- Email: [email protected]
Ponte Osteotomies With Pedicle Screw Instrumentation in the Treatment of Adolescent Idiopathic Scoliosis
Spine Deformity 1 (2013) 196e204 www.spine-deformity.org
Ponte Osteotomies With Pedicle Screw Instrumentation in the Treatment of Adolescent Idiopathic Scoliosis Suken A. Shah, MDa,*, Arjun A. Dhawale, MDa, Jon E. Oda, MDb, Petya Yorgova, MSa, Geraldine I. Neiss, PhDa, Laurens Holmes, Jr., PhD, DrPHa, Peter G. Gabos, MDa a
Department of Orthopaedic Surgery, Nemours/Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, USA b Department of Orthopaedic Surgery, Central California Children’s Hospital, 9300 Children’s Blvd, Madera, CA 93636, USA Received 24 May 2012; revised 12 March 2013; accepted 15 March 2013
Abstract Study design: Review of prospective database. Objectives: To report the results of Ponte osteotomy with pedicle screw instrumentation for major thoracic adolescent idiopathic (AIS) curves. Summary of background data: Ponte osteotomy for achieving coronal and sagittal correction of major thoracic curves in AIS with pedicle screw instrumentation is a widespread technique, but results have not been well described. Methods: Review of 87 consecutive AIS patients with Lenke 1e4 curves who underwent Ponte osteotomies and pedicle screw instrumentation by 2 surgeons at a single institution. Surgical details, blood loss, and complications were recorded. We evaluated coronal and sagittal radiological measurements and Scoliosis Research Societye22 (SRS-22) questionnaire scores over 2-year follow-up. Results: The mean preoperative thoracic coronal Cobb angle was 57 9.7 , fulcrum flexibility was 47.2%, and lateral Cobb angle was 17.8 4 . The mean estimated blood loss (EBL), expressed as percent estimated blood volume, was 35.8 20.5 mL. There was significant improvement in coronal thoracic Cobb angle, percent correction, and apical vertebral translation over 2-year follow-up (p ! .05). In hypokyphotic curves, there was a significant increase in lateral thoracic T5eT12 kyphosis from 8.1 to 18.3 (p ! .001). In hyperkyphotic curves, mean lateral thoracic T5eT12 kyphosis improved from 45 to 26 (p ! .001). Median SRS-22 domains were higher after treatment (p ! .05). Complications included significant hypotension (1), EBL greater than 75% estimated blood volume (2), and wound infection needing drainage (2). There were neuromonitoring signal changes in 7 patients but no significant neurological complications. Conclusions: In this case series of major thoracic AIS curves treated with segmental pedicle screw instrumentation and Ponte osteotomies, there was an improvement in the coronal and sagittal radiological parameters. A prospective controlled study is needed to determine whether pedicle screw instrumentation and Ponte osteotomies influence outcomes and complications. Ó 2013 Scoliosis Research Society. Keywords: Thoracic; Spinal deformity; Sagittal; Coronal; Curve correction
Author disclosures: SAS (grant from the Setting Scoliosis Straight Foundation; provision of writing assistance, medicines, equipment, or administrative support from Harms Study Group; board membership with Scoliosis Research Society and Setting Scoliosis Straight Foundation; patents from DePuy Synthes Spine; AAD (none); JEO (none); PY (none); GIN (none); LH (none); PGG (payment for lecture from DePuy Synthes Spine). Corporate or industry funds were received to support this work. Although one or more of the author(s) has/have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this manuscript, benefits will be directed solely to a research fund, foundation, educational institution, or other nonprofit organization that the author(s) has/have been associated with. *Corresponding author. Department of Orthopaedic Surgery, Nemours/ Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, USA. Tel.: (302) 651-5904; fax: (302) 651-5951. E-mail address: [email protected] (S.A. Shah). 2212-134X/$ - see front matter Ó 2013 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2013.03.002
Introduction The technique of partial facetectomy for the purpose of obtaining a solid posterior spinal fusion in scoliosis surgery was first described by Hibbs [1] in 1924 and later by Moe [2]. In 1945, Smith-Petersen et al. [3] described a singlelevel posterior column osteotomy for correction of fixed lumbar flexion deformity in the fused spine resulting from rheumatoid arthritis and/or ankylosing spondylitis. Indications for performing the Smith-Petersen osteotomy were expanded for correcting kyphotic deformities with other etiologies in the thoracic and lumbar spine [4-8]. In 1984, Alberto Ponte et al. [9,10] advocated a similar procedure consisting of wide posterior releases, complete facet joint
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
