Impact of cervical range of motion on the global spinal alignment in ankylosing spondylitis patients with thoracolumbar kyphosis following pedicle subtraction osteotomy

Journal Pre-proof

Impact of cervical range of motion on the global spinal alignment in ankylosing spondylitis patients with thoracolumbar kyphosis following pedicle subtraction osteotomy Shi-zhou Zhao MD , Bang-ping Qian MD , Yong Qiu MD , Bin Wang MD , Ji-chen Huang MD , Mu Qiao MD PII: DOI: Reference:

S1529-9430(19)30973-8 https://doi.org/10.1016/j.spinee.2019.09.009 SPINEE 58018

To appear in:

The Spine Journal

Received date: Revised date: Accepted date:

22 December 2018 7 September 2019 10 September 2019

Please cite this article as: Shi-zhou Zhao MD , Bang-ping Qian MD , Yong Qiu MD , Bin Wang MD , Ji-chen Huang MD , Mu Qiao MD , Impact of cervical range of motion on the global spinal alignment in ankylosing spondylitis patients with thoracolumbar kyphosis following pedicle subtraction osteotomy, The Spine Journal (2019), doi: https://doi.org/10.1016/j.spinee.2019.09.009

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.

Impact of cervical range of motion on the global spinal alignment in ankylosing spondylitis patients with

thoracolumbar

kyphosis

following

pedicle

subtraction osteotomy Shi-zhou Zhao1,2, MD, Bang-ping Qian1,2, MD*, Yong Qiu1,2, MD, Bin Wang 1,2, MD, Ji-chen Huang1,2, MD, Mu Qiao1, MD 1. Spine Surgery, Drum Tower Hospital of Nanjing University Medical School, Nanjing, China 2. Medical School of Nanjing University, Nanjing, China *To whom correspondence should be addressed. E-mail: [email protected] Address: Drum Tower Hospital of Nanjing University, Zhongshan Road 321,Nanjing 210008, China; Fax: 0086-25-6818-2202; Telephone number: 0086-25-6818-2202

Conflicts of interest and source of funding: All authors declare there is nothing to disclose.

Acknowledgements: Thank Wei-yi Diao, Bin-wen Xuan and Zhuo-jie Liu for proof reading the article.

Impact of cervical range of motion on the global spinal alignment in ankylosing spondylitis patients with

thoracolumbar

kyphosis

following

pedicle

subtraction osteotomy

Abstract Background Context: The head’s center of gravity (COG) plumb line (PL) and C7 PL could be simultaneously positioned over the pelvis in adult spinal deformity (ASD) with normal cervical mobility. However, the position of the head in relation to the global spinal alignment has yet to be investigated in ankylosing spondylitis (AS) patients with thoracolumbar kyphosis. Purpose: The objective of this study was to analyze the position of head in relation to the global spinal alignment in AS-related thoracolumbar kyphosis. Study Design/Setting: Retrospective single-center study. Patient Sample: AS patients who underwent lumbar pedicle subtraction osteotomy (PSO) for thoracolumbar kyphosis from January 2010 to August 2016 were reviewed. Only patients with a visible ear canal on the preoperative, immediate postoperative and final follow-up radiographs were included. Outcome Measures: The chin-brow angle (CBVA), cervical range of motion (ROM), lumbar lordosis (LL), thoracolumar angle (TLA), thoracic kyphosis (TK), L1 pelvic angle (L1PA), pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), TK+LL+PI, PI-LL, maximal kyphosis (MK), deformity angular rate (DAR), T1 pelvic angle

(T1PA), T1 tilt, spinosacral angle (SSA), sagittal vertical axis (SVA) COG-C7, SVA COG-femoral head (FH) and SVA C7-S were evaluated. Data regarding the health-related quality of life, including the Oswestry disability index (ODI) and visual analogue scale (VAS)-back score, were also collected preoperatively, 2 years postoperatively and at the latest follow-up. Methods: The cohort was divided into patients with occiput-trunk (OT) concordance or with OT discordance according to the SVA COG-C7 ≤ 30mm or > 30mm, respectively. There was no funding in this study and there are no conflict of interest-associated biases. Results: A total of 43 patients (36 males and 7 females) with a mean age of 34.2 years (range, 18 to 59 years) were identified. There were 17 patients accompanied with OT concordance and 26 patients with OT discordance preoperatively. The cervical ROM was significantly lower (24.0° vs 56.1°, P < 0.001) and SVA COG-C7 was significantly larger (71.7mm vs 7.4mm, P < 0.001) in patients with OT discordance. Furthermore, the PT was larger (41.0° vs 33.5°, P = 0.010) in patients with OT discordance. After surgery, the whole cohort showed an improvement in LL (-8.6° vs -52.8°, P < 0.001). Moreover, the CBVA (25.4° vs 1.3°, P < 0.001) and SVA COG-C7 (46.2mm vs 21.6mm, P < 0.001) were significantly decreased following lumbar PSO. There were 13 patients accompanied with OT discordance postoperatively, and the cervical ROM was still lower (22.5° vs 42.8°, P = 0.024) in these patients. Postoperative PT was larger (26.5° vs 20.1°, P = 0.033) in the patients with OT discordance. At the latest follow-up, there were 17 patients accompanied with OT

discordance. In these 17 patients, the cervical ROM was significantly lower (21.0° vs 47.0°, P = 0.001) and PT was significantly higher (26.2° vs 19.2°, P = 0.012). The ODI and VAS-back scores demonstrated no significant differences between the 2 groups preoperatively, 2 years postoperatively or at the latest follow-up. Conclusions: OT discordance in AS-related thoracolumbar kyphosis could be caused by the reduced cervical ROM. To maintain global spinal balance, the pelvis rotated further backward in response to the larger SVA COG-C7. Moreover, the larger SVA COG-C7 could be decreased after the lumbar PSO. Although there were radiographic differences between the patients with OT concordance and with OT discordance, there was no difference in clinical outcomes, and that a larger sample size and longer follow-up is needed. Key words: ankylosing spondylitis, thoracolumbar kyphosis, cervical range of motion, global spinal alignment, pelvic tilt, sagittal vertical axis.

