INTRODUCTION
Lumbar spondylolisthesis refers to the abnormal connection between adjacent vertebrae due to congenital dysplasia, trauma, degeneration, and other factors of the lumbar spine, which leads to partial or total slippage of the upper vertebrae relative to the lower vertebrae. It is one of the common diseases leading to lumbar and leg pain in clinical practice. Wiltse et al. [1] classified lumbar spondylolisthesis into dysplastic, isthmic, degenerative, traumatic, and pathologic spondylolisthesis based on the cause of occurrence, of which isthmic and degenerative lumbar spondylolisthesis are the most common. Degenerative lumbar spondylolisthesis was first identified and described by Junghannsyu in 1930, who named it pseudospondylolisthesis, mainly due to the occurrence of slippage of the upper vertebrae relative to the lower vertebrae because of degeneration without pars interarticularis defect. As discovered and reported by Killam, isthmic spondylolisthesis (IS) generally refers to vertebral slippage of the diseased vertebrae due to loss of traction from the posterior facet joints (FJs) due to pars defects. However, the pathogenesis of isthmic and degenerative lumbar spondylolisthesis remains unclear [2,3].
Previous studies have reported that the spinopelvic parameters of the 2 types of spondylolisthesis are significantly different compared with those of normal populations; that is, patients with lumbar spondylolisthesis have greater pelvic incidence (PI) and sacral slope (SS) [4-7]. Recently, relevant biomechanical studies have shown that under different spinopelvic types, there are significant differences in the distribution of stress in the lower lumbar spine [8], and the differences in stress will lead to differences in the morphology of the local bone structure of the lumbar spine [9-11]. To date, few studies have analyzed isthmic and degenerative lumbar spondylolisthesis from the perspective of bony morphology combined with biomechanics under different spinopelvic types. Therefore, we hypothesize that isthmic and degenerative lumbar spondylolisthesis, due to their unique spinopelvic morphology, result in an abnormal distribution of local stress in the lower lumbar spine, and we want to determine whether this stress abnormality is related to their pedicle-facet joint (PFJ) morphological characteristics.
The objectives of this study were to: (1) measure the spinopelvic parameters and morphological parameters of PFJs using radiographs and 3-dimensional (3D) reconstructed computed tomography (CT) images of patients with IS, those with degenerative spondylolisthesis (DS), and those without lumbar spondylolysis; (2) analyze the differences in the spinopelvic parameters and morphological parameters of PFJs between patients with spondylolysis and those without spondylolysis; (3) analyze the correlation between the spinopelvic parameters and morphological characteristics of the PFJs in combination with biomechanical principles; and (4) investigate the mechanism of the onset and development of lumbar spondylolysis and evaluate the clinical significance.
MATERIALS AND METHODS
1. Study Design and Participants
We conducted a retrospective cross-sectional analysis of the medical records and imaging data of all patients (Asian, from Southwest China) who visited our hospital from August 1, 2018, to May 1, 2023. This study was approved by the Institutional Review Board (IRB) of the Affiliated Hospital of Southwest Medical University (IRB No. KY2202058). Informed consent was not required, as the study was retrospective in nature. Based on the inclusion and exclusion criteria (Fig. 1), 194 patients with L5 spondylolysis or L5–S1 low-grade IS and 172 patients with L4–5 DS were included in the IS group and the DS group, respectively. A random numerical table method was used to select 366 patients from 682 patients with normal L4–5 but lumbar disc herniation in other segments as the control group (i.e., nonlumbar spondylolysis [NL] group). All of the aforementioned patients were diagnosed by spine surgeons (CT, YL) on the basis of lumbosacral frontal and lateral radiographs and 3D reconstructed lumbar CT images, and when there was disagreement between the diagnoses of these 2 surgeons, the final decision was made by a third, more senior spine surgeon (DZ).
2. Data Source
Lumbosacral radiographs of all patients were obtained in the standing neutral position, and the patients were scanned in the supine position using a Siemens 128-slice spiral CT scanner with scan parameters of a 1.0-mm thickness and 1.0 mm-spacing. The original image data of all patients were stored in the postprocessing workstation for reconstruction, and the relevant parameters were measured using the measurement tool in the Siemens Syngo software workstation, with accuracy values of 0.1 mm and 0.1° for distance and angle measurements, respectively.