excisions, and posterior column shortening at multiple levels for correcting sagittal-plane deformity in the unfused spine in Scheurmann’s kyphosis. Shufflebarger and Clark [11] and Shufflebarger et al. [12] first described the use of this procedure for the correction of lumbar and thoracolumbar adolescent idiopathic scoliosis (AIS) and showed better curve correction with use of both hooks and pedicle screws as anchors. Geck et al. [13] reported the use of all thoracic pedicle screw instrumentation with the Ponte procedure for Scheuermann’s kyphosis, with good results. Although the use of all posterior pedicle screw constructs in AIS has provided better correction of the coronal plane deformity than hook and hybrid instrumentation [14,15], results with respect to sagittal plane deformity correction have been more variable; some authors reported loss of thoracic kyphosis with pedicle screw constructs [14,16,17]. Preservation and restoration of thoracic kyphosis are important for maintaining lumbar lordosis and preventing flatback and loss of sagittal balance with aging [18]. Ponte osteotomy for achieving coronal, rotational, and sagittal correction of major thoracic curves in AIS with pedicle screw instrumentation is a widespread technique, but results have not been well described in the literature. Halanski and Cassidy [19] recently reported the use of Ponte osteotomies in 18 patients with thoracic idiopathic scoliosis (Lenke 1 and 2) and compared the results with 19 patients who underwent inferior facetectomy. They concluded that the Ponte osteotomy group had increased blood loss and operative time with no significant differences in the coronal and sagittal correction. The limitations of this retrospective series were the inadequate follow-up (only 6-week to 4-month radiographs were used for radiographic evaluation); the small number of patients, which resulted in an underpowered study; the significant baseline differences in Cobb angles with differences in curve flexibility, which made postoperative comparison difficult; and non-randomization with surgeon pre-selection of patients. These major limitations precluded any conclusion. Pizones et al. [20] reported a retrospective study of idiopathic scoliosis patients treated with hybrid instrumentation and sublaminar wires. A total of 21 patients underwent a wide posterior release and 25 patients underwent a standard posterior release. The authors found better coronal curve correction in the wide posterior release group, although there were no significant differences in the kyphosis correction or complications. This series included no patients treated only with pedicle screw constructs, and the number of patients treated with wide posterior release was relatively small. There is an important difference between the posterior column shortening osteotomy described by previous authors for Scheuermann’s kyphosis, major thoracolumbar scoliosis, and lumbar scoliosis [9-13] and the osteotomy performed for achieving restoration of kyphosis in major thoracic scoliosis with hypokyphosis: The latter involves lengthening the posterior column to produce kyphosis.
197
In the current study, we reviewed a prospectively collected, single-center database of consecutive AIS patients with major thoracic curves, who were treated with pedicle screw instrumentation and the modified Ponte osteotomy involving wide posterior release, complete facetectomy, and posterior column lengthening (in hypokyphotic curves). Patients had a minimum postoperative follow-up of 2 years. Our primary hypothesis was that the modified Ponte osteotomy with pedicle screw instrumentation resulted in postoperative restoration of thoracic kyphosis and sagittal balance. We also evaluated the coronal curve correction, blood loss, neuromonitoring signal changes, complications, and Scoliosis Research Society (SRS)-22 questionnaire outcomes. Materials and Methods After we obtaining institutional review board approval, we queried a prospectively collected database of AIS patients with major thoracic curves (Lenke 1e4) who underwent modified Ponte osteotomies and pedicle screw instrumentation consecutively by 2 surgeons at a single institution between November 2005 and August 2009, with minimum 2-year follow-up data available. Patients with Lenke 5 and 6 curves, patients with Lenke 1e4 curves with follow-up less than 2 two years, and patients who also underwent thoracoscopic anterior release were excluded. A total of 87 patients met the inclusion criteria and had adequate radiographs. Broadly, indications for the Ponte osteotomy were preoperative coronal Cobb angle greater than 50 with fulcrum flexibility less than 45%, hypokyphosis (lateral Cobb angle less than 20 ) or hyperkyphosis (lateral Cobb angle greater than 40 ). Technique With the patient prone on the Jackson table, the spine was exposed subperiosteally from the proximal to distal levels of planned fusion with an electrocautery blade and adequate hemostasis was achieved. We exposed The transverse processes at each level. The spinous processes were removed, and the ligamentum flavum was exposed. We osteotomized a portion of the inferior articular process at each level to expose the superior articular process of the lower vertebra. The cartilage of the superior articular process was decorticated. We carefully removed the ligamentum flavum with a Kerrison rongeur across the dorsal extent of the spinal canal (Fig. 1A). Using a narrow, double-action Leksell rongeur or Kerrison rongeur, we removed the superior articular facet, disarticulating the superior vertebrae from the inferior vertebrae posteriorly (Fig. 1B). We used a Penfield dissector to palpate the neural foramen and ensure complete facet resection. Gelfoam was laid across the vertebral interspace to tamponade epidural bleeding (Fig. 1C). The procedure was repeated on the opposite facet and at adjacent levels as planned. We placed freehand pedicle screws [21] at all levels on the concave side
