Objectives:To analyze the effect of vertebral or intraspinal abnormalities on the efficacy of pos-terior corrective surgery for scoliosis patients with arthrogryposis multiplex congenita(AMC).Methods:A retro-spective study was conducted on 30 scoliosis patients with AMC who underwent posterior corrective surgery in the Department of Spine Surgery of Drum Tower Hospital between August 2001 and November 2021.There were 18 males and 12 females with a mean age of 15.9±5.8(6-32)years.The patients were divided into ab-normal group(15 cases)and control group(15 cases)according to with or without vertebral or intraspinal ab-normalities.The types of vertebral or intraspinal abnormalities in the abnormal group were recorded,and the number of fusion segments,operative time and intraoperative blood loss were compared between groups.The complications during follow-up were also collected.The flexibility of major curve was assessed on Bending radiographs,and the main curve Cobb angle,the distance between C7 plumb line and center sacral vertical line(C7PL-CSVL),the sagittal vertical axis(SVA),the thoracic kyphosis(TK),and the lumbar lordosis(LL)were measured on the standing whole spine anteroposterior and lateral X-ray images at pre-operation,postoperative two weeks and two years,and the correction rate of major curve was calculated at 2 weeks after surgery and the final follow-up.Results:There were 10 cases of simple poor segmentation(66.6％),2 cases of poor seg-mentation combined with tethered cord(13.3％),and 1 case of poor segmentation combined with arachnoid cyst,simple hemivertebra,and simple wedge-shaped vertebra(6.7％)each in the abnormal group.There were no significant differences between abnormal group and control group in the number of fusion segments,opera-tive time and intraoperative blood loss(P＞0.05).No complication was observed during operation in the abnor-mal group,and 3 complications were observed during follow-up,including 2 cases with screw misplacements and 1 case with thoracic effusion and the right brachial plexus paralysis;5 cases of complications in the control group were observed,including 1 case with malignant hyperthermia and cardiac arrest during the surgery,3 cases with screw misplacements and 1 with thoracic effusion and screw placement failure.The in-cidence of complications was not statistically different between the two groups(P=0.628).The flexibility of ma-jor curve before operation was not statistically different between the two groups(P＞0.05);The major curve Cobb angle,C7PL-CSVL,SVA,TK and LL at pre-operation,post-operative two weeks and 2 months were not statistically different between groups(P＞0.05);The correction rate of major curve at 2 weeks and 2 months after surgery were not significantly different as well(P＞0.05).Conclusions:Vertebral or intraspinal abnormali-ties have no obvious effects on the efficacy of posterior corrective surgery for the treatment of scoliosis pa-tients with AMC,and there is no significant increase in the incidence of intraoperative and postoperative com-plications.

Objective:To investigate the feasibility and effects of modified sequential correction technique combined 3-columns osteotomy for severe kyphoscoliosis.Methods:A retrospective analysis was performed on 18 patients (7 males and 11 females) with severe kyphosis who received modified sequential correction technique combined 3-columns osteotomy in our hospital from June 2019 to April 2020. Preoperative, postoperative and final follow-up clinical and imaging outcomes were evaluated.Results:In this cohort, the average fixed segment was 11.2±3.8. The average operative duration was 401.9±68.9 min and the average intraoperative blood loss was 2 418.8±736.9 ml. The Cobb angle was improved significantly from 65.0°±16.4° pre-operatively to 41.6°±14.1° post-operatively. At final follow-up, it was 41.4°±14.3°, which was not significantly different from that after operation. Global kyphosis (GK) was 65.5°±20.8° pre-operatively and 28.1°±13.8° post-operatively with correction rate of 57.8%±17.8%. However, GK was 29.3°±14 .2° at postoperative 1 year , which was not significantly different from that after operation. There was no significant difference in C 7PL-CSVL ( F=0.449 , P=0.642) or SVA ( F=3.519, P=0.058) among the three time points. There was no alter of SEP and MEP observed during operation. Four patients had temporary lower limb numbness after operation, while the symptoms disappeared at 6 months after operation. There was no instrumental failure during the follow-up. Conclusion:Patients with severe kyphoscoliosis can obtain satisfied local correction by undergoing modified sequential correction technique combined 3-columns osteotomy without significant loss of correction at 1 year after operation. It can effectively avoid instability and dislocation of the osteotomy site and massive bleeding during the operation. As a simplified surgical procedure, it can reduce the difficulty of rod loading without prolonged operation duration. Further, this technique can ensure lower incidence of neurological complications and rod failure.