3. Measurements
Vertebral body slip percentage (SP) and spinopelvic parameters, i.e., PI, pelvic tilt (PT), SS, and lumbar lordosis (LL), were measured on radiographs according to the method proposed by Legaye et al. [12] and Bourassa-Moreau et al. [13] (Fig. 2). Lumbar spondylolisthesis was then graded in accordance with the Meyerding classification (grades I–IV).
The morphological parameters of the L4–5 and L5–S1 FJs, including the facet joint angle (FJA), pedicle-facet joint angle (PFA), facet joint osteoarthritis (FJOA), and facet joint tropism (FT), were measured on 3D reconstructed lumbar CT images according to the method proposed by Grobler et al. [14] and Boden et al. [15] (Fig. 2). In addition, FJOA was evaluated using the FJOA grading method suggested by Kalichman et al. [16], which classifies FJOA into 4 grades based on subchondral cysts, subarticular erosions, joint space, the vacuum phenomenon, sclerosis, and articular process hypertrophy (grade I, normal; grade II, mild; grade III, moderate; and grade IV, severe).
Two spine surgeons (CT, YL) performed each measurement independently, and the final results were averaged for statistical analysis. In addition, 2 spine surgeons separately evaluated the results for FJOA grading, and in cases of disagreement, the results were evaluated by a third spine surgeon (DZ) who made a decision based on the combined evaluations. General information about all patients, including age, sex, weight, height, and body mass index (BMI), was obtained from their medical records.
4. Statistical Analysis
Statistical analysis was performed using IBM SPSS Statistics ver. 25.0 (IBM Co., Armonk, NY, USA). Independent-samples t-test was used to analyze differences in the spinopelvic parameters and PFJ morphological parameters by sex (men vs. women) and to analyze the differences in SP between the IS and DS groups. Paired-samples t-test was used to analyze differences in the PFJ morphological parameters between different segments. Analysis of variance was used to analyze the differences in general information (age, height, weight, and BMI) among the 3 groups. With age as a covariate, analysis of covariance was used to analyze the differences in the spinopelvic and PFJ morphological parameters among the 3 groups, and all between-group comparisons were performed using Bonferroni post hoc tests. Differences in Meyerding grade between the IS and DS groups and differences in FJOA severity between groups and between segments were analyzed using nonparametric tests, and between-group comparisons were performed. Categorical variables were compared using the chi-square test. Correlation analysis was performed using the Pearson correlation coefficient. At p< 0.05, differences were considered statistically significant.
RESULTS
General information and clinical features of the patients in the IS, DS, and NL groups are compared in Table 1. The differences in age and sex between the 3 groups were statistically significant (p< 0.001). The SP values in the IS group (27.3%± 10.8%) were significantly higher than those in the DS group (19.8%± 5.0%) (p < 0.001). Based on the Meyerding classification of lumbar spondylolisthesis, there were 57 cases (47.1%) of grade I and 64 cases (52.9%) of grade II in the IS group, while there were 136 cases (79.1%) of grade I and 36 cases (20.9%) of grade II in the DS group, and the difference was statistically significant (p<0.001). There were no significant differences in height, weight, and BMI among the IS, DS, and NL groups (p> 0.05).
1. Comparison of the Spinopelvic Parameters and PFJ Morphological Parameters of L4–5 and L5–S1 Among the IS, DS, and NL Groups
The results of the comparisons between the 3 groups of patients were consistent irrespective of whether the confounding factor of age was considered or not (Table 2). The differences in the spinopelvic parameters among the 3 groups were statistically significant. PI (56.7± 11.2, 58.3± 9.9), SS (38.6± 9.0, 33.2± 10.5), and LL (54.4± 12.0, 47.8± 14.6) in the IS and DS groups were significantly higher than those in the NL group (48.8± 9.2, 30.2± 8.3, 30.2± 8.3) (p< 0.001). While SS and LL were significantly higher in the IS group than in the DS group (p< 0.001), there was no significant difference in PI between the IS and DS groups (p> 0.999). PT values in both the IS and NL groups (18.3± 8.5, 18.6± 7.6) were significantly lower than those in the DS group (25.1± 9.2) (p< 0.001), but there was no significant difference in PT between the IS and NL groups (p= 0.573).
