Introduction

Although posterior lumbar fusion with pedicle screws is a “gold-standard” for the treatment of lumbar degenerative diseases, the long-term effects of rigid fixed fusion accelerate the degeneration of adjacent segments, such as adjacent segment instability, spinal stenosis, disc degeneration, fracture, and other long-term degenerative manifestations [1,2,3]. After intervertebral fusion, the increased range of motion (ROM) and abnormal stresses in adjacent segments cause abnormalities in one or more structures, including vertebrae, disc, facet joint, peripheral ligament, and paravertebral muscle, resulting in adjacent segment degeneration (ASD) [4, 5]. As a result, a variety of mobility-preserving devices and surgical approaches, such as non-fusion and hybrid, have gradually emerged clinically to reduce interference with the biomechanical environment of spine in order to prevent or slow down the progression of ASD [6, 7].

In recent years, clinical practitioners have paid attention to the role of the paravertebral muscles in the management and prognosis of the lumbar and cervical spines, and paravertebral muscle degeneration is gradually being recognized as an important cause of ASD [8, 9], and fatty infiltration of the paravertebral muscles is closely associated with degeneration of other structures of the spine [10,11,12,13,14]. Onesti et al [15]. found that the fatty infiltration of the paravertebral muscles that occurs after spinal fusion was found to cause failed back syndrome, suggesting an important association between persistent and recurrent low back pain after lumbar spine surgery and fatty infiltration of the paravertebral muscles. In addition, postoperative stiffness of the low back is a common problem after lumbar spine surgery [16]. Although numerous predisposing factors are known, such as ischemia and paravertebral muscle denervation caused by long incisions, extensive dissection, and long intraoperative muscle strains [17,18,19,20]. However, it is now thought that abnormal mobility and stress transmission after fusion causes paravertebral muscle overload, which is the primary cause of persistent, recurring postoperative pain and postoperative stiffness [21,22,23]. When the ROM of fixed segments is lost, muscle wasting atrophy and fatty infiltration of muscle tissue occur during the healing process [24], causing an increase in compensatory mobility of the adjacent segment and fatty infiltration of the paravertebral muscle of the adjacent vertebrae caused by stresses on the adjacent segment's paravertebral muscle exceeding its limits. Lin et al [36]. However, clinical research is currently divided on how much mobility should be retained and whether higher mobility is better for dynamic stabilization devices. The Isobar Dynamic Stabilization System has undergone five generations of design since its inception: Isolock (1993), Isobar TTL (1998), Aladyn (2002), Isobar Duo (2008), and Isobar EVO (2010), and has become a mature internal fixation device. The original design concept was based on Wolff's law to promote intervertebral fusion for fusion surgery. Since the 1980s, non-fusion stabilization systems such as the posterior lumbar interspinous spine bracing device, interspinous joint device, and posterior transforaminal dynamic internal fixation have been improved and evolved. Some clinicians have also used the Isobar dynamic stabilization system to improve clinical outcomes in non-fusion and hybrid technology [37,38,39]. This system provides spinal stability while preserving a certain degree of mobility of the operated segment, and the load transfer center of the surgical segment is close to the anterior-middle column of the spine, which results in less compressive stress than traditional rigid fixation devices. Given the mere mobility and subtle contour differences between EVO system and the TTL system, we believe it is more scientific to explore mobility through the EVO system and the TTL system for low back stiffness.

In this study, by comparing the C-LSDI before and after surgery in the PITF, TTL and EVO groups, we found significant differences in the C-LSDI of patients in the three groups, with the final follow-up score of the PITF group (29.76 ± 5.39), the TTL group (19.71 ± 6.29) and the EVO group (12.61 ± 1.78), indicating that the postoperative fusion group Patients had the most pronounced low back stiffness, and patients in the dynamic fixation group had lower low back stiffness than those in the rigid group (PITF group), and patients in the EVO group had the lowest low back stiffness scores. This implies that both non-fusion and fusion procedures reduce the mobility of the fixed segment, which causes an increase in the mobility of the superior adjacent segment. In a comparison between groups, the surgical segment mobility was significantly higher in the EVO and TTL groups than in the PITF group, and the upper adjacent segment mobility was lower than in the PITF group, indicating that the Isobar non-fusion procedure has an advantage in preserving surgical segment mobility. Subsequently, we performed a correlation analysis between mobility and low back stiffness and found a significant positive correlation between fixed segment mobility and C-LSDI (r = -0.7968, P < 0.001). This suggests that there is indeed a close relationship between mobility and low back stiffness, and that increasing the mobility of the surgical segment can effectively reduce the stiffness of the patient's low back after surgery, so it is believed that increasing the mobility of the dynamic stabilization system within a certain range may lead to better clinical outcomes.