Objective:To investigate the clinical and imaging outcomes of Lenke 5 idiopathic scoliosis posterior selective fusion with "Last Touching Vertebra-1" as the lower instrumented vertebra (LIV).Methods:A total of 103 patients with Lenke 5 idiopathic scoliosis who underwent posterior selective fusion orthopedic surgery from April 2009 to March 2020 were analyzed retrospectively. The LIV was the last touching vertebra (LTV) in 45 cases (LTV group) and the LTV-1 in 58 cases (LTV-1 group). The follow-up duration was more than 2 years. SRS- 22 questionnaire was used to evaluate the clinical effects at 2 years after operation. The anterior and lateral radiographs of the whole spine were measured preoperatively, postoperatively and at 2 years after operation to obtain all the following imaging parameters, including scoliosis Cobb angle, apical vertebral translation (AVT), coronal balance, LIV tilt, LIV lower intervertebral disc angle, LIV translation, LIV lower vertebral translation, LTV/LIV rotation degree, lumbar lordosis angle, pelvic incidence angle, sagittal balance. The complications were summarized and were analyzed for investigating potential risk factors.Results:At 2 years after operation, the correction rates of main Cobb in LTV group and the LTV-1 group were 60.2%±11.1% and 55.3%±14.1%, respectively. The coronal balance was 3.5±9.8 mm and 4.9±10.6 mm respectively. The sagittal balance was -15.5±18.1 mm and -19.6±22.6 mm respectively. There was no significant difference between the two groups ( t=2.305, P=0.085; t=-0.695, P=0.489; t=0.992, P=0.324). The incidence of proximal junction kyphosis in the two groups was 2.2% (1/45) and 8.6% (5/58), respectively. The incidence of significant loss of main Cobb correction and distal adding-on was 13.3% (6/45) and 25.9% (15/58) respectively without significant difference (χ 2=1.891, P=0.169; χ 2=2.451, P=0.117). Compared with non-complication patients (39 cases), 19 patients with complications in LTV-1 group had a greater degree of coronal balance to the convex side (23.9±9.5 mm vs. 14.6±11.5 mm, t=3.06, P=0.003), a greater LIV tilt (29.2°±3.7° vs. 25.3°± 5.3°, t=2.85, P=0.006), and a greater degree of LTV rotation (1.0(1, 1) vs. 0.6(0, 1), Z=-2.97, P=0.003). Logistic regression analysis showed that large preoperative LIV tilt and large preoperative coronal balance were the risk factors of complications during follow-up. Conclusion:The selection of LTV and LTV-1 as LIV in patients with Lenke 5 adolescent idiopathic scoliosis could obtain satisfied coronal, sagittal balance and low incidence of mechanical related complications during follow-up. For patients with preoperative coronal balance >17.0 mm or LIV tilt >25.3°, the risk of mechanical related complications might be higher than that when "LTV-1" was selected as LIV.