198
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Fig. 1. Technique of performing Ponte osteotomy. (A) Removal of the ligamentum flavum with a Kerrison rongeur. (B) Removal of the facet with a narrow, double-action Leksell rongeur. (C) Completed osteotomy with Gelfoam in vertebral interspace.
of the major thoracic curve and at most levels on the convex side (implant density was typically greater than 85%). The Ponte osteotomies were performed before the screws were placed, to allow access laterally through the foramen. We used a burr, gearshift probe, ball-tipped feeler, and tap to prepare the pedicle screw path. To confirm position, we stimulated the screws directly using electromyographic
thresholds after fluoroscopic examination. Differential rod contouring (more kyphosis on concave sided rod versus less kyphosis on convex sided rod) was performed [22]. In hypokyphotic deformities, posterior column distraction was performed, and in hyperkyphotic deformities, posterior column compression was performed. Segmental and/or en bloc derotation [22] were performed. All cases used the Expedium Spine System (DePuy Synthes Spine, Inc., a Johnson & Johnson Company, Rayhnam, MA) posterior segmental instrumentation. We paid meticulous attention to decortication of the laminae and transverse processes at each vertebral level to obtain a good posterior fusion. We used local autograft, b-tricalcium sulfate, or allograft bone for fusion. At the levels of Ponte osteotomy, we placed longitudinal bone slivers across the superior and inferior laminae and transverse processes. Precautions were taken to place no small, morselized pieces of graft posteriorly at the levels of the osteotomies, because these could encroach on the spinal canal. All cases were performed under somatosensory evoked potentials (SSEP), transcranial motor evoked potentials (TcMEP), and direct pedicle screw electromyographic monitoring. Surgical details including operating time, instrumentation, levels of osteotomy, blood loss, neuromonitoring signal changes, hospital stay duration, and complications were reviewed. We reviewed clinical outpatient visits for rib hump scoliometer measurements and any complications. A single observer who was not associated with the surgeries evaluated standard preoperative and postoperative coronal and sagittal radiological measurements that were recorded on standing radiographs at first-erect (the first postoperative visit, usually 3e4 weeks after surgery), 1-, and 2-year follow-up. Figure 2A-D presents a radiographic case example. Coronal measurements included main thoracic and thoracolumbar/lumbar Cobb angle, fulcrum flexibility rate, apical vertebral translation, percent correction, and fulcrum bending correction index [23]. Sagittal measurements included Cobb angles at levels T2eT12, T5eT12, T10eL2, T2eT5, and T12eS1; junctional kyphosis; and global sagittal balance. In this study, as defined by the SRS, the groups were classified on the lateral T5e12 Cobb angle as follows: thoracic hypokyphosis less than 20 , normal kyphosis 20 to 40 , and hyperkyphosis grater than 40 . Criteria for radiological evaluation of pseudarthrosis were loss of Cobb angle correction of greater than 10 and/or instrumentation failure. We evaluated patient self-reported SRS-22 individual domain and total scores recorded at preoperative and follow-up visits when available. Statistical analysis We tested data for normality and performed descriptive and comparative statistics as applicable using parametric and nonparametric tests. We used repeated measures analysis of variance to compare preoperative and postoperative
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
199
Fig. 2. Case illustration of a 14-year-old girl with a Lenke 2B curve, who underwent 5-level Ponte osteotomy with pedicle screw instrumentation. (A) Preoperative posteroanterior (PA) radiograph: main thoracic curve, 74 ; fulcrum bending, 47 ; fulcrum flexibility, 36%; and lumbar curve, 35 . (B) Preoperative lateral radiograph: lateral thoracic T2eT12 kyphosis, 17 , and T5eT12 kyphosis, 4 . (C) Postoperative 2-year follow-up PA radiograph: main thoracic curve, 20 (73% correction), and lumbar curve, 10 . (D) Postoperative 2-year follow-up lateral radiograph: lateral thoracic T2eT12 kyphosis, 25 , and T5eT12 kyphosis, 15 .