Introduction Ankylosing spondylitis (AS) is a chronic inflammatory disease and patients’ daily activities are severely restricted by rigid thoracolumbar kyphosis [1, 2]. With the progression of thoracolumbar kyphosis, hyperlordosis occurs in the flexible cervical spine [3, 4]. Inversely, the cervical hyperlordotic compensation relaxes following thoracolumbar realignment [3, 4]. Both pre- and postoperative changes in the cervical spine aim to maintain the global spinal balance [3, 4]. As previously reported, the spine is considered balanced when the head’s center of gravity (COG) plumb line (PL) is positioned over the femoral head (FH) [5]. However, due to the high degree of mobility of the cervical spine in most adult spinal deformity (ASD) patients, the head’s COG and C7 PLs can both be positioned over the pelvis simultaneously [6]. Therefore, evaluation of the C7 PL could be considered a standard method for assessing global spinal alignment without considering cervical alignment [7]. To date, the position of the head in relation to the C7 PL has not been investigated in AS patients with thoracolumbar kyphosis. In the context of thoracolumbar kyphosis, the position of head may be positioned well related to the C7 PL while the cervical mobility is normal. However, in the natural history of AS, the cervical spine is characterized by the ossification and syndesmophyte formation with the loss of mobility [8, 9]. Therefore, we hypothesized that the position of head would be located more anteriorly to C7 PL in the AS patients with less degrees of cervical flexibility. Thus, the current study aimed to investigate the relationship between the cervical

range of motion (ROM) and global spinal alignment in AS patients with thoracolumbar kyphosis following pedicle subtraction osteotomy (PSO). Methods Patient Population A retrospective review was performed among AS patients with thoracolumbar kyphosis who underwent lumbar PSO from January 2010 to August 2016. All data in this single-center study were collected with the approval of the institutional review committee. The inclusion criteria were: (1) age greater than 18 years; (2) treatment with one-level PSO; (3) available cervical flexion and extension radiographs; (4) a minimum of 2 years of follow-up; and (5) visible ear canal on the preoperative, immediate postoperative and final follow-up lateral standing full-length radiographs. The cervical flexion and extension radiographs were taken preoperatively with the patients instructed to flex and extend their neck as far as possible, respectively. A vertical line drawn through the ear canal was measured as the head’s COG PL [10]. A freestanding posture with elbows flexed at 45° was used for standing image acquisition. Additionally, each subject was asked to maintain their gaze as close to horizontal as possible. The exclusion criteria were: (1) AS patients with cervical kyphosis and (2) prior spinal surgery. The included patients were divided into patients with occiput-trunk (OT) concordance or with OT discordance. OT concordance was defined as the sagittal vertical axis (SVA) COG-C7 ≤ 30mm while OT discordance defined as SVA COG-C7 > 30mm [10].

Surgical decisions were made based on the standing full-length radiographs [11]. The osteotomy site was selected at the apex in principle to reconstruct spinal curve harmony. Moreover, the planned osteotomy angle was designed to restore a relatively normal sagittal alignment. Notably, intraoperative verification should be performed to avoid over- or under-correction. After the initial closure of osteotomy site, intraoperative lateral view fluoroscopy of osteotomy vertebra would be routinely taken to confirm the acquired correction magnitude. Preoperative and postoperative pedicle subtraction angle (PSA) formed between the cranial endplate of 1 suprajacent vertebra above the osteotomy vertebra and caudal endplate of 1 infrajacent vertebra below the osteotomy vertebra would be measured on the intraoperative fluoroscopy and preoperative lateral standing full-length radiographs, respectively. The obtained correction angle was defined as the difference between preoperative and postoperative PSA. If the obtained correction angle was less than the planned osteotomy angle, further correction should be gained through manually pushing the osteotomy site and compressing the pedicle screws adjacent to the osteotomy vertebra until the planned osteotomy angle was reached. In this procedure, anterior cortex of the osteotomy vertebra body could be fractured to pursue larger deformity correction. Additionally, the spatial relationship of occiput, shoulder and pelvis could be served as an adjunctive reference standard of the intraoperative correction magnitude, the maximal limitation of the correction magnitude should not exceed the standard that occiput, shoulders and pelvis was repositioned in the same horizontal line [12]. Data collection

Standing full-length radiographs and cervical flexion and extension radiographs were analyzed using the Surgimap Software (Nemaris, Inc., New York, NY, USA). Cervical lordosis was measured as the angle between the inferior C2 endplate and the inferior C7 endplate [13]. Cervical ROM was evaluated as the difference in the cervical lordosis angles measured on the cervical flexion and extension radiographs (Figure 1). Cervical kyphosis was defined as the C2-7 Cobb angle of 0° or more on the lateral standing full-length radiographs regardless of the C2-7 Cobb angle on cervical extension or flexion radiographs[13]. The following radiographic parameters including morphological and positional parameters were measured on the standing full-length radiographs preoperatively, immediate postoperatively and at the latest follow-up. Morphological parameters: (1) spinosacral angle (SSA): the angle between the sacral plate and the line connecting C7 and the midpoint of sacral plate [14]; (2) lumbar lordosis (LL): the Cobb angle between L1 and S1 [3]; (3) thoracolumbar angle (TLA): the Cobb angle between T10 and L2 ; (4) thoracic kyphosis (TK): the Cobb angle between T5 and T12 [3]; (5) L1 pelvic angle (L1PA): the angle between the line drawn from femoral head axis to L1 and the line drawn from femoral head axis to the midpoint of the sacral plate [15]; (6) pelvic incidence (PI)：the angle between the line drawn from the femoral head axis to the midpoint of the sacral plate and the line perpendicular to the sacral plate [3]; (7) maximal kyphosis (MK) was the Cobb angle between the maximally tilted end vertebra [16]; (8) deformity angular rate (DAR) was calculated as the MK divided by the number of vertebral levels involved [16]; (9) T1 pelvic angle (T1PA) was the angle between the line from femoral head axis to T1 and