Objective:To investigate the imaging features and surgical strategies of late-onset spinal deformity after myelomeningocele (MMC) repair.Methods:A total of 23 patients with late-onset spinal deformity after MMC repair from January 2006 to December 2019 were retrospectively analyzed, including 16 males and 7 females, aged 15.4±5.9 years (range, 6-28 years). All patients underwent MMC resection and repair in infancy (0-4 years). The complications of MMC, imaging characteristics of spinal deformity (Cobb angle of scoliosis, coronal balance, regional kyphosis), surgical methods, clinical outcomes and incidence of complications were analyzed. The Scoliosis Research Society-22 (SRS-22) score and Oswestry disability index (ODI) were used to evaluate the quality of life.Results:All patients were followed up for 2.4±0.8 years (range, 1-4 years). Among 23 patients, MMC occurred in the upper thoracic segment in 3 cases, thoracic segment in 1 case, thoracolumbar segment in 13 cases, and lumbosacral segment in 6 cases. 16 patients had scoliosis or kyphosis with the apex of the spine in the same segment as the MMC lesion. Among 13 patients with MMC located in thoracolumbar segment, 12 patients had scoliosis and 9 patients had kyphosis. Among 6 patients with MMC located in lumbosacral segment, 3 patients had pelvic tilt. Vertebral deformities included widening of pedicle space in 21 cases, enlargement of spinal canal in 19 cases, absence of spinous process in 17 cases, malsegmentation in 17 cases, and hemivertebra deformity in 9 cases. Intramedullary lesions included split cord in 6 cases and tethered cord in 9 cases. The overall implant density was 57.2%±17% (range, 16.6%-100%). At the last follow-up, the Cobb angle of scoliosis was 40.9°±19.1°, which was significantly smaller than 71.5°±28.2° before operation ( P<0.001). The local kyphosis angle was 26.7°±12.9°, which was significantly lower than that before operation (40.4°±21.5°), the difference was statistically significant ( P<0.001).The coronal balance was 16.1±13.6 mm, which was smaller than that before operation 28.5± 23.7 mm, the difference was statistically significant ( P<0.001). The total score of SRS-22 was 18.7±0.7, which was higher than that before operation 17.7±0.9, and the difference was statistically significant ( t=-9.74, P<0.001); ODI was 25.5%±6.2% after operation, which was significantly lower than that before operation (44.8%±10.1%), the difference was statistically significant ( t=13.66, P<0.001). Dural rupture occurred in 4 patients, including postoperative cerebrospinal fluid leakage in 2 cases; postoperative pleural effusion in 1 patient; and screw malposition in 2 patients. Three patients had broken rods and one had deep infection at final follow-up. Conclusion:About 70% of MMC patients who underwent resection and repair in early childhood developed late-onset spinal deformity in adulthood with the lesion at the parietal vertebrae. Posterior correction can obtain satisfactory clinical results. If the posterior element of the apical vertebral body is hypoplastic, the implant density can be increased by anterior vertebral screw, lamina hook fixation, and S 2 sacroiliac screw.

Objective:To investigate the clinical outcome and complications associated with utilizing sagittal plane stable vertebra-1 (SSV-1) as the distal instrumented vertebra (LIV) in posterior fusion of thoracic kyphosis with Scheuermann's Disease kyphosis (STK).Methods:A longitudinal study on patients with STK who underwent posterior correction and fusion surgery from January 2018 to June 2021 were conducted. All participants had a follow-up duration over two years. Patients were divided into two groups according to the segment of LIV: the SSV group, where LIV was located in SSV; and the SSV-1 group, where LIV was located in the vertebral body above SSV. The radiographic parameters, including global kyphosis (GK), lumbar lordosis (LL), and sagittal plane (SVA), LIV offset distance (LIV translation), pelvic incidence (PI), pelvic tilt (PT) and sacral slope (SS), were compared between the two groups. The SRS-22 scale was used to evaluate health-related quality of life at pre-operation and last follow-up, and the incidence of postoperative distal junctional kyphosis (DJK) was also recorded. Analytical techniques, such as Analysis of Variance and Mann-Whitney tests, were employed to compare inter-group differences.Results:A total of 57 patients were included in the study, 36 in the SSV group and 21 in