Abstract
OBJECTIVE
Use finite element analysis to evaluate the biomechanical effects of spinal decompression procedures in healthy Beagle dogs, comparing individualized mini-hemilaminectomy–corpectomy (iMHC), mini-hemilaminectomy, partial lateral corpectomy (PLC), and hemilaminectomy.
METHODS
A finite element model of the L1-L2 functional spinal unit was generated using CT data. For each decompression model, loads were applied in 0.2-Nm steps (maximum, 2.0 Nm) in 6 directions: flexion, extension, right and left lateral bending, and right and left axial rotation. The L1 spinous process tip displacement angle was quantified numerically.
RESULTS
Among the 4 techniques, mini-hemilaminectomy exhibited the smallest displacement angles across all directions. Hemilaminectomy exhibited the largest displacement angles in extension, flexion, right rotation, and left rotation across all techniques. Left and right lateral bending displacement angles were marginally larger for iMHC than for hemilaminectomy at 0.4 Nm; however, at 2.0 Nm, displacement angles were similar.
CONCLUSIONS
Mini-hemilaminectomy minimizes functional spinal unit instability to the greatest extent. Hemilaminectomy is more unstable than iMHC and PLC in flexion, extension, and rotation. Mini-hemilaminectomy–corpectomy and PLC are more unstable than hemilaminectomy in lateral bending, with iMHC being slightly more unstable than PLC or nearly equal.
CLINICAL RELEVANCE
Mini-hemilaminectomy minimizes instability to the greatest extent in cases of ventrolateral spinal compression. In cases of ventral spinal compression, iMHC may be preferable to PLC for providing equivalent stability without impeding spinal cord visualization, but both techniques can cause instability depending on loading direction, so careful attention to postoperative instability is necessary when excessive vertebral body resection is involved.
Numerous surgical techniques have been documented in the literature1–6 for achieving spinal cord decompression in dogs with thoracolumbar spinal surgical diseases. The optimal technique involves removing deviant material to expose the spinal canal, thereby preventing iatrogenic spinal cord injury while preserving the postoperative biomechanical stability of the intervertebral region.7–9 Epidural compression within the ventrolateral or ventral spinal canal commonly occurs due to intervertebral disc disease.7–9 Hemilaminectomy (HL) and mini-hemilaminectomy (mHL) are commonly used for ventrolateral approaches because they expose the lateral to ventrolateral aspects of the spinal canal. However, these techniques limit ventral access,5,10,11 and attempting to remove ventrally positioned epidural material using HL or mHL may require excessive spinal manipulation, resulting in postoperative neurological deterioration.12,13
Depending on the specific circumstances, HL or mHL may be combined with partial lateral corpectomy (PLC), which offers an adequate ventral approach to the spinal canal, enabling the removal of protruding or deviated disc material with minimal spinal manipulation.6–14 Although reports6 indicate that PLC alone does not compromise spinal stability, the concurrent use of PLC and HL for the same intervertebral region significantly deteriorates spinal stability and is not recommended without attempting vertebral fixation.15 Individualized mini-hemilaminectomy–corpectomy (iMHC) combines PLC and mHL. Although iMHC has been reported16 to improve the effectiveness of surgical spinal decompression in canines without causing spinal instability or collapsing the intervertebral space, an in vitro biomechanical study17 suggested that iMHC decreases spinal stability. Variability in the reported biomechanical spinal effects of iMHC across different studies may be due to interindividual differences between clinical cases and cadaveric dogs. In addition, maintaining a consistent corpectomy range across multiple cases in clinical scenarios can be challenging. To overcome the challenging ethical considerations associated with clinical cases and substantial interindividual differences, biomechanical investigations18 have increasingly used finite element analysis. Kikuchi et al19 reported that the L1-L2 outcomes of mechanical tests conducted using finite element models of the lumbar spine region were similar to the outcomes obtained in cadaveric dogs.
In biomechanical studies, the initial increase in the range of motion observed when a microload is applied to the spine is called the neutral zone; the subsequent gradual increase in the range of motion is known as the elastic zone.20 These zones exhibit distinct spinal motion characteristics. The neutral zone encompasses the physiological range of motion associated with comfortable, low-risk, flexible, and natural movements typical in everyday activities, such as standing and walking, which are facilitated by the elastic properties of muscles and ligaments.21,22 By contrast, the elastic zone encompasses the range of motion provided by the structural integrity of the intervertebral discs and bones,20 aligning closely with physiological motion limits encountered during more forceful and extensive movements, such as jumping and abrupt directional changes.21,22
This study developed an L1-L2 segment model using CT data obtained from a healthy Beagle dog and used this model to replicate the iMHC, PLC, mHL, and HL techniques. Finite element analysis was then used to compare and quantify the mechanical stability and spinal segment behaviors in the neutral and elastic zones in response to loads applied in the directions of flexion, extension, lateral bending, and rotation after each surgical technique. Based on previous findings, we hypothesized that HL would result in reduced stability compared with the other techniques and that iMHC would result in reduced stability relative to PLC and mHL.