radiological measurements at defined intervals. We used median-based chi-square test to compare normalized selfreported SRS scores. All tests were 2-tailed with significance set at p ! .05. We used SPSS version 17.0 (Chicago, IL) and STATA version 11.0 (College Station, TX) for statistical analyses. Results There were 23 males and 64 females. The mean age at surgery was 14.7 2.3 years (range, 10e19 years). Preoperative Risser grades were as follows: grade 0, 12 patients; grade 1, 7; grade 2, 10; grade 3, 17; grade 4, 29; and grade 5, 12 patients. Lenke classification of the curves was as follows: Lenke 1, 50 patients; Lenke 2, 29; Lenke 3, 5; and Lenke 4, 3 patients. Lumbar modifier was A in 39 cases, B in 25, and C in 23 cases. Thoracic kyphosis according to the Lenke classification was negative (less than 10 kyphosis) in 25 patients, neutral (10 e40 ) in 53 patients, and positive (greater than 40 kyphosis) in 9 patients. According to the study definition, 51 of 87 patients (59%) had preoperative thoracic hypokyphosis (less than 20 ), 27 (31%) had normal kyphosis (20 e40 ), and 9 (10%) had hyperkyphosis (greater than 40 ). The mean preoperative thoracic coronal Cobb angle was 57 9.7 (range, 40 e81 ), fulcrum flexibility was 47.2%, and lateral Cobb angle was 17.8 4 (range, 19 e49 ).
The mean number of Ponte osteotomies performed was 4 1 (range, 1e7) per patient, and the mean number of levels fused was 11 1.7 (range, 6e15). The mean operating time was 321 63 minutes (range, 210e475 minutes). Operating time was less than 300 minutes in 34 patients, 300e399 minutes in 42 patients, and greater than 400 minutes in 11 patients. Estimated blood loss (EBL) was 1,508 874 mL (range, 412e4,810 mL). The mean EBL, expressed as percent estimated blood volume (EBV), was 35.8 20.5 mL (range, 9.4e129.6 mL) and was less than 25 mL in 31 patients, 26e50 mL in 41, 51e75 mL in 13, and greater than 75 mL in 2. The mean volume of cell saver transfusion was 476 375 mL (range, 0e2,125 mL), and the mean volume of other blood products transfused was 315 346 mL (range, 0e1,717 mL). The mean length of hospital stay was 5.2 0.7 days (range, 4e9 days), including the day of surgery. Rod material used was stainless steel in 76 patients, cobalt chrome in 9, and titanium in 2. Rod diameter was 5.5 mm in 85 patients and 6.35 mm in 2 patients. We recorded details of derotation maneuvers [22] in 85 patients and included segmental derotation in 54 patients, en bloc maneuvers in 11, and both in 20 patients. There was significant improvement in coronal main thoracic Cobb angle, percent correction, fulcrum bending correction index, and apical vertebral translation over 2-year follow-up (p ! .05) (Table 1). The mean coronal thoracic
200
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Table 1 Coronal radiological measurements. Variable Coronal main thoracic Cobb angle Preoperatively First erect 1 year 2 years Thoracic Cobb % correction First erect 1 year 2 years Thoracic fulcrum bend correction index First erect 1 year 2 years Thoracic apicaleC7 plumbline translation Preoperatively First erect 1 year 2 years Thoracic apicaleCSVL translation Preoperatively First erect 1 year 2 years Thoracolumbar/lumbar Cobb angle Preoperatively First erect 1 year 2 years Thoracolumbar/lumbar Cobb % correction First erect 1 year 2 years Thoracolumbar/lumbar fulcrum bend correction index First erect 1 year 2 years Thoracolumbar/lumbar apical translation Preoperatively First erect 1 year 2 years
Mean
Standard deviation
p !.001
57.1 15.7 18.3 20.2
9.7 5.1 4.9 5.3
71.5 69.6 63.9
9.8 12.1 11.1
183.5 179 165
122.5 122.7 113.5
4.6 1.0 1.5 1.5
2.6 1.0 0.9 1.0
4.4 0.3 0.8 1.0
3.3 1.5 1.3 1.4
35.1 11.4 11.7 12.3
11.3 6.7 5.7 6.5
67.1 68.3 65.1
17.5 15.4 16.3
!.001
!.001
!.001
Complications
!.001
!.001
.16
!.001 105.8 214.7 21.4
105.6 84.2 28.0
1.2 0.9 0.8 0.6
1.5 1.3 1.1 1.2
