7th International Congress on Early Onset Scoliosis and Growing Spine (ICEOS)
November 21-22, 2013
Rancho Bernardo, CA
FREE PAPERS
Disclosure Key:
A. Grants/Research Support
B. Consulting Fees
C. Speakers’ Bureau
D. Ownership Interest/Shareholder
E. Salary
F. Royalty /Patent Holder
Free Paper #22: Proximal Hooks In Growing Rod Systems: Can They Prevent Proximal Junctional Failures?
Francesco Lolli, MD; Konstantinos Martikos, MD; Elena Maredi, MD; Mario Di Silvestre, MD; Francesco Vommaro, MD; Andrea Baioni, MD; Tiziana Greggi, MD
Introduction: Dual growing rod systems are routinely used for the treatment of early and late onset scoliosis. However, the complications incidence is still high, especially in terms of proximal anchors mobilization. The aim of our study is to compare the use of hooks and pedicle screws as proximal anchors to prevent proximal junctional failures.
Methods: We retrospectively reviewed 21 patients, affected by early onset scoliosis and surgically treated with growing rod system between 2006 and 2011 (minimum follow up 2 years). The etiologies were: idiopathic scoliosis (9 cases), kyphosis in Morquio disease (1) Pott’s disease (1), congenital scoliosis (3), scoliosis in Escobar syndrome (1), NF1 (2), arthrogryposis (1), Prader Willi (1), trisomy 8 (1) and myopathy (1).
Results: There were 14 females and 7 males, with a mean age of 7.6 years (range 5 to 11). All cases had distal anchors pedicle screws. Proximal pedicle screws were used in 7 patients (all dual rod constructs) and proximal hooks in 14 (13 dual rod, 1 single rod). At a mean follow up of 40 months (minimum 24 months), the main scoliosis was corrected from 60.9° to 36.0° with proximal pedicle screws (mean correction 40.9%) and from 51.9° to 24.2° with proximal hooks (53.4%); thoracic kyphosis was corrected from 52.1° to 45.8° and from 51.3° to 24.3°, respectively. Proximal anchors mobilization occurred in 6 patients (28.6%), 5 in case of pedicle screws (71.4%) and 1 in case of hooks (7.1%), always requiring revision surgery (in 3 cases performed during lengthening procedure). Moreover, the only case in which hooks mobilization occurred was a single rod construct. The proximal junctional disease was always symptomatic, with subcutaneous rod prominence, pain and skin suffering. No neurological complications occurred.
Conclusions: These results showed that hooks used as proximal anchors seem to have a protective role versus proximal junctional failures, if compared to pedicle screws (7.1% vs 71.4%; p<0.05). Obviously, other parameters can play a role, such as etiology, upper thoracic kyphosis, sagittal balance, posterior ligaments preservation, and further studies are required.
Disclosures: F. Lolli: None. K. Martikos: None. E. Maredi: None. M. Di Silvestre: None. F. Vommaro: None. A. Baioni: None. T. Greggi: None.
Free Paper #23: Proximal Junctional Kyphosis Measurement Variability in Patients with Growing Rods
Kody K. Barrett, BA; Lindsay M. Andras, MD; Paul D. Choi, MD; Vernon T. Tolo, MD; David Skaggs
Introduction: The amount of variability that exists within and between observers in the measurement of proximal junctional kyphosis (PJK) in patients with growing rods is unknown.
Methods: Four pediatric orthopaedic spine surgeons measured the PJK in ten patients with growing rod instrumentation using 2 methods. In method 1, measurements were made from the inferior endplate of the upper instrumented vertebrae (UIV) to the superior endplate of one level above the UIV. In method 2, measurements were made from the inferior endplate 2 levels below the UIV to the superior endplate two levels above the UIV. These measurements were repeated one week later.