Methods
Finite element model of intact lumbar functional spinal unit
A 3-D finite element model of the L1-L2 functional spinal unit was developed using CT data obtained from a healthy Beagle dog (age, 2 years, 3 months; female; weight, 12.8 kg) with a normal neurological examination, normal blood test results, and an overall normal general condition. Approval for animal experiments was obtained from the Animal Experimentation Committee and Bioethics Committee of Nippon Veterinary and Life Science University (approval No. 2020K-10). The CT scan was performed using an 80-row, 160-slice CT system (Aquilion PRIME; Toshiba Medical Systems). A surface model of the L1-L2 functional spinal unit was generated from CT data using bone strength evaluation software (MECHANICAL FINDER, version 13.0; Research Center of Computational Mechanics). With the use of this surface model, material properties were assigned, boundary conditions and load constraints were established, and an analysis was conducted to meticulously recreate a solid model of the L1 and L2 vertebrae using tetrahedral meshes ranging from 0.3 to 3.0 mm. These vertebrae were designated as different regions of interest to distinguish them as separate bones. The intervertebral disc, comprising the nucleus pulposus and annulus fibrosus, was detailed with a 0.1- to 0.4-mm tetrahedral mesh to accurately visualize its boundaries, and the disc was reproduced as closely as possible using varying contrast levels.23,24 The L1 vertebra includes 63,676 elements, the L2 vertebra includes 66,912 elements, and the intervertebral disc contains 13,768 elements. Articular surfaces of facet joints were joined by surface-to-surface friction contact (coefficient of friction = 0.01).25,26 The vertebral body was attached to the disc by tie constraints. Five ligaments of the motion segment were reproduced: the anterior longitudinal ligament, posterior longitudinal ligament, supraspinous ligament (SSL), interspinous ligament (ISL), and ligamentum flavum. Using 3-D modeling software (Metasequoia 4; Tetraface), the anterior longitudinal ligament and posterior longitudinal ligament were each modeled as a single 0.3-mm-thick band along the ventral and dorsal surfaces of the vertebral body, respectively. The SSL, ISL, and ligamentum flavum were modeled as axial springs with tension connector behavior. The anatomical positioning and trajectory of these ligaments were established based on previous anatomical data reports27 (Supplementary Figure S1). The material properties of the various components were differentiated based on variations in Young modulus and Poisson ratio, referencing previously reported values for both canine and human lumbar spine models28–33 (Supplementary Table S1).
Model for simulating spinal decompression surgery
The untreated L1-L2 model was considered intact (Figure 1). Four spinal decompression procedures, iMHC, mHL, HL, and PLC, were simulated on the left side of the model. For the mHL model, the caudal one-fourth of the L1 vertebra was designated the cephalad margin, and the caudal one-fourth of the L2 vertebra was designated the caudal margin, extending dorsally from the bases of the L1 and L2 vertebral pedicles. In this model, the accessory process was removed, and the articular processes were preserved. For the PLC model, the caudal one-fourth of the L1 vertebra was designated the cranial margin, and the cranial one-fourth of the L2 vertebra was designated the caudal margin. The depth was set at half the width of the L1 and L2 vertebrae. The ventral margin was established at half the height of the L1 and L2 vertebrae, whereas the dorsal margin was aligned with the base of the spinal canal. For the iMHC model, the range combined the PLC and mHL parameters, adhering to the criteria specified above. For the HL model, the caudal one-fourth of the L1 vertebra was designated the cranial margin, and the cranial one-fourth of the L2 vertebra was designated the caudal margin. The ventral margin was the base of the spinal canal. The accessory process and the articular processes were resected, and the dorsal margins were up to the base of the spinous process.
Schematic of each decompression model and 6-axis degrees of freedom setup and loading conditions. A—Intact model. B—Mini-hemilaminectomy model (mHL). C—Partial lateral corpectomy (PLC) model. D—Individualized mini-hemilaminectomy–corpectomy model (iMHC). E—Hemilaminectomy (HL) model. F—Six axes of freedom. G—Flexion and extension test. H—Right and left lateral bending test. I—Right and left rotation test.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.08.0244
Axis settings and loading conditions
A cross-shaped jig was positioned to adhere to the cranial side of the L1 vertebra, ensuring consistent loading conditions across all tests. The center of the jig was aligned as an extension of the vertebral body axis, determined by the intersection between the longest transverse diameter and the shortest longitudinal diameter on both the cranial surface of the L1 vertebral body and the caudal surface of the L2 vertebral body. The line connecting these 2 intersections defined the vertebral body axis. To analyze the 3-D biomechanics of the functional spinal unit, the center of the jig and the contact point on the vertebral body axis were designated as the origin. Six degrees of freedom were established, following the framework proposed by Panjabi et al.20 In the coordinate system defined for this study, the positive x-axis is ventral, the positive y-axis is right, and the positive z-axis is caudal; negative indicates the opposite directions for each axis. From the origin, a clockwise rotation is defined as positive (+), and a counterclockwise rotation is defined as negative (−). Consequently, +x represents flexion, −x represents extension, +y represents right lateral bending, −y represents left lateral bending, +z represents right axial rotation, and −z represents left axial rotation (Figure 1). To model flexion and extension, lateral bending, and rotation, the caudal aspect of the L2 vertebral body was immobilized in all degrees of freedom, and pure moments of 2.0 Nm were applied to the cranial aspect of the L1 vertebral body using a jig. This load value was based on the biomechanical load typically applied to the functional spinal unit of a dog weighing approximately 10 kg, as described by de Vicente et al.17
Validation
Numerical and visual analyses were used to assess study outcomes. Incrementally larger loads (0.2-Nm steps; maximum, 2.0 Nm) were applied in each direction, and the resulting angular displacement was calculated using the dot product of 2-position vectors at the top of the spinous process obtained before and after loading. Primary translation and revolution movements along 1 axis in vivo are commonly accompanied by translation and rotation movements along other axes, a phenomenon known as “coupled motion” that can be quantified as a numerical value.34 The elastic and neutral zone displacement points, regions with distinct spinal motion characteristics, were analyzed using load-displacement curves.20 Contour plots were used to visualize equivalent stresses in the L1 vertebra, L2 vertebra, and intervertebral disc at a load of 2.0 Nm. This approach facilitated the comparison of stress distributions across different decompression techniques and loading directions.