T5eT12 Cobb angle increased significantly from 8.1 7.5 to 18.3 6.0 (p ! .001) (Table 2), which was 2.5 kyphosis improvement per osteotomy level. The mean sagittal balance improved from 0.6 3.8 cm (range, 10.8 to 9.2 cm) preoperatively to 0.1 2.8 cm (range, 7.5 to 8.8 cm) at 2-year follow-up (p 5 .01) (Table 2). Radiological data are summarized in Figures 3A and 3B. Mean thoracic rib hump scoliometer measurements reduced from 15 4.4 preoperatively to 7.4 3.6 at 1-year follow-up and to 6.9 3.9 at 2-year follow-up. Mean lumbar scoliometer measurements decreased from 8 4.6 preoperatively to 3.7 2.7 at 1-year follow-up and to 4.2 3.2 at 2-year follow-up. Median SRS-22 individual domain and total scores improved after treatment, as shown in Table 3 and Figure 4.
!.001
First erect, first standing follow-up visit; CSVL, central sacral vertical line.
Cobb angle improved from 57.1 9.7 (range, 40 e81 ) to 20.2 5.3 (range, 6 e38 ). The thoracic apical eC7 plumbline translation improved from 4.6 2.6 cm ( 6.6 to 8.8 cm) to 1.5 1.0 cm (range, 1.8 to 3.8 cm). The mean lateral T2eT12 Cobb angle measurement increased from 27 14.5 (range 9 to 59 ) to 30.5 10.5 (range, 6 e62 ) (p5.03) (Table 2). The mean lateral thoracic T5eT12 Cobb angle measurement increased from 17.8 4.0 (range, 19 to 49 ) to 19.5 7.4 (range, 2 e41 ) (p 5 .3) (Table 2). In hyperkyphotic curves (greater than 40 ), the mean T5eT12 kyphosis improved from 44.6 2.8 to 25.8 8.6 (p ! .001) (Table 2). In hypokyphotic curves (less than 20 ), the mean lateral thoracic
Intraoperative complications included a significant episode of hypotension in 1 patient and EBL greater than 75% EBV in 2 patients (2.3%). We observed intraoperative neuromonitoring signal changes in 7 patients (8%). These included TcMEP changes in 3 patients (3.4%), SSEP changes in 2 (2.3%), and both TcMEP and SSEP changes in 2 patients (2.3%). The intraoperative monitoring signal changes occurred during exposure in 1 patient, during or after pedicle screw insertion in 3, and during or after curve correction maneuvers after instrumentation in 3s. The signal changes were temporary. Correction maneuvers included pausing surgery for a few minutes, raising mean arterial pressure, changing the position of the extremities, and removing and reinserting screws. None of the surgeries required staging. The cause of the TcMEP changes was hypotension in 2 cases, screw malposition in 1, and corrective maneuvers in 2. There were no significant neurological complications in any patient, except 1 who developed axillary nerve palsy (Parsonage-Turner syndrome, or brachial neuritis). The patient recovered 3 months postoperatively; we believed that this was not related to the Ponte osteotomy. Two patients (2.3%) developed a wound infection needing surgical drainage. One patient developed a superior mesenteric artery syndrome 2 weeks postoperatively that was treated with hospitalization with no food or drink, nasogastric/nasojejunal tubes, and total parenteral nutrition. One patient with a positive family history developed deep vein thrombosis 12 days after surgery and was treated with warfarin and enoxaparin. There was 1 instrumentationrelated complication: a distal pedicle screw loosened in the vertebra. We noted no cases of obvious pseudarthrosis at the 2-year follow-up. Discussion Wide facetectomy consists of removing the inferior articular process of the cephalad level, releasing the capsule, and removing articular cartilage. The Ponte
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
201
Table 2 Sagittal radiographic measurements. Variable Lateral T2eT12 Cobb angle Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (hypokyphosis group!20 ) Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (normal kyphosis group 20 e40 ) Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (hyperkyphosis groupO40 ) Preoperatively First erect 1 year 2 years Lateral T10eL2 Cobb angle Preoperatively First erect 1 year 2 years Lateral T2eT5 Cobb angle Preoperatively First erect 1 year 2 years Proximal junctional kyphosis First erect 1 year 2 years Distal junctional kyphosis First erect 1 year 2 years Lordosis (T12eS1) Preoperatively First erect 1 year 2 years Sagittal balance (C7 plumbline to sacrum) Preoperatively First erect 1 year 2 years
Mean
Standard deviation
p .03
27.0 29.0 30.3 30.5
14.5 9.2 9.7 10.5
17.8 18.5 18.9 19.5
4.0 5.8 6.3 7.4
.3
!.001 8.1 16.9 17.5 18.3 25.7
