A Pilot Study on Posterior Polyethylene Tethers to Prevent Proximal Junctional Kyphosis After Multilevel Spinal Instrumentation for Adult Spinal Deformity

A Pilot Study on Posterior Polyethylene Tethers to Prevent Proximal Junctional Kyphosis After... Abstract BACKGROUND Proximal junctional kyphosis (PJK) is a common problem after multilevel spine instrumentation. OBJECTIVE To determine if junctional tethers reduce PJK after multilevel instrumented surgery for adult spinal deformity (ASD). METHODS ASD patients who underwent posterior instrumented fusion were divided into 3 groups: no tether (NT), polyethylene tether-only (TO; tied securely through the spinous processes of the uppermost instrumented vertebra [UIV] + 1 and UIV-1), and tether with crosslink (TC; passed through the spinous process of UIV+1 and tied to a crosslink between UIV-1 and UIV-2). PJK was defined as proximal junctional angle ≥ 10° and ≥ 10° greater than the corresponding preoperative measurement. RESULTS One hundred eighty-four (96%) of 191 consecutive patients achieved minimum 3-mo follow-up (mean = 20 mo [range:3-56 mo]; mean age = 66 yr; 67.4% female). There were no significant differences between groups based on demographic, surgical, and sagittal radiographic parameters. PJK rates were 45.3% (29/64), 34.4% (22/64), and 17.9% (10/56) for NT, TO, and TC, respectively. PJK rate for all tethered patients (TO + TC; 26.7% [32/120]) was significantly lower than NT (P = .011). PJK rate for TC was significantly lower than NT (P = .001). Kaplan-Meier analysis showed significant time-dependent PJK reduction for TC vs NT (log rank test, P = .010). Older age and greater change in lumbar lordosis were independent predictors of PJK, while junctional tethers had a significant protective effect. CONCLUSION Junctional tethers significantly reduced occurrence of PJK. This difference was progressive from NT to TO to TC, but only reached pairwise significance for NT vs TC. This suggests potential benefit of tethers to reduce PJK, and that future prospective studies are warranted. Adjacent segment disease, Adult spinal deformity, Complication, Proximal junctional angle, Proximal junctional kyphosis, Scoliosis, Spinal fusion ABBREVIATIONS ABBREVIATIONS ANOVA analysis of variance ASD adult spinal deformity BMI body mass index CI confidence interval LL lumbar lordosis NT no tether OR odds ratio PJA proximal junctional angle PJF proximal junctional failure PJK proximal junctional kyphosis PI pelvic incidence PT pelvic tilt SPSS Statistical Package for Social Science SVA sagittal vertical axis TC tether with crosslink TO tether-only UIV uppermost instrumented vertebra Proximal junctional kyphosis (PJK) continues to be problematic after multilevel spinal instrumentation for adult spinal deformity (ASD). The reported incidence ranges from 11.0% to 52.9%.1-8 Several risk factors have been identified and include high body mass index (BMI), older age at surgery, fusion to sacrum, low bone mineral density, larger preoperative sagittal vertical axis (SVA), postoperative proximal junctional angle (PJA) >5° and thoracic kyphosis >30°, and greater correction of lumbar lordosis.4,7,9,10 Early reports of PJK were primarily focused on adolescent idiopathic scoliosis.11 Authors used varying radiographic criteria, with abnormal PJA ranging from 5° to 15°.11-15 More recently, for ASD, definitions of abnormal PJK have been more consistent.1 PJA is measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). Abnormal PJK is defined as a proximal junction sagittal Cobb angle that is ≥10° and ≥10° greater than the corresponding preoperative measurement.1 Although there is increasing emphasis on PJK detection, as well as various classification systems to guide its management,16 the question remains whether PJK is a true complication, especially in the absence of clinical symptoms.5 Glattes and colleagues have reported no significant difference in outcome scores (Scoliosis Research Society-24 outcomes measure) between PJK and non-PJK groups.1,2 However, a subset of patients with PJK may still require surgical revision; therefore, various preventative techniques have been proposed and include vertebral augmentation, multilevel stabilization screws, transverse process hooks at the UIV, proximal transition rods of reduced diameter, and hybrid constructs.3,17,18 The purpose of this study was to determine if junctional tethers reduce the incidence of postoperative PJK after multilevel instrumented surgery for ASD. METHODS We performed a single-center, retrospective evaluation of a prospectively maintained database of patients who underwent multilevel spine instrumentation at the University of Virginia Health System (Charlottesville, Virginia). All operations were performed by the 2 senior authors (CIS, JSS) between December 31, 2011 and June 2, 2016, and the database was reviewed on December 2, 2016. The decision to use junctional tethers was not randomized; instead, CIS and JSS started utilizing junctional tethers from September 2013 and January 2015 onwards, respectively. No other technique such as vertebral augmentation, multilevel stabilization screws, transitional rods, or fusion with hybrid constructs was used that might confound PJK rates.3,17,18 The study was approved by the University of Virginia Institutional Review Board. Patient consent was waived since data collection and analysis was retrospective and all patient-specific identifiers were removed for publication. We searched for ASD patients with diagnosis of scoliosis and/or global sagittal malalignment (ie, abnormal SVA, thoracic kyphosis, lumbar lordosis, and/or pelvic incidence to lumbar lordosis mismatch). Inclusion criteria for the study were patient age > 18 yr, deformity correction with instrumented segmental posterior spine fusion (may have also had anterior approach procedure) at a minimum > 6 motion segments, thoracic UIV, pedicle screw instrumentation without transitional rods or hooks at the UIV, and complete radiographic data with preoperative, postoperative, and final standing long-cassette films at minimum follow-up of 3 mo. We excluded patients with preoperative diagnosis of (1) degenerative spine disease without significant scoliosis or global sagittal malalignment, and/or (2) vertebral osteomyelitis/discitis. Patients who underwent a revision spine surgery without change in UIV were also excluded from the study. Radiographic Measurements Subjects had standing posteroanterior and lateral long-cassette (14 × 36 inch) radiographs preoperatively, postoperatively, and at final follow-up. PJA was the sagittal Cobb angle measured from the caudal endplate of the UIV to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV) (Figure 1). Abnormal PJK was defined as a PJA that was ≥ 10° and ≥ 10° greater than the corresponding preoperative measurement.1 Figure 2 depicts measurement technique for pelvic tilt (PT), pelvic incidence (PI), and lumbar lordosis (LL). FIGURE 1. View largeDownload slide A, B, Preoperative and postoperative standing long-cassette lateral radiographs of a 73-yr-old female with spinal deformity show the proximal junctional angle (PJA) measuring 16.2° and 29.3°, respectively. PJA was the sagittal Cobb angle measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). For this study, abnormal proximal junctional kyphosis (PJK) was defined as a PJA (1) greater than or equal to 10° and (2) at least 10° greater than the corresponding preoperative measurement. The presence of both criteria was necessary to be considered abnormal. B, This patient developed postoperative PJK. FIGURE 1. View largeDownload slide A, B, Preoperative and postoperative standing long-cassette lateral radiographs of a 73-yr-old female with spinal deformity show the proximal junctional angle (PJA) measuring 16.2° and 29.3°, respectively. PJA was the sagittal Cobb angle measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). For this study, abnormal proximal junctional kyphosis (PJK) was defined as a PJA (1) greater than or equal to 10° and (2) at least 10° greater than the corresponding preoperative measurement. The presence of both criteria was necessary to be considered abnormal. B, This patient developed postoperative PJK. FIGURE 2. View largeDownload slide Pelvic tilt is the angle between the vertical line from the center of the femoral head and the line from the center of the femoral head to the center of the sacral endplate. Pelvic incidence is the angle between the line perpendicular to the sacral endplate and the line connecting the midpoint of the sacral endplate to the center of the femoral head. Lumbar lordosis is the Cobb angle between the inferior endplate of T12 and superior endplate of the sacrum. FIGURE 2. View largeDownload slide Pelvic tilt is the angle between the vertical line from the center of the femoral head and the line from the center of the femoral head to the center of the sacral endplate. Pelvic incidence is the angle between the line perpendicular to the sacral endplate and the line connecting the midpoint of the sacral endplate to the center of the femoral head. Lumbar lordosis is the Cobb angle between the inferior endplate of T12 and superior endplate of the sacrum. Tethering Technique Our proximal junction posterior tether consisted of a 5-mm woven polyethylene Mersilene tape (Ethicon, Somerville, New Jersey) on a blunt needle, and in 56 patients, this was anchored to a standard crosslink. The tape-only technique first involves using a high-speed drill to create holes in the base of the UIV+1 and UIV-1 spinous processes. Then polyethylene tape is passed through these and tightened securely. An assistant holds the first knot in the polyethylene tape while the surgeon completes tying the tape for maximal tension. Our tape-crosslink technique is similar, but first involves placement of a standard crosslink between the UIV-1 and UIV-2 spinous processes. Again, using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. Polyethylene tape is then passed around the crosslink and through the UIV+1 spinous process, and tied securely. Finally, the crosslink is distracted inferiorly to fully tension the polyethylene tape for junctional support (Figure 3). FIGURE 3. View largeDownload slide The polyethylene tape-crosslink tethering technique first involves placement of a standard crosslink between the base of UIV-1 and UIV-2 spinous processes. Using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. A, Next, a 5-mm woven polyethylene tape is passed around the crosslink and through the UIV+1 spinous process, and B, then tightened securely (the black line traces the polyethylene tether). C, The single arrow and double arrows indicate the superior and inferior attachments of the polyethylene tape. D, E, Finally, the crosslink is distracted inferiorly to fully tension the tape for junctional support. F, The final result is aimed to reduce the risk of developing PJK. FIGURE 3. View largeDownload slide The polyethylene tape-crosslink tethering technique first involves placement of a standard crosslink between the base of UIV-1 and UIV-2 spinous processes. Using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. A, Next, a 5-mm woven polyethylene tape is passed around the crosslink and through the UIV+1 spinous process, and B, then tightened securely (the black line traces the polyethylene tether). C, The single arrow and double arrows indicate the superior and inferior attachments of the polyethylene tape. D, E, Finally, the crosslink is distracted inferiorly to fully tension the tape for junctional support. F, The final result is aimed to reduce the risk of developing PJK. Statistical Analysis Patients were divided into 3 groups based on tethering technique, which include: (1) no tether (NT), (2) polyethylene tether-only (TO), and (3) polyethylene tether with crosslink (TC). Data are presented as mean and standard deviation for continuous variables, and as frequency for categorical variables. Baseline characteristics of the cohorts, including patient demographics, surgical data, and sagittal plane radiographic parameters, were compared using Pearson's Chi square test, Fisher's exact test, and 1-way analysis of variance (ANOVA). Rates of PJK and revision surgery for proximal junctional failure (PJF), time-to-PJK, change in postoperative PJA, and total radiographic follow-up duration were analyzed using Pearson's Chi square test, Fisher's exact test, and 1-way ANOVA followed by Bonferroni post hoc test, when appropriate. Odds ratios (ORs) were calculated from 2 × 2 contingency tables. A Kaplan-Meier