Results
Validity of the intact model
At a 2.0-Nm load, the displacement angles of the intact model measured 21.80° for extension and flexion, 32.24° for lateral bending, and 4.72° for rotation. These values are generally consistent with the results reported by Julie et al35 for ex vivo experiments (extension and flexion, 23.16 ± 4.87°; lateral bending, 33.47 ± 6.67°; and rotation 5.04 ± 1.21°; Supplementary Figure S2) in Beagle dogs under identical loading conditions. Therefore, the intact model accurately approximates the behavior of the functional spinal unit. Contour plots (Figure 2) illustrate the deformation displacement of the intact model.
Displacement contour plot of the functional spinal unit for each load test. The shift from blue to red indicates an increase in displacement. The smooth gradation of displacement in the loading direction confirms the validity of the intact model's behavior.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.08.0244
Displacement angles for each decompression technique (numerical analysis)
The variation in displacement angles obtained from the simulation across decompression techniques for each load applied is illustrated (Figure 3). Each technique, including the intact model, displayed sharp increases in displacement angles in response to loads up to 0.4 Nm in all directions, followed by more gradual increases in displacement angles in response to loads ranging from 0.4 to 2.0 Nm. The range of motion up to 0.4 Nm was classified as the neutral zone for this study, and the range of motion beyond 0.4 Nm was designated as the elastic zone. The displacement angles measured in response to 0.4- and 2.0-Nm loads in each direction and for each decompression technique were extracted, and the results are presented (Table 1).
Load-displacement angle curves of the functional spinal unit for each model. For flexion, extension, and rotation, displacement angles for all load ranges increased in the following order: intact < mHL < PLC < iMHC < HL. For bending, in the 0- to 0.4-Nm range, displacement angles increased in the following order: intact < mHL < HL < PLC < iMHC; in the 0.4- to 2.0-Nm range, displacement angles increased in the following order: intact < mHL < PLC < HL < iMHC.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.08.0244
Displacement angle (°) of each decompression technique at 0.4- and 2.0-Nm loads.
| Intact | mHL | PLC | iMHC | HL | |
|---|---|---|---|---|---|
| 0.4 Nm | |||||
| Flexion | 9.40 | 10.03 | 10.67 | 10.99 | 12.45 |
| Extension | 7.22 | 7.40 | 7.61 | 8.13 | 8.43 |
| Right bending | 14.04 | 15.26 | 17.01 | 17.95 | 16.19 |
| Left bending | 13.52 | 14.30 | 15.83 | 16.81 | 15.03 |
| Right rotation | 1.54 | 1.60 | 1.88 | 1.98 | 2.52 |
| Left rotation | 1.34 | 1.41 | 1.61 | 1.78 | 2.27 |
| 2.0 Nm | |||||
| Flexion | 12.91 | 13.24 | 13.80 | 14.06 | 15.52 |
| Extension | 8.89 | 9.18 | 9.28 | 9.59 | 9.84 |
| Right bending | 16.35 | 17.18 | 19.18 | 19.82 | 19.56 |
| Left bending | 15.89 | 16.75 | 18.68 | 19.55 | 19.21 |
| Right rotation | 2.47 | 2.60 | 2.92 | 3.08 | 3.64 |
| Left rotation | 2.25 | 2.38 | 2.74 | 2.95 | 3.39 |
HL = Hemilaminectomy. iMHC = Individualized mini-hemilaminectomy–corpectomy. mHL = Mini-hemilaminectomy. PLC = Partial lateral corpectomy.
Displacement angles consistently increased when loads were applied to the extension and flexion directions. The measured displacement angles were lowest for the intact model, and the angle sizes for the decompression techniques can be ordered as follows: mHL < PLC < iMHC < HL. Relative to the intact model, in response to a 2.0-Nm flexion load, the displacement angles were 2.6% larger for mHL, 6.9% larger for PLC, and 8.9% larger for iMHC. The displacement angle for HL, which showed the most pronounced increase in displacement angle of all techniques, was 20.2% larger than the intact angle.