7.5 4.9 5.4 6.0 4.9
19.4 20 19.6
6.1 7.0 8.3
!.001
!.001 44.6 24.8 23.8 25.8
2.8 5.8 6.3 8.6
3.5 4.3 4.8 4.4
10.8 8.2 8.1 8.0
.47
.08 9.6 11.2 11.3 11.0
6.6 6.9 6.9 6.9
7.2 7.9 8.0
4.7 5.5 5.3
6.6 9.4 9.1
8.2 8.9 9.3
58.3 54.2 58.1 55.9
13.7 12.6 10.8 12.0
.2
!.001
.006
.01 0.6 0.76 0.71 0.14
3.8 4.0 3.1 2.8
osteotomy differs by removing the interspinous ligament and ligamentum flavum (an important tether to posterior lengthening) and resecting the entire facet (superior and
Fig. 3. Radiological data summarized in graphs. (A) Changes in mean coronal mid-thoracic (MT) Cobb angle in 87 patients and changes in mean sagittal T5e12 Cobb angle in 51 patients with preoperative hypokyphosis. (B) Changes in mean thoracic apical-C7 plumbline translation, mean thoracic apicalecentral sacral vertical line (CSVL), and translation and mean sagittal balance (C7 plumblineesacrum) in 87 patients.
inferior articular process) and ventral portion of the facet capsule and pars out to the foramen. The Ponte osteotomy was originally described for Scheuermann’s kyphosis [9,10], and Shufflebarger et al. [11,12] demonstrated its role in lumbar and thoracolumbar scoliosis. Indications for this procedure in thoracic scoliosis have not been well defined. Although a recent review article [24] mentioned that scoliosis greater than 70 to 75 that does not bend down to less than 40 is an indication in current practice, no reference or data are provided. Some surgeons have argued for the routine use of this osteotomy in AIS; consequently, this study provides valuable evidence-based information for spinal deformity surgeons for decision making. Because there is considerable variability in obtaining bending radiographs, we have been using the fulcrum bending film to assess the curve flexibility [23]. In thoracic hypokyphosis, normal kyphosis could also be restored effectively with an anterior release [25]; however, this would entail a combined anterior and posterior procedure. Imrie et al. [17] found that in Lenke 1 curves where high coronal plane correction (greater than 80%) was achieved, patients had a loss of kyphosis compared with the low coronal plane correction (less than 40%) group, in which there was a gain in kyphosis. In hyperkyphosis, there was a significant reduction of almost 20 in the mean kyphosis, and in
202
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Table 3 Scoliosis Research Society (SRS)e22 patient outcomes. SRS domain Pain Preoperatively 1 year 2 years Self-image Preoperatively 1 year 2 years General function Preoperatively 1 year 2 years Mental health Preoperatively 1 year 2 years Satisfaction Preoperatively 1 year 2 years Total Preoperatively 1 year 2 years
Median
Interquartile range
4.2 4.6 4.6
0.6 0.7 0.8
3.4 4.6 4.6
0.8 0.6 0.6
4.8 4.8 5.0
0.8 0.5 0.5
4.2 4.6 4.4
0.8 0.8 1.0
3.5 5.0 5.0
1.5 0.5 0.5
4.1 4.6 4.5
0.6 0.4 0.4
p !.001
!.001
.13 Fig. 4. Median SRS outcomes preoperatively and at 1 and 2 years. !.001
!.01
!.001
hypokyphosis we observed a mean increase in thoracic kyphosis of 10 . The Ponte osteotomy allows better segmental vertebral rotational correction, although we were unable to objectively assess this in our series because placement of thoracic pedicle screws prevented comparison of vertebral rotation by the Nash-Moe method on radiographs. A radiographic method of assessment of pedicle screw rotation would not allow comparison of the preoperative and postoperative vertebral rotation because the Nash-Moe and pedicle screw grading systems are different. Computed tomography scans were not routinely performed in our series, but there was an improvement in clinical scoliometer measurements. We have sought to improve thoracic restoration of kyphosis (alleviation of lordosis) in these patients, not just for cosmesis, but more importantly because an improvement in thoracic kyphosis results in an improvement in lumbar lordosis for the long term in these patients. Cho et al. [8] compared the results of Smith-Petersen versus pedicle subtraction osteotomy for the correction of fixed sagittal imbalance in adults with kyphotic spinal deformities, and reported that a Smith-Petersen osteotomy could achieve 10.7 of correction at each level. In our series, we found a significantly lower kyphosis correction rate of 2.5 per osteotomy level in the hypokyphotic curves. One reason could be the different patient populations. The series of Cho et al. included adult patients with a heterogeneous group of lumbar kyphotic deformities, who underwent primary or revision surgeries. Our series included AIS patients who underwent primary surgeries;
