Determining the vertebra for pedicle subtraction osteotomy in surgical correction for ankylosing spondylitis with thoracolumbar kyphosis

Objective To compare the effects of Smith-Petersen osteotomies (SPOs) and pedicle subtraction osteotomy (PSO) on sagittal alignment in patients with thoracolumbar kyphosis secondary to anky losing spondylitis (AS).Methods The data of the 39 patients with ankylosing spondylitis of thoracic and lumbar kyphosis were retrospectively reviewed,who received SPOs or PSO orthopedic internal fixation from August 2000 to June 2010.According to the types of osteotomies,the 39 patients were categorized into 2groups.Of them,15 patients including 13 males and 2 females comprised the SPOs group,whose average age was 28.1±7.1 years,ranging from 18 to 42,whereas the other 24 patients including 20 males and 4 females comprised the PSO group,whose average age was 38.3±7.9 years,ranging from 21 to 53.Radiographic parameters were measured and compared between the two groups including thoracic kyphosis (TK),lumbar lordosis (LL),global kyphosis (GK),sagittal imbalance (SVA),pelvic incidence (PI),sacral slope (SS),and pelvic tilting (PT) before the operation,3 months after the operation,and at the last follow-up,respectively.Results Significant differences were found in terms of the improvement of LL,PT,SS,SVA and GK except that of TK and PI.Preoperatively,no significant difference was found between the two groups in TK,SVA,PT and SS except PI,GK and LL.Postoperatively,significant differences were found between the two groups in LL,GK,PT and SS.Significant differences were found between the two groups in the mean loss of correction of GK,LL,SVA,PT and SS at the last follow-up.Conclusion For patients with ankylosing spondylitis of thoracic and lumbar kyphosis,satisfactory reconstruction of spino-pelvic alignment can be obtained by PSO,whereas loss of correction occurred frequently in SPOs. Key words: Spondylitis, ankylosing; Kyphosis; Osteotomy

To evaluate whether acetabular orientation (abduction and anteversion) can be restored by lumbar pedicle subtraction osteotomy (PSO) in ankylosing spondylitis (AS) patients with thoracolumbar kyphosis. A total of 33 consecutive AS patients with thoracolumbar kyphosis undergoing one-level lumbar PSO were retrospectively reviewed. Radiographical measurements included sagittal vertical axis, global kyphosis, thoracic kyphosis, local kyphosis, lumbar lordosis, pelvic incidence, sacral slope, and pelvic tilt. Acetabular abduction and anteversion were measured on CT scans of the pelvis before and after lumbar PSO. The preoperative and postoperative parameters were compared by the paired samples t test. Pearson's correlation analysis was conducted to determine the correlations between the changes in acetabular abduction and anteversion and the changes in sagittal spinopelvic parameters. After lumbar PSO, sagittal vertical axis, global kyphosis, and pelvic tilt were corrected from 15.7±6.7cm, 66.8°±17.5°, and 38.6°±9.0° to 2.9±4.9cm, 21.3°±8.2°, and 23.2°±8.2°, respectively (p<0.001). Of note, acetabular abduction and anteversion decreased from 59.6°±4.6° to 31.4°±6.5° before surgery to 51.4°±6.5° and 20.2°±4.4° after surgery, respectively (p<0.001). Moreover, the changes in acetabular abduction and anteversion were observed significantly correlated with the change in pelvic tilt (r=0.527, p=0.002; r=0.586, p<0.001). Abnormal acetabular abduction and anteversion could be corrected by lumbar PSO in AS patients with thoracolumbar kyphosis. Consequently, a relatively normal acetabular orientation could be achieved after lumbar PSO, which might decrease the potential risk of dislocation in AS patients with spine and hip deformities requiring subsequent THR surgery.

Is the pelvic incidence a determinant factor for kyphosis curve patterns of ankylosing spondylitis patients?

Does the preoperative lumbar sagittal profile affect the selection of osteotomy level in pedicle subtraction osteotomy for thoracolumbar kyphosis secondary to ankylosing spondylitis?