The displacement angles recorded when loads were applied in the lateral bending directions did not differ significantly between right and left bending for any of the tested techniques, and displacement behaviors were fairly consistent across all loads. The measured displacement angles from 0 to 0.4 Nm can be ordered as follows: intact < mHL < HL < PLC < iMHC; however, from 0.4 to 2.0 Nm, the order of angle sizes was slightly different: intact < mHL < PLC < HL < iMHC. At the 0.4-Nm level, the percentage increases in displacement angles relative to the intact model in the right and left directions, respectively, were 8.7% and 5.8% for mHL, 21.2% and 17.1% for PLC, 27.2% and 24.3% for iMHC, and 15.3% and 11.2% for HL. At the 2.0-Nm level, the percentage increases in displacement angles relative to the intact model in the right and left directions, respectively, were 5.1% and 5.4% for mHL, 17.3% and 17.6% for PLC, 21.2% and 23.0% for iMHC, and 20.0% and 20.9% for HL. The differences in displacement angles between iMHC and HL were approximately 12% to 13% at 0.4 Nm and approximately 1% to 2% at 2.0 Nm.
When loads were applied in the rotational direction, no differences in displacement angles were observed between the left and right rotational directions for any decompression technique. Displacement angles consistently increased in response to increasing loads, and the angles could be ordered as follows: intact < mHL < PLC < iMHC < HL. At the 2.0-Nm level, the percent increases in displacement angles relative to the intact model in the right and left directions, respectively, were 5.3% and 5.8% for mHL, 18.2% and 21.8% for PLC, 24.7% and 31.1% for iMHC, and 47.3% and 50.7% for HL.
Equivalent stress for each decompression technique (visual analysis)
Equivalent stress contour plots (Figures 4 and 5) were generated for each decompression technique at 2.0 Nm on the L1 and L2 vertebrae and intervertebral discs using simulation results. A change from blue to red indicates an increase in equivalent stress. The contour plot for the intact model is predominantly blue, indicating greater stability than any tested decompression technique.
Equivalent stress contour diagrams of the L1-L2 vertebrae. Equivalent stresses in the L1 and L2 vertebrae for each decompression technique and loading test are presented. Blue indicates less stress, and red indicates increased stress.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.08.0244
Equivalent stress contour diagrams of the intervertebral disc. The equivalent stresses in the intervertebral disc for each decompression technique and loading test are presented. Blue indicates less stress, and red indicates increased stress.
Citation: American Journal of Veterinary Research 85, 12; 10.2460/ajvr.24.08.0244
Under flexion load, increased stress is observed for the intact model at the articular process joint, the L2 vertebral root, and the caudal side of the L2 vertebral body. These regions displayed increases in equivalent stress for all models in the following order: intact < mHL < PLC < iMHC < HL. A trend toward escalating equivalent stress was observed for the L2 vertebra in the cephalad direction. In the HL model, a minimal increase in equivalent stress was observed at the L2 left vertebral root, but a notable increase in equivalent stress was observed in and around the surface of the preserved right articular. The increase in equivalent stress at the L2 vertebral body was more pronounced for the HL model than for the other decompression models. An increase in equivalent stress was observed in the ventral fiber ring of the intervertebral disc in the following order: mHL = PLC < iMHC < HL. Additionally, an increase in equivalent stress was observed on both the dorsal and ventral fiber rings in the HL model but not in the other decompression models.
The equivalent stress contour plots for lateral bending showed little variation between the left and right loading directions. Because equivalent stresses are scalar values and do not provide information on the distribution of tensile and compressive stresses, contour plots were used to identify the maximum and minimum principal stresses (Supplementary Figure S3). Increased stress was observed for the intact model at the articular process joint, the base of the L1 posterior articular process, and the caudal aspect of the L2 vertebral body. In addition to increased stress in these regions, the mHL, PLC, and iMHC models also resulted in increased stress at the caudal aspect of the L1 vertebral arch and the root of the L2 vertebral arch in the following order: mHL < HL< PLC < iMHC. Conversely, the HL model exhibited a minimal increase in stress at the L1 vertebral arch and the left-sided vertebral root of L2. However, an increase in stress at the right-sided vertebral root was observed, albeit to a lesser extent than in the iMHC model. The HL model also resulted in increased stress at the L1 vertebra, which was not observed for the other models, accompanied by a more extensive increase in stress at the L2 vertebra. Increased stress was observed on both the right and left annulus fibrosus of the intervertebral disc in the following order: intact < mHL < PLC < iMHC < HL.
The equivalent stress contour plots for rotation showed little variation between the left and right loading directions. Therefore, the maximum and minimum principal stresses were also identified using contour plots for rotation (Supplementary Figure S3). Increased stress was observed at the articular process joint, the base of the L1 posterior articular process, the cephalad endplate of the L1 vertebral body, and the caudal portion of the L2 vertebral body in the following order: intact < mHL < PLC < iMHC < HL. The HL model showed noticeably increased stress on the articular surface, the base of the preserved right articular process, and the caudal side of the L2 vertebral body. Increased stress was noted in the ventral fiber ring of the intervertebral disc in the following order: intact < mHL < PLC < iMHC < HL. In the HL model, a marked increase in stress was observed on the left fibrous ring where the articular process was resected.
Discussion
This study employed finite element analysis to evaluate the biomechanical effects of various spinal decompression techniques on the treated intervertebral segments. The findings indicate that HL reduced stability as hypothesized, whereas mHL had a smaller impact on stability. There was little difference in stability between iMHC and PLC, showing almost similar behavior.