more than half had hypokyphosis. Hypokyphosis correction needs to be more controlled because excessive posterior distraction could result in spinal cord deficit and might increase the risk of pseudarthrosis. In the absence of a control group, it is difficult to draw definite conclusions. We reviewed our AIS patients who did not undergo Ponte osteotomies, but we found that there were significant baseline differences and lower preoperative mean Cobb angles, which made comparison subject to potential inaccuracies and bias. Lehman et al. [26] reported a single-center series of 114 consecutive AIS patients treated with posterior pedicle screw constructs with a mean age of 14.9 years, mean main thoracic coronal Cobb angle of 59.2 , mean preoperative lateral T5eT12 Cobb angle of 25.8 , and mean of 10 levels fused. The mean main thoracic coronal Cobb angle decreased from 59.2 to 16.8 postoperatively, the mean lateral Cobb angle decreased from 26.2 to 14.5 postoperatively, and the C7 sagittal plumbline decreased from 7.1 to 15.6 mm postoperatively [26]. Lehman et al. acknowledged the decrease in thoracic kyphosis, which they attributed to the use of the 5.5-mm stainless-steel rod, the use of monoaxial screws, and the absence of posterior soft tissue release in the lordotic segment. Suk et al. [27] reported improved thoracic kyphosis with the use of a stiffer, 7-mm, stainless-steel rod. Although we used the 5.5 rods in 85 of 87 patients, we were able to achieve kyphosis correction, presumably because of the Ponte osteotomies, pre-contouring of kyphosis in an ultra-high-strength, stainless-steel rod, and distraction of the concave apex. We cannot comment on the co-relation of the improved sagittal contour and the rod diameter based on the study data, because we used 6.35-mm rods in only 2 patients, whereas 85 patients had 5.5-mm rods. However Suk et al. showed better restoration of the sagittal balance with use of the 7-mm stainless-steel rods. The procedure of pedicle screw instrumentation and Ponte osteotomies may be associated with a relatively longer operating time and increased blood loss; however, we have no way to separate out the variables in the surgical technique that may affect the outcome without a true control group. Our institution’s practice of calculating EBL
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
is to add the weight of sponges rinsed in saline and squeezed for cell salvage to the cell salvage blood obtained, plus any estimates of blood on the drapes, confirmed by the anesthesiologist. There may be variation in the calculation of blood loss over multiple centers, and this may reflect institutional practices and not just surgeon or technical factors. We saw intraoperative neuromonitoring changes in 8% of our patients, although there were no significant or permanent neurological complications. The cause of the TcMEP changes was hypotension in 2 cases, screw malposition in 1, and corrective maneuvers (translation and distraction) in 2. All changes were reversed and signals returned to baseline with appropriate corrective maneuvers. Our monitoring team is well qualified and experienced, and prides itself on a spectacular track record of sensitivity and specificity of neurological alerts. Without doing a complete facetectomy on the concave side, it is not possible to obtain further sagittal and coronal plane correction. We believe that the contracted facet and soft tissues on the concave side are an important structural tether to correction of scoliosis and restoration of kyphosis. However, there will be a gap after correction. One danger of releasing the concave side of a thoracic curve is that the spinal cord hugs the concave pedicle