the line from femoral head axis to the middle of the S1 superior end plate [17]; All positive angles indicate kyphosis. (10) TK+LL+PI and (11) PI-LL mismatch were also calculated. Positional parameters: (1) pelvic tilt (PT): the angle between the vertical and the line drawn from the femoral head axis to the midpoint of the sacral plate [3]; (2) sacral slope (SS): the angle between the sacral plate and the horizontal [3]; (3) T1 tilt was the angle between the line drawn from T1 to femoral head axis and the vertical [17]; (4) Sagittal vertical axis (SVA) COG-C7: the distance between the head’s COG and C7 PL [6]; (5) SVA COG-FH: the distance between the head’s COG and femoral head axis and (6) SVA C7-S: the distance between the C7 PL and the superior posterior corner of the sacrum [3]. All positive distances indicate an anterior position. Measuring methods of the radiographic parameters are presented in Figure 2. Chin-brow vertical angle (CBVA) was defined as the angle between a line drawn from the brow to the chin and the vertical while the lower extremities extended [18]. CBVA was measured on the photographs preoperatively, immediate postoperatively and at the latest follow-up. Clinical outcome data, including the Oswestry disability index (ODI) and visual analogue scale (VAS)-back scores with a range of 0 cm (no pain) to 10cm (worst pain imaginable), were also collected preoperatively, 2 years postoperatively and at the latest follow-up. Statistical analysis

Data analyses were performed using SPSS 18.0 software (SPSS, Chicago, IL). Wilcoxon signed-rank test would be performed if the variable was not normally distributed. Student's t-tests were used to assess differences in the variables distributed normally. Multivariate linear regression was used between the cervical ROM and sagittal radiographic parameters. P values less than 0.05 were considered statistically significant. Results Of the 121 AS patients with thoracolumbar kyphosis underwent lumbar PSO between January 2010 and August 2016, 9 patients underwent 2-level PSO, 15 patients without cervical flexion or extension radiographs, 27 patients without visible ear canal on the preoperative lateral standing full-length radiographs, 21 patients without a completed follow-up of 2 years were excluded. Additionally, 5 patients with cervical kyphosis and 1 patients with prior spinal surgery were excluded. Finally, forty-three patients consisted of 36 males and 7 females were included in the current study. There was no significant difference regarding the clinical demographic and surgical data between the patients with OT concordance and with OT discordance immediately after surgery (Table 1). Mean cervical ROM was 33.8°, ranging from 0 to 78°. During the follow-up, there was one patient presented with cerebrospinal fluid leakage and one patient with coronal decompensation. However, the cerebrospinal fluid leakage did not lead to any neurological complications or infection and the coronal decompensation did not affect the quality of life as well. Thus, no patient received the revision surgery.

For the whole cohort, mean LL was significantly improved after the lumbar PSO (-8.6° vs -52.8°, P < 0.001). Additionally, the CBVA (25.4° vs 1.3°, P < 0.001) and SVA COG-C7 (46.2mm vs 21.6mm, P < 0.001) were significantly decreased following lumbar PSO (Table 2). Before surgery, there were 17 patients with OT concordance and 26 patients with OT discordance. CBVA and SVA COG-C7 were significantly larger in the patients with OT discordance. Cervical ROM was significantly higher (56.1° vs 24.0°, P < 0.001) and age was significantly lower (29.5 vs 37.3, P = 0.009) in the patients with OT concordance. There was no significant difference in the morphological parameters between the patients with OT concordance and with OT discordance. T1 tilt was significantly larger (6.8° vs 2.5°, P = 0.009), SVA COG-FH was significantly lower (83.8mm vs 119.2mm, P = 0.008) and SVA C7-S was significantly larger (160.0mm vs 128.0mm, P = 0.043) in the patients with OT concordance. Furthermore, PT was significantly higher in the patients with OT discordance (41.0° vs 33.5°, P = 0.010) (Table 3). Immediately after PSO, there were 30 patients accompanied with OT concordance and 13 patients with OT discordance. Cervical ROM was significantly lower in the patients with OT discordance (22.5° vs 42.8°, P = 0.024). Moreover, SVA COG-C7, T1PA and PT were significantly larger in the patients with OT discordance (Table 4). Regarding the perioperative change in CBVA and the sagittal parameters, only the change in T1PA was significantly different between the patients with OT concordance

