CLP has some potential complications, such as LCL, decreased neck motion range, axial neck pain, C5 nerve root palsy, and lamina closure. Among them, LCL is a significant issue. Having sufficient postoperative cervical lordosis is a prerequisite for CLP to obtain the indirect anterior decompression effect. LCL has been reported to be associated with poor outcomes after laminoplasty in many studies[6–8, 16, 22, 23]. Kim et al.[16] and Miyazaki et al.[18] reported that preoperative higher T1S was a risk factor for LCL. T1S-CL[7, 16, 21] and CL/T1S[24] have also been considered as predictors for LCL after CLP. However, there are some studies showing different results regarding the correlation between T1S and LCL [15, 25–27]. Michael et al.[11] and Seo et al.[7] emphasize the importance of cephalad vertebral level and cervical foraminal stenosis in LCL after laminoplasty. Some studies also report other factors for LCL, such as cSVA[14, 16, 21], C7-SVA[19, 20], CGH-C7 SVA[6, 8] and age[5]. In the present study, we evaluated regional static parameters to identify possible risk factors for postoperative kyphotic alignment change. However, no significant correlations were observed between the static parameters and LCL in our study. These static parameters have their own limitations in accounting for LCL after surgery. According to our knowledge, there are no theories with consensus that explain why these static parameters can affect LCL after surgery. Many researchers think that the posterior neck muscular-ligament complex may play an important role in these processes[5, 11, 15–17, 21, 28, 29].
Recently, some studies have shown the relationship between preoperative dynamic cervical sagittal alignment and LCL after CLP. Lee et al[15] reported the extension function of the cervical spine as an indicator to predict kyphotic change after CLP, and showed that significant kyphotic change occurred in patients whose Ext ROM was < 14°. Moreover, some studies have shown that higher Flex ROM results in greater LCL after CLP[10, 17, 28–30]. The present study showed similar results to those studies: preoperative Ext ROM (β = -0.421) and Flex ROM (β = 0.208) were predictors for postoperative LCL. Our study reported a high negative correlation between Ext ROM and LCL, which implies that enough Ext ROM is a highly reliable factor in preventing LCL after CLP. Similar to the results of previous studies[10, 17, 28–30], our study shows the positive correlation between Flex ROM and LCL. Cervical flexion mobility is blocked by degenerative structures, such as bone, ligaments, or muscles. Fujishiro et al.[28] speculated that increased motion in the flexional direction indicates that such structural forces restricting motion toward the kyphotic position are weak. Because of the surgical injury, the equilibrium necessary to maintain cervical sagittal alignment is disrupted and results in a higher prevalence of LCL.
In our study, we discovered that different degrees of postoperative LCL implied different degrees of neurological recovery. Worse JOA recovery rate was reported in patients in the severe loss group compared with the stability group. Similar tendencies were also shown between the stability group and the mild loss group; however, there was no evidence of statistical significance (p > 0.05). Postoperative mild LCL occurred in patients with a low level of Ext ROM and the influences of LCL on postoperative neurological recovery were limited. Preoperative high levels of Flex ROM aggravate postoperative LCL for patients with low Ext ROM, and severe LCL implies poor clinical outcomes (Table 4, Figs. 4 and 5).
Some researchers have speculated that the degree of cervical extension mobility indicates the cervical constriction reservoir[15] and cervical flexion mobility indicates the forces inhibiting cervical kyphosis[28]. Both Flex ROM and Ext ROM were important factors for LCL after CLP. Ono et al.[10] proposed CL ratio (100 × Flex ROM / total ROM) as a novel predictor for the loss of cervical lordosis after laminoplasty and reported the cut off value of CL ratio for predicting postoperative LCL. Compared with Flex ROM, Ext ROM had a greater influence on postoperative LCL in our study. Therefore, we reported EXR (100 × Ext ROM / total ROM) as a predictor, and EXR showed better prediction in severe lordosis loss than Ext ROM or Flex ROM alone. The optimal cutoff value of EXR to discriminate between severe LCL and not severe LCL was 16.8% (Fig. 3). For patients with a preoperative EXR less than 16.8%, more cervical exercises should be encouraged after surgery due to the high prevalence of severe postoperative LCL. Multilevel posterior cervical fusion or anterior cervical fusion surgery can also be considered, if necessary.
This study has several limitations. First, because our study was retrospective, a selection bias may exist. Second, the number of patients was low. Only 17 cases were assigned to the severe loss group. Third, the follow-up period was 1 year. Choi et al.[31] reported that changes in cervical sagittal alignment generally reach a plateau at 6 months after CLP. Thus, the follow-up period was enough to investigate the risks for LCL after CLP. Finally, only the JOA score was used to evaluate clinical outcomes in the present study.