Results: Method 1 had an intraobserver variability of ± 13.2° and interobserver variability of ± 21.6°, while method 2 had an intraobserver variability of ± 18.3° and interobserver variability of ± 20.7°. The intraclass correlation coefficient (ICC) was calculated for each method, with method 1 having an ICC of 0.311 (0.085-0.576) while method 2 showed an ICC of 0.822 (0.685-0.916). In comparing the reliability of each method from the first measurement to the second measurement, the ICC for method 1 was 0.728 (0.539-0.847) and for method 2 was 0.840 (0.716-0.913).
Conclusions: Overall, both interobserver and intraobserver variability was high, with ±15° of error to be expected in each. This amount of variability makes it challenging to evaluate PJK in the setting of growing rod instrumentation on the basis of radiographs alone.
Disclosures: K.K. Barrett: None. L.M. Andras: None. P.D. Choi: B; Stryker, Integra. C; Stryker. V.T. Tolo: C; Editor for the Journal of Bone and Joint Surgery. D. Skaggs: A; Institutional support from Medtronic, POSNA and SRS; both paid to Columbia University. B; Biomet; Medtronic; BeachBody LLC. C; Biomet; Medtronic; Stryker. F; Biomet; Medtronic/Biomet (osteotome).
Free Paper #24: Reliability of Proximal Junctional Kyphosis Measurements for Early Onset Scoliosis
Ammar Al Khudairy; Luke Gauthier; Jacob Matz; John Heflin; Ron El-Hawary
Introduction: Distraction-based surgery has been associated with the development of proximal junctional kyphosis (PJK). The development of PJK may lead to pre-mature implant failure and may affect level of upper instrumented vertebrae (UIV); however, PJK is not clearly defined in the literature. Recent studies have used various definitions resulting in widely varying rates of PJK. As a first step towards defining risk factors that may lead to clinically significant PJK, an evaluation of definitions of PJK should be performed. Our purpose was to use three recently used definitions to report the rates of PJK for a group of children treated with growth friendly surgery and to define the variability associated with these measurements.
Methods: Radiographs were analyzed for 36 children with EOS who were treated with posterior distraction-based surgery. Three definitions were accessed. Definition A: angle between the caudle endplate of UIV to the cephalad endplate 2 vertebrae above UIV ≥10° and at least 10° more than pre-op, B: 2 levels below UIV to 2 above UIV >10° and at least 10° more than pre-op, C: UIV to 1 level above UIV > 20°. Two observers each measured the radiographs 2 weeks apart. Wilcoxian Signed Ranks Test and Kappa analysis were used to assess agreement.
Results: At 2-year follow-up, rates of PJK measured using each definition were: A=8%, B=33%, C=3%, with definition B being significantly different than A and C. The rates of PJK were similar between rib and spine-based groups. Inter-observer agreement was moderate for A (50%) and poor for B and C (41-47%). Intra-observer agreement was moderate for C (68%) and poor for A and B (31-47%).
Conclusions: Different definitions for PJK resulted in different rates of PJK (3%-33%). Poor inter-rater agreement was found except for definition A (moderate) and poor intra-rater agreement was found except for definition C (moderate).
Disclosures: A. Al Khudairy: None. L. Gauthier: None. J. Matz: None. J. Heflin: None. R. El-Hawary: A; Depuy-Synthes, Medtronic. B; Depuy-Synthes, Medtronic.
Free Paper #25: Radiographic Analysis of Cervicothoracic and Spinopelvic Changes After Dual Growing Rod Surgery for Early Onset Scoliosis
Daniel Barba, MD; Gregory M. Mundis Jr, MD; Burt Yaszay, MD; Jeff Pawelek, BS; Stacie Nguyen, MPH; Nima Kabirian, MD; Suken Shah, MD; John B. Emans, MD; Charles E. Johnston, MD; Behrooz A. Akbarnia, MD; Growing Spine Study Group
Introduction: Recent studies on growing rod (GR) surgery have focused on coronal plane improvement and avoidance of PJK while global changes in the sagittal plane remain poorly understood. We specifically aimed to investigate the sagittal reciprocal changes of cervicothoracic (C-T) and spinopelvic parameters.