A retrospective study. The aim of this study was to make a thorough comparison of clinical and radiographic outcomes between ankylosing spondylitis (AS) patients with severe kyphosis who underwent one- or two-level pedicle subtraction osteotomy (PSO) and to determine the indications of one-level PSO. Traditionally, one-level PSO was considered being able to obtain 35° to 40° correction. However, in our practice, one-level PSO might achieve satisfied clinical and radiographic outcomes in AS patients with severe thoracolumbar kyphosis defined as global kyphosis (GK) ≥80°. Fifty-five AS-related severe thoracolumbar kyphosis patients undergoing one- or two-level PSO from January 2007 to November 2016 were reviewed. The radiographic parameters included thoracic kyphosis (TK), lumbar lordosis (LL), GK, pelvic tilt (PT), sacral slope (SS), pelvic incidence (PI), sagittal vertical axis (SVA), and femoral obliquity angle (FOA). Clinical outcomes were evaluated by Oswestry Disability Index (ODI) and Visual Analogue Scale (VAS). The mean follow-up period was 39.7 ± 20.2 months (range, 24-120 months). Patients who underwent one-level PSO have significantly smaller preoperative GK, SVA, FOA, and larger preoperative LL and SS compared to those who underwent two-level PSO (P < 0.05). The optimal cutoff points of preoperative radiographic parameters for selecting one-level PSO were: GK <94°, SVA <18.0 cm, and LL <18°. No significant difference was observed between the two groups with regard to preoperative ODI and VAS (P > 0.05), and the improvement of ODI and VAS (P > 0.05). Significantly more operative time, blood loss, and fusion levels were found in two-level PSO group (P < 0.05). One-level PSO might be appropriate for selected severe AS-related kyphosis patients with GK <94°, SVA <18.0 cm, and LL <18°. This finding might be beneficial for surgical decision-making in performing one-level PSO, a relatively less risky procedure, to reconstruct the ideal sagittal alignment in AS patients with severe thoracolumbar kyphosis.Level of Evidence: 2.

What is the optimal postoperative sagittal alignment in ankylosing spondylitis patients with thoracolumbar kyphosis following one-level pedicle subtraction osteotomy?

This article is a comparative study. The aim of the study is to investigate the difference of sagittal alignment of the pelvis and spine between patients with thoracolumbar kyphosis secondary to ankylosing spondylitis (AS) and thoracolumbar fracture, and to evaluate the role of sacropelvic component in AS patients’ adaption to the changes in sagittal alignment.Advanced stages of AS are often associated with thoracolumbar kyphosis, resulting in an abnormal spinopelvic balance and pelvic morphology, whereas thoracolumbar fractures may lead to major kyphosis with a potential compromise of the spinal canal, which can cause an abnormal spinopelvic balance. Until now, the comparison of that sagittal alignment between AS and thoracolumbar fracture is not found in the literature.This study included 30 cases of AS and 30 cases of thoracolumbar fracture. Sagittal spinal and pelvic parameters were measured from the standing lateral radiograph, and the following 11 radiological parameters were measured, including global kyphosis (GK), thoracic kyphosis (TK), C7 tilt (C7T), sagittal vertical axis (SVA), spino-pelvic angle (SSA), lumbar lordosis (LL), upper arc of lumbar lordosis (ULL), lower arc of lumbar lordosis (LLL), pelvic incidence (PI), sacrum slope (SS), pelvic tilt (PT), and T9 tilt (T9T). Analysis of variance was used in the comparison of each dependent variable between the 2 cohorts. The relationship between sagittal spinal alignment and pelvic morphology of AS patients was determined via Pearson correlation coefficient (r).Compared with the thoracolumbar fracture group, AS patients had significantly lower C7T, SSA, LL, LLL and SS (78.3° ± 9.3° vs 88.0° ± 2.7°, P < 0.001 for C7T; 91.6° ± 22.7° vs 119.1° ± 9.0°, P < 0.001 for SSA; 20.7° ± 21.0° vs 36.3° ± 16.8°, P = 0.001 for LL; 18.1° ± 11.9° vs 29.0° ± 9.7°, P < 0.001 for LLL; and 18.1° ± 11.9° vs 29.0° ± 9.7°, P < 0.001 for SS), whereas in terms of SVA and PT, AS patients had an obviously higher value than those of thoracolumbar fracture patients (94.5mm ± 58.4 mm vs 8.0mm ± 23.3 mm, P < 0.001 for SVA; and 26.5° ± 10.3° vs 17.5° ± 6.6°, P < 0.001 for PT). In AS patients, SS were found to be significantly correlated with SVA, SSA, and LL (r = −0.312, P < 0.05 for SVA; r = 0.475, P < 0.05 for SSA; r = 0.809, P < 0.001 for LL).In our study, there were significant differences in sagittal alignment of the pelvis and spine between patients with AS and thoracolumbar fracture, and changes in pelvic morphology compensated more in AS patients for a thoracolumbar kyphosis. These findings may be helpful for better understanding of sagittal alignment in patients with thoracolumbar kyphosis secondary to AS.