the SSV-1 group. The average age for patients were 16.1±2.3 years (range 13-20 years), and the average follow-up time was 32.8±6.8 months (range 24-53 months). There were no statistically significant differences between the two groups in terms of gender, age, follow-up time, surgical time, intraoperative bleeding volume, and fusion level. Before surgery, the LIV deviation distance in the SSV group was significantly lower than that in the SSV-1 group (-7.9±11.0 mm vs. 31.5±11.5 mm, t=7.64, P<0.001). In the SSV group, the preoperative GK was 79.3°±10.5°, and the last follow-up GK was 44.4°±8.5°, which was significantly improved compared to preoperative value ( t=28.28, P<0.001); in the SSV-1 group, the preoperative GK was 81.1°±10.6°, and the value at 1-week post-operative was 44.9°±7.8°, which was significantly improved compared to pre-operative value ( t=22.23, P<0.001). At the last follow-up, it was 45.1°±8.7°, with a correction rate of 44.3%±8.5%. No significant difference was observed between the two groups in terms of GK, LL, SVA, PI, PT and SS at pre-operative, 1-week post-operative and last follow-up ( P>0.05). All patients had no intraoperative complications of nerve injury. During the follow-up period, one patient (1/21, 4.8%) developed DJK without complications such as proximal kyphosis, pseudarthrosis, or failed internal fixation. At the last follow-up, the functional score of SRS-22 in SSV-1 group improved from preoperative (3.5±0.54) to postoperative (4.1±0.62), with an average improvement rate of 19.2%±3.2%, and the difference was statistically significant ( t=3.74, P=0.001). These results indicating that the surgical treatment was effective in relieving the symptoms of the patients. Conclusion:Selecting SSV-1 as LIV in corrective surgeries for STK appears to produce commendable clinical results with minimal implant-associated complications over a two-year observation period.

Objective:To investigate the clinical outcomes and complication of posterior surgery for Scheuermann kyphosis fusing to different distal fusion levels.Methods:From January 2012 to December 2017, a consecutive cohort of 34 patients who were treated with posterior spinal instrumented correction and satisfied the inclusion criteria were retrospectively reviewed, including 29 males and 5 females, aged 17.1±4.3 years (range, 12-30 years). All of the patients had a minimum follow-up of 2 years. According to the distal fusion level, patients were divided into 2 groups. Group sagittal stable vertebra (SSV) (22 cases) included patients whose lowest instrumented vertebra (LIV) was SSV; Group SSV-1 (12 cases) included patients who had a LIV one level above the SSV. Radiographic parameters including global kyphosis (GK), lumbar lordosis (LL), sagittal vertical axis (SVA), pelvic incidence (PI), pelvic tilt (PT), and sacral slope (SS) were measured in the standing radiographs before and after operation and at the latest follow up. Intraoperative and postoperative complications were recorded. The Scoliosis Research Society-22 questionnaire (SRS-22) were conducted at pre-operation and the final follow up to evaluate the clinical outcomes. The sagittal radiographic parameters and the incidence of distal junctional kyphosis (DJK) were compared between the two groups.Results:There were no significant differences in terms of age, sex, radiographic measurements and scores of SRS-22 between two groups preoperatively ( P>0.05). The correction rates of GK in the SSV group and the SSV-1 group were 42.8%±7.6% and 43.2%±8.4% ( t=0.151, P=0.881) respectively. While the correction rates loss were 1.2%±5.2% and 3.9%±7.2% ( t=0.767, P=0.449) at the latest follow up. No significant difference was observed in terms of other radiographic parameters ( P>0.05). During the postoperative follow up period, 3 patients (16.7%) in SSV group and 2 patients (13.6%) in SSV-1 group developed DJK. The incidence of DJK did not show any significant difference between two groups ( χ2=0.057, P=0.812). At the final follow-up, the function scores of SRS-22 in SSV-1 group (4.1±0.6) was significantly higher than SSV group (3.7±0.5) ( t=2.300, P=0.028) and there was no significant difference in the rest of the domain ( P>0.05). Conclusion:Compared with stopping at SSV, fusion to SSV-1 could achieve comparable curve correction with the preservation of more lumbar motility. Moreover, it would not increase the risk of DJK. As a result, we recommend selecting SSV-1 as the ideal LIV for SK patients.