and with OT discordance. Change in T1PA was significantly larger in the patients with OT concordance (29.2° vs 24.8°, P = 0.043). At the latest follow-up, there were 26 patients presented with OT concordance while the residual 17 with OT discordance. Cervical ROM was significantly lower (21.0° vs 47.0°, P = 0.001) and age was significantly larger (39.9 vs 30.5, P = 0.001) in the patients with OT discordance. SVA COG-C7 and PT were significantly larger in the patients with OT discordance (P < 0.05 for both) (Table 5). Regarding the change in CBVA and the sagittal parameters during the follow-up, only the change in SVA COG-C7 was significantly different between the patients with OT concordance and with OT discordance. Change in SVA COG-C7 was significantly larger in the patients with OT discordance (19.2mm vs -8.4mm, P = 0.001). Compared with that preoperatively, ODI (23.2 vs 10.7, P < 0.001) and VAS-back (5.8 vs 2.0, P < 0.001) questionnaires were significantly improved 2 years after surgery for the whole cohort. At the latest follow-up, there was no change in ODI (10.7 vs 10.5, P = 0.259), and VAS showed a significant decrease (2.0 vs 1.7, P = 0.049). Furthermore, there were no significant differences regarding the ODI or VAS-back scores between the patients with OT concordance and with OT discordance preoperatively, 2 years postoperatively or at the latest follow-up (Table 6). Correlations between the cervical ROM and preoperative radiographic parameters were performed using the multivariate linear regression analysis, cervical ROM was correlated with PI, SSA and SVA C7-S while controlling for age (Table 7).

Discussion The importance of global spinal alignment has been well recognized; the reversed sagittal alignment aimed to maintain a balanced spine with the head could be positioned over the FH [19]. The concept of OT concordance/discordance was proposed to evaluate the position of the head in relation to the C7 PL. Due to the greater mobility in the cervical spine, OT concordance could be achieved in most ASD patients; then, the C7 PL could be used to assess global spinal alignment [7, 10]. In the patients with OT discordance, the C7 PL was not appropriate for evaluating the true global spinal alignment [5, 6, 20]. Moreover, the C7 PL was reported to be located more posteriorly in the patients with OT discordance [6]. In the AS patients with ankylosed cervical spine, cervical hyperlordosis to compensate for the thoracolumbar kyphosis would be restricted [8]. As a result, the capacity to maintain OT concordance may be compromised [21]. Therefore, the current study aimed to investigate the relationship between the cervical ROM and global spinal alignment, including the head’s COG PL and the C7 PL, in AS patients with thoracolumbar kyphosis following PSO. As previously reported, while OT discordance is caused by the primary rigid cervical kyphosis, thoracolumbar sagittal alignment could be normal [6]. And, cervical reconstruction surgery is necessary to restore the sagittal alignment [5, 6]. On the other hand, OT discordance could be also attributed to the severe thoracolumbar deformity in ASD [10]. Due to the severely high T1 slope in these patients, OT

concordance could not be maintained even with the maximal extension of cervical spine. In the current study, CBVA and SVA COG-C7 was significantly reduced following lumbar PSO. This phenomenon is supported by a previous study that demonstrated simultaneous improvement in the CBVA following thoracolumbar PSO [18]. Both results demonstrate a positive effect of thoracolumbar realignment on craniocervical alignment. In clinical practice, with regard to the OT discordance in AS patients with thoracolumbar kyphosis, we propose that surgical correction be performed first at thoracolumbar spine. If OT discordance is still present following the thoracolumbar realignment, then cervical reconstruction surgery can be considered. Additionally, cervical ROM in the patients with OT concordance was higher preoperatively, postoperatively and at the latest follow-up (Figure 3). The reduced cervical ROM in the patients with OT discordance could be attributed to the larger age. AS is a chronic inflammatory disease characterized by ossification, and the cervical spine would be rigid following syndesmophyte formation and facet joint fusion [8, 22]. Due to the cervical ankylosis, cervical hyperlordosis required to compensate the thoracolumbar kyphosis could not be achieved; consequently, the SVA COG-C7 was significantly larger. Thus, we assumed that OT discordance in AS-related thoracolumbar kyphosis was the result of the combined effect of thoracolumbar kyphosis and cervical ankylosis.

In ASD, various compensatory mechanisms would be recruited for the regional spinal malalignment to maintain the global spinal balance [19]. As previously reported, thoracic hypokyphosis has been reported to compensate for OT discordance [6]. In the patients with OT discordance preoperatively, the lower SVA C7-S and T1 tilt (positional parameters) revealed that C7 PL moved posteriorly to position the COG of the head over the FH (Figure 4). However, instead of the decreased TK observed in the patients with cervical primary deformity, the posterior movement of C7 PL was achieved through the increased PT in AS patients with a rigid thoracolumbar spine [6]. As both the backward pelvic rotation and cervical hyperlordosis were considered compensatory mechanisms of the thoracolumbar kyphosis, the results of the current study indicate that one of the compensatory mechanisms would be heightened if the other one is restricted. After the lumbar PSO, perioperative change in T1PA was lower and the postoperative T1PA was larger in the patients with OT discordance. Since the deformity under-correction was more common in the patients with OT discordance, more careful intraoperative verification should be performed in the patients with reduced cervical ROM. According to the intraoperative lateral view fluoroscopy, further strengthen of the lordosis should be obtained through compressing the adjacent pedicle screws if the planned lordosis angle was not reached after the initial closure of osteotomy site. Additionally, pelvic backward rotation occurs in response to the OT discordance. As well known, sagittal parameters were widely reported to be correlated with health-related quality of life and achieving a PT < 20° is considered

the objective of spinopelvic realignment [23]. Furthermore, in a previous study with larger patients sample, PT was found to be significantly correlated with VAS-back, ODI, Scoliosis Research Society-22 and Bath Ankylosing Spondylitis Disease Activity Index scores in AS [24]. According to these, we proposed that the physiological PT should be restored intraoperatively in AS patients with thoracolumbar kyphosis. However, in the current study with a relatively smaller patients sample, although the PT was different between the patients with OT concordance and with OT discordance, there were no differences regarding the clinical outcomes, and that a larger sample size and longer follow-up is needed. In clinical practice, greater lumbar lordosis correction and a more caudal PSO level could be performed to provide a larger change in PT [25, 26]. Therefore, when planning PSO for the AS-related thoracolumbar kyphosis accompanied with a reduced cervical ROM, more extensive PSO procedures or more caudal PSO locations could be considered in order to achieve a more ideal PT. Moreover, intraoperative verification of the osteotomy angle as well as the occiput, shoulder and pelvis should be positioned in the same horizontal line must be kept in mind to avoid overcorrection. Limitations of the current study include the single center and retrospective design. Furthermore, syndesmophyte formation and facet joint fusion progress during the natural history of AS. Theoretically, the preoperative cervical ROM would be different from that at the latest follow-up. However, cervical flexion and extension radiographs were not performed during the follow-up period. Finally, the patients