As for the morphological parameters of the L4–5 FJs, FJA in the IS group (47.4± 4.6) was significantly larger than that in the DS group (30.8± 5.9) and the NL group (46.2± 4.1) (p< 0.001), while FJA in the DS group was significantly smaller than that in the NL group (p< 0.001); PFA in the IS and DS groups (106.1± 3.4, 114.6± 4.6) was significantly larger than that in the NL group (104.2± 3.3) (p< 0.001), while PFA in the IS group was significantly smaller than that in the DS group (p< 0.001). The degree of FJOA in the DS group was significantly more severe than the degree of FJOA in the IS and NL groups (p< 0.001), while there was no significant difference in the degree of FJOA between the IS and NL groups (p> 0.999).
As for the morphological parameters of the L5–S1 FJs, FJA in the IS and NL groups (51.3 ± 7.5, 51.2 ± 3.2) was significantly larger than that in the DS group (49.5± 4.2) (p= 0.006), while there was no significant difference in FJA between the IS and NL groups (p> 0.999); PFA in the DS group (106.0± 4.5) was significantly larger than that in the NL group (102.7± 3.5) (p<0.001). The degree of FJOA in the IS and DS groups was significantly more severe than that in the NL group (p< 0.001), while there was no significant difference in the degree of FJOA between the IS and DS groups (p> 0.999).
In addition, to exclude the interference of gender, we stratified the 3 groups of patients according to gender before comparing the imaging parameters (Table 3), and the comparative results were largely consistent with the aforementioned results.
2. Correlation Between the Spinopelvic Parameters and PFJ Morphological Parameters Within the IS, DS, and NL Groups
Pearson correlation analysis showed correlations between the spinopelvic parameters and lumbar PFJ morphological parameters within the IS, DS, and NL groups (Table 4). SS negatively correlated with L4 FJA in the IS group (r = -0.446, p < 0.001), DS group (r= -0.642, p< 0.001), and NL group (r= -0.563, p< 0.001), but it positively correlated with L4 PFA in the IS group (r = 0.571, p < 0.001), DS group (r = 0.677, p < 0.001), and NL group (r = 0.654, p < 0.001). LL negatively correlated with L4 FJA in the IS group (r= -0.343, p< 0.001), DS group (r= -0.574, p< 0.001), and NL group (r= -0.443, p< 0.001), and positively correlated with L4 PFA in the IS group (r= 0.437, p< 0.001), DS group (r= 0.625, p< 0.001), and NL group (r= 0.590, p< 0.001).
3. Comparison of the PFJ Morphological Parameters Between the L4–5 and L5–S1 Segments Within the IS, DS, and NL Groups
Within the IS, DS, and NL groups, L4 FJA was smaller than L5 FJA (p< 0.001), while L4 PFA was larger than L5 PFA (p<0.001). L5 FJOA was more severe than L4 FJOA in the IS group (p< 0.001); L4 FJOA was more severe than L5 FJOA in the DS group (p<0.001); and there was no significant difference between L4 FJOA and L5 FJOA in the NL group (p= 0.968) (Table 5).
DISCUSSION
The lumbar PFJs refer to a bony region consisting of the pedicle, the superior and inferior articular processes, and the isthmus between them. They play an important role in maintaining spinal stability and transmitting stresses [17,18]. First, under the influence of LL, the lower lumbar spine is subjected to a forward shear force that can be transmitted through the intact pedicles and the pars interarticularis to the FJs [18]. Second, through biomechanical studies, Yang and King [17] found that the distribution of spinal axial loads shared by the vertebral body and FJs was significantly different due to the differences in lumbar spine morphology and that these axial loads could be transmitted between the superior and inferior articular processes through the lumbar pars interarticularis (LPI) [18]. In a histological study, Sagi et al. [19] found the pars interarticularis of the lower lumbar spine to be a stress-weak area. Thus, both the structural features and biomechanical properties of the lower lumbar PFJs contribute to their susceptibility to morphological changes. Recently, multiple biomechanical studies have demonstrated that different spinopelvic types result in significantly different stress distribution patterns in the lumbar spine [8-11]. A large number of prior studies have found differences in spinopelvic parameters in patients with isthmic and degenerative lumbar spondylolisthesis compared with the normal population, but it has never been possible to determine whether this is related to the onset and progression of lumbar spondylolisthesis [20,21]. Considering that few studies have analyzed the PFJ morphological characteristics of patients with lumbar spondylolisthesis from a biomechanical view under different spinopelvic types, our study aimed to explore this issue based on spinopelvic parameters.