The load-displacement curves revealed that the displacement angle increased rapidly from 0 Nm to approximately 0.4 Nm; from 0.4 to 2.0 Nm, the displacement angle increased more slowly. This pattern suggests that the range of motion observed in response to loads as high as 0.4 Nm corresponds to the neutral zone, whereas the range of motion observed in response to loads larger than 0.4 Nm corresponds to the elastic zone. Importantly, the model developed in this study successfully reproduced the behavior of the canine functional spinal unit in these 2 zones. Although the mechanical loads experienced by the canine spine during normal walking and exercise have not been definitively established, the findings of the present study suggest that loads at or below 0.4 Nm or less can be used to evaluate the mechanical properties of the functional spinal unit in Beagle dogs during typical daily activities, whereas loads exceeding 0.4 Nm can be used to assess responses to excessive external forces.
The load-displacement curves and equivalent stress contour plots indicate that the functional spinal unit is more flexible in response to flexion loads than in response to extension loads. Minimal differences were observed between left and right displacement in response to lateral bending or rotation loads, even though all tested decompression techniques were performed on the left side. Increases in maximum principal stress indicate tensile stress, whereas increases in minimum principal stress indicate compressive stress (Supplementary Figure S3). Although the contour plot indicated similar levels of tensile and compressive stresses, the relationships between tensile and compressive stresses for the left and right loads were almost opposite. For example, during right lateral bending, tensile stress is generated in the left half of the functional spinal unit, while compressive stress is generated in the right half. Although the equivalent stress contour plots for lateral bending and rotation show little variation in the left-to-right direction, tensile and compressive stresses are actually occurring separately on the left and right sides but at similar levels, resulting in no observed differences in the ranges of motion for the functional spinal unit as a whole between the left and right sides.
The present study demonstrates that the 4 tested spinal decompression techniques, mHL, PLC, iMHC, and HL, each have distinct impacts on spinal stability in dogs. Among the tested techniques, mHL was associated with greater functional spinal unit stability, exhibiting consistently small displacement angles across all loads. The mHL procedure preserves both the articular process joint and the intervertebral disc, 2 important structures that contribute to functional spinal unit stability, and requires less bone removal than the other tested techniques.5,36 The displacement angles measured for all tested loads were larger for HL than for mHL, which is unsurprising because HL extends the dorsal cutting beyond that used in mHL and includes articular process joint resection. However, previous studies9,37–39 have indicated that HL does not induce clinical spinal instability in any loading direction. Therefore, the increased displacement angles observed for HL in the present study may not result in a level of functional spinal unit instability that would be considered problematic in clinical practice. Therefore, displacement angles smaller than those measured for HL should indicate that the mechanical differences between HL and other decompression techniques minimize functional spinal unit instability.
The displacement angles in response to all extension, flexion, and rotation loads were consistently smaller for PLC and iMHC than for HL. The displacement angles in response to lateral bending were larger for PLC than for HL in the neutral zone, but the displacement angles were smaller for PLC than for HL in the elastic zone. Larger displacement angles were observed for iMHC than for HL in both the neutral and elastic zones, but the angles for iMHC approached equivalence with those for HL at increasing loads.
The increased displacement angles observed for HL in response to extension, flexion, and rotation loads and those observed for PLC and iMHC in response to lateral bending loads are likely due to the functional spinal unit anatomy and the characteristics of each technique. The functional spinal unit is broadly classified into dorsal elements, such as the articular process joints, SSL, and ISL, and ventral elements, including vertebral bodies, intervertebral discs, and longitudinal ligaments.40 Takeuchi et al41 reported that the dorsal elements primarily contribute to stability in response to extension, flexion, and rotation loads, whereas the ventral elements contribute to axial spinal compression and provide stability during lateral bending.42 In the present study, the equivalent stress contour plots indicated that loads applied in the extension, flexion, and rotation directions resulted in increased stress at the left and right articular process joints, their bases, and the vertebral arch region but were confined to the L1 cephalad endplate and the caudal third of the vertebral body. Conversely, loads applied in the lateral bending directions increased stress at the left and right articular process joints, over the L1 caudal vertebral body endplate, and throughout the entire L2 vertebral body.
These stress distributions corroborate the findings of a mechanical study conducted by Takeuchi et al.41 The HL procedure involves the resection of one side of the articular process joint, which diminishes dorsal element stabilization, whereas the PLC and iMHC procedures involve only the partial resection of the disc and vertebral body, diminishing ventral element stabilization. These effects are clearly reflected in the contour plots. The removal of one articular process during HL results in increased stress on and around the articular process joints in response to loads applied to the flexion, extension, and rotation directions. Conversely, PLC and iMHC distribute stress more evenly to the bilateral articular processus joints, resulting in relatively minimal increases in stress. The partial removal of the intervertebral disc and vertebral body during PLC and iMHC prevents stress distribution to the vertebral bodies in response to lateral bending loads, leading to notable stress concentration in the articular process and its surroundings. Conversely, stress was more evenly distributed across the L1 and L2 vertebral bodies during HL, with relatively minor stress concentration in the preserved articular process joint. In other words, PLC and iMHC are techniques in which the annulus fibrosus of the intervertebral disc, which provides stabilization against intervertebral rotational and compressive loading, is partially resected, resulting in the loss of a small amount of its function.