The relationship between structural damage and inflammation of the spine and the sagittal imbalance in ankylosing spondylitis (AS) is not well understood. The present study aimed to investigate the correlation between structural damage and inflammation of the lumbar spine and the sagittal imbalance in AS patients with thoracolumbar kyphosis. Forty-five AS patients with thoracolumbar kyphosis were retrospectively reviewed. Six sagittal spinal parameters, including the C7 tilt (C7T), spino-sacral angle (SSA), global kyphosis (GK), the sagittal vertical axis (SVA), thoracic kyphosis (TK), and lumbar lordosis (LL), were measured. Structural damage of the lumbar spine was assessed by the modified Stoke AS Spine Score (mSASSS) on radiographs. Lumbar spinal inflammation was evaluated by the AS spinal magnetic resonance imaging (MRI) activity (ASspiMRI-a) on MRI. Correlation analysis was performed using the paired sample t-test. Multivariable linear regression models were constructed to analyze the contributions of mSASSS and ASspiMRI-a to the sagittal parameters. The average values of the sagittal parameters C7T, SSA, GK, SVA, TK, and LL were 68.1°, 80.1°, 77.3°, 168.7 mm, 47.7°, and -0.7°, respectively. The average mSASSS and ASspiMRI-a scores were 9.8 and 10.8, respectively. Correlation analysis showed that the mSASSS and ASspiMRI-a were correlated with C7T, SSA, SVA, and LL (the Spearman correlation coefficients were -0.439, -0.390, 0.424, and 0.530 for mSASSS; -0.406, -0.402, 0.378, and 0.486 for ASspiMRI-a; P<0.05). The C7T, SSA, and SVA were significantly correlated with LL (r=-0.696, -0.779, and 0.633, respectively; P<0.05). There was a weak correlation between the mSASSS and ASspiMRI-a (β=0.299, P=0.046). The multivariable regression models indicated that the sagittal imbalance was determined to a greater extent by the mSASSS than ASspiMRI-a (the β values were -1.550 vs. -0.649 for C7T, -1.865 vs. -1.231 for SSA, 9.161 vs. 3.823 for SVA, and 3.128 vs. 1.717 for LL). Both structural damage and inflammation of the lumbar spine contributed to the sagittal imbalance in AS patients with thoracolumbar kyphosis. In the late stages of AS, the sagittal imbalance was more attributable to the structural damage than the inflammation of the lumbar spine.

Spontaneous Remodeling of Spinal Canal After Sagittal Translation in Pedicle Subtraction Osteotomy for Correction of Thoracolumbar Kyphosis in Ankylosing Spondylitis

Does solid fusion eliminate rod fracture after pedicle subtraction osteotomy in ankylosing spondylitis-related thoracolumbar kyphosis?

Retrospective comparison of database patients. To evaluate the difference of loss of correction between Smith-Petersen osteotomies (SPOs) and pedicle subtraction osteotomy (PSO) in patients with thoracolumbar kyphosis secondary to ankylosing spondylitis (AS). SPOs and PSO are reported to be the 2 major techniques for correction of thoracolumbar kyphosis resulting from AS. Previous studies have tried to compare the indication, technical aspects, correction obtained, and complication rates between the aforementioned 2 techniques. However, reports addressing a comparison of loss of correction between SPOs and PSO are limited. On the basis of the types of osteotomies, 50 patients were divided into 2 groups: (1) SPOs group (n=19) including 16 male and 3 female patients, with an age range from 21 to 40 years (mean 27 y). The preoperative global kyphosis (GK) ranged from 41 to 99 degrees (average 64.6±25.6 degrees); (2) PSO group (n=31) consisted of 26 male and 5 female patients, with an age range from 22 to 54 years (mean 36 y). The preoperative GK was 50 to 96 degrees (average 73.7±23.6 degrees). Radiographic parameters including sagittal vertical axis, T12-S1 lordosis, GK, and angle of fusion levels were measured. Both groups showed similar preoperative and postoperative thoracic kyphosis, lumbar lordosis, and sagittal vertical axis. The average GK was corrected to 25.5 degrees and 31.4 degrees in SPOs group and PSO group, respectively. All cases were followed for a minimum of 2 years. At the last follow-up, mean loss of correction in the fusion levels were 6.1 degrees in SPOs group and 1.3 degrees in PSO group, respectively. The difference was statistically significant (P=0.034). Loss of correction of >5 degrees occurred in 4 cases (21.1%) in SPOs group, and 5 cases (16.1%) in PSO group. Both SPOs and PSO showed similar effect in correcting the thoracolumbar kyphosis secondary to AS. However, patients treated with the SPOs technique showed higher risk in loss of correction in the instrumented region.