sample in the current study was too small, future study with longer follow-up, more patients and more comprehensive analysis is certainly warranted. Conclusions OT discordance was more common in the AS patients with thoracolumbar kyphosis accompanied with reduced cervical ROM. In response to the OT discordance, increased PT occurred to achieve global spinal balance. Moreover, OT discordance could be improved after the lumbar PSO. The positional radiographic parameters were different between the patients with OT concordance and with OT discordance. However, there was no difference in clinical outcomes and a larger sample size and longer follow-up is needed.

Reference 1. Liu C, Zheng G, Zhang Y, et al. The radiologic, clinical results and digestive function improvement in patients with ankylosing spondylitis kyphosis after pedicle subtraction osteotomy. Spine J 2015;15(9):1988-93. 2. Qian BP, Qiu Y, Wang B, et al. Pedicle subtraction osteotomy through pseudarthrosis to correct thoracolumbar kyphotic deformity in advanced ankylosing spondylitis. Eur Spine J 2012;21(4):711-8. 3. Obeid I, Boniello A, Boissiere L, et al. Cervical spine alignment following lumbar pedicle

subtraction

2015;24(6):1191-8.

osteotomy

for

sagittal

imbalance.

Eur

Spine

J

4. Cecchinato R, Langella F, Bassani R, et al. Variations of cervical lordosis and head alignment after pedicle subtraction osteotomy surgery for sagittal imbalance. Eur Spine J 2014;23 Suppl 6:644-9. 5. Sugrue PA, McClendon J Jr, Smith TR, et al. Redefining global spinal balance: normative values of cranial center of mass from a prospective cohort of asymptomatic individuals. Spine (Phila Pa 1976) 2013;38(6):484-9. 6. Mizutani J, Verma K, Endo K, et al. Global spinal alignment in cervical kyphotic deformity: the importance of head position and thoracolumbar alignment in the compensatory mechanism. Neurosurgery 2018;82(5):686-94. 7. Glassman SD, Berven S, Bridwell K, et al. Correlation of radiographic parameters and clinical symptoms in adult scoliosis. Spine (Phila Pa 1976) 2005;30(6):682-8. 8. Tan S, Yao J, Flynn JA, et al. Zygapophyseal joint fusion in ankylosing spondylitis assessed by computed tomography: associations with syndesmophytes and spinal motion. J Rheumatol 2017;44(7):1004-10. 9. Qian J, Qiu Y, Qian BP, et al. Compensatory modulation for severe global sagittal imbalance: significance of cervical compensation on quality of life in thoracolumbar kyphosis secondary to ankylosing spondylitis. Eur Spine J 2016;25(11):3715-22. 10. Yagi M, Takeda K, Machida M, et al. Discordance of gravity line and C7PL in patient with adult spinal deformity-factors affecting the occiput-trunk sagittal discordance. Spine J 2015;15(2):213-21.

11. Liu ZJ, Qian BP, Qiu Y, et al. Does postoperative PI-LL mismatching affect surgical outcomes in thoracolumbar kyphosis associated with ankylosing spondylitis patients? Clin Neurol Neurosurg 2018;169:71-6. 12. Chang KW, Chen HC, Chen YY, et al. Sagittal translation in opening wedge osteotomy for the correction of thoracolumbar kyphotic deformity in ankylosing spondylitis. Spine (Phila Pa 1976) 2006;31(10):1137-42. 13. Machino M, Yukawa Y, Imagama S, et al. Age-related and degenerative changes in the osseous anatomy, alignment, and range of motion of the cervical spine: a comparative study of radiographic data from 1016 patients with cervical spondylotic myelopathy and 1230 asymptomatic subjects. Spine (Phila Pa 1976) 2016;41(6):476-82. 14. Zhang YP, Qian BP, Qiu Y, et al. Sagittal vertical axias, spinosacral angle, spinopelvic angle, and T1 pelvic angle: which parameters may effectively predict the quality of life in ankylosing spondylitis patients with thoracolumbar kyphosis? Clin Spine Surg 2017;30(7):E871-6. 15. Protopsaltis TS, Lafage R, Smith JS, et al. The lumbar pelvic angle, the lumbar component of the T1 pelvic angle, correlates with HRQOL, PI-LL mismatch, and it predicts global alignment. Spine (Phila Pa 1976) 2018;43(10):681-7. 16. Lewis ND, Keshen SG, Lenke LG, et al. The deformity angular ratio: does it correlate with high-risk cases for potential spinal cord monitoring alerts in pediatric 3-column thoracic spinal deformity corrective surgery? Spine (Phila Pa 1976) 2015;40(15):E879-85.