Based on the results of the current study, PLC and iMHC may reduce functional spinal unit stability to a greater degree than HL, particularly in the lateral bending directions that occur in daily life. However, PLC and iMHC are less likely than HL to reduce the anatomic stability of the functional spinal unit under conditions of excessive motion exposure to external forces. In an ex vivo biomechanics study of the L1-L2 functional spinal unit from Beagle dogs, de Vicente et al17 found that PLC did not increase spinal instability in the extension and flexion directions, but some instability was induced in the lateral bending direction (17), which is consistent with the findings of the present study. However, clinical reports16 indicate no observed collapse of the intervertebral space following PLC or iMHC.
Discrepancies between clinical reports and the findings of mechanical and finite element analysis studies may be due to the inability of laboratory biomaterials to replicate functional spinal unit stabilization attributable to the healing process, which is unique to clinical cases. In canines, fibrocartilage fills in disc defects over time following disc excision.40,42 Functional spinal unit instability caused by partial disc and vertebral body resection during PLC and iMHC may decrease over time as fibrous connective tissue fills the resultant space. Consequently, caution is warranted regarding the potential for PLC or iMHC to cause functional spinal unit instability during the early postoperative period, and excessive resection could also induce instability.
de Vicente et al17 reported that iMHC was not associated with a marked increase in instability compared with PLC alone, despite iMHC extending resection to the base of the articular process. In the present study, the differences in displacement angles at 2.0 Nm between PLC and iMHC and between PLC and HL were similar, suggesting that iMHC does not negatively impact functional spinal unit stability compared to PLC. Although iMHC extends resection dorsally relative to PLC, the critical structures of the dorsal elements remain unchanged, and functional spinal unit stability is maintained. Because iMHC, which represents a combination of PLC and mHL, offers better visibility of the ventral spinal canal than PLC alone, surgeons may be able to remove disc material ventral to the spinal cord more safely and accurately without significantly affecting functional spinal unit stability during iMHC than during PLC alone, reducing the risks of postoperative complications related to postoperative spinal instability. However, care should be taken not to over-resect the vertebral body, as PLC and IMHC may cause more instability than HL, if only with respect to loading in the lateral flexion direction.
This study has several limitations. First, this study used only a single dog model. The behavior of the model itself is similar to that reported in the past, so its validity is somewhat assured, but it does not take into account the variation caused by different solids. Second, the simple, single-direction loading test does not account for the complex movements that occur during exercise.43,44 Studies45,46 in humans and sheep show that intervertebral discs are constantly subjected to axial compressive stresses in daily life, and the same stresses are likely experienced by dogs. Therefore, to better evaluate the behavior of the functional spinal unit following each decompression technique, a loading test that applies a constant axial compressive load to the intervertebral disc would be ideal. However, the internal disc pressures experienced by dogs in daily life are not known. Finite element analysis cannot fully replicate the complex environment and dynamic physiological changes that occur in vivo. Therefore, multiple perspectives are necessary to understand the mechanical properties and therapeutic effects of each decompression technique in the canine lumbar spine, which may be achieved by combining the results of biomaterials studies with those from studies following the postoperative courses of clinical cases. Understanding the unique advantages and limitations of each method and how they complement one another is essential for developing more effective treatment strategies for the lumbar spine. Third, all of our validation is theoretical modeling that does not account for the degree of disc dehydration and degeneration in patients with actual disc herniation or the impact of surgical approaches on soft tissue. To produce an analysis that recapitulates a clinical situation, it may be necessary to infer the degree of degeneration of the intervertebral disc from MRI signal intensity and CT value and insert materials with matching properties. Also, human spine studies use finite element models that include muscle tissue, such as the erector spinae and multifidus muscle fibers,47 and, hopefully, this approach will also be extended to veterinary studies for appropriate muscle strength representation.
In conclusion, consistent with previous reports, this study demonstrated that mHL minimizes functional spinal unit instability to the greatest extent among the 4 tested decompression techniques. Since PLC and iMHC can cause instability depending on the direction of loading, careful attention may need to be paid to postoperative instability associated with excessive vertebral body resection when these techniques are applied in clinical cases.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
None reported.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
The authors have nothing to disclose.
ORCID
M. Shimada https://orcid.org/0000-0002-9233-2496
References
- 1.↑
Lubbe A, Kirberger R, Verstraete F. Pediculectomy for thoracolumbar spinal decompression in the dachshund. J Am Anim Hosp Assoc. 1994;30:233–238.
- 2.
Flo G, Brinker W. Lateral fenestration of thoracolumbar disks. J Am Anim Hosp Assoc. 1975;11:619–626.
- 3.
Redding R. Laminectomy in the dog. Am J Vet Res. 1951;12(43):123–128.
- 4.
Schulman A, Lippincott C. Dorsolateral hemilaminectomy in the treatment of thoracolumbar intervertebral disk disease in dogs. Compend Contin Educ Pract Vet. 1987;9:305–310.