survivorship analysis was performed to analyze PJK as a time-dependent variable. After dichotomizing patients based on development of PJK at 6 mo postop, independent variables were entered in a stepwise, multivariable model using binary logistic analysis. The fit of the final model was assessed using the Hosmer-Lemeshow goodness-of-fit test.19 For time-dependent regression, a Cox proportional hazards model was implemented. For this study, P-values < .05 were considered statistically significant, and a post hoc Bonferroni adjustment was made for multiple pairwise comparisons. Statistical Package for Social Science (SPSS) version 24.0 (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. IBM Corp, Armonk, New York) was used to perform all analyses. RESULTS Patient Demographics and Surgical Data One hundred eighty-four (96%) of 191 consecutive patients, between December 31, 2011 and June 2, 2016, met inclusion criteria and achieved minimum 3-mo follow-up (mean = 20 mo [range: 3-56 mo]; mean age = 66 yr; 67.4% female). Patient demographics and surgical data are presented in Table 1. There were 64 NT, 64 TO, and 56 TC patients. For these 3 cohorts, there were no significant differences in patient age at surgery, gender, preoperative BMI, number of instrumented vertebrae, use of pelvic fixation, use of 3-column osteotomy, or type of approach (posterior-only vs 2-stage anterior-posterior). The majority of cases involved at least 10 motion segments with pelvic fixation, did not have 3-column osteotomies (vertebral column resection and/or pedicle subtraction osteotomy), and were posterior-only approaches. A total of 12 patients were treated via a 2-stage, combined anterior-posterior approach. No intraoperative or postoperative complication in this study could be attributed to posterior polyethylene tethers. TABLE 1. Patient Demographics and Surgical Data Separated by Tethering Techniquea No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 aValues are expressed as the mean ± standard deviation or frequency for continuous and categorical variables, respectively. bComparison of the 3 tethering groups analyzed with Pearson's chi square test, Fisher's exact test, or one-way analysis of variance. View Large TABLE 1. Patient Demographics and Surgical Data Separated by Tethering Techniquea No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 aValues are expressed as the mean ± standard deviation or frequency for continuous and categorical variables, respectively. bComparison of the 3 tethering groups analyzed with Pearson's chi square test, Fisher's exact test, or one-way analysis of variance. View Large Sagittal Plane Radiographic Parameters Sagittal plane radiographic parameters are presented in Table 2. There were no significant differences in the pre- and postoperative SVA, PT, PI, LL, PI-LL mismatch, and thoracic kyphosis among cohorts. The amount of deformity correction, measured by the postoperative change in SVA and LL, was comparable among cohorts. TABLE 2. Sagittal Plane Radiographic Parametersa No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 PI = pelvic incidence, LL = lumbar lordosis. aValues are expressed as the mean ± SD for continuous variables. bComparison of the 3 tethering groups analyzed with one-way analysis of variance. View Large TABLE 2. Sagittal Plane Radiographic Parametersa No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 PI = pelvic incidence, LL = lumbar lordosis. aValues are expressed as the mean ± SD for continuous variables. bComparison of the 3 tethering groups analyzed with one-way analysis of variance. View Large PJK in Non-tethered vs Tethered Patients Univariate comparison of non-tethered vs all tethered patients for the primary outcome of postoperative PJK is presented in Table 3. The overall PJK rate for non-tethered patients (NT) was significantly higher compared to the tethered cohort (TO + TC): 45.3% (29/64) vs 26.7% (32/120; P = .011). TABLE 3. Postoperative Development of PJK in 184 ASD Patients Dichotomized Based on Use of Polyethylene Tethersa No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 aAnalyzed using Pearson's chi square test with 2 × 2 contingency table. bTether patients with polyethylene tape only (TO) or tape with crosslink (TC) were grouped into a single cohort. View Large TABLE 3. Postoperative Development of PJK in 184 ASD Patients Dichotomized Based on Use of Polyethylene Tethersa No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 aAnalyzed using Pearson's chi square test with 2 × 2 contingency table. bTether patients with polyethylene tape only (TO) or tape with crosslink (TC) were grouped into a single cohort. View Large Secondary Analyses After Stratifying Tethered Patients Based on Use of an Anchoring Crosslink Secondary analyses after stratification of tethered patients into TO and TC subgroups is presented in Tables 4-6. Overall PJK rates were 45.3% (29/64), 34.4% (22/64), and 17.9% (10/56) for NT, TO, and TC patients, respectively (Table 4). The PJK rate for TC tethers was significantly lower than NT at the Bonferroni-adjusted alpha level (P = .001). Based on the OR, the odds of developing PJK was 3.81 times higher for NT compared to TC patients. There was a nonsignificant trend in PJK reduction for TC compared to TO, and TO compared to NT. The rate of revision surgery for PJF was lowest for TC (3.6% [2/56]) compared to TO (9.4% [6/64]) and NT (4.7% [3/64]); however, pairwise comparisons did not reach statistical significance. TABLE 4. PJK and Revision Surgery for PJF: NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 PJK, proximal junctional kyphosis; PJF, proximal junctional failure. aAnalyzed with Pearson's chi square test with 2 × 2 contingency tables. bAnalyzed with Fisher's exact test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .001 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large TABLE 4. PJK and Revision Surgery for PJF: NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 PJK, proximal junctional kyphosis; PJF, proximal junctional failure. aAnalyzed with Pearson's chi square test with 2 × 2 contingency tables. bAnalyzed with Fisher's exact test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .001 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large Table 5 presents pairwise comparisons of the postoperative change in PJA between NT, TO, and TC subgroups. The postoperative change in PJA was calculated using preoperative and final long-cassette lateral radiographs. If revision surgery for PJF was performed, the final PJA was measured using the long-cassette radiograph taken just prior to revision surgery. Postoperative change in PJA for TC was significantly less than both TO and NT using post hoc ANOVA with Bonferroni correction (P = .026 and P = .024, respectively). There was no significant difference in postoperative PJA change for TO compared to NT (P = 1.000). TABLE 5. Postoperative Change in Proximal Junctional Angle (PJA): NT vs TO vs TCa No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* aChange in PJA at last follow-up (or prior to revision for proximal junctional failure) compared to preoperative PJA. bComparison of the 3 tethering groups analyzed with one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. *Post hoc ANOVA with Bonferroni correction TO compared to NT (P = 1.000), TC compared to TO (P = .026), TC compared to NT (P = .024). View Large TABLE 5. Postoperative Change in Proximal Junctional Angle (PJA): NT vs TO vs TCa No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* aChange in PJA at last follow-up (or prior to revision for proximal junctional failure) compared to preoperative PJA. bComparison of the 3 tethering groups analyzed with one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. *Post hoc ANOVA with Bonferroni correction TO compared to NT (P = 1.000), TC compared to TO (P = .026), TC compared to NT (P = .024). View Large Time-Dependent PJK and Kaplan-Meier Survivorship Analysis Time-to-PJK among NT, TO, and TC subgroups was not significantly different (P = .530). NT, TO, and TC patients developed PJK at 17.02 ± 23.44, 11.59 ± 12.58, and 11.40 ± 14.88 wk, respectively. Comparison of postoperative radiographic follow-up duration was significantly different among all cohorts using post hoc ANOVA with Bonferroni-adjusted analysis (P < .001). NT, TO, and TC patients had total radiographic follow-up of 115.59 ± 50.91, 79.47 ± 41.66, and 41.42 ± 18.48 wk, respectively. Owing to the difference in follow-up duration among NT, TO, and TC subgroups, 3- and 6-mo PJK rates were calculated and Kaplan-Meier analysis was performed. Table 6 presents PJK rates at 3- and 6-mo following multilevel spine instrumentation. At 3 mo, there is a nonsignificant trend in PJK reduction with TC compared to TO and NT. At 6 mo, there is significant PJK reduction for TC compared to NT at the Bonferroni-adjusted alpha level (16.1% compared to 39.1%, P = .005). TABLE 6. Time-Dependent Proximal Junctional Kyphosis (PJK): NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 aPearson's chi square test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .005 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large TABLE 6. Time-Dependent Proximal Junctional Kyphosis (PJK): NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 aPearson's chi square test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .005 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large Kaplan-Meier analysis demonstrated a significant difference in the time-dependent rate of PJK (log rank test, P = .030). Using pairwise comparisons, Kaplan-Meier analysis demonstrated a significant reduction in PJK for TC compared to NT (log rank test, P = .010) at the Bonferroni-corrected alpha level (Figure 4). Although not significant, there was a trend suggestive of PJK reduction for TC compared to TO (log rank test, P = .074). There was no significant difference in PJK rates for TO compared to NT (log rank test, P = .337). FIGURE 4. View largeDownload slide Kaplan-Meier curves demonstrate the PJK-free probability among the 3 cohorts as a time-dependent variable in the weeks following multilevel posterior spine instrumentation for adult spinal deformity. Pairwise comparisons (log rank test): TC compared to TO (P = .074), TO compared to NT (P = .337), TC compared to NT (P = .010). PJK = proximal junctional kyphosis. TC = tether-crosslink. TO = tether-only. NT = no tether. FIGURE 4. View largeDownload slide Kaplan-Meier curves demonstrate the PJK-free probability among the 3 cohorts as a time-dependent variable in the weeks following multilevel posterior spine instrumentation for adult spinal deformity. Pairwise comparisons (log rank test): TC compared to TO (P = .074), TO compared to NT (P = .337), TC compared to NT (P = .010). PJK = proximal junctional kyphosis. TC = tether-crosslink. TO = tether-only. NT = no tether. Multivariable Analyses With Patients Dichotomized Based on PJK Variables included in multivariable binary logistic and Cox regression models were age at surgery, female gender, BMI, number of instrumented vertebrae, use of pelvic fixation, use of 3-column osteotomy, use of polyethylene tether (including both TO and TC), preoperative SVA, change in SVA, and amount of LL correction. Results of binary logistic regression are presented in Table 7. Older age at surgery and greater correction of LL were independent predictors of PJK. Use of a polyethylene tether was a significant predictor of reduced PJK risk (OR = 0.422, 95% confidence interval [CI] 0.205-0.868, P = .019). Based on the Hosmer-Lemeshow goodness-of-fit test, the regression was not statistically different from a null model without the explanatory variables, indicating a good fit (P = .963). TABLE 7. Multivariable Analysis of ASD Patients Dichotomized Based on Development of PJK 6 mo After the Index Operationa OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 CI, confidence interval; OR, odds ratio. aExplanatory variables were entered in a stepwise, multivariable model using binary logistic regression. bPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large TABLE 7. Multivariable Analysis of ASD Patients Dichotomized Based on Development of PJK 6 mo After the Index Operationa OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 CI, confidence interval; OR, odds ratio. aExplanatory variables were entered in a stepwise, multivariable model using binary logistic regression. bPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large Results of Cox proportional hazards regression are presented in Table 8. Use of a