In general, decrease of thoracic kyphosis and increase of pelvic retroversion work as compensatory mechanisms for decreased lumbar lordosis in ASD patients 13. Static sagittal spino-pelvic parameters measured by full-spine lateral standing radiographs are influenced by transient compensations for spinal deformity and may underestimate spinal balance 5. As previously reported, 3D gait analysis demonstrated that spinal kyphosis and pelvic anteversion significantly increased while walking and detected a failure of compensation for ASD 6. Although 3D gait analysis could be optimal for quantitatively evaluating spinal balance, it has a high initial cost, making its broad use difficult. Thus, another method of static evaluation that can predict the dynamic change of spinal balance is needed. We focused on analyzing fatty degeneration of the lumbar back muscles using MRI, which is both convenient and static. As a result, the current analysis revealed significant positive correlation between increased spinal kyphosis during gait and fat infiltration in the lumbar erector spinae muscles (Fig. 4).
There has been increasing recognition of the benefits of evaluating the back muscles in patients with spinal disorders such as spinal deformity. Recent studies have mainly reported the relationship between fatty degeneration of the back muscles and static radiological spino-pelvic parameters. Fat infiltration in the lumbar back muscles has been correlated with several static spino-pelvic parameters including SVA 14, PT 15 and LL 16. Furthermore, it also has been reported that fatty degeneration of lumbar back muscles is correlated with postoperative outcomes for ASD. The increase of fat infiltration in back muscles assessed by preoperative MRI may predict proximal junctional kyphosis 17 and nonunion of the lumbosacral junction 18. In addition, a very limited number of studies have reported on the correlation between fatty degeneration of the back muscles and the dynamic spinal balance while walking. Lee et al. reported that dynamic spinal balance parameters assessed using 3D gait analysis showed greater correlation with fat infiltration of the lumbar paraspinal extensor muscles than with static radiological parameters 19. It seems that the correlation between fatty degeneration of the back muscles and the dynamic spinal balance while walking might be important because symptoms such as back pain and gait difficulty in ASD patients increase while walking.
In the present study, the fat infiltration rate of the lumbar erector spinae muscles at L1-2 was significantly and positively correlated with thoracic kyphosis and pelvic tilt. These observations led us to speculate that increased fatty degeneration in lumbar erector spinae muscles could weaken spinal compensation for lumbar kyphosis, preventing decreased thoracic kyphosis. Pelvic compensation was not prevented, resulting in the increase of pelvic tilt. On the other hand, the increase of spinal kyphosis during walking was positively correlated with the fat infiltration rate of the lumbar erector spinae muscles at both L1-2 and L4-5. The increased fatty degeneration in lumbar erector spinae muscles could cause severe failure of spinal compensation during walking in ASD. Thus, evaluation of the fat infiltration of lumbar erector spinae muscles has the potential to reflect the dynamic change in spinal sagittal balance in ASD patients. In addition to conventional static radiological examination, it might be useful to detect individual differences in dynamic spinal compensation by analyzing the fatty condition of the lumbar erector spinae muscles to better understand the individual pathophysiology of each ASD patient. This may result in better clinical outcomes following corrective surgery for ASD.
There are several limitations to this study. First, the study is limited by the small sample size. Additionally, we evaluated fat infiltration only of the lumbar back muscles. It has been reported that fatty degeneration of hip and knee joint muscles is correlated with static radiological spino-pelvic parameters 20. Hence, fat infiltration of hip and knee joint muscles may be correlated with dynamic spino-pelvic balance. Further, back muscle strength and sarcopenia, which may affect the evaluation of back muscles as well, were not assessed in this study. A prospective study with a larger sample using these variables should be completed in the future to elucidate the role of those muscles for dynamic spino-pelvic balance.