Objective:To establish age- and gender-based normative values of sagittal spinal-pelvic alignment in Chinese adult population, and to investigate influence of age, gender and ethnicity on sagittal spinal-pelvic alignment in Chinese normal adults.Methods:A total of 786 asymptomatic Chinese adult volunteers aged between 20 and 89 years were prospectively recruited from different spine centers. The inclusion criteria were: 1) age between 20 to 89 years old; and 2) Oswestry disability index (ODI) scored lower than 20. The exclusion criteria were: 1) previous history of spinal, pelvic or lower limb pathologies that could affect the spine; 2) presence of recent and/or regular back pain; 3) previous surgeries on spine, pelvic and/or lower limb; and 4) pregnancy. Demographic characteristics of these subjects including age, gender, body weight and height were recorded. During the enrollment of volunteers, 16 groups were defined based on the age (20 s, 30 s, 40 s, 50 s, 60 s, 70 s and 80 s) and gender. Whole body biplanar standing EOS X-ray radiographs were acquired to evaluate the sagittal alignment. Spinal-pelvic parameters including pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), thoracic kyphosis (T 5-T 12, TK), lumbar lordosis (L 1-S 1, LL), lower lumbar lordosis (L 4-S 1, LLL), global tilt (GT), T1 pelvic angle (TPA) and sagittal vertical axis (SVA) were measured. Values of PI-LL and lordosis distribution index (LLL/LL, LDI) were calculated. Radiographic measurements of 100 subjects were randomly selected to determine the intra- and inter-observer reliabilities using inter- and intra-class correlation coefficients (ICC). The spinal-pelvic parameters were compared among volunteers between different age and gender groups. The comparison was also made among various ethnic population. Results:The mean value was 23.7±7.1 kg/m 2 for BMI and 6.9%±2.5% (range, 0-18%) for ODI score. Each sagittal spinal-pelvic parameter was presented with mean value and standard deviationbased on age and gender. The ICCs of radiographic measurements ranged from 0.89 to 0.95, suggesting good to excellent intra- and inter-observer reliabilities. Significant differences were observed between males and females in multiple sagittal parameters (all P values <0.05). Compared to the male subjects, significantly higher values of PI (41.4° for male vs. 45.0° for female, P<0.001), PT (10.7° for male vs. 13.9° for female, P<0.001), PI-LL (-0.5° for male vs. 1.8° for female, P<0.001), and GT (10.9° for male vs. 13.5° for female, P<0.001) were documented in female subjects. Males had significantly higher values of LLL (28.6° for male vs. 26.6° for female, P<0.001) and LDI (0.68 for male vs. 0.63 for female, P<0.001). PI-LL, SVA, GT and TPA increased with aging from Group 40 s to Group 80 s, while LL, LLL and LDI decreased gradually, and TK decreased slowly with aging. Comparison of sagittal spinal-pelvic parameters between different ethnic subjects showed that Chinese adult population presented lower PI, SS, TK and LL as compared with American population; lower PI, SS and LL as compared with Japanese population. But the variation trend with aging tended to be consistent among different ethnic populations. Conclusion:Age- and gender-based normative values of sagittal spinal-pelvic alignment were established in asymptomatic Chinese adult population. Sagittal spinal-pelvic alignment varies with age and gender, and presented different compensation mechanism among different ethnic populations. Therefore, to achieve balanced sagittal alignment, age, gender and ethnicity should be take intoconsideration when planning spine correction surgery.