polyethylene tether remained a significant independent factor for avoiding PJK (hazards ratio = 0.532, 95% CI 0.318-0.892, P = .017). Cumulative hazard functions for tethered and nontethered patients are plotted in Figure 5. FIGURE 5. View largeDownload slide A Cox regression model was implemented, and hazard functions were plotted for NT and TO + TC patients. The hazard functions demonstrate a lower cumulative risk of PJK for TO+TC vs NT patients (HR = 0.532, 95% CI 0.318-0.892, P = .017). NT = no tether. TO = tether-only. TC = tether-crosslink. PJK = proximal junctional kyphosis. HR = hazards ratio. CI = confidence interval. FIGURE 5. View largeDownload slide A Cox regression model was implemented, and hazard functions were plotted for NT and TO + TC patients. The hazard functions demonstrate a lower cumulative risk of PJK for TO+TC vs NT patients (HR = 0.532, 95% CI 0.318-0.892, P = .017). NT = no tether. TO = tether-only. TC = tether-crosslink. PJK = proximal junctional kyphosis. HR = hazards ratio. CI = confidence interval. TABLE 8. Cox Proportional Hazards Regression for ASD Patients and Postoperative PJK HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 CI, confidence interval; HR, hazards ratio. aPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large TABLE 8. Cox Proportional Hazards Regression for ASD Patients and Postoperative PJK HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 CI, confidence interval; HR, hazards ratio. aPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large DISCUSSION PJK is common following multilevel spine instrumentation for ASD, with a reported incidence of 11.0% to 52.9%.1-8 The exact pathophysiological mechanism of PJK is unclear; however, most authors agree it is likely multifactorial.20,21 PJK may in part be secondary to an iatrogenic effect of altered spinal biomechanics, which produce pathologic flexion-loading forces at the proximal fusion segment. This includes compensatory changes following operative correction of thoracic kyphosis and sagittal malalignment, which are further exacerbated by posterior tension band disruption and the creation of a moment arm at the proximal segment of long, rigid fusion constructs.13,22,23 PJF is considered a distinct entity in the spectrum of PJK that involves mechanical failure at the UIV or just above and/or proximal junctional posterior discoligamentous failure.20,24 The most common modes of failure in the upper thoracic spine compared to the thoracolumbar region are soft tissue disruption and mechanical fracture, respectively.25 Patients with PJF may present with progressively worsening pain, focal neurological deficits, and ambulatory difficulty—resulting in significantly worse standardized outcome scores.16 Therefore, PJK and PJF prevention is, and should be, among the principal goals for optimizing clinical outcomes after spinal deformity surgery. Posterior polyethylene junctional tethers at the UIV in multilevel spine instrumentation may be an effective anti-PJK technique. Bess and colleagues,26 using finite element analysis, demonstrated more gradual changes in segmental range-of-motion and reduction in proximal segment intradiscal pressures, pedicle screw loads, and posterior ligament complex forces with increasing number of posterior tethers. The authors hypothesized that posterior polyethylene tethers may reduce biomechanical risk factors for PJK. A literature search for posterior tethers, and their clinical use as an anti-PJK device after spine surgery, produced only a single clinical report by Zaghloul and colleagues.27 The authors reported no case of PJK in 18 patients with polyethylene tape–based strap stabilization at the proximal end of fusion constructs. However, this relatively small case series, with average 11.9 mo of follow-up, lacked specific inclusion and exclusion criteria for spinal deformity surgery or long-segment fusion—12 patients had fusion constructs of 6 or less motion segments, and 4 patients were fused at a single segment.27 We present the first large clinical study of posterior polyethylene junctional tethers as an anti-PJK device for multilevel spine instrumentation (>6 fused segments) in ASD patients with scoliosis and/or global sagittal malalignment. Our results show that junctional tethers (both TO and TC) are a significant negative predictor of PJK, and that there are significantly reduced overall crude and time-dependent PJK rates when polyethylene tape is anchored to a crosslink. Additionally, TC tethers significantly reduced the postoperative change in PJA at last follow-up. Pairwise comparisons showed no significant effect of polyethylene tethering without utilization of an anchoring crosslink, although trends still suggested a reduction in PJK. To explain this difference, we hypothesize that anchoring polyethylene tape to a crosslink between UIV-1 and UIV-2, and then distracting the crosslink inferiorly, produces additional tension and increased junctional support compared to polyethylene tape-only tethers. In this study, the decision to use junctional tethers was not randomized; instead, the 2 senior authors (CIS, JSS) started utilizing junctional tethers from September 2013 and January 2015 onwards, respectively. No other change in surgical technique (vertebral augmentation, multilevel stabilization screws, transverse process hooks at the UIV, proximal transition rods of reduced diameter, hybrid constructs)3,17,18 was identified that may confound PJK rates. However, we still attempted to limit potential bias from practice variability and nonrandomization by controlling for baseline patient demographic, surgical data, as well as pre- and postoperative sagittal plane radiographic parameters in multivariable models (Tables 7 and 8) and performing time-dependent analyses (Figures 4 and 5). In addition, the confounding effect of possible practice variability was limited by our decision not to extend the inclusion start date to an earlier time. This produced relatively balanced cohorts with 64 NT, 64 TO, and 56 TC patients. Limitations The limitations of this study include its nonblinded retrospective design, the experience level and technical ability of the surgeons at a single center, the different follow-up duration among cohorts, and the relatively short minimum follow-up duration of 3 mo. To account for differences in follow-up, we calculated 3- and 6-mo PJK rates, performed multivariable regression using 6-mo PJK rates, implemented a Cox regression model, and performed a life-table survival analysis (Kaplan-Meier estimation). Kaplan-Meier analysis was congruent with the overall crude rates and demonstrated a significant time-dependent reduction in PJK for TC compared to NT (log rank test, P = .010) at the Bonferroni-corrected alpha level (Figure 4). Although minimum radiographic follow-up of 3 mo may seem relatively short, prior studies have reported up to 76% of PJK occurs within 3 mo postoperatively, and that new PJK after 6 mo may be rare.2,3,22,28,29 Also, Kim and colleagues2 showed that the average PJA increase at 2 mo postoperation accounts for the majority of total PJA increase at 5 yr postoperation. In our study, 93% (171/184) of patients had at least 6 mo of follow-up, and the vast majority of patients who developed PJK did so within 3 mo of the index operation. Although TC tethers significantly reduced PJK, rates of revision surgery for PJF were comparable across all cohorts. However, the small number of patients who underwent revision surgery limits meaningful comparison of revision rates; therefore, a larger tether study with longer follow-up duration may be warranted. We are currently accruing more patients in a prospectively maintained database, with longer follow-up and clinical outcome scores, to further elucidate the clinical impact of posterior polyethylene tethers in spinal deformity surgery. Additionally, future studies are necessary to determine if our findings are generalizable to other spine surgery patients without ASD, particularly those without scoliosis and/or global sagittal malalignment, or who have 6 or less fused motion segments. CONCLUSION PJK is a common problem after multilevel spine instrumentation for spinal deformity correction, and in a subset of patients, may require surgical revision. Posterior junctional tethers, consisting of polyethylene tape anchored to a crosslink at the superior aspect of fusion constructs, may represent a novel anti-PJK device for ASD patients. In this study, use of junctional tethers for long-segment posterior fusion for ASD significantly reduced the occurrence of PJK. This difference was progressive from NT to TO to TC cohorts, but only reached statistical significance on pairwise comparisons for NT vs TC. These findings suggest potential benefit of junctional tethers to reduce PJK, and that future prospective studies with longer-term follow-up are warranted. Disclosures Dr Bess receives research support from K2M, Innovasis, Nuvasive, Medtronic, DePuy Synthes, Stryker, and Zimmer-Biomet; is a consultant for K2M and Allosource; and holds patents with K2M and Innovasis. Mr Line, BSME, is a consultant for ISSGF. Dr Ames is a consultant for Medtronic, DePuy Synthes, and Stryker; receives royalties from Zimmer-Biomet and Stryker; has research support from UCSF; and holds a patent with Fish & Richardson PC. Dr Schwab is a consultant for Zimmer-Biomet, K2M, Nuvasive, Medicrea, and MSD; receives research support from SRS, AOSpine, DePuy Spine Synthesis, and ISSGF; and is a stock holder in Nemaris Inc. Dr Lafage is a consultant for Nemaris Inc, Nuvasive, Medicrea, and DePuy Synthes; is a stock holder in Nemaris Inc; and receives non-study-related support from SRS, NIH, DePuy Synthes, and ISSGF. Dr Shaffrey is a consultant for Medtronic, Nuvasive, Zimmer-Biomet, and K2M; receives royalties from Medtronic, Nuvasive, and Zimmer-Biomet; is a stock holder in Nuvasive; and has grants from NIH, DOD, and NACTN. Dr Smith receives royalties from Zimmer Biomet, is a consultant for Zimmer Biomet, Cerapedics, Nuvasive, K2M, and AlloSource; receives honoraria from Zimmer Biomet, Nuvasive, and K2M; has research support from DePuy Synthes, and ISSGF; and receives fellowship support from NREF and AOSpine. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Glattes RC , Bridwell KH , Lenke LG , Kim YJ , Rinella A , Edwards C 2nd . Proximal junctional kyphosis in adult spinal deformity following long instrumented posterior spinal fusion . Spine (Phila Pa 1976) . 2005 ; 30 ( 14 ): 1643 - 1649 . Google Scholar CrossRef Search ADS PubMed 2. Kim YJ , Bridwell KH , Lenke LG , Glattes CR , Rhim S , Cheh G . Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion . Spine (Phila Pa 1976) . 2008 ; 33 ( 20 ): 2179 - 2184 . Google Scholar CrossRef Search ADS PubMed 3. Ha Y , Maruo K , Racine L et al. Proximal junctional kyphosis and clinical outcomes in adult spinal deformity surgery with fusion from the thoracic spine to the sacrum: a comparison of proximal and distal upper instrumented vertebrae . J Neurosurg Spine . 2013 ; 19 ( 3 ): 360 - 369 . Google Scholar CrossRef Search ADS PubMed 4. Park SJ , Lee CS , Chung SS , Lee JY , Kang SS , Park SH . Different risk factors of proximal junctional kyphosis and proximal junctional failure following long instrumented fusion to the sacrum for adult spinal deformity: survivorship analysis of 160 patients . Neurosurgery . 2016 ; 80 ( 2 ): 279 - 286 . 5. Smith JS , Klineberg E , Lafage V et al. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery . J Neurosurg Spine . 2016 ; 25 ( 1 ): 1 - 14 . Google Scholar CrossRef Search ADS PubMed 6. Mummaneni PV , Park P , Fu KM et al. Does minimally invasive percutaneous posterior instrumentation reduce risk of proximal junctional kyphosis in adult spinal deformity surgery? a propensity-matched cohort analysis . Neurosurgery . 2016 ; 78 ( 1 ): 101 - 108 . Google Scholar CrossRef Search ADS PubMed 7. Reames DL , Kasliwal MK , Smith JS , Hamilton DK , Arlet V , Shaffrey CI . Time to development, clinical and radiographic characteristics, and management of proximal junctional kyphosis following adult thoracolumbar instrumented fusion for spinal deformity . J Spinal Disord Tech . 2015 ; 28 ( 2 ): E106 - E114 . Google Scholar CrossRef Search ADS PubMed 8. Yan C , Li Y , Yu Z . Prevalence and consequences of the proximal junctional kyphosis after spinal deformity surgery . Medicine (Baltimore) . 2016 ; 95 ( 20 ): e3471 . Google Scholar CrossRef Search ADS PubMed 9. Liu FY , Wang T , Yang SD , Wang H , Yang DL , Ding WY . Incidence and risk factors for proximal junctional kyphosis: a meta-analysis . Eur Spine J . 