A retrospective radiographical study. To identify the radiographical predictors for sagittal imbalance in patients with thoracolumbar kyphosis secondary to ankylosing spondylitis (AS) after 1-level lumbar pedicle subtraction osteotomy (PSO). Few studies had correlated the preoperative sagittal parameters with postoperative sagittal alignments to determine the radiographical predictors for postoperative sagittal imbalance in patients with AS after 1-level lumbar PSO. Thirty-six patients with thoracolumbar kyphosis secondary to AS who underwent 1-level lumbar PSO were recruited with a minimal follow-up of 24 months (mean = 27.4 mo; range, 24-53 mo). Correlation analysis and subsequent stepwise multiple regression analysis were used to evaluate the correlations between preoperative parameters, including global kyphosis, local kyphosis, thoracic kyphosis, thoracolumbar Cobb angle, lumbar lordosis, pelvic incidence (PI), pelvic tilt, sacral slope, and sagittal vertical axis (SVA), as well as SVA at the last follow-up. All these patients were further divided into 2 groups according to the PI value (group A: PI >50°; group B: PI ≤50°). The correction outcomes were compared between these 2 groups. The preoperative SVA was not significantly different between group A and group B (157.6 mm vs. 124.5 mm; P> 0.05), and both groups had similar magnitudes of kyphosis corrections at the last follow-up (global kyphosis: 42.9° vs. 46.1°; local kyphosis: 42.7° vs. 40.5°; lumbar lordosis: 35.7° vs. 43.0°). However, group A patients had significantly larger SVA at the last follow-up (73.2 mm vs. 28.7 mm; P< 0.05) and a higher incidence of postoperative sagittal imbalance (77.8% vs. 25.9%; P< 0.05) than those in group B. The stepwise multiple regression analysis demonstrated that both preoperative SVA and PI were significant independent predictors of postoperative sagittal alignments, which explained 52.0% and 9.7% of the variability of SVA at the last follow-up, respectively. Patients with AS with either larger preoperative SVA or larger PI are more likely to experience failed sagittal realignments after 1-level lumbar PSO. For these patients, additional osteotomies may be recommended for satisfactory correction outcomes. 4.

Objective To investigate anterior longitudinal ligaments (ALL) ossified surrounding osteotomy vertebra impact the lordosing effect of pedicle subtraction osteotomy (PSO) in patients with thoracolumbar kyphosis secondary to ankylosing spondylitis (AS). Methods We retrospectively reviewed 102 AS patients with thoracolumbar kyphosis treated with single-level PSO at our institution from September 2007 to August 2015. There were 92 male and 10 female. The average age was (35.6±11.8) years old (range from 17 to 65 years old). Patients were stratified into ossified group (54 cases) and non-ossified group (48 cases) based on the presence of ALL ossification adjacent to osteotomy vertebra. Compared the contribution of adjacent disc wedging to total correction of each PSO segment between the ossified and non-ossified groups. The long-term correction loss of spine and pelvic sagittal morphology were also evaluated and compared between the 2 groups. Results Patients in the ossified group accomplished significantly lower amount of correction in single level segment of PSO (36.3°±6.9° vs. 41.5°±6.9°), and there was significant difference between the two groups. The contribution of adjacent disc wedging to total correction of PSO was significantly larger in the non-ossified group (22.9% vs. 7.8%, P<0.001). For subgroups with a minimum 2 year follow-up, loss of corrections concerned sagittal vertical axis (SVA), which was (1.7±4.5) cm vs. (-0.2±4.0) cm in ossified group and non-ossified group, and there was significant difference between the two groups. Pelvic tilt (PT) was 3.5°±8.2° vs. 2.0°±10.4°, lumbar lordosis (LL) was -7.9°±11.9° vs. -0.1°±11.9° and sacral slope (SS) was 4.5°±9.3° vs. 1.6°±7.9°, and there were all significant differences between the two groups. The change of adjacent disc wedging angle was marginally higher in the unossified group (-2.1°±6.2° vs. -0.1°±3.7°, P=0.09), but there was no significant difference between the two groups. No significant correction loss of osteotomy angle was observed in both groups. Conclusion Osteotomy vertebrae accompanied by unossified adjacent ALL in PSO of AS were prone to create more disc-originated lordosing effect immediately after surgery. However, a correction loss might occur more commonly during a long term follow-up. Key words: Vertebroplasty; Spondylitis, ankylosing; Kyphosis; Orthopedic procedures