- 5.↑
Jeffery N. Treatment of acute and chronic thoracolumbar disk disease by ‘mini’ hemilaminectomy. J Small Anim Pract. 1988;29:611–616. doi:10.1111/j.1748-5827.1988.tb02181.x
- 6.↑
Moissonnier P, Meheust P, Carozzo C. Thoracolumbar lateral corpectomy for treatment of chronic disk herniation: technique description and use in 15 dogs. Vet Surg. 2005;33(6):620–628.
- 7.↑
Toombs J, Waters D. Intervertebral disc disease. In: Slatter D, ed. Textbook of Small Animal Surgery. 3rd ed. Saunders; 2002, p. 1193–1209.
- 8.
Sharp NJH, Wheeler SJ. Thoracolumbar disc disease. In: Small Animal Spinal Disorders: Diagnosis and Surgery. 2nd ed. Elsevier; 2005, p. 121–159.
- 9.↑
Corse M, Renberg W, Friis E. In vitro evaluation of biomechanical effects of multiple hemilaminectomies on the canine lumbar vertebral column. Am J Vet Res. 2003;64(9):1139–1145. doi:10.2460/ajvr.2003.64.1139
- 10.↑
Schulman A, Lippincott C. Dorsolateral hemilaminectomy in the treatment of thoracolumbar intervertebral disk disease in dogs. Comp Cont Educ Pract Vet. 1987;9:305–310.
- 11.↑
Forterre F, Spreng D, Rytz U, Jaggy A, Schawalder P. Thoracolumbar dorsolateral laminectomy with osteotomy of the spinous process in fourteen dogs. Vet Surg. 2007;36(5):458–463. doi:10.1111/j.1532-950X.2007.00292.x
- 12.↑
Seim HB 3rd. Surgery of the thoracolumbar spine. In: Fossum TW, ed. Small Animal Surgery. Mosby; 1997:1101–1130.
- 13.↑
Cudia SP, Duval JM. Thoracolumbar intervertebral disk disease in large, nonchondrodystrophic dogs: a retrospective study. J Am Anim Hosp Assoc. 1997;33(5):456–460. doi:10.5326/15473317-33-5-456
- 14.↑
Flegel T, Boettcher I, Ludewig E, Kiefer I, Oechtering G, Böttcher P. Partial lat- eral corpectomy of the thoracolumbar spine in 51 dogs: assessment of the slot morphometry and spinal cord decompression. Vet Surg. 2011;40(1):14–21. doi:10.1111/j.1532-950X.2010.00747.x
- 15.↑
Reves NV, Bürki A, Ferguson S, Geissbühler U, Stahl C, Forterre F. Influence of partial lateral corpectomy with and without hemilaminectomy on canine thoracolumbar stability: a biomechanical study. Vet Surg. 2011;41(2):228–234. doi:10.1111/j.1532-950X.2011.00912.x
- 16.↑
Susanne C, Sven R, Nicola S. Individualized mini-hemilaminectomy-corpectomy (iMHC) for treatment of thoracolumbar intervertebral disc herniation in large breed dogs. Vet Surg. 2017;46(3):422–432. doi:10.1111/vsu.12616
- 17.↑
de Vicente F, Bernard F, FitzPatrick D, Moissonnier P. In vitro radiographic characteristics and biomechanical properties of the canine lumbar vertebral motion unit after lateral corpectomy, mini-hemilaminectomy and hemilaminectomy. Vet Comp Orthop Traumatol. 2013;26(1):19–26. doi:10.3415/VCOT-12-02-0016
- 18.↑
Bonelli MA, Shah A, Goel V, Costa FS, da Costa RC. Development of a finite element model of the ligamentous cervical vertebral column of a Great Dane. Res Vet Sci. 2018;118:97–100. doi:10.1016/j.rvsc.2018.01.011
- 19.↑
Kikuchi Y, Shimada M, Yamaguchi S, Hara Y. Finite element analysis predictions in the canine lumbar spine are useful and correlate with ex vivo biomechanical studies. Am J Vet Res. 2023; 84(11):ajvr.23.06.0125. doi:10.2460/ajvr.23.06.0125
- 20.↑
Panjabi MM, White AA, Brand RA. A note on defining body parts configurations. J Biomech. 1974;7(4):385–387. doi:10.1016/0021-9290(74)90034-7
- 21.↑
Widmer J, Fornaciari P, Senteler M, Roth T, Snedeker JG, Farshad M. Kinematics of the spine under healthy and degenerative conditions: a systematic review. Ann Biomed Eng. 2019;47(7):1491–1522. doi:10.1007/s10439-019-02252-x
- 22.↑
Suk SI, Kim WJ. Biomechanics of posterior instrumentation for spinal arthrodesis. In: Poitout D, ed. Biomechanics and Biomaterials in Orthopedics. Vol 5. Springer; 2016:437–467.
