3.7°) and PSO (25.1° ± 3.9°) was similar. Bilateral fixation significantly reduced the RoM (FE: 31%, LB: 2%, AR: 18%), providing a stability consistent with PSO constructs (p > 0.05); however, it demonstrates significantly higher rod strains compared to PSO constructs with lateral accessory rods and interbody cages in FE and AR (p < 0.05), while being comparable in FE or slightly higher in AR compared to PSO constructs with two and three rods.

CONCLUSION

Bilateral posterior fixation is highly recommended following ACR to provide adequate primary stability. However, primary rod strains in ACR were found comparable or higher than weak PSO construct associated with frequent rod failure; therefore, caution is recommended. These slides can be retrieved under Electronic Supplementary Material.

Computational and Experimental Fatigue Analysis of Contoured Spinal Rods

Posterior fixation with contoured rods is an established methodology for the treatment of spinal deformities. Both uniform industrial preforming and intraoperative contouring introduce tensile and compressive plastic deformations, respectively, at the concave and at the convex sides of the rod. The purpose of this study is to develop a validated numerical framework capable of predicting how the fatigue behavior of contoured spinal rods is affected by residual stresses when loaded in lordotic and kyphotic configurations. Established finite element models (FEM) describing static contouring were implemented as a preliminary simulation step and were followed by subsequent cyclical loading steps. The equivalent Sines stress distribution predicted in each configuration was compared to that in straight rods (SR) and related to the corresponding experimental number of cycles to failure. In the straight configuration, the maximum equivalent stress (441 MPa) exceeds the limit curve, as confirmed by experimental rod breakage after around 1.9 × 105 loading cycles. The stresses further increased in the lordotic configuration, where failure was reached within 2.4 × 104 cycles. The maximum equivalent stress was below the limit curve for the kyphotic configuration (640 MPa), for which a run-out of 106 cycles was reached. Microscopy inspection confirmed agreement between numerical predictions and experimental fatigue crack location. The contouring technique (uniform contouring (UC) or French bender (FB)) was not related to any statistically significant difference. Our study demonstrates the key role of residual stresses in altering the mean stress component, superposing to the tensile cyclic load, potentially explaining the higher failure rate of lordotic rods compared to kyphotic ones.