Both unchanged upper cervical lordosis combined with decreased lower cervical lordosis and decreased upper cervical lordosis combined with decreased lower cervical lordosis have been reported to occur after correction surgery for adult spinal deformity. However, variations in cervical alignment after correction surgery in patients with ankylosing spondylitis (AS) have not been investigated. The current study aimed to investigate the variations in cervical alignment following the correction surgery in AS patients with thoracolumbar kyphosis. Patients with AS who underwent pedicle subtraction osteotomy (PSO) for thoracolumbar kyphosis from June 2016 to June 2019 with a minimum of 1-year follow-up were reviewed. Patients were grouped according to the presence (ossified group) and absence (non-ossified group) of total ossification of the anterior longitudinal ligament (ALL) in the lower cervical spine. Radiographic parameters, including thoracolumbar, craniocervical, and global radiographic parameters, were measured on lateral sitting EOS images. Thirty-two patients (27 males and 5 females) with a mean follow-up of 1.5 years were identified. There were 21 patients in the non-ossified group and 11 patients in the ossified group. After PSO, both groups showed a decrease in the occiput-C7 angle (p < 0.001 for both). In the non-ossified group, the C2-7 angle decreased significantly (p < 0.001), while the occiput-C2 angle remained unchanged (p = 0.570). In the ossified group, the occiput-C2 angle decreased significantly (p < 0.001), while C2-7 angle remained unchanged (p = 0.311). In addition, the change in occiput-C2 was correlated with the osteotomy angle in the ossified group (R = 0.776, p = 0.005). The variation patterns of cervical alignment following correction surgery for AS-related thoracolumbar kyphosis were different based on patients with or without total ossification of ALL in the lower cervical spine. When planning PSO for patients in the ossified group, restoration of the physiological upper cervical lordosis angle could be achieved by adjusting the osteotomy angle.

The aim of this study was to investigate the factors affecting postoperative quality of life in patients with ankylosing spondylitis (AS) and thoracolumbar kyphosis (TLK), and establish a personalized sagittal reconstruction strategy. Patients with AS and TLK who underwent pedicle subtraction osteotomy (PSO) from February 2009 to May 2019 were retrospectively included. Quality of life and spinal sagittal radiographic parameters were collected before surgery and at the last follow-up. Patients were divided into two groups based on the attainment of minimal clinically important difference (MCID) on the Bath Ankylosing Spondylitis Functional Index and Oswestry Disability Index. Comparisons of radiographic parameters and clinical outcomes were conducted between and within groups. Regression analysis was used to identify the risk factors within the missing MCID cohort. Sagittal reconstruction equations were established using the pelvic incidence (PI) and thoracic inlet angle (TIA) in the reached MCID cohort. The study comprised 82 participants. Significant improvements were observed in most radiographic parameters and all quality-of-life indicators during the final follow-up compared with the preoperative measures (p < 0.05). Factors including cervical lordosis (CL) ≥ 18° (OR 9.75, 95% CI 2.26-58.01, p = 0.005), chin-brow vertical angle (CBVA) ≥ 25° (OR 14.7, 95% CI 3.29-91.21, p = 0.001), and pelvic tilt (PT) ≥ 33° (OR 21.77, 95% CI 5.92-103.44, p < 0.001) independently correlated with a failure to attain MCID (p < 0.05). Sagittal realignment targets were constructed as follows: sacral slope (SS) = 0.84 PI - 17.4° (R2 = 0.81, p < 0.001), thoracic kyphosis (TK) = 0.51 PI + 10.8° (R2 = 0.46, p = 0.002), neck tilt (NT) = 0.52 TIA - 5.8° (R2 = 0.49, p < 0.001), and T1 slope (T1S) = 0.48 TIA + 5.8° (R2 = 0.45, p = 0.002). PSO proved efficacious in treating AS complicated by TLK, yielding favorable outcomes. CBVA ≥ 25°, CL ≥ 18°, and PT ≥ 33° were the primary factors affecting postoperative quality of life in patients with AS. The personalized sagittal reconstruction strategy in this study focused on the subjective sensations and daily needs of patients with AS, which were delineated by the equations SS = 0.84 PI - 17.4°, TK = 0.51 PI + 10.8°, NT = 0.52 TIA - 5.8°, and T1S = 0.48 TIA + 5.8°.