The relationship between mobility and fatty infiltration of paravertebral muscles

The paravertebral muscles are an important group of muscles for maintaining normal trunk posture and spinal stability, although it has been shown in anatomy and physiology that the multifidus muscle (MM) is the most important for maintaining lumbar joint stability, the erector spinae (ES), and the psoas (PS) and quadratus lumborum (QL) are also indispensable for maintaining lumbar spine stability [39]. Through our clinical observations and the results of most studies [23, 40, 41], fat infiltration in patients with lumbar degeneration is mainly found in MM and ES. Therefore, the overall level of fat infiltration in LM and ES was used in this study to reflect the level of fat infiltration in the paravertebral muscles. There is lots of evidence that there is a correlation between fatty infiltration of the paravertebral muscles and lumbar spine degeneration of multiple structures [42,43,44]. Faur et al. [45] noted a low correlation (R = 0.37) and significant association (ANOVA, P = 0.001, 95% CI: 2.07 to 8.14) between the grade of disc degeneration and lumbar multifidus fat atrophy and concluded that the degree of paravertebral muscle fat infiltration was mildly negatively correlated with the level of the disc. This suggests that intramuscular fat infiltration also bridges the gap between disc degeneration and paravertebral sarcopenia. Paravertebral muscle degeneration also progresses with age, including muscle size reduction and muscle fat infiltration [46]. However, there are no reports in the literature on the relationship between posterior lumbar internal fixation surgical segment mobility and upper adjacent segment mobility and fatty degeneration of the paravertebral muscles in adjacent segments. Because of the prevalence of ASD in the upper adjacent segment and the varying degrees of paravertebral muscle destruction in the fixed segment, the paravertebral muscles of the upper adjacent segment were chosen as the target muscle group for this study.

Current commonly used measures of paravertebral muscle degeneration include paravertebral muscle cross-sectional area (CSA) and fat infiltration grade. Among them, both CSA and fat infiltration grade are associated with low back pain, radiculopathy and spinal stenosis [47,48,49,50]. However, some reported results of CSA were not related to fatty infiltration of the paravertebral muscle [51, 52] and indicated that fatty infiltration is more reflective of paravertebral muscle degeneration compared to CSA [53, 54]. Kader et al. [55] and Parkkola et al. [56] found significant muscle atrophy of the paravertebral muscle even in the absence of reduced CSA, with the atrophied muscle being replaced by fatty and fibrous tissue. Therefore, the fatty infiltration of the paravertebral muscles is more sensitive to early paravertebral degeneration when analyzing whether paravertebral muscle degeneration is occurring. The commonly used grading of fat infiltration is the Goutallier grading, and Battaglia et al [57]. reported a significant correlation between the Goutallier classification and the rate of fat infiltration in the paravertebral muscles, which has also been shown to be a reliable grading system for fat infiltration in several studies [57, 58]. Therefore, this grading system was selected for the evaluation index of paravertebral muscle degeneration in this study. In addition, Lee et al [29]. found the most significant changes in the L4/5 and L3/4 segments by measuring paravertebral muscle CSA and fat infiltration rates at each segmental level in patients with lumbar paravertebral muscle degeneration. Jun et al [59]. found that the degree of fatty infiltration of the paravertebral muscles was negatively correlated with the lumbar lordosis by analyzing the imaging data of 50 elderly patients [60]. Accordingly, in order to minimize confounding factors, only patients with single-segment fixation at L4/5 and L5/S1 and no significant difference in lumbar lordosis at the last follow-up were included in this study.

The MRI results showed that the TTL group had a significantly lower Goutallier grade at the last follow-up than the PITF group, but because of the variability in intraoperative bleeding and operative time between the two (164.67 ± 17.99 mL vs. 144.58 ± 18.65 mL, 145.71 ± 21.11 min vs. 129.17 ± 8.80 min), in order to eliminate the effect of these factors, we therefore compared the EVO and TTL groups again. The results showed that the fixed segment mobility in the EVO group was significantly higher than that in the TTL group, the upper adjacent segment mobility in the EVO group was significantly lower than that in the TTL group, and the Goutallier grading of the paravertebral muscles in the upper adjacent segment in the EVO group was significantly lower than that in the TTL group. The Goutallier classification showed a significant negative correlation with the mobility of the fixed segment (r = -0.8092, P < 0.001) and a positive correlation with the mobility of the upper adjacent segment (r = 0.6703, P < 0.001), indicating that when the intervertebral mobility exceeds the normal level, the paravertebral muscles will show significant fatty infiltration, so increasing the mobility of the dynamic fixation device can reduce the fatty infiltration of the paravertebral muscles in the upper adjacent segment by reducing the mobility of the upper adjacent segment.

Conclusion

Compared with decompression combined with rigid fusion, decompression combined with dynamic fixation for single-segment lumbar degenerative disease can significantly reduce postoperative low back stiffness. And a certain range of increased mobility of the dynamic stabilization device can effectively reduce the compensatory mobility of the upper adjacent segment and slow down the fatty infiltration of the paravertebral muscle in the adjacent segment.

Limitation

In the current study, we established relatively strict inclusion and exclusion criteria, excluding as much as possible other factors that may induce paravertebral muscle degeneration, such as intervertebral discs, small joints, age, BMI, and anterior lumbar convexity angle; secondly, during the surgical operation, the intervertebral discs and synovial joints of the operated segments were protected as much as possible to reduce the interference with the biomechanical environment of the spinal unit; finally, for the fusion group we chose PITF because its surgical procedure is more similar to the Isobar system procedure, which better ensures the scientific validity of the controlled trial. However, there were also two main confounding factors in this study: first, the fusion group wore a postoperative lumbar brace for 3 months and the dynamic fixation group wore a brace for only 1 month after surgery, because the purpose of dynamic stabilization was to allow patients to exercise their lumbar back muscles earlier, so the early paravertebral muscle recovery would be more advantageous in the dynamic fixation group compared with the fusion group, which may also be one of the reasons why the paravertebral muscle fat infiltration grade was lower in the dynamic fixation group than in the fusion group at the final follow-up; second, there was a difference in the inclusion criteria, with lumbar spondylolisthesis limited to degree I in the dynamic fixation group and degree II in the fusion group, which may also explain the lower fat infiltration grade in the dynamic fixation group at the final follow-up.