2016 ; 25 ( 8 ): 2376 - 2383 . Google Scholar CrossRef Search ADS PubMed 10. Annis P , Lawrence BD , Spiker WR et al. Predictive factors for acute proximal junctional failure after adult deformity surgery with upper instrumented vertebrae in the thoracolumbar spine . Evid-Based Spine Care J . 2014 ; 5 ( 2 ): 160 - 162 . Google Scholar CrossRef Search ADS PubMed 11. Lee GA , Betz RR , Clements DH 3rd , Huss GK . Proximal kyphosis after posterior spinal fusion in patients with idiopathic scoliosis . Spine (Phila Pa 1976) . 1999 ; 24 ( 8 ): 795 - 799 . Google Scholar CrossRef Search ADS PubMed 12. Kim YJ , Lenke LG , Bridwell KH et al. Proximal junctional kyphosis in adolescent idiopathic scoliosis after 3 different types of posterior segmental spinal instrumentation and fusions: incidence and risk factor analysis of 410 cases . Spine (Phila Pa 1976) . 2007 ; 32 ( 24 ): 2731 - 2738 . Google Scholar CrossRef Search ADS PubMed 13. Rhee JM , Bridwell KH , Won DS , Lenke LG , Chotigavanichaya C , Hanson DS . Sagittal plane analysis of adolescent idiopathic scoliosis: the effect of anterior versus posterior instrumentation . Spine (Phila Pa 1976) . 2002 ; 27 ( 21 ): 2350 - 2356 . Google Scholar CrossRef Search ADS PubMed 14. Kim YJ , Bridwell KH , Lenke LG , Kim J , Cho SK . Proximal junctional kyphosis in adolescent idiopathic scoliosis following segmental posterior spinal instrumentation and fusion: minimum 5-year follow-up . Spine (Phila Pa 1976) . 2005 ; 30 ( 18 ): 2045 - 2050 . Google Scholar CrossRef Search ADS PubMed 15. Helgeson MD , Shah SA , Newton PO et al. Evaluation of proximal junctional kyphosis in adolescent idiopathic scoliosis following pedicle screw, hook, or hybrid instrumentation . Spine (Phila Pa 1976) . 2010 ; 35 ( 2 ): 177 - 181 . Google Scholar CrossRef Search ADS PubMed 16. Lau D , Funao H , Clark AJ et al. The clinical correlation of the hart-issg proximal junctional kyphosis severity scale with health-related quality-of-life outcomes and need for revision surgery . Spine (Phila Pa 1976) . 2016 ; 41 ( 3 ): 213 - 223 . Google Scholar CrossRef Search ADS PubMed 17. Aubin CE , Cammarata M , Wang X , Mac-Thiong JM . Instrumentation strategies to reduce the risks of proximal junctional kyphosis in adult scoliosis: a detailed biomechanical analysis . Spine Deform . 2015 ; 3 ( 3 ): 211 - 218 . Google Scholar CrossRef Search ADS PubMed 18. Sandquist L , Carr D , Tong D , Gonda R , Soo TM . Preventing proximal junctional failure in long segmental instrumented cases of adult degenerative scoliosis using a multilevel stabilization screw technique . Surg Neurol Int . 2015 ; 6 ( 1 ): 112 . Google Scholar CrossRef Search ADS PubMed 19. Steyerberg EW , Vickers AJ , Cook NR et al. Assessing the performance of prediction models: a framework for traditional and novel measures . Epidemiology . 2010 ; 21 ( 1 ): 128 - 138 . Google Scholar CrossRef Search ADS PubMed 20. Nguyen NL , Kong CY , Hart RA . Proximal junctional kyphosis and failure-diagnosis, prevention, and treatment . Curr Rev Musculoskelet Med . 2016 ; 9 ( 3 ): 299 - 308 . Google Scholar CrossRef Search ADS PubMed 21. Arlet V , Aebi M . Junctional spinal disorders in operated adult spinal deformities: present understanding and future perspectives . Eur Spine J . 2013 ; 22 ( S2 ): 276 - 295 . Google Scholar CrossRef Search ADS 22. Yagi M , King AB , Boachie-Adjei O . Incidence, risk factors, and natural course of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis. Minimum 5 years of follow-up . Spine (Phila Pa 1976) . 2012 ; 37 ( 17 ): 1479 - 1489 . Google Scholar CrossRef Search ADS PubMed 23. Cammarata M , Aubin CE , Wang X , Mac-Thiong JM . Biomechanical risk factors for proximal junctional kyphosis: a detailed numerical analysis of surgical instrumentation variables . Spine (Phila Pa 1976) . 2014 ; 39 ( 8 ): E500 - E507 . Google Scholar CrossRef Search ADS PubMed 24. Fradet L , Wang X , Lenke LG , Aubin CE . Biomechanical analysis of proximal junctional failure following adult spinal instrumentation using a comprehensive hybrid modeling approach . Clin Biomech (Bristol, Avon) . 2016 ; 39 : 122 - 128 . Google Scholar CrossRef Search ADS PubMed 25. Hostin R , McCarthy I , O’Brien M et al. Incidence, mode, and location of acute proximal junctional failures after surgical treatment of adult spinal deformity . Spine (Phila Pa 1976) . 2013 ; 38 ( 12 ): 1008 - 1015 . Google Scholar CrossRef Search ADS PubMed 26. Bess S , Harris JE , Turner AW et al. The effect of posterior polyester tethers on the biomechanics of proximal junctional kyphosis: a finite element analysis . J Neurosurg Spine . 2017 ; 26 ( 1 ): 125 - 133 . Google Scholar CrossRef Search ADS PubMed 27. Zaghloul KM , Matoian BJ , Denardin NB , Patel VV . Preventing proximal adjacent level kyphosis with strap stabilization . Orthopedics . 2016 ; 39 ( 4 ): e794 - e799 . Google Scholar CrossRef Search ADS PubMed 28. Yagi M , Akilah KB , Boachie-Adjei O . Incidence, risk factors and classification of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis . Spine (Phila Pa 1976) . 2011 ; 36 ( 1 ): E60 - E68 . Google Scholar CrossRef Search ADS PubMed 29. Lau D , Clark AJ , Scheer JK et al. Proximal junctional kyphosis and failure after spinal deformity surgery: a systematic review of the literature as a background to classification development . Spine (Phila Pa 1976) . 2014 ; 39 ( 25 ): 2093 - 2102 . Google Scholar CrossRef Search ADS PubMed COMMENTS The authors present the results of their retrospective, single-center, comparative cohort study of 184 patients who underwent surgery for adult spinal deformity between 2012 and 2016. During that period, the authors' surgical techniques at the proximal junction changed from not using a polyethylene tether, to placing a tether that was secured to the spinous processes, and finally to placing a tether that was secured to a crosslink. In their primary analysis, they found that the incidence of PJK without a tether was 45% and with a tether, using either technique, it was 27%. Secondary analyses were generally consistent with the primary finding. The crude odds ratio (OR) for developing PJK with a tether was 0.44; controlling for factors such as BMI, pelvic fixation, and amount of correction using multivariable regression, the odds ratio for PJK with the use of a tether was 0.53. This study includes 3 cohorts made up of patients treated by different techniques as the surgical practices at a single center evolved over time. The 2 surgeons may have changed their techniques in response to a perceived problem with PJK, perhaps as the result of a particularly high incidence of PJK in the period prior to instituting the technical change. There is evidence that this may, in fact, be the case, as the “no tether” cohort had a 45% incidence of PJK, which is toward the upper end of the range reported in the literature.1 This raises the possibility that some of the observed difference in the incidence of PJK is due to regression to the mean. If aspects of the surgical technique (reduced application of corrective forces near the UIV, more limited proximal soft tissue dissection) or decision-making (selection of the construct levels) were changed in response to a particularly high incidence of PJK, one would expect the incidence to subsequently decline independent of other changes. Assessing “statistical significance” is only 1 part of analyzing a study; one should also critically evaluate the estimate of the effect size, the degree of uncertainty regarding the estimate, and how the study design may affect the estimate.2 In this study, the absolute reduction of 18% in the incidence of PJK with tether use is very likely to be an overestimate of the effect size. Regression to the mean and other changes in management, as discussed above, are 2 likely explanations for some of the observed difference in the incidence of PJK between the subgroups in this case series. These considerations must be weighed against the potential risks of the intervention. In this case the procedure likely adds little risk to the procedure itself and probably does not put the patient at increased risk of delayed adverse events, although these aspects need further study. Even if the effect size of 0.18 is exaggerated by a factor of 3, with a standard error of approximately 0.05, it is unlikely (less than 1% probability) that the incidence of PJK with a tether is actually higher than without a tether.3 If revision for PJK is necessary, the presence of a tether probably would not interfere with the subsequent surgery or alter its outcome. This technique may therefore be an option that spinal deformity surgeons should consider but with more modest expectations for outcome than reported in this paper. Follow-up reports and studies by other groups should be performed help to further refine our knowledge of the effectiveness and safety of this procedure. Peter D. Angevine New York 1. Nguyen N-L Kong CY Hart RA . Proximal junctional kyphosis and failure- diagnosis, prevention, and treatment . Curr Rev Musculoskelet Med . 2016 ; 9 ( 3 ): 299 - 308 . Google Scholar CrossRef Search ADS PubMed 2. Leek J McShane BB Gelman A Colquhoun D Nuijten MB Goodman SN . Five ways to fix statistics . Nature . 2017 ; 551 ( 7682 ): 557 - 559 . Google Scholar CrossRef Search ADS PubMed 3. Gelman A Carlin J . Beyond power calculations: assessing type S (sign) and type M (magnitude) errors . Perspect Psychol Sci . 2014 ; 9 ( 6 ): 641 - 651 . Google Scholar CrossRef Search ADS PubMed Proximal junctional kyphosis (PJK) is a common entity that develops after long-segment fusions for adult spinal deformity (ASD). The authors propose an interesting technique to reduce the incidence of PJK by incorporating junctional tether at the proximal junction. For this purpose, the authors retrospectively reviewed the data and radiographic images of 184 patients that underwent posterior or anteroposterior fixation at a single institute for ASD between 2012 and 2016. The authors compared between 3 different techniques: no proximal tether, 64 patients; polyethylene tether in isolation, 64 patients; and polyethylene tether in conjunction with a crosslink, 56 patients. After a mean follow-up of 20 months, the authors found that the rate of PJK was significantly higher in the nontethered group compared to the tethered group (45% vs 27%, P = .011). Based on Kaplan-Meier survivorship analysis, there was a significant reduction in PJK rate for tethered group vs nontethered group (log-rank test, P = .010). The revision surgery rate for proximal junctional failure (PJF) was lower in the tethered group as well, although this difference did not reach a statistical significance. This article is well-written and novel in terms of using a new, safe, and effective technique to improve the surgical outcomes of long-segment fusions in ASD. The authors should be commended for this important addition to our understanding of the junctional tethers and their effect on PJK. Moving forward, it will be interesting to utilize the findings of this study in future prospective randomized studies, with a longer follow-up duration, to investigate long-term outcomes, such as PJF, associated with junctional tethers. Seba Ramhmdani Ali Bydon Baltimore, Maryland The authors have presented initial data on the novel use of polyethelene tethers placed 1 level past the proximal end of a fusion construct in an attempt to help prevent proximal junctional kyphosis (PJK). The idea of stress sharing is not new, but this particular use of an “elastic band” is of interest to anybody who has had to deal with PJK. As with many studies such as this, there are potential systemic biases in how the 2 cohorts were created and in surgical technique that may have adapted over time. Though this paper offers an attractive adjunct device to implant with many fusions, it still needs more study and investigation to justify its use. John Chi Boston, Massachusetts Copyright © 2018 by the Congress of Neurological Surgeons http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Operative Neurosurgery Oxford University Press

A Pilot Study on Posterior Polyethylene Tethers to Prevent Proximal Junctional Kyphosis After Multilevel Spinal Instrumentation for Adult Spinal Deformity

Loading next page...
 
/lp/ou_press/a-pilot-study-on-posterior-polyethylene-tethers-to-prevent-proximal-SiDJePrjlA
Publisher
Congress of Neurological Surgeons
Copyright
Copyright © 2018 by the Congress of Neurological Surgeons
ISSN
2332-4252
eISSN
2332-4260
D.O.I.