- 23.↑
Schmidt H, Heuer F, Simon U, et al. Application of a new calibration method for a three-dimensional finite element model of a human lumbar annulus fibrosus. Clin Biomech. 2006;21(4):337–344. doi:10.1016/j.clinbiomech.2005.12.001
- 24.↑
Fagan MJ, Julian S, Siddall DJ, Mohsen AM. Patient-specific spine models. Part 1: finite element analysis of the lumbar intervertebral disc—a material sensitivity study. Proc Inst Mech Eng H. 2002;216(5):299–314. doi:10.1243/09544110260216577
- 25.↑
Bellini CM, Galbusera F, Raimondi MT, Mineo GV, Brayda-Bruno M. Biomechanics of the lumbar spine after dynamic stabilization. J Spinal Disord Tech. 2007;20(6):423–429. doi:10.1097/BSD.0b013e318031af6f
- 26.↑
Fantigrossi A, Galbusera F, Raimondi MT, Sassi M, Fornari M. Biomechanical analysis of cages for posterior lumbar interbody fusion. Med Eng Phys. 2007;29(1):101–109. doi:10.1016/j.medengphy.2006.02.007
- 28.↑
Lim TH, Goel VK, Weinstein JN, Wong K. Stress analysis of a canine spinal motion segment using the finite element technique. J Biomech. 1994;27(10):1259–1269. doi:10.1016/0021-9290(94)90279-8
- 29.
Villarraga ML, Anderson RC, Hart RT, Dinh DH. Contact analysis of a posterior cervical spine plate using a three-dimensional canine finite element model. J Biomech Eng. 1999;121(2):206–214. doi:10.1115/1.2835105
- 30.
Urban JPG, Roberts S. Degeneration of the intervertebral disc. Arthritis Res Ther. 2003;5(3):120–130. doi:10.1186/ar629, 10.1186/ar921
- 31.
Shirazi-Adl A. Strain in fibers of a lumbar. Analysis of the role of lifting in producing disc prolapse. Spine. 1989;14(1):96–103.
- 32.
Chazal J, Tanguy A, Bourges M, et al. Biomechanical properties of spinal ligaments and a histological study of the supraspinous ligament in traction. J Biomec. 1985;18(3):167–176. doi:10.1016/0021-9290(85)90202-7
- 33.↑
Yoganandan N, Kumaresan S, Pintar FA. Geometric and mechanical properties of human cervical spine ligaments. J Biomec. 2001;34(6):771–776.
- 34.↑
Benninger MI, Seiler GS, Robinson LE, et al. Three-dimensional motion pattern of the caudal lumbar and lumbosacral portions of the vertebral column of dogs. Am J Vet Res. 2004;65(5):544–551. doi:10.2460/ajvr.2004.65.544
- 35.↑
Julie S, Benjamin V, Franck F, et al. Comparison of the biomechanical performance of a customized unilateral locking compression plate with and without an intervertebral spacer applied to the first and second lumbar vertebrae after intervertebral diskectomy in canine cadaveric specimens. Am J Vet Res. 2020;81:915–921.
- 36.↑
Arthurs G. Spinal instability resulting from bilateral mini-hemilaminectomy and pediculectomy. Vet Comp Orthop Traumatol. 2009;22(5):422–426. doi:10.3415/VCOT-08-09-0092
- 37.↑
Shires P, Waldron D, Hedlund C, et al. A biomechanical study of rotational instability in unaltered and surgically altered canine thoracolumbar vertebral motion units. Prog Vet Neurol. 1991;2:6–14.
- 38.
Smith G, Walter M. Spinal decompressive procedures and dorsal compartment injuries: comparative biomechanical study in canine cadavers. Am J Vet Res. 1988;49(2):266–273.
- 39.↑
Schulz K, Waldron D, Grant J, Shell L, Smith G, Shires PK. Biomechanics of the thoracolumbar vertebral column of dogs during lateral bending. Am J Vet Res. 1996;57(8):1228–1232. doi:10.2460/ajvr.1996.57.08.1228
- 40.↑
Smith J, Walmsley R. Experimental incision of the intervertebral disc. J Bone Joint Surg Br. 1951;33(4):612–625. doi:10.2106/00004623-195133030-00007
- 41.↑
Takeuchi T, Abumi K, Shono Y, Oda I, Kaneda K. Biomechanical role of the intervertebral disc and costovertebral joint in stability of the thoracic spine. Spine. 1999;15(14):1414–1420. doi:10.1097/00007632-199907150-00005
- 42.↑
Wagner S, Ferguson H, Leipold H, Guffy MM, Butler HC. Radiographic and histologic changes after thoracolum-bar disc curettage. Vet Surg. 1987;6(1):65–69. doi:10.1111/j.1532-950X.1987.tb00915.x
- 43.↑
Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement. J Spinal Disord. 1992;5(4):383–389.
- 44.↑
Panjabi MM. The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis. J Spinal Disord. 1992;5(4):390–397.
- 45.↑
Fernando HN, Czamanski J, Yuan TY, Gu W, Salahadin A, Huang CY. Mechanical loading affects the energy metabolism of intervertebral disc cells. J Orthop Res. 2011;29(11):1634–1641. doi:10.1002/jor.21430
- 46.↑
Reitmaier S, Schmidt H, Ihler R, Kocak T, Graf N, Ignatius A. Preliminary investigations on intradiscal pressures during daily activities: an in vivo study using the merino sheep. PLoS One. 2013;8(7):24.
- 47.↑
Kumaran Y, Shah A, Katragadda A, et al. Iatrogenic muscle damage in transforaminal lumbar interbody fusion and adjacent segment degeneration: a comparative finite element analysis of open and minimally invasive surgeries. Eur Spine J. 2021;30(9):2622–2630. doi:10.1007/s00586-021-06909-x