Objective:To evaluate the clinical outcomes and complications of second sacral alar-iliac (S 2AI) technique utilized in adult patients with neuromuscular scoliosis, and to evaluate the impact on patients' quality of life. Methods:All of 11 patients (6 males and 5 females) applying S 2AI technique from January 2014 to December 2016 were retrospectively reviewed. The average age of the patients was 39.6±12.7 years. Among them, 8 cases were poliomyelitis, 2 cases were spinal muscular atrophy and 1 case was muscular dystrophy. All of 11 patients underwent posterior spinal fusion and utilized S 2AI screws for pelvic fixation. All patients were taken anteroposterior and lateral radiographs of the entire spine. Cobb's angle, spinal pelvic obliquity (SPO), regional kyphosis (RK), sagittal vertical axis (SVA) were recorded at pre-operation, post-operation and last follow-up. The Scoliosis Research Society (SRS)-22 questionnaires and Oswestry disability index (ODI) were utilized to evaluate the patient-reported outcomes. All complications were also recorded. Repeated measurement analysis of variance, t-test or non-parametric test was used to analyzed the data, respectively. Results:The average follow-up period was 62.4±10.8 months. The pre-operative Cobb angle was 98.0°±24.0°, and the post-operative Cobb angle was 60.7°±20.8°, of which difference was significant ( Z=3.015, P=0.003). The correction rate of Cobb angles was 57.2%±17.7%. 1-year after operation, the Cobb angle was 62.8°±23.6°, no loss of correction was found ( Z=0.294, P=0.797). And at last follow-up, the Cobb angle was 61.6°±21.7°, the correction maintained well ( Z=0.603, P=0.594). The pre-operative, post-operative, 1-year post-operative and last follow-up spinal pelvic obliquity were 37.0°±11.8°, 21.5°±11.6°, 23.2°±10.1° and 21.1°±8.6°. The significant improvement was obtained ( Z=2.934, P=0.003) and no loss of correction was found ( Z=0.690, P=0.519; Z=0.000, P=1.000). The pre-operative, post-operative, 1-year post-operative and last follow-up regional kyphosis were 46.8°±23.6°, 18.6°±10.6°, 18.9°±11.4° and 19.5°±9.8°. The significant improvement was obtained postoperatively ( Z=4.364, P<0.001) and remained stable at the last follow-up ( Z=0.074, P=0.945; Z=0.271, P=0.838). When compared the pre- and post-operative sagittal vertical axis, no significant difference was detected. In these patients, one patient had rod breakage and underwent revision, one patient suffered deep infection, and recovered by debridement surgery, one patient suffered from severe pain in the lower back and relieved with conservative treatment. Conclusion:The S 2AI technique utilized in patients with neuromuscular scoliosis could obtain satisfying clinical outcomes and provides safe, durable fixation with low rates of complications.

Objectives:To investigate the feasibility of second sacral alar-iliac (S 2AI) screw placement and trajectories index in patients with neuromuscular scoliosis with severe pelvic obliquity; and to explore the accuracy of S 2AI screw placement by O-arm three-dimensional (3D) CT navigation (Medtronic, Minneapolis, MN, USA). Methods:All of 28 patients with neuromuscular scoliosis who underwent posterior long fusion with S 2AI between January 2017 and August 2020 were reviewed, with an average age of 22.2 years old (ranged from 10 to 51 years), and the pelvic obliquity angle was 27.54±9.90° (ranged from 16.2° to 53.6°). Based on 3D CT reconstruction of these specimens, virtual S 2AI screw channels were identified and measured. Entry point was determined by 1 mm inferior and 1 mm lateral to the S 1 dorsal foramen, and virtual S 2AI screw trajectories with maximum length and width were explored by rotating 3D pelvis. The parameters of the determined channels were measured including caudal angulation on the sagittal plane (sagittal angle, SA), lateral angulation on the transverse plane (transverse angle, TA) and the maximal length of the channel (maximal length, ML). The accuracy of screw placement was evaluated by postoperative pelvic CT scan. Results:All of the virtual S 2AI screw trajectories can be reconstructed. The screw trajectory parameters were shown as follows: SA was 30.20°±21.94° and 50.94°±16.02° on the high and low sides of pelvis, respectively, and the difference was statistically significant ( t=3.990 , P<0.001). SA was 