Methods: A multicenter database was queried and identified 94 patients who met the inclusion criteria: <10 years of age, ambulatory, dual GR surgery and minimum 2-year follow up. 33 of 94 patients had adequate x-rays for analysis. Pre-op, post-op, and 2-year full spine lateral x-rays were analyzed. Cervical lordosis, T1 slope, T1 thoracic inclination, thoracic kyphosis (T2-T12 and T5-T12), lumbar lordosis, pelvicparameters, C2-S1 SVA, and C7-S1 SVA were measured by a single observer. Thoracic inlet angle (TIA), described by Lee S, et al., was used as an equivalent to pelvic incidence of thoracolumbar spine.
Results: 33 patients with a mean age of 5.2 years (range, 0.7 to 10.6) at index surgery were included. Etiologies included congenital (n=5), idiopathic (n=13), neuromuscular (n=2) and syndromic (n=13). Mean length of follow-up was 4.5 years (range, 1.9 to 12.3) and mean number of lengthenings was 5.8 (range, 3 to 14). There was a large variation in cervical lordosis and T1 slope across all time points (Table). TIA was constant from pre-op to 2-year. There were significant changes in T2-T12 kyphosis and T5-T12 kyphosis between all time points; both parameters decreased after index surgery (40° to 32°; 36° to 24°, respectively) (p<0.05) and increased during the lengthening period (32° to 39°; 24° to 30°, respectively) (p<0.05). C2-S1 SVA and C7-S1 SVA significantly improved during the lengthening period (39 mm to 22 mm; 27 mm to 10 mm, respectively)(p<0.05). There were no significant changes in C2-C7 SVA. There was a significant decrease in PI-LL during the lengthening period (-5 to -10; p<0.05). There were no significant changes in the other spinopelvic parameters.
Conclusion: There is a growing interest in how distracted-based surgery effects sagittal alignment in EOS patients. Our findings suggest that patients do not experience unwanted reciprocal changes of the C-T spine and key spinopelvic parameters (SVA and PI-LL) may improve after two years of treatment.
Disclosures: D. Barba: None. G.M. Mundis Jr: A; Nuvasive, Depuy, OREF. B; K2M, Nuvasive. C; Nuvasive, K2M. F; K2M, Nuvasive. B. Yaszay: A; DePuy, Harms Study Group. B; K2M, Orthopaediatrics, Medtronic. C; DePuy, K2M. F; Orthopaediatrics, K2M. J. Pawelek: None. S. Nguyen: None. N. Kabirian: None. S. Shah: A; DePuy Spine. B; DePuy Spine, K Spine, Orthopaediatrics. D; Globus Medical. F; Arthrex, DePuy Spine. J.B. Emans: B; Medtronic, Synthes. F; Synthes. C.E. Johnston: F; Saunders/Mosby-Elsevier, Medtronic. B.A. Akbarnia: A; DePuy Spine, Nuvasive. B; Nuvasive, K2M, Ellipse, K Spine. D; Nuvasive, Ellipse, K Spine, Nocimed. F; DePuy Spine. G. Study Group: A; Growing Spine Foundation.
Free Paper #26: Growing Spinal Systems And Early Onset Deformities: Is Hyperkhyphosis A Contraindication?
Tiziana Greggi, MD; Francesco Lolli, MD; Elena Maredi, MD; Konstantinos Martikos, MD; Mario Di Silvestre, MD; Andrea Baioni, MD
Introduction: Growing spinal systems are actually used for the treatment of early onset scoliosis. However, they are distraction based systems, so the hyperkyphosis is not considered as a correct indication. Aim of our study is to show if those systems can be effectively used in the treatment of spinal kyphotic deformities.
Methods: We retrospectively reviewed 16 paediatric patients affected by kyphotic spinal deformity (T3-T12 kyphosys > 60°) surgically treated with Growing Rod or VEPTR-like systems from 2006 to 2011. There were 8 males and 8 females, with a mean age of 7 years (range, 4 to 11). The aetiology was: idiopathic scoliosis (5 cases), kyphosis in Morquio disease (1) and in Pott disease (1), congenital scoliosis (3), trisomy 8 (1), Escobar syndrome (1), Prader Willi (1), spondylocostal dysplasia (1), arthrogryposis (2). Dual growing rod was implanted in 9 cases, VEPTR in 9 (always rib to spine construct).