In this study, we first compared the spinopelvic parameters among the IS, DS, and NL groups. The results indicated that there were significant differences in the spinopelvic parameters among the 3 groups, and SS and LL were significantly higher in the IS group than in the DS and NL groups, and they were also significantly higher in the DS group than in the NL group. Initially, the lower lumbar spine was mechanically analyzed based on the lumbar static force equation [9,10,22]. The contact force (CF) acting on the lumbar spine is the sum of gravity and posterior spinal muscle forces. The anterior shear force (F1) is equal to CF × sinα, and the vertical compression force (F2) is equal to CF× cosα, where α is the angle between F2 and CF, and α positively correlated with SS. In this study, the patients in the IS group with higher SS were subjected to a greater forward shear force in the lower lumbar spine, particularly at L5, compared with the DS and NL groups, which could lead to the onset or even progression of lumbar spondylolisthesis (Fig. 3). At the same time, when the forward shear force is transmitted to the posterior FJs, the superior articular process of the inferior vertebra prevents the slippage by exerting a reverse resistance to the inferior articular process of the slipping vertebra, and the entire morphology of the PFJ undergoes stress-related remodeling changes or even stress fracture of the LPI under long-term stress. In addition, the stress distribution of the lumbar spine is also affected by lumbar curvature; as lumbar curvature increases, the lumbar force line gradually moves backward, and the load exerted on the posterior FJs and LPI gradually increases [9,18,23]. According to the biomechanical studies reported by Schendel et al. [23] and Dunlop et al. [24], the FJs are subjected to greater axial compressive loads in the hyperextension position of the lumbar spine, and this compressive load can be transmitted primarily through the articular surfaces, tip contacts, and LPI. Moreover, Terai et al. [25] found that when the lumbar spine was in the hyperextension position, the stress exerted on the LPI was the greatest and more likely to cause the “nutcracker” mechanism, leading to microtrauma or even fracture of the LPI. Therefore, with the increase of lumbar curvature, the LPI and FJs are subjected to greater forward shear and axial compression forces (Fig. 3), making the LPI more susceptible to fracture at greater stress and the morphological changes of the PFJs more pronounced in response to long-term stress. In this study, compared with the NL group, the PFJ morphology of the patients in the IS group with the highest SS and LL manifested pars defects, while the patients in the DS group with the second highest SS and LL showed significant morphological changes of the PFJs, i.e., a smaller FJA, a larger PFA, and more severe FJOA, due to the traction of the stress through the intact pedicle and LPI (Fig. 4). Meanwhile, SP was significantly higher in the IS group than in the DS and NL groups and was higher in the DS group than in the NL group. Moreover, within the IS, DS, and NL groups, the patients with higher SS and LL had more pronounced morphological changes of the PFJs and more severe vertebral body slippage, i.e., a smaller FJA, a larger FPA, more severe FJOA, and higher SP (Fig. 5). These findings are in agreement with our previously mentioned hypotheses, i.e., that the differences in the spinopelvic parameters among the IS, DS, and NL groups result in the entire region of the PFJs showing different structural characteristics under different stresses (Fig. 4).