10.1093/ons/opy065
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Proximal junctional kyphosis (PJK) is a common problem after multilevel spine instrumentation. OBJECTIVE To determine if junctional tethers reduce PJK after multilevel instrumented surgery for adult spinal deformity (ASD). METHODS ASD patients who underwent posterior instrumented fusion were divided into 3 groups: no tether (NT), polyethylene tether-only (TO; tied securely through the spinous processes of the uppermost instrumented vertebra [UIV] + 1 and UIV-1), and tether with crosslink (TC; passed through the spinous process of UIV+1 and tied to a crosslink between UIV-1 and UIV-2). PJK was defined as proximal junctional angle ≥ 10° and ≥ 10° greater than the corresponding preoperative measurement. RESULTS One hundred eighty-four (96%) of 191 consecutive patients achieved minimum 3-mo follow-up (mean = 20 mo [range:3-56 mo]; mean age = 66 yr; 67.4% female). There were no significant differences between groups based on demographic, surgical, and sagittal radiographic parameters. PJK rates were 45.3% (29/64), 34.4% (22/64), and 17.9% (10/56) for NT, TO, and TC, respectively. PJK rate for all tethered patients (TO + TC; 26.7% [32/120]) was significantly lower than NT (P = .011). PJK rate for TC was significantly lower than NT (P = .001). Kaplan-Meier analysis showed significant time-dependent PJK reduction for TC vs NT (log rank test, P = .010). Older age and greater change in lumbar lordosis were independent predictors of PJK, while junctional tethers had a significant protective effect. CONCLUSION Junctional tethers significantly reduced occurrence of PJK. This difference was progressive from NT to TO to TC, but only reached pairwise significance for NT vs TC. This suggests potential benefit of tethers to reduce PJK, and that future prospective studies are warranted. Adjacent segment disease, Adult spinal deformity, Complication, Proximal junctional angle, Proximal junctional kyphosis, Scoliosis, Spinal fusion ABBREVIATIONS ABBREVIATIONS ANOVA analysis of variance ASD adult spinal deformity BMI body mass index CI confidence interval LL lumbar lordosis NT no tether OR odds ratio PJA proximal junctional angle PJF proximal junctional failure PJK proximal junctional kyphosis PI pelvic incidence PT pelvic tilt SPSS Statistical Package for Social Science SVA sagittal vertical axis TC tether with crosslink TO tether-only UIV uppermost instrumented vertebra Proximal junctional kyphosis (PJK) continues to be problematic after multilevel spinal instrumentation for adult spinal deformity (ASD). The reported incidence ranges from 11.0% to 52.9%.1-8 Several risk factors have been identified and include high body mass index (BMI), older age at surgery, fusion to sacrum, low bone mineral density, larger preoperative sagittal vertical axis (SVA), postoperative proximal junctional angle (PJA) >5° and thoracic kyphosis >30°, and greater correction of lumbar lordosis.4,7,9,10 Early reports of PJK were primarily focused on adolescent idiopathic scoliosis.11 Authors used varying radiographic criteria, with abnormal PJA ranging from 5° to 15°.11-15 More recently, for ASD, definitions of abnormal PJK have been more consistent.1 PJA is measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). Abnormal PJK is defined as a proximal junction sagittal Cobb angle that is ≥10° and ≥10° greater than the corresponding preoperative measurement.1 Although there is increasing emphasis on PJK detection, as well as various classification systems to guide its management,16 the question remains whether PJK is a true complication, especially in the absence of clinical symptoms.5 Glattes and colleagues have reported no significant difference in outcome scores (Scoliosis Research Society-24 outcomes measure) between PJK and non-PJK groups.1,2 However, a subset of patients with PJK may still require surgical revision; therefore, various preventative techniques have been proposed and include vertebral augmentation, multilevel stabilization screws, transverse process hooks at the UIV, proximal transition rods of reduced diameter, and hybrid constructs.3,17,18 The purpose of this study was to determine if junctional tethers reduce the incidence of postoperative PJK after multilevel instrumented surgery for ASD. METHODS We performed a single-center, retrospective evaluation of a prospectively maintained database of patients who underwent multilevel spine instrumentation at the University of Virginia Health System (Charlottesville, Virginia). All operations were performed by the 2 senior authors (CIS, JSS) between December 31, 2011 and June 2, 2016, and the database was reviewed on December 2, 2016. The decision to use junctional tethers was not randomized; instead, CIS and JSS started utilizing junctional tethers from September 2013 and January 2015 onwards, respectively. No other technique such as vertebral augmentation, multilevel stabilization screws, transitional rods, or fusion with hybrid constructs was used that might confound PJK rates.3,17,18 The study was approved by the University of Virginia Institutional Review Board. Patient consent was waived since data collection and analysis was retrospective and all patient-specific identifiers were removed for publication. We searched for ASD patients with diagnosis of scoliosis and/or global sagittal malalignment (ie, abnormal SVA, thoracic kyphosis, lumbar lordosis, and/or pelvic incidence to lumbar lordosis mismatch). Inclusion criteria for the study were patient age > 18 yr, deformity correction with instrumented segmental posterior spine fusion (may have also had anterior approach procedure) at a minimum > 6 motion segments, thoracic UIV, pedicle screw instrumentation without transitional rods or hooks at the UIV, and complete radiographic data with preoperative, postoperative, and final standing long-cassette films at minimum follow-up of 3 mo. We excluded patients with preoperative diagnosis of (1) degenerative spine disease without significant scoliosis or global sagittal malalignment, and/or (2) vertebral osteomyelitis/discitis. Patients who underwent a revision spine surgery without change in UIV were also excluded from the study. Radiographic Measurements Subjects had standing posteroanterior and lateral long-cassette (14 × 36 inch) radiographs preoperatively, postoperatively, and at final follow-up. PJA was the sagittal Cobb angle measured from the caudal endplate of the UIV to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV) (Figure 1). Abnormal PJK was defined as a PJA that was ≥ 10° and ≥ 10° greater than the corresponding preoperative measurement.1 Figure 2 depicts measurement technique for pelvic tilt (PT), pelvic incidence (PI), and lumbar lordosis (LL). FIGURE 1. View largeDownload slide A, B, Preoperative and postoperative standing long-cassette lateral radiographs of a 73-yr-old female with spinal deformity show the proximal junctional angle (PJA) measuring 16.2° and 29.3°, respectively. PJA was the sagittal Cobb angle measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). For this study, abnormal proximal junctional kyphosis (PJK) was defined as a PJA (1) greater than or equal to 10° and (2) at least 10° greater than the corresponding preoperative measurement. The presence of both criteria was necessary to be considered abnormal. B, This patient developed postoperative PJK. FIGURE 1. View largeDownload slide A, B, Preoperative and postoperative standing long-cassette lateral radiographs of a 73-yr-old female with spinal deformity show the proximal junctional angle (PJA) measuring 16.2° and 29.3°, respectively. PJA was the sagittal Cobb angle measured from the caudal endplate of the uppermost instrumented vertebra (UIV) to the cephalad endplate of UIV+2 (2 supra-adjacent levels above the UIV). For this study, abnormal proximal junctional kyphosis (PJK) was defined as a PJA (1) greater than or equal to 10° and (2) at least 10° greater than the corresponding preoperative measurement. The presence of both criteria was necessary to be considered abnormal. B, This patient developed postoperative PJK. FIGURE 2. View largeDownload slide Pelvic tilt is the angle between the vertical line from the center of the femoral head and the line from the center of the femoral head to the center of the sacral endplate. Pelvic incidence is the angle between the line perpendicular to the sacral endplate and the line connecting the midpoint of the sacral endplate to the center of the femoral head. Lumbar lordosis is the Cobb angle between the inferior endplate of T12 and superior endplate of the sacrum. FIGURE 2. View largeDownload slide Pelvic tilt is the angle between the vertical line from the center of the femoral head and the line from the center of the femoral head to the center of the sacral endplate. Pelvic incidence is the angle between the line perpendicular to the sacral endplate and the line connecting the midpoint of the sacral endplate to the center of the femoral head. Lumbar lordosis is the Cobb angle between the inferior endplate of T12 and superior endplate of the sacrum. Tethering Technique Our proximal junction posterior tether consisted of a 5-mm woven polyethylene Mersilene tape (Ethicon, Somerville, New Jersey) on a blunt needle, and in 56 patients, this was anchored to a standard crosslink. The tape-only technique first involves using a high-speed drill to create holes in the base of the UIV+1 and UIV-1 spinous processes. Then polyethylene tape is passed through these and tightened securely. An assistant holds the first knot in the polyethylene tape while the surgeon completes tying the tape for maximal tension. Our tape-crosslink technique is similar, but first involves placement of a standard crosslink between the UIV-1 and UIV-2 spinous processes. Again, using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. Polyethylene tape is then passed around the crosslink and through the UIV+1 spinous process, and tied securely. Finally, the crosslink is distracted inferiorly to fully tension the polyethylene tape for junctional support (Figure 3). FIGURE 3. View largeDownload slide The polyethylene tape-crosslink tethering technique first involves placement of a standard crosslink between the base of UIV-1 and UIV-2 spinous processes. Using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. A, Next, a 5-mm woven polyethylene tape is passed around the crosslink and through the UIV+1 spinous process, and B, then tightened securely (the black line traces the polyethylene tether). C, The single arrow and double arrows indicate the superior and inferior attachments of the polyethylene tape. D, E, Finally, the crosslink is distracted inferiorly to fully tension the tape for junctional support. F, The final result is aimed to reduce the risk of developing PJK. FIGURE 3. View largeDownload slide The polyethylene tape-crosslink tethering technique first involves placement of a standard crosslink between the base of UIV-1 and UIV-2 spinous processes. Using a high-speed drill, a hole is created in the base of the UIV+1 spinous process. A, Next, a 5-mm woven polyethylene tape is passed around the crosslink and through the UIV+1 spinous process, and B, then tightened securely (the black line traces the polyethylene tether). C, The single arrow and double arrows indicate the superior and inferior attachments of the polyethylene tape. D, E, Finally, the crosslink is distracted inferiorly to fully tension the tape for junctional support. F, The final result is aimed to reduce the risk of developing PJK. Statistical Analysis Patients were divided into 3 groups based on tethering technique, which include: (1) no tether (NT), (2) polyethylene tether-only (TO), and (3) polyethylene tether with crosslink (TC). Data are presented as mean and standard deviation for continuous variables, and as frequency for categorical variables. Baseline characteristics of the cohorts, including patient demographics, surgical data, and sagittal plane radiographic parameters, were compared using Pearson's Chi square test, Fisher's exact test, and 1-way analysis of variance (ANOVA). Rates of PJK and revision surgery for proximal junctional failure (PJF), time-to-PJK, change in postoperative PJA, and total radiographic follow-up duration were analyzed using Pearson's Chi square test, Fisher's exact test, and 1-way ANOVA followed by Bonferroni post hoc test, when appropriate. Odds ratios (ORs) were calculated from 2 × 2 contingency tables. A Kaplan-Meier survivorship analysis was performed to analyze PJK as a time-dependent variable. After dichotomizing patients based on development of PJK at 6 mo postop, independent variables were entered in a stepwise, multivariable model using binary logistic analysis. The fit of the final model was assessed using the Hosmer-Lemeshow goodness-of-fit test.19 For time-dependent regression, a Cox proportional hazards model was implemented. For this study, P-values < .05 were considered statistically significant, and a post hoc Bonferroni adjustment was made for multiple pairwise comparisons. Statistical Package for Social Science (SPSS) version 24.0 (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. IBM Corp, Armonk, New York) was used to perform all analyses. RESULTS Patient Demographics