Pre-operative main thoracic scoliosis averaged 64° (range, 10° to 100°), lumbar scoliosis 55° and thoracic kyphosis 71° (60° to 90°), 67° in patients treated with growing rod and 77° for those treated with VEPTR.
Results: Mean follow-up was 30 months (range, 18 to 67). After the first surgery, thoracic kyphosis was corrected from a mean value of 71° (range, 60° to 90°) to 52° (21° to 80°) (p<0.05); in cases treated with growing rod, kyphosis was corrected from 67° to 44° (p<0.05), in cases treated with VEPTR from 77° to 60° (p<0.05).
At final follow up, after 31 lengthening procedures, a loss of correction occurred on sagittal plane: thoracic kyphosis increased from 52° to 59° (p<0.05); in case of growing rod, from 44° to 50° (p<0.05), in case of VEPTR from 60° to 70° (p<0.05).
15 complications occurred in 8 patients, requiring revision surgery in 7 (4 cases of proximal junctional failure).
Conclusions: Our results showed that growing spinal implants can be safely used in the treatment of kyphotic deformities. Due to distraction procedures, a loss of correction on sagittal plane is commonly observed at follow up. The final result is mostly related to kyphosis correction obtained during first surgery: growing rods, through cantilever manoeuvre, seem to grant a better sagittal plane restoration compared to VEPTR.
Disclosures: T. Greggi: None. F. Lolli: None. E. Maredi: None. K. Martikos: None. M. Di Silvestre: None. A. Baioni: None.
Free Paper #27: Do Thoracolumbar/lumbar Curves Respond Differently To Growing Rod Surgery Compared To Thoracic Curves?
Navid R. Arandi, BS; Jeff Pawelek, BS; Nima Kabirian, MD; George H. Thompson, MD; John B. Emans; John M. Flynn, MD; John P. Dormans, MD; Behrooz A. Akbarnia, MD; Growing Spine Study Group
Introduction: Previous studies have shown the benefits and efficacy of growing rod (GR) surgery for progressive early onset scoliosis (EOS). The purpose of this study was to compare the radiographic response of primary thoracic versus primary lumbar curves following GR surgery.
Methods: A multicenter EOS database query identified 175 patients who met the following inclusion criteria: non-congenital etiology, GR surgery, ≤ 10 years old at index surgery, minimum 2-year follow up and at least 3 lengthenings. Patients were categorized into two groups based on the SRS definition of the anatomical location of primary curves: Group 1 included thoracic apices (T2 to T11/12 disc) and Group 2 included thoracolumbar (T12-L1) and lumbar (L1/2 disc to L4) apices. Primary curve size, maximum thoracic kyphosis, T1-S1 spinal height, lumbar lordosis (LL), and coronal balance were measured at three time points: pre-index, post-index and latest follow up prior to “final” fusion.
Results: Of the 175 patients, 139 (79%) had primary thoracic (Group 1) and 36 (21%) had primary thoracolumbar or lumbar curves (Group 2). Mean pre-op age (5.8 years vs. 6.3 years), no. of lengthenings (5.9 vs. 5.8), length of follow-up (5.0 years vs 5.8 years) and male to female ratio (0.8 for both) was not significantly different between Group 1 and 2. Mean number of levels instrumented in Group 2 (14.9) was significantly greater than Group 1 (13.6) (p<0.001). Group 2 had significantly better mean curve correction than Group 1 following the index GR surgery (51% and 44%, respectively; p=0.03); however there was no significant difference in curve correction at latest follow-up (45% and 37%, respectively; p=0.123). Increase in T1-S1 was similar between the two groups across all time points. While pre-op coronal balance was significantly worse for Group 2, initial and overall improvement in coronal balance was significantly greater in Group 2 than in Group 1 (p<0.001) (Table).