In addition, previous studies have also reported that isthmic lumbar spondylolisthesis occurs predominantly in the L5 vertebra, while degenerative lumbar spondylolisthesis tends to occur more commonly in the L4 vertebra. We believe that this difference is due to the differences in local structure and stress. First, compared with L1–4 vertebrae, the L5 vertebra has the greatest difference between the upper and lower parts of the LPI as well as cephalad and caudal lateral inclinations, and there is the narrowest lateral buttress in the L5 vertebra, so the LPI of the L5 vertebra is anatomically the most fragile [26,27]. Second, the L5 vertebra is located at the lumbosacral junction, and according to the static force equation, the forward shear force (F1) is equal to CF× sin α, where CF is the largest and α= SS, so the L5 vertebra is subjected to the greatest forward shear force. Furthermore, compared with the L4–5 FJs, the articular surfaces of the L5–S1 FJs are wider, coronally oriented in FJA, and vertically oriented in PFA (Fig. 6). In addition, there is a strong iliolumbar ligament at the posterior aspect of L5, as a result of which when the L5 vertebra is subjected to forward shear force, the posterior FJs and ligaments can generate greater reverse resistance, making the L5 vertebra less susceptible to slippage. Therefore, when faced with a strong forward shear force, the strong reverse resistance generated by the L5 vertebra due to these special anatomical structures results in a greater shear stress acting on the LPI, which may lead to its fracture or even spondylolisthesis when the stress that exceeds the bone strength (Fig. 4A). In contrast, the L4 vertebra, with its FJs oriented more sagittally in FJA and horizontally in PFA, and with smaller articular surfaces (Fig. 5), can generate a weaker resistance to a forward shear force than the L5 vertebra in both the posterior FJs and the ligaments, and is therefore more prone to pseudospondylolisthesis with an intact LPI, often resulting in significant morphological changes in the FJs with repetitive traction of shear force (Fig. 4B). In this study, L5 FJA was larger and L5 PFA was smaller compared with those of L4 in the 3 groups. Degeneration of the L4–5 FJs was significantly more severe in the DS group than in the IS and NL groups. In addition, the correlations of SS with FJA and PFA were stronger in the L4 than in the L5 vertebra in all 3 groups. These results were consistent with our analysis described above.
Finally, in this study, we often observed significant disc degeneration below the slipped segment in both the IS and DS groups compared with the patients in the NL group. The mechanisms may be as follows. First, patients in the IS and DS groups have a greater forward shear force of the vertebral body compared with those in the NL group, which may lead to lumbar slippage under the traction of the shear stress; at the same time, the intervertebral discs show significant degeneration under the effect of shear stress during lumbar slippage. Second, because of the LPI defect in patients in the IS group, the role of the posterior FJs in resisting the forward shear force disappears and sharing the axial load of the spine is significantly weakened; therefore, the forward shear force and greater axial compression loads exerted on the disc increase. All of these abnormal stresses can accelerate disc degeneration (Fig. 4A and B). On imaging, we also observed that patients in the IS group tended to have more severe disc degeneration than patients of the same age in the DS group. As the disc degenerates further, its ability to maintain the stability of the lumbar spine is greatly reduced, potentially leading to the onset and progression of spondylolisthesis. Therefore, for patients with lumbar spondylolysis or low-grade isthmic lumbar spondylolisthesis presenting with high SS and LL and surgical indication, the surgical protocol recommends internal fixation treatment to correct the biomechanical imbalance and thereby restore the stability of the lumbar spine caused by the defect of the LPI. This is because direct repair of the LPI, decompression alone, or fusion without internal fixation cannot correct the biomechanical imbalance triggered by the pars defect, which can easily lead to the subsequent progression of slippage due to accelerated degeneration of the remaining stabilizing structures, such as the intervertebral disc, or can even cause severe sagittal imbalance of the lumbar spine or nerve damage.
This study has some limitations. This is a single-center, retrospective, cross-sectional study, and the conclusions need to be further validated by multicenter, large-sample, prospective cohort studies.
CONCLUSION
In this study, we found a correlation between the PFJ morphological characteristics of patients with lumbar spondylolisthesis with the spinopelvic parameters. This was evident from the fact that, compared with NL patients, patients with IS had a larger SS and LL, followed by patients with degenerative lumbar spondylolisthesis, and as SS and LL increased, the changes in the morphology of PFJs became more pronounced, and the lumbar slippage became more severe. This result suggests that different spinopelvic morphologies may result in different morphological characteristics in the PFJ region due to stress differences and the progression of lumbar spondylolisthesis. The present study provides new clues and ideas for further study into the mechanism of the onset and development of lumbar spondylolisthesis.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