and Surgical Data One hundred eighty-four (96%) of 191 consecutive patients, between December 31, 2011 and June 2, 2016, met inclusion criteria and achieved minimum 3-mo follow-up (mean = 20 mo [range: 3-56 mo]; mean age = 66 yr; 67.4% female). Patient demographics and surgical data are presented in Table 1. There were 64 NT, 64 TO, and 56 TC patients. For these 3 cohorts, there were no significant differences in patient age at surgery, gender, preoperative BMI, number of instrumented vertebrae, use of pelvic fixation, use of 3-column osteotomy, or type of approach (posterior-only vs 2-stage anterior-posterior). The majority of cases involved at least 10 motion segments with pelvic fixation, did not have 3-column osteotomies (vertebral column resection and/or pedicle subtraction osteotomy), and were posterior-only approaches. A total of 12 patients were treated via a 2-stage, combined anterior-posterior approach. No intraoperative or postoperative complication in this study could be attributed to posterior polyethylene tethers. TABLE 1. Patient Demographics and Surgical Data Separated by Tethering Techniquea No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 aValues are expressed as the mean ± standard deviation or frequency for continuous and categorical variables, respectively. bComparison of the 3 tethering groups analyzed with Pearson's chi square test, Fisher's exact test, or one-way analysis of variance. View Large TABLE 1. Patient Demographics and Surgical Data Separated by Tethering Techniquea No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Age at surgery (years) 66.06 ± 8.47 66.27 ± 10.92 66.64 ± 8.59 .944 Gender (female/male) 42/22 43/21 39/17 .895 (χ2 = 0.221) Body mass index (kg/m2) 28.83 ± 5.88 30.23 ± 4.76 30.76 ± 6.02 .144 Number of instrumented vertebrae 11.75 ± 3.21 11.20 ± 2.76 12.23 ± 3.30 .192 Pelvic fixation (yes/no) 62/2 56/8 52/4 .136 3-column osteotomy (yes/no) 18/46 17/47 8/48 .153 (χ2 = 3.753) Approach  Posterior-only 57 61 54 .252  Anterior-posterior 7 3 2 aValues are expressed as the mean ± standard deviation or frequency for continuous and categorical variables, respectively. bComparison of the 3 tethering groups analyzed with Pearson's chi square test, Fisher's exact test, or one-way analysis of variance. View Large Sagittal Plane Radiographic Parameters Sagittal plane radiographic parameters are presented in Table 2. There were no significant differences in the pre- and postoperative SVA, PT, PI, LL, PI-LL mismatch, and thoracic kyphosis among cohorts. The amount of deformity correction, measured by the postoperative change in SVA and LL, was comparable among cohorts. TABLE 2. Sagittal Plane Radiographic Parametersa No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 PI = pelvic incidence, LL = lumbar lordosis. aValues are expressed as the mean ± SD for continuous variables. bComparison of the 3 tethering groups analyzed with one-way analysis of variance. View Large TABLE 2. Sagittal Plane Radiographic Parametersa No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) Pb Preoperative  Sagittal vertical axis (cm) 8.65 ± 6.43 9.23 ± 7.87 8.66 ± 7.33 .876  Pelvic tilt (°) 24.97 ± 10.16 28.14 ± 9.85 25.71 ± 9.39 .167  Pelvic incidence (°) 51.91 ± 11.46 53.92 ± 10.70 51.21 ± 11.07 .374  Lumbar lordosis (°) 30.47 ± 20.09 26.92 ± 25.14 30.71 ± 16.78 .534  PI-LL mismatch (°) 21.03 ± 19.07 25.81 ± 24.41 20.05 ± 19.42 .273  Thoracic kyphosis (°) 25.34 ± 21.37 32.13 ± 18.96 33.41 ± 20.60 .063 Postoperative  Sagittal vertical axis (cm) 3.85 ± 4.36 4.26 ± 4.66 3.20 ± 4.06 .412  Pelvic tilt (°) 22.14 ± 8.47 23.91 ± 9.54 22.20 ± 8.86 .458  Pelvic incidence (°) 52.38 ± 11.21 51.77 ± 10.82 48.91 ± 10.08 .179  Lumbar lordosis (°) 48.39 ± 9.90 47.66 ± 17.17 46.93 ± 8.79 .819  PI-LL mismatch (°) 3.98 ± 11.33 4.11 ± 18.78 1.98 ± 12.79 .682  Thoracic kyphosis (°) 36.98 ± 12.09 40.14 ± 17.22 39.14 ± 17.30 .509 Amount of correction  Sagittal vertical axis (cm) 4.80 ± 5.46 4.32 ± 7.86 5.46 ± 6.78 .653  Lumbar lordosis (°) 17.92 ± 19.31 19.52 ± 25.20 16.21 ± 17.07 .691 PI = pelvic incidence, LL = lumbar lordosis. aValues are expressed as the mean ± SD for continuous variables. bComparison of the 3 tethering groups analyzed with one-way analysis of variance. View Large PJK in Non-tethered vs Tethered Patients Univariate comparison of non-tethered vs all tethered patients for the primary outcome of postoperative PJK is presented in Table 3. The overall PJK rate for non-tethered patients (NT) was significantly higher compared to the tethered cohort (TO + TC): 45.3% (29/64) vs 26.7% (32/120; P = .011). TABLE 3. Postoperative Development of PJK in 184 ASD Patients Dichotomized Based on Use of Polyethylene Tethersa No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 aAnalyzed using Pearson's chi square test with 2 × 2 contingency table. bTether patients with polyethylene tape only (TO) or tape with crosslink (TC) were grouped into a single cohort. View Large TABLE 3. Postoperative Development of PJK in 184 ASD Patients Dichotomized Based on Use of Polyethylene Tethersa No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 No tether Tetherb (NT, n = 64) (TO + TC, n = 120) P Overall PJKa (%) 45.3 26.7 .011 (χ2 = 6.548)  Yes 29 32  No 35 88 aAnalyzed using Pearson's chi square test with 2 × 2 contingency table. bTether patients with polyethylene tape only (TO) or tape with crosslink (TC) were grouped into a single cohort. View Large Secondary Analyses After Stratifying Tethered Patients Based on Use of an Anchoring Crosslink Secondary analyses after stratification of tethered patients into TO and TC subgroups is presented in Tables 4-6. Overall PJK rates were 45.3% (29/64), 34.4% (22/64), and 17.9% (10/56) for NT, TO, and TC patients, respectively (Table 4). The PJK rate for TC tethers was significantly lower than NT at the Bonferroni-adjusted alpha level (P = .001). Based on the OR, the odds of developing PJK was 3.81 times higher for NT compared to TC patients. There was a nonsignificant trend in PJK reduction for TC compared to TO, and TO compared to NT. The rate of revision surgery for PJF was lowest for TC (3.6% [2/56]) compared to TO (9.4% [6/64]) and NT (4.7% [3/64]); however, pairwise comparisons did not reach statistical significance. TABLE 4. PJK and Revision Surgery for PJF: NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 PJK, proximal junctional kyphosis; PJF, proximal junctional failure. aAnalyzed with Pearson's chi square test with 2 × 2 contingency tables. bAnalyzed with Fisher's exact test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .001 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large TABLE 4. PJK and Revision Surgery for PJF: NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 Overall PJKa (%) 45.3 34.4 17.9 .206, .041, .001  Yes 29 22 10 1.597, 4.167, 10.263  No 35 42 46 Revision for PJFb (%) 4.7 9.4 3.6 .492, .281, 1.00  Yes 3 6 2  No 61 58 54 PJK, proximal junctional kyphosis; PJF, proximal junctional failure. aAnalyzed with Pearson's chi square test with 2 × 2 contingency tables. bAnalyzed with Fisher's exact test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .001 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large Table 5 presents pairwise comparisons of the postoperative change in PJA between NT, TO, and TC subgroups. The postoperative change in PJA was calculated using preoperative and final long-cassette lateral radiographs. If revision surgery for PJF was performed, the final PJA was measured using the long-cassette radiograph taken just prior to revision surgery. Postoperative change in PJA for TC was significantly less than both TO and NT using post hoc ANOVA with Bonferroni correction (P = .026 and P = .024, respectively). There was no significant difference in postoperative PJA change for TO compared to NT (P = 1.000). TABLE 5. Postoperative Change in Proximal Junctional Angle (PJA): NT vs TO vs TCa No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* aChange in PJA at last follow-up (or prior to revision for proximal junctional failure) compared to preoperative PJA. bComparison of the 3 tethering groups analyzed with one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. *Post hoc ANOVA with Bonferroni correction TO compared to NT (P = 1.000), TC compared to TO (P = .026), TC compared to NT (P = .024). View Large TABLE 5. Postoperative Change in Proximal Junctional Angle (PJA): NT vs TO vs TCa No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* No tether Tape-only Tape-crosslink (NT n = 64) (TO, n = 64) (TC, n = 56) Pb Change in PJA (°) 12.68 ± 9.34 12.63 ± 9.94 8.18 ± 8.01 .011* aChange in PJA at last follow-up (or prior to revision for proximal junctional failure) compared to preoperative PJA. bComparison of the 3 tethering groups analyzed with one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. *Post hoc ANOVA with Bonferroni correction TO compared to NT (P = 1.000), TC compared to TO (P = .026), TC compared to NT (P = .024). View Large Time-Dependent PJK and Kaplan-Meier Survivorship Analysis Time-to-PJK among NT, TO, and TC subgroups was not significantly different (P = .530). NT, TO, and TC patients developed PJK at 17.02 ± 23.44, 11.59 ± 12.58, and 11.40 ± 14.88 wk, respectively. Comparison of postoperative radiographic follow-up duration was significantly different among all cohorts using post hoc ANOVA with Bonferroni-adjusted analysis (P < .001). NT, TO, and TC patients had total radiographic follow-up of 115.59 ± 50.91, 79.47 ± 41.66, and 41.42 ± 18.48 wk, respectively. Owing to the difference in follow-up duration among NT, TO, and TC subgroups, 3- and 6-mo PJK rates were calculated and Kaplan-Meier analysis was performed. Table 6 presents PJK rates at 3- and 6-mo following multilevel spine instrumentation. At 3 mo, there is a nonsignificant trend in PJK reduction with TC compared to TO and NT. At 6 mo, there is significant PJK reduction for TC compared to NT at the Bonferroni-adjusted alpha level (16.1% compared to 39.1%, P = .005). TABLE 6. Time-Dependent Proximal Junctional Kyphosis (PJK): NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 aPearson's chi square test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .005 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large TABLE 6. Time-Dependent Proximal Junctional Kyphosis (PJK): NT vs TO vs TC No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 No tether Tape-only Tape-crosslink (NT, n = 64) (TO, n = 64) (TC, n = 56) P*P**P*** χ2 3-mo PJKa (%) 29.7 26.6 14.3 .694, .099, .044  Yes 19 17 8 .155, 2.729, 4.063  No 45 47 48 6-mo PJKa (%) 39.1 32.8 16.1 .461, .035, .005  Yes 25 21 9 .543, 4.464, 7.775  No 39 43 47 aPearson's chi square test with 2 × 2 contingency tables. *TO compared to NT. **TC compared to TO. ***TC compared to NT. P = .005 is the only significant pairwise comparison at the Bonferroni-adjusted alpha level. View Large Kaplan-Meier analysis demonstrated a significant difference in the time-dependent rate of PJK (log rank test, P = .030). Using pairwise comparisons, Kaplan-Meier analysis demonstrated a significant reduction in PJK for TC compared to NT (log rank test, P = .010) at the Bonferroni-corrected alpha level (Figure 4). Although not significant, there was a trend suggestive of PJK reduction for TC compared to TO (log rank test, P = .074). There was no significant difference in PJK rates for TO compared to NT (log rank test, P = .337). FIGURE 4. View largeDownload slide Kaplan-Meier curves demonstrate the PJK-free probability among the 3 cohorts as a time-dependent variable in the weeks following multilevel posterior spine instrumentation for adult spinal deformity. Pairwise comparisons (log rank test): TC compared to TO (P = .074), TO compared to NT (P = .337), TC compared to NT (P = .010). PJK = proximal junctional kyphosis. TC = tether-crosslink. TO = tether-only. NT = no tether. FIGURE 4. View largeDownload slide Kaplan-Meier curves demonstrate the PJK-free probability among the 3 cohorts as a time-dependent variable in the weeks following multilevel posterior spine instrumentation for adult spinal deformity. Pairwise comparisons (log rank test): TC compared to TO (P = .074), TO compared to NT (P = .337), TC compared to NT (P = .010). PJK = proximal junctional kyphosis. TC = tether-crosslink. TO = tether-only. NT = no tether. Multivariable Analyses With Patients Dichotomized Based on PJK Variables included in multivariable binary logistic and Cox regression models were age at surgery, female gender, BMI, number of instrumented vertebrae, use of pelvic fixation, use of 3-column osteotomy, use of polyethylene tether (including both TO and TC), preoperative SVA, change in SVA, and amount of LL correction. Results of binary logistic regression are presented in Table 7. Older age at surgery and greater correction of LL were independent predictors of PJK. Use of a polyethylene tether was a significant predictor of reduced PJK risk (OR = 0.422, 95% confidence interval [CI] 0.205-0.868, P = .019). Based on the Hosmer-Lemeshow goodness-of-fit test, the regression was not statistically different from a null model without the explanatory variables, indicating a good fit (P = .963). TABLE 7. Multivariable Analysis of ASD Patients Dichotomized Based on Development of PJK 6 mo After the