17. Protopsaltis T, Schwab F, Bronsard N, et al. The T1 pelvic angle, a novel radiographic measure of global sagittal deformity, accounts for both spinal inclination and pelvic tilt and correlates with health-related quality of life. J Bone Joint Surg Am 2014;96(19):1631-40. 18. Suk KS, Kim KT, Lee SH, et al. Significance of chin-brow vertical angle in correction of kyphotic deformity of ankylosing spondylitis patients. Spine (Phila Pa 1976) 2003;28(17):2001-5. 19. Diebo BG, Gammal I, Ha Y, et al. Role of ethnicity in alignment compensation: propensity matched analysis of differential compensatory mechanism recruitment patterns for sagittal malalignment in 288 ASD patients from Japan, Korea, and United States. Spine (Phila Pa 1976) 2017;42(4):E234-40. 20. Mizutani J, Strom R, Abumi K, et al. How cervical reconstruction surgery affects global spinal alignment. Neurosurgery 2018; [Epub ahead of print]. 21. Garrido-Castro JL, Medina-Carnicer R, Schiottis R, et al. Assessment of spinal mobility in ankylosing spondylitis using a video-based motion capture system. Man Ther 2012;17(5):422-6. 22. Baraliakos X, Listing J, von der Recke A, et al. The natural course of radiographic progression in ankylosing spondylitis: differences between genders and appearance of characteristic radiographic features. Curr Rheumatol Rep 2011;13(5):383-7. 23. Schwab F, Patel A, Ungar B, et al. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in

assessing alignment and planning corrective surgery. Spine (Phila Pa 1976) 2010;35(25):2224-31. 24. Shin JK, Lee JS, Goh TS, et al. Correlation between clinical outcome and spinopelvic parameters in ankylosing spondylitis. Eur Spine J 2014;23(1):242-7. 25. Suk KS, Kim KT, Lee SH, et al. Preoperative planning for pedicle subtraction osteotomy: does pelvic tilt matter? Spine Deform 2014;2(5):358-66. 26. Lafage V, Schwab F, Vira S, et al. Does vertebral level of pedicle subtraction osteotomy correlate with degree of spinopelvic parameter correction? J Neurosurg Spine 2011;14(2):184-91.

Figure Legends:

Figure 1

Figure 1. Lateral cervical flexion and extension radiographs showing measurements of the cervical range of motion (ROM). Cervical lordosis was considered as the angle between the inferior endplates of C2 and C7. The cervical ROM was calculated as the difference between the values of cervical lordosis measured on the cervical flexion (A) and extension (B) radiographs.

Figure 2 Figure 2. Lateral standing full-length radiographs showing measurements of the radiographic parameters. A, Sagittal vertical axis (SVA) C7-S was measured as the horizontal distance between the C7 plumb line and the posterior superior corner of S1 (point a).

B, T1 tilt was the angle

between the line drawn from T1 to femoral head axis (point b) and the vertical. L1PA was the angle between the line drawn from point b to L1 and the line drawn from point b to the midpoint of the sacral plate (point c).

C, SVA center of gravity (COG)-C7 was the horizontal distance

between C7 and the vertical line through the ear canal (point d). Spinosacral angle (SSA) indicates the angle between the sacral plate and a line from C7 to point c. SVA COG-FH was the horizontal distance between the vertical line through point d and the point b. T1 pelvic angle (T1PA) was the angle between the line from point b to T1 and the line from point b to the point c. D, Pelvic tilt (PT) was considered as the angle between the vertical and the line drawn from the point a to femoral head axis (point d).

Figure 3 Figure 3. A 25-year-old male with a cervical range of motion (ROM) of 45° underwent PSO at L2 for AS-related thoracolumbar. C, Before surgery, the SVA center of gravity (COG)-C7 and SVA C7-S were -5mm and 170mm, respectively. D, After surgery, the PT was reduced from 33° to 18° and the LL was improved from -19° to -60°. T1 tilt was decreased from 9.7° preoperatively to 1.4° postoperatively. E, At the 2 years follow-up, the LL was -60° and PT was 24°.

Figure 4 Figure 4. A 42-year-old man with a cervical range of motion (ROM) of 3° underwent L2 PSO for the thoracolumbar secondary to AS. C, The preoperative SVA center of gravity (COG)-C7 and SVA C7-S were 65mm and 170mm, respectively. D, PT was reduced from 65° to 43° and LL was improved from 2° to -40° following the PSO. The SVA COG-C7 and SVA C7-S were decreased to 25mm and 55mm, respectively. T1 tilt was decreased from 6.8° preoperatively to -3.8° postoperatively. E, At the 2 years follow-up, LL was -36°. The PT and SVA C7-S were 40° and 60mm.

Table 1. Comparison of the demographic and surgical data between the patients with OT concordance and with OT discordance immediately after surgery* Whole cohort (N=43)

Range

OT concordance (N=30)

Range

OT discordance (N=13)

Range

P value

83.7

-

83.3

-

84.6

-

0.918

Age (years)

34.2 ± 9.9

18-59

33.2 ± 10.3

18-59

36.5 ± 8.6

20-46

0.315

Follow-up time (years)

2.4 ± 0.8

2-5

2.4 ± 0.8

2-4.9

2.5 ± 0.9

2-5

0.909

Fusion levels

8.4 ± 0.8

8-13

8.5 ± 0.9

8-11

8.4 ± 0.8

8-10

0.769

Sex (male, %)

UIV level

1.000

Above T9

11

T7-T9

8

T7-T9

3

T7-T9

Below T10

32

T10-T11

22

T10-T11

10

T10-T11

LIV level

1.000

L5 or above

34

L4-L5

24

L4-L5

10

L4-L5

Sacrum

9

S1

6

S1

3

S1

PSO level

0.412

L1

7

-

4

-

3

-

L2

29

-

22

-

7

-

L3

7

-

4

-

3

-

* Mean values are presented ± SD. Bold type represents statistical significance. UIV, upper instrumented vertebra; LIV, lower instrumented vertebra; PSO, pedicle subtraction osteotomy.