30.14°±21.93° on the anterior side of the pelvis and 51.00°±15.96° on the posterior side, respectively, with statistical significance ( t=4.027, P<0.001). TA was 43.67°±12.86° on the high side of pelvic tilt and 31.95°±13.80° on the low side, with statistical significance ( t=2.834, P=0.009). TA was 42.56°±12.52° on the anterior side of the pelvis and 33.05°±14.94° on the posterior side, respectively, and the differences were statistically significant ( t=2.192, P=0.037). ML was 97.12±12.44 mm and 92.28±11.04 mm on the high and low side of pelvis, and there was no significant difference ( t=0.963 , P=0.060). ML was 97.72±12.41 mm on the anterior sides of the pelvis and 91.68±10.57 mm on the posterior side, and the difference was statistically significant ( t=2.556 , P=0.017). SA tended to be smaller on the high side of pelvic tilt ( r=0.474, P<0.01) and TA tended to be higher on the anterior side of pelvis ( r=-0.419, P<0.01) . Only 2 screws (3.6%) showed screw breaches after surgery, with no clinically notable neurovascular or visceral complications. Conclusion:In patients of neuromuscular scoliosis with severe pelvic obliquity, the virtual S 2AI screw trajectory can be found in 3D CT reconstruction of the pelvis. But the parameters are very discrete at SA and TA. In these patients, the O-arm 3D CT navigation can be used to make sure the direction and length of the S 2AI screw, greatly improving the accuracy of screw placement and effectively descending the ratio of poor screw.

Objective:To investigate how pelvic incidence (PI) would change during the follow-up in degenerative scoliosis (DS) patients who underwent second sacralalar-iliac (S 2AI) fixation and identify the possible factors associated with the changes in PI. Methods:The DS patients who underwent long fusion to pelvis with S 2AI fixation with a minimum follow-up of two years between November 2014 to January 2017 were retrospectively reviewed in this study. The following sagittal radiographic parameters were measured, including pelvic incidence (PI), lumbar lordosis (LL), pelvic tilt (PT), PI minus LL (PI-LL), and sagittal vertical axis (SVA) at pre-operation, post-operation and 2-year follow-up. Patients were divided into two groups at immediate post-operation: PI decreased less than 5° or increased (Group PI stabilization); PI decreased larger than 5° (Group PI activity). Descriptive statistics were calculated for all patients in the form of mean value and standard deviation (SD). Comparisons of means between variables were performed using an unpaired Student's t test. Pearson correlation coefficienttest was performed to determine the correlations between all radiographic variables. Inter- and intra-observer reliability was assessed using intraclass correlation coefficient (ICC). The internal consistency of the measurements was characterized as excellent ( ICC≥0.9), good (0.7≤ ICC<0.9), acceptable (0.6< ICC≤0.7), poor (0.5≤ ICC<0.6), or unpredictable ( ICC<0.5). Results:There were no significant differences in terms of age, sex, radiographic measurements and scores of SRS-22 between twogroups preoperatively ( P>0.05). 80 DS patients with a mean age of 55.3±16.2 years were enrolled in this study with a mean follow-up period of 34.6±8.7 months. At post-operation, 39 patients (38.8%) were in group PI stabilization whose PI decreased from 45.7°±11.4° to 45.3°±11.2° with no significant difference; while the other 41 (61.2%) were in group PI activity whose PI significantly decreased from 51.6°±14.5° to 40.9°±14.0°. At the last follow-up, 24 patients (49%) in group PI activity had PI returned with an increase of larger than 5°; while the other 25 (51%) showed no increase with a mean ΔPI change of -4.2°. Subgroup comparison revealed that ΔPI, post-operation PI, post-operation PT and age were significantly different between the two subgroups. Pre-operation PI, post-operation PI, post-operation PT, post-operation PI-LL were significantly correlated with ΔPI at the last follow-up. Logistic regression analysis showed that post-operation PI was the associated factor ( OR=0.87, P=0.024). Conclusion:PI decreased in more than half of DS patients after spinal surgery using S 2AI screws, while returned among 48% of them during 2-year follow-up. Lower pre-operation PI, post-operation PI and PT were strongly associated with the return of PI.