Index Operationa OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 CI, confidence interval; OR, odds ratio. aExplanatory variables were entered in a stepwise, multivariable model using binary logistic regression. bPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large TABLE 7. Multivariable Analysis of ASD Patients Dichotomized Based on Development of PJK 6 mo After the Index Operationa OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 OR (95% CI) P Age at surgery 1.068 (1.020-1.119) .005 Female 1.975 (0.888-4.392) .095 Body mass index 1.014 (0.953-1.078) .671 Number of instrumented vertebrae 1.050 (0.930-1.185) .434 Pelvic fixation 1.605 (0.303-8.506) .578 3-column osteotomy 1.059 (0.456-2.458) .895 Use of polyethylene tetherb 0.422 (0.205-0.868) .019 Preoperative sagittal vertical axis 1.007 (0.935-1.085) .849 Amount of correction  Sagittal vertical axis 1.050 (0.967-1.141) .246  Lumbar lordosis 1.030 (1.006-1.055) .014 CI, confidence interval; OR, odds ratio. aExplanatory variables were entered in a stepwise, multivariable model using binary logistic regression. bPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large Results of Cox proportional hazards regression are presented in Table 8. Use of a polyethylene tether remained a significant independent factor for avoiding PJK (hazards ratio = 0.532, 95% CI 0.318-0.892, P = .017). Cumulative hazard functions for tethered and nontethered patients are plotted in Figure 5. FIGURE 5. View largeDownload slide A Cox regression model was implemented, and hazard functions were plotted for NT and TO + TC patients. The hazard functions demonstrate a lower cumulative risk of PJK for TO+TC vs NT patients (HR = 0.532, 95% CI 0.318-0.892, P = .017). NT = no tether. TO = tether-only. TC = tether-crosslink. PJK = proximal junctional kyphosis. HR = hazards ratio. CI = confidence interval. FIGURE 5. View largeDownload slide A Cox regression model was implemented, and hazard functions were plotted for NT and TO + TC patients. The hazard functions demonstrate a lower cumulative risk of PJK for TO+TC vs NT patients (HR = 0.532, 95% CI 0.318-0.892, P = .017). NT = no tether. TO = tether-only. TC = tether-crosslink. PJK = proximal junctional kyphosis. HR = hazards ratio. CI = confidence interval. TABLE 8. Cox Proportional Hazards Regression for ASD Patients and Postoperative PJK HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 CI, confidence interval; HR, hazards ratio. aPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large TABLE 8. Cox Proportional Hazards Regression for ASD Patients and Postoperative PJK HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 HR (95% CI) P Age at surgery 1.051 (1.016-1.088) .004 Female 1.570 (0.855-2.883) .145 Body mass index 1.019 (0.976-1.063) .395 Number of instrumented vertebrae 1.035 (0.948-1.131) .438 Pelvic fixation 1.213 (0.365-4.023) .753 3-column osteotomy 0.949 (0.515-1.752) .868 Use of polyethylene tethera 0.532 (0.318-0.892) .017 Preoperative sagittal vertical axis 0.990 (0.935-1.048) .737 Amount of correction  Sagittal vertical axis 1.030 (0.966-1.099) .363  Lumbar lordosis 1.022 (1.005-1.039) .013 CI, confidence interval; HR, hazards ratio. aPolyethylene tape only and tape with crosslink tethers were grouped into a single cohort. View Large DISCUSSION PJK is common following multilevel spine instrumentation for ASD, with a reported incidence of 11.0% to 52.9%.1-8 The exact pathophysiological mechanism of PJK is unclear; however, most authors agree it is likely multifactorial.20,21 PJK may in part be secondary to an iatrogenic effect of altered spinal biomechanics, which produce pathologic flexion-loading forces at the proximal fusion segment. This includes compensatory changes following operative correction of thoracic kyphosis and sagittal malalignment, which are further exacerbated by posterior tension band disruption and the creation of a moment arm at the proximal segment of long, rigid fusion constructs.13,22,23 PJF is considered a distinct entity in the spectrum of PJK that involves mechanical failure at the UIV or just above and/or proximal junctional posterior discoligamentous failure.20,24 The most common modes of failure in the upper thoracic spine compared to the thoracolumbar region are soft tissue disruption and mechanical fracture, respectively.25 Patients with PJF may present with progressively worsening pain, focal neurological deficits, and ambulatory difficulty—resulting in significantly worse standardized outcome scores.16 Therefore, PJK and PJF prevention is, and should be, among the principal goals for optimizing clinical outcomes after spinal deformity surgery. Posterior polyethylene junctional tethers at the UIV in multilevel spine instrumentation may be an effective anti-PJK technique. Bess and colleagues,26 using finite element analysis, demonstrated more gradual changes in segmental range-of-motion and reduction in proximal segment intradiscal pressures, pedicle screw loads, and posterior ligament complex forces with increasing number of posterior tethers. The authors hypothesized that posterior polyethylene tethers may reduce biomechanical risk factors for PJK. A literature search for posterior tethers, and their clinical use as an anti-PJK device after spine surgery, produced only a single clinical report by Zaghloul and colleagues.27 The authors reported no case of PJK in 18 patients with polyethylene tape–based strap stabilization at the proximal end of fusion constructs. However, this relatively small case series, with average 11.9 mo of follow-up, lacked specific inclusion and exclusion criteria for spinal deformity surgery or long-segment fusion—12 patients had fusion constructs of 6 or less motion segments, and 4 patients were fused at a single segment.27 We present the first large clinical study of posterior polyethylene junctional tethers as an anti-PJK device for multilevel spine instrumentation (>6 fused segments) in ASD patients with scoliosis and/or global sagittal malalignment. Our results show that junctional tethers (both TO and TC) are a significant negative predictor of PJK, and that there are significantly reduced overall crude and time-dependent PJK rates when polyethylene tape is anchored to a crosslink. Additionally, TC tethers significantly reduced the postoperative change in PJA at last follow-up. Pairwise comparisons showed no significant effect of polyethylene tethering without utilization of an anchoring crosslink, although trends still suggested a reduction in PJK. To explain this difference, we hypothesize that anchoring polyethylene tape to a crosslink between UIV-1 and UIV-2, and then distracting the crosslink inferiorly, produces additional tension and increased junctional support compared to polyethylene tape-only tethers. In this study, the decision to use junctional tethers was not randomized; instead, the 2 senior authors (CIS, JSS) started utilizing junctional tethers from September 2013 and January 2015 onwards, respectively. No other change in surgical technique (vertebral augmentation, multilevel stabilization screws, transverse process hooks at the UIV, proximal transition rods of reduced diameter, hybrid constructs)3,17,18 was identified that may confound PJK rates. However, we still attempted to limit potential bias from practice variability and nonrandomization by controlling for baseline patient demographic, surgical data, as well as pre- and postoperative sagittal plane radiographic parameters in multivariable models (Tables 7 and 8) and performing time-dependent analyses (Figures 4 and 5). In addition, the confounding effect of possible practice variability was limited by our decision not to extend the inclusion start date to an earlier time. This produced relatively balanced cohorts with 64 NT, 64 TO, and 56 TC patients. Limitations The limitations of this study include its nonblinded retrospective design, the experience level and technical ability of the surgeons at a single center, the different follow-up duration among cohorts, and the relatively short minimum follow-up duration of 3 mo. To account for differences in follow-up, we calculated 3- and 6-mo PJK rates, performed multivariable regression using 6-mo PJK rates, implemented a Cox regression model, and performed a life-table survival analysis (Kaplan-Meier estimation). Kaplan-Meier analysis was congruent with the overall crude rates and demonstrated a significant time-dependent reduction in PJK for TC compared to NT (log rank test, P = .010) at the Bonferroni-corrected alpha level (Figure 4). Although minimum radiographic follow-up of 3 mo may seem relatively short, prior studies have reported up to 76% of PJK occurs within 3 mo postoperatively, and that new PJK after 6 mo may be rare.2,3,22,28,29 Also, Kim and colleagues2 showed that the average PJA increase at 2 mo postoperation accounts for the majority of total PJA increase at 5 yr postoperation. In our study, 93% (171/184) of patients had at least 6 mo of follow-up, and the vast majority of patients who developed PJK did so within 3 mo of the index operation. Although TC tethers significantly reduced PJK, rates of revision surgery for PJF were comparable across all cohorts. However, the small number of patients who underwent revision surgery limits meaningful comparison of revision rates; therefore, a larger tether study with longer follow-up duration may be warranted. We are currently accruing more patients in a prospectively maintained database, with longer follow-up and clinical outcome scores, to further elucidate the clinical impact of posterior polyethylene tethers in spinal deformity surgery. Additionally, future studies are necessary to determine if our findings are generalizable to other spine surgery patients without ASD, particularly those without scoliosis and/or global sagittal malalignment, or who have 6 or less fused motion segments. CONCLUSION PJK is a common problem after multilevel spine instrumentation for spinal deformity correction, and in a subset of patients, may require surgical revision. Posterior junctional tethers, consisting of polyethylene tape anchored to a crosslink at the superior aspect of fusion constructs, may represent a novel anti-PJK device for ASD patients. In this study, use of junctional tethers for long-segment posterior fusion for ASD significantly reduced the occurrence of PJK. This difference was progressive from NT to TO to TC cohorts, but only reached statistical significance on pairwise comparisons for NT vs TC. These findings suggest potential benefit of junctional tethers to reduce PJK, and that future prospective studies with longer-term follow-up are warranted. Disclosures Dr Bess receives research support from K2M, Innovasis, Nuvasive, Medtronic, DePuy Synthes, Stryker, and Zimmer-Biomet; is a consultant for K2M and Allosource; and holds patents with K2M and Innovasis. Mr Line, BSME, is a consultant for ISSGF. Dr Ames is a consultant for Medtronic, DePuy Synthes, and Stryker; receives royalties from Zimmer-Biomet and Stryker; has research support from UCSF; and holds a patent with Fish & Richardson PC. Dr Schwab is a consultant for Zimmer-Biomet, K2M, Nuvasive, Medicrea, and MSD; receives research support from SRS, AOSpine, DePuy Spine Synthesis, and ISSGF; and is a stock holder in Nemaris Inc. Dr Lafage is a consultant for Nemaris Inc, Nuvasive, Medicrea, and DePuy Synthes; is a stock holder in Nemaris Inc; and receives non-study-related support from SRS, NIH, DePuy Synthes, and ISSGF. Dr Shaffrey is a consultant for Medtronic, Nuvasive, Zimmer-Biomet, and K2M; receives royalties from Medtronic, Nuvasive, and Zimmer-Biomet; is a stock holder in Nuvasive; and has grants from NIH, DOD, and NACTN. Dr Smith receives royalties from Zimmer Biomet, is a consultant for Zimmer Biomet, Cerapedics, Nuvasive, K2M, and AlloSource; receives honoraria from Zimmer Biomet, Nuvasive, and K2M; has research support from DePuy Synthes, and ISSGF; and receives fellowship support from NREF and AOSpine. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Glattes RC , Bridwell KH , Lenke LG , Kim YJ , Rinella A , Edwards C 2nd . Proximal junctional kyphosis in adult spinal deformity following long instrumented posterior spinal fusion . Spine (Phila Pa 1976) . 2005 ; 30 ( 14 ): 1643 - 1649 . Google Scholar CrossRef Search ADS PubMed 2. Kim YJ , Bridwell KH , Lenke LG , Glattes CR , Rhim S , Cheh G . Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion . Spine (Phila Pa 1976) . 2008 ; 33 ( 20 ): 2179 - 2184 . Google Scholar CrossRef Search ADS PubMed 3. Ha Y , Maruo K , Racine L et al. Proximal junctional kyphosis and clinical outcomes in adult spinal deformity surgery with fusion from the thoracic spine to the sacrum: a comparison of proximal and distal upper instrumented vertebrae . J Neurosurg Spine . 2013 ; 19 ( 3 ): 360 - 369 . Google Scholar CrossRef Search ADS PubMed 4. Park SJ , Lee CS , Chung SS , Lee JY , Kang SS , Park SH . Different risk factors of proximal junctional kyphosis and proximal junctional failure following long instrumented fusion to the sacrum for adult spinal deformity: survivorship analysis of 160 patients . Neurosurgery . 2016 ; 80 ( 2 ): 279 - 286 . 5. Smith JS , Klineberg E , Lafage V et al. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery . J Neurosurg Spine . 