Table 2. CBVA and radiographic parameters of the whole cohort* P Value† Latest follow-up P value&

Baseline

After PSO

CBVA (°)

25.4 ± 17.4

1.3 ± 6.6

<0.001

2.7 ± 7.2

0.238

SVA COG-C7 (mm)

46.2 ± 38.6

21.6 ± 21.3

<0.001

24.1 ± 30.2

0.545

SVA COG-FH (mm) 105.2 ± 43.9

-1.7 ± 56.8

<0.001

-1.7 ± 46.9

1.000

SVA C7-S (mm)

140.6 ± 51.0

23.0 ± 39.5

<0.001

25.5 ± 40.8

0.718

SSA (°)

83.4 ± 11.8

116.4 ± 9.4

<0.001

112.7 ± 8.9

<0.001

T1 tilt (°)

4.2 ± 5.4

-3.0 ± 4.2

<0.001

-3.9 ± 3.4

0.163

T1PA (°)

42.7 ± 10.4

14.9 ± 9.1

<0.001

18.0 ± 8.9

<0.001

MK (°)

67.3 ± 14.2

51 ± 9.3

<0.001

54.5 ± 10.5

<0.001

5.8 ± 1.6

5.3 ± 1.1

0.035

5.5 ± 1.3

<0.001

TK+LL+PI (°)

82.1 ± 16.4

35.0 ± 15.2

<0.001

41.7 ± 17.3

<0.001

TK (°)

43.0 ± 16.5

41.4 ± 13.7

0.155

44.4 ± 13.8

0.008

TLA (°)

30.1 ± 11.3

-10.7 ± 15.1

<0.001

-9.5 ± 15.0

0.026

LL (°)

-8.6 ± 17.7

-52.8 ± 15.4

<0.001

-49.9 ± 13.8

0.019

L1PA (°)

15.7 ± 7.3

9.7 ± 5.7

<0.001

10.8 ± 5.7

0.009

PI-LL (°)

39.1 ± 16.7

-6.4 ± 15.0

<0.001

-2.7 ± 14.7

0.001

PI (°)

47.7 ± 10.8

46.4 ± 9.8

0.081

47.2 ± 9.4

0.265

SS (°)

9.7 ± 9.8

28.7 ± 9.9

<0.001

25.2 ± 7.6

0.003

PT (°)

38.1 ± 9.6

17.7 ± 10.2

<0.001

22.0 ± 9.1

<0.001

DAR

* Mean values are presented ± SD. Bold type represents statistical significance. PSO, pedicle subtraction osteotomy; CBVA, chin-brow vertical angle; SVA, sagittal vertical axis; COG, center of gravity; FH, femoral head; SSA, spinosacral angle; T1PA, T1 pelvic angle; MK, maximal kyphosis; DAR, deformity angular rate; TK, thoracic kyphosis; LL, lumbar lordosis; PI, pelvic incidence; TLA, thoracolumbar angle; L1PA, L1 pelvic angle; SS, sacral slope; PT, pelvic tilt. †The p value represents the comparison between the baseline value and immediate postoperative value, &The p value represents the comparison between the immediate postoperative value and the latest follow-up value.

Table 3. Comparison of the preoperative CBVA and radiographic parameters between the patients with OT concordance and with OT discordance* OT concordance (N = 17)

OT discordance (N = 26)

P Value

CBVA (°)

18.2 ± 18.1

30.2 ± 15.6

0.026

ROM (°)

56.1 ± 22.4

24.0 ± 22.6

<0.001

SVA COG-C7 (mm)

7.4 ± 17.9

71.7 ± 24.4

<0.001

SVA COG-FH (mm)

83.8 ± 45.6

119.2 ± 37.2

0.008

SVA C7-S (mm)

160.0 ± 44.9

128.0 ± 51.6

0.043

SSA (°)

83.8 ± 11.6

83.2 ± 12.1

0.858

T1 tilt (°)

6.8 ± 4.3

2.5 ± 5.4

0.009

T1PA (°)

40.8 ± 9.2

44.0 ± 11.0

0.332

MK (°)

62.2 ± 16.2

70.7 ± 11.9

0.055

5.6 ± 1.9

5.9 ± 1.3

0.450

TK+LL+PI (°)

79.1 ± 17.3

84.1 ± 15.8

0.332

TK (°)

37.9 ± 18.9

46.3 ± 14.2

0.105

TLA (°)

30.7 ± 11.8

26.7 ± 11.2

0.778

LL (°)

-6.2 ± 16.8

-10.2 ± 18.4

0.470

L1PA (°)

14.7 ± 6.1

16.4 ± 8.0

0.441

PI-LL (°)

41.1 ± 12.5

37.8 ± 19.0

0.526

PI (°)

47.3 ± 12.1

48.0 ± 10.1

0.837

DAR

SS (°)

13.8 ± 9.4

7.0 ± 9.2

0.024

PT (°)

33.5 ± 8.9

41.0 ± 9.0

0.010

* Mean values are presented ± SD. Bold type represents statistical significance. OT, occiput trunk; CBVA, chin-brow vertical angle; ROM, range of motion; SVA, sagittal vertical axis; COG, center of gravity; FH, femoral head; SSA, spinosacral angle; T1PA, T1 pelvic angle; MK, maximal kyphosis; DAR, deformity angular rate; TK, thoracic kyphosis; LL, lumbar lordosis; PI, pelvic incidence; TLA, thoracolumbar angle; L1PA, L1 pelvic angle; SS, sacral slope; PT, pelvic tilt.