2016 ; 25 ( 1 ): 1 - 14 . Google Scholar CrossRef Search ADS PubMed 6. Mummaneni PV , Park P , Fu KM et al. Does minimally invasive percutaneous posterior instrumentation reduce risk of proximal junctional kyphosis in adult spinal deformity surgery? a propensity-matched cohort analysis . Neurosurgery . 2016 ; 78 ( 1 ): 101 - 108 . Google Scholar CrossRef Search ADS PubMed 7. Reames DL , Kasliwal MK , Smith JS , Hamilton DK , Arlet V , Shaffrey CI . Time to development, clinical and radiographic characteristics, and management of proximal junctional kyphosis following adult thoracolumbar instrumented fusion for spinal deformity . J Spinal Disord Tech . 2015 ; 28 ( 2 ): E106 - E114 . Google Scholar CrossRef Search ADS PubMed 8. Yan C , Li Y , Yu Z . Prevalence and consequences of the proximal junctional kyphosis after spinal deformity surgery . Medicine (Baltimore) . 2016 ; 95 ( 20 ): e3471 . Google Scholar CrossRef Search ADS PubMed 9. Liu FY , Wang T , Yang SD , Wang H , Yang DL , Ding WY . Incidence and risk factors for proximal junctional kyphosis: a meta-analysis . Eur Spine J . 2016 ; 25 ( 8 ): 2376 - 2383 . Google Scholar CrossRef Search ADS PubMed 10. Annis P , Lawrence BD , Spiker WR et al. Predictive factors for acute proximal junctional failure after adult deformity surgery with upper instrumented vertebrae in the thoracolumbar spine . Evid-Based Spine Care J . 2014 ; 5 ( 2 ): 160 - 162 . Google Scholar CrossRef Search ADS PubMed 11. Lee GA , Betz RR , Clements DH 3rd , Huss GK . Proximal kyphosis after posterior spinal fusion in patients with idiopathic scoliosis . Spine (Phila Pa 1976) . 1999 ; 24 ( 8 ): 795 - 799 . Google Scholar CrossRef Search ADS PubMed 12. Kim YJ , Lenke LG , Bridwell KH et al. Proximal junctional kyphosis in adolescent idiopathic scoliosis after 3 different types of posterior segmental spinal instrumentation and fusions: incidence and risk factor analysis of 410 cases . Spine (Phila Pa 1976) . 2007 ; 32 ( 24 ): 2731 - 2738 . Google Scholar CrossRef Search ADS PubMed 13. Rhee JM , Bridwell KH , Won DS , Lenke LG , Chotigavanichaya C , Hanson DS . Sagittal plane analysis of adolescent idiopathic scoliosis: the effect of anterior versus posterior instrumentation . Spine (Phila Pa 1976) . 2002 ; 27 ( 21 ): 2350 - 2356 . Google Scholar CrossRef Search ADS PubMed 14. Kim YJ , Bridwell KH , Lenke LG , Kim J , Cho SK . Proximal junctional kyphosis in adolescent idiopathic scoliosis following segmental posterior spinal instrumentation and fusion: minimum 5-year follow-up . Spine (Phila Pa 1976) . 2005 ; 30 ( 18 ): 2045 - 2050 . Google Scholar CrossRef Search ADS PubMed 15. Helgeson MD , Shah SA , Newton PO et al. Evaluation of proximal junctional kyphosis in adolescent idiopathic scoliosis following pedicle screw, hook, or hybrid instrumentation . Spine (Phila Pa 1976) . 2010 ; 35 ( 2 ): 177 - 181 . Google Scholar CrossRef Search ADS PubMed 16. Lau D , Funao H , Clark AJ et al. The clinical correlation of the hart-issg proximal junctional kyphosis severity scale with health-related quality-of-life outcomes and need for revision surgery . Spine (Phila Pa 1976) . 2016 ; 41 ( 3 ): 213 - 223 . Google Scholar CrossRef Search ADS PubMed 17. Aubin CE , Cammarata M , Wang X , Mac-Thiong JM . Instrumentation strategies to reduce the risks of proximal junctional kyphosis in adult scoliosis: a detailed biomechanical analysis . Spine Deform . 2015 ; 3 ( 3 ): 211 - 218 . Google Scholar CrossRef Search ADS PubMed 18. Sandquist L , Carr D , Tong D , Gonda R , Soo TM . Preventing proximal junctional failure in long segmental instrumented cases of adult degenerative scoliosis using a multilevel stabilization screw technique . Surg Neurol Int . 2015 ; 6 ( 1 ): 112 . Google Scholar CrossRef Search ADS PubMed 19. Steyerberg EW , Vickers AJ , Cook NR et al. Assessing the performance of prediction models: a framework for traditional and novel measures . Epidemiology . 2010 ; 21 ( 1 ): 128 - 138 . Google Scholar CrossRef Search ADS PubMed 20. Nguyen NL , Kong CY , Hart RA . Proximal junctional kyphosis and failure-diagnosis, prevention, and treatment . Curr Rev Musculoskelet Med . 2016 ; 9 ( 3 ): 299 - 308 . Google Scholar CrossRef Search ADS PubMed 21. Arlet V , Aebi M . Junctional spinal disorders in operated adult spinal deformities: present understanding and future perspectives . Eur Spine J . 2013 ; 22 ( S2 ): 276 - 295 . Google Scholar CrossRef Search ADS 22. Yagi M , King AB , Boachie-Adjei O . Incidence, risk factors, and natural course of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis. Minimum 5 years of follow-up . Spine (Phila Pa 1976) . 2012 ; 37 ( 17 ): 1479 - 1489 . Google Scholar CrossRef Search ADS PubMed 23. Cammarata M , Aubin CE , Wang X , Mac-Thiong JM . Biomechanical risk factors for proximal junctional kyphosis: a detailed numerical analysis of surgical instrumentation variables . Spine (Phila Pa 1976) . 2014 ; 39 ( 8 ): E500 - E507 . Google Scholar CrossRef Search ADS PubMed 24. Fradet L , Wang X , Lenke LG , Aubin CE . Biomechanical analysis of proximal junctional failure following adult spinal instrumentation using a comprehensive hybrid modeling approach . Clin Biomech (Bristol, Avon) . 2016 ; 39 : 122 - 128 . Google Scholar CrossRef Search ADS PubMed 25. Hostin R , McCarthy I , O’Brien M et al. Incidence, mode, and location of acute proximal junctional failures after surgical treatment of adult spinal deformity . Spine (Phila Pa 1976) . 2013 ; 38 ( 12 ): 1008 - 1015 . Google Scholar CrossRef Search ADS PubMed 26. Bess S , Harris JE , Turner AW et al. The effect of posterior polyester tethers on the biomechanics of proximal junctional kyphosis: a finite element analysis . J Neurosurg Spine . 2017 ; 26 ( 1 ): 125 - 133 . Google Scholar CrossRef Search ADS PubMed 27. Zaghloul KM , Matoian BJ , Denardin NB , Patel VV . Preventing proximal adjacent level kyphosis with strap stabilization . Orthopedics . 2016 ; 39 ( 4 ): e794 - e799 . Google Scholar CrossRef Search ADS PubMed 28. Yagi M , Akilah KB , Boachie-Adjei O . Incidence, risk factors and classification of proximal junctional kyphosis: surgical outcomes review of adult idiopathic scoliosis . Spine (Phila Pa 1976) . 2011 ; 36 ( 1 ): E60 - E68 . Google Scholar CrossRef Search ADS PubMed 29. Lau D , Clark AJ , Scheer JK et al. Proximal junctional kyphosis and failure after spinal deformity surgery: a systematic review of the literature as a background to classification development . Spine (Phila Pa 1976) . 2014 ; 39 ( 25 ): 2093 - 2102 . Google Scholar CrossRef Search ADS PubMed COMMENTS The authors present the results of their retrospective, single-center, comparative cohort study of 184 patients who underwent surgery for adult spinal deformity between 2012 and 2016. During that period, the authors' surgical techniques at the proximal junction changed from not using a polyethylene tether, to placing a tether that was secured to the spinous processes, and finally to placing a tether that was secured to a crosslink. In their primary analysis, they found that the incidence of PJK without a tether was 45% and with a tether, using either technique, it was 27%. Secondary analyses were generally consistent with the primary finding. The crude odds ratio (OR) for developing PJK with a tether was 0.44; controlling for factors such as BMI, pelvic fixation, and amount of correction using multivariable regression, the odds ratio for PJK with the use of a tether was 0.53. This study includes 3 cohorts made up of patients treated by different techniques as the surgical practices at a single center evolved over time. The 2 surgeons may have changed their techniques in response to a perceived problem with PJK, perhaps as the result of a particularly high incidence of PJK in the period prior to instituting the technical change. There is evidence that this may, in fact, be the case, as the “no tether” cohort had a 45% incidence of PJK, which is toward the upper end of the range reported in the literature.1 This raises the possibility that some of the observed difference in the incidence of PJK is due to regression to the mean. If aspects of the surgical technique (reduced application of corrective forces near the UIV, more limited proximal soft tissue dissection) or decision-making (selection of the construct levels) were changed in response to a particularly high incidence of PJK, one would expect the incidence to subsequently decline independent of other changes. Assessing “statistical significance” is only 1 part of analyzing a study; one should also critically evaluate the estimate of the effect size, the degree of uncertainty regarding the estimate, and how the study design may affect the estimate.2 In this study, the absolute reduction of 18% in the incidence of PJK with tether use is very likely to be an overestimate of the effect size. Regression to the mean and other changes in management, as discussed above, are 2 likely explanations for some of the observed difference in the incidence of PJK between the subgroups in this case series. These considerations must be weighed against the potential risks of the intervention. In this case the procedure likely adds little risk to the procedure itself and probably does not put the patient at increased risk of delayed adverse events, although these aspects need further study. Even if the effect size of 0.18 is exaggerated by a factor of 3, with a standard error of approximately 0.05, it is unlikely (less than 1% probability) that the incidence of PJK with a tether is actually higher than without a tether.3 If revision for PJK is necessary, the presence of a tether probably would not interfere with the subsequent surgery or alter its outcome. This technique may therefore be an option that spinal deformity surgeons should consider but with more modest expectations for outcome than reported in this paper. Follow-up reports and studies by other groups should be performed help to further refine our knowledge of the effectiveness and safety of this procedure. Peter D. Angevine New York 1. Nguyen N-L Kong CY Hart RA . Proximal junctional kyphosis and failure- diagnosis, prevention, and treatment . Curr Rev Musculoskelet Med . 2016 ; 9 ( 3 ): 299 - 308 . Google Scholar CrossRef Search ADS PubMed 2. Leek J McShane BB Gelman A Colquhoun D Nuijten MB Goodman SN . Five ways to fix statistics . Nature . 2017 ; 551 ( 7682 ): 557 - 559 . Google Scholar CrossRef Search ADS PubMed 3. Gelman A Carlin J . Beyond power calculations: assessing type S (sign) and type M (magnitude) errors . Perspect Psychol Sci . 2014 ; 9 ( 6 ): 641 - 651 . Google Scholar CrossRef Search ADS PubMed Proximal junctional kyphosis (PJK) is a common entity that develops after long-segment fusions for adult spinal deformity (ASD). The authors propose an interesting technique to reduce the incidence of PJK by incorporating junctional tether at the proximal junction. For this purpose, the authors retrospectively reviewed the data and radiographic images of 184 patients that underwent posterior or anteroposterior fixation at a single institute for ASD between 2012 and 2016. The authors compared between 3 different techniques: no proximal tether, 64 patients; polyethylene tether in isolation, 64 patients; and polyethylene tether in conjunction with a crosslink, 56 patients. After a mean follow-up of 20 months, the authors found that the rate of PJK was significantly higher in the nontethered group compared to the tethered group (45% vs 27%, P = .011). Based on Kaplan-Meier survivorship analysis, there was a significant reduction in PJK rate for tethered group vs nontethered group (log-rank test, P = .010). The revision surgery rate for proximal junctional failure (PJF) was lower in the tethered group as well, although this difference did not reach a statistical significance. This article is well-written and novel in terms of using a new, safe, and effective technique to improve the surgical outcomes of long-segment fusions in ASD. The authors should be commended for this important addition to our understanding of the junctional tethers and their effect on PJK. Moving forward, it will be interesting to utilize the findings of this study in future prospective randomized studies, with a longer follow-up duration, to investigate long-term outcomes, such as PJF, associated with junctional tethers. Seba Ramhmdani Ali Bydon Baltimore, Maryland The authors have presented initial data on the novel use of polyethelene tethers placed 1 level past the proximal end of a fusion construct in an attempt to help prevent proximal junctional kyphosis (PJK). The idea of stress sharing is not new, but this particular use of an “elastic band” is of interest to anybody who has had to deal with PJK. As with many studies such as this, there are potential systemic biases in how the 2 cohorts were created and in surgical technique that may have adapted over time. Though this paper offers an attractive adjunct device to implant with many fusions, it still needs more study and investigation to justify its use. John Chi Boston, Massachusetts Copyright © 2018 by the Congress of Neurological Surgeons

Journal

Operative NeurosurgeryOxford University Press

Published: Apr 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off