Table 4. Comparison of the immediate postoperative CBVA and radiographic parameters between the patients with OT concordance and with OT discordance* OT concordance (N = 30)

OT discordance (N = 13)

P Value

CBVA (°)

1.1 ± 6.5

1.8 ± 7.1

0.764

ROM (°)

42.8 ± 28.4

22.5 ± 18.9

0.024

SVA COG-C7 (mm)

14.9 ± 13.5

37.0 ± 27.9

0.001

SVA COG-FH (mm)

-10.3 ± 44.7

18.0 ± 76.4

0.135

SVA C7-S (mm)

19.4 ± 37.3

31.4 ± 44.5

0.367

SSA (°)

117.5 ± 7.7

114.1 ± 12.5

0.282

T1 tilt (°)

-2.9 ± 4.1

-3.1 ± 4.7

0.901

T1PA (°)

12.8 ± 7.9

19.9 ± 10.1

0.018

MK (°)

50.9 ± 10.3

52.5 ± 6.7

0621

5.2 ± 1.3

5.3 ± 0.8

0.832

TK+LL+PI (°)

32.3 ± 15.6

41.4 ± 12.8

0.071

TK (°)

41.2 ± 14.9

41.9 ± 10.9

0.876

TLA (°)

-12.1 ± 13.3

-7.5 ± 18.7

0.371

LL (°)

-53.8 ± 13.6

-50.3 ± 19.4

0.498

L1PA (°)

8.7 ± 4.9

12.1 ± 6.8

0.077

PI-LL (°)

-8.9 ± 12.6

-0.5 ± 18.8

0.093

DAR

PI (°)

44.9 ± 9.5

49.8 ± 9.9

0.135

SS (°)

29.7 ± 7.8

26.2 ± 13.7

0.298

PT (°)

20.1 ± 8.0

26.5 ± 10.3

0.033

* Mean values are presented ± SD. Bold type represents statistical significance. OT, occiput trunk; CBVA, chin-brow vertical angle; ROM, range of motion; SVA, sagittal vertical axis; COG, center of gravity; FH, femoral head; SSA, spinosacral angle; T1PA, T1 pelvic angle; MK, maximal kyphosis; DAR, deformity angular rate; TK, thoracic kyphosis; LL, lumbar lordosis; PI, pelvic incidence; TLA, thoracolumbar angle; L1PA, L1 pelvic angle; SS, sacral slope; PT, pelvic tilt.

Table 5. Comparison of the CBVA and radiographic parameters between the patients with OT concordance and with OT discordance at the latest follow-up* OT concordance (N = 26)

OT discordance (N = 17)

P Value†

CBVA (°)

1.2 ± 6.4

4.9 ± 8.0

0.097

ROM (°)

47.0 ± 26.9

21.0 ± 20.0

0.001

SVA COG-C7 (mm)

8.9 ± 23.1

47.9 ± 23.9

<0.001

SVA COG-FH (mm)

-7.5 ± 46.5

7.2 ± 47.5

0.321

SVA C7-S (mm)

24.1 ± 45.8

27.6 ± 32.8

0.786

SSA (°)

113.5 ± 8.5

111.5 ± 9.8

0.487

T1 tilt (°)

-3.3 ± 3.7

-4.8 ± 2.6

0.160

T1PA (°)

16.0 ± 8.8

21.1 ± 8.5

0.063

MK (°)

54.4 ± 11.4

54.7 ± 9.1

0.937

5.5 ± 1.4

5.5 ± 1.2

0.928

TK+LL+PI (°)

40.5 ± 20.3

43.5 ± 11.8

0.584

TK (°)

44.9 ± 14.9

43.7 ± 12.3

0.777

TLA (°)

-12.2 ± 12.4

-5.3 ± 17.8

0.139

LL (°)

-50.8 ± 12.3

-48.7 ± 16.2

0.628

L1PA (°)

9.7 ± 5.1

12.4 ± 6.2

0.132

PI-LL (°)

-4.4 ± 13.7

-0.18 ± 16.3

0.362

DAR

PI (°)

46.4 ± 10.5

48.47 ± 7.3

0.473

SS (°)

27.1 ± 6.0

22.2 ± 8.9

0.037

PT (°)

19.2 ± 8.5

26.2 ± 8.6

0.012

* Mean values are presented ± SD. Bold type represents statistical significance. OT, occiput trunk; CBVA, chin-brow vertical angle; ROM, range of motion; SVA, sagittal vertical axis; COG, center of gravity; FH, femoral head; SSA, spinosacral angle; T1PA, T1 pelvic angle; MK, maximal kyphosis; DAR, deformity angular rate; TK, thoracic kyphosis; LL, lumbar lordosis; PI, pelvic incidence; TLA, thoracolumbar angle; L1PA, L1 pelvic angle; SS, sacral slope; PT, pelvic tilt.

Table 6. Comparison of clinical outcomes between the patients with OT concordance and with OT discordance* OT concordance

OT discordance

P Value

Baseline

22.4 ± 15.1

23.8 ± 14.2

0.766

2 years follow-up

10.6 ± 9.1

11.0 ± 11.1

0.902

ODI

Latest follow-up

10.1 ± 9.0

11.1 ± 10.9

0.749

Baseline

5.4 ± 2.5

6.1 ± 1.3

0.258

2 years follow-up

1.9 ± 1.7

2.1 ± 1.7

0.801

Latest follow-up

1.5 ± 1.2

2.1 ± 2.1

0.230

VAS

* Bold type represents statistical significance. OT, occiput trunk; ODI, Oswestry disability index; VAS, visual analogue scale. Table 7. Multivariate linear regression analysis between the preoperative sagittal radiographic parameters and cervical range of motion while controlling for age* Coefficient

T Value

P Value

PI

1.294

2.442

0.022

SSA

-1.474

-2.652

0.013

SVA C7-S

-0.997

-2.320

0.028

* Bold type represents statistical significance. PI, pelvic incidence; SSA, spinosacral angle; SVA, sagittal vertical axis.