at the apex, and if the curve apex is hypokyphotic or lordotic, the cord is posterior along the dorsal lamina and articular process. The surgeon must exercise extreme caution in this area when positioning the rongeur and resecting bone or soft tissue as part of the release. Our infection rate was 2.3% in this series of patients. The study population was a consecutive series of patients during a time when we had a cluster of infections. Since that time, our infection rate has returned to near 0 for AIS, even with the routine use of the Ponte osteotomy. Increased operative time and bleeding may be implicated as factors, but this has not been our experience in subsequent years. Although we observed no cases of pseudarthrosis over the 2-year followup period, this is a potential complication that will need to be re-evaluated on further follow-up. Because facetectomy is a critical component of obtaining a posterior fusion, removal of the facets along with posterior distraction of the laminae could theoretically impede fusion. Due attention should be paid to decortication of the laminae and transverse processes and use of bridging autograft along with allograft to obtain a posterolateral fusion. Excessive distraction should never be performed because of the risks of spinal cord deficit and pseudarthrosis. The use of TcMEP with this technique is important to improve detection of impending neurological deficit. Coe et al. [28] reported neurological complications in 14 patients (0.32%), wound infections in 59 patients (1.35%), deep vein thrombosis in 2 patients (0.05%), fatal blood loss in 1 patient (0.02%), and implant-related problems in 28 patients (0.64%), in a total of 4,369 patients who underwent posterior spinal instrumentation and fusion. Pseudarthrosis rates and neuromonitoring changes were not reported. Lehman et al. [26] reported 5 complications in 114 patients, including 3
203
deep wound infections and 2 cases of adding on extension to the primary fusion; there were no short-term or longterm neurological complications. Blood loss and neuromonitoring changes were not reported.
Limitations Although this is the first large, consecutive, single-center case series of Ponte osteotomies with pedicle screw instrumentation performed for thoracic AIS, this study has limitations. First, in the absence of a control group, we can make no conclusions about the increased operative time, blood loss, or neuromonitoring changes, because there is no way to separate out the variables in the surgical technique that may affect the outcome. Historical controls are not sufficient to make conclusions. Second, our follow-up of 2 years is not sufficient to evaluate pseudarthrosis, which is a potential risk with any spinal osteotomy. Third, SRS-22 questionnaire results were not available for all patients. When they were available, they showed that patients were satisfied with their appearance postoperatively. Fourth, we do not have the ability to perform body surface contour assessments, and suggest that these do not always correlate with Cobb angle measurement or degree of postoperative correction. Threedimensional analysis would probably enable better understanding; however, we did not have this information. We are currently collecting prospective data on 3-dimensional alignment changes preoperatively and postoperatively from EOS images (EOS Imaging, Paris, France), but this will require longer follow-up. The contour of the thoracic cage has been measured with the scoliometer and captured with digital photography. Scoliometer measurements have been reported for assessment of the horizontal plane and thoracic cage contour. Pree and postecomputed tomography scans of these patients were not available, because this practice exposes patients to a tremendous amount of radiation and is not our current clinical practice. We have presented detailed radiological data using universally accepted standard deformity measurements on lateral and coronal projections. Apart from Cobb angle, thoracic apicalC7 plumbline translation, thoracic apicalecentral sacral vertical line translation, and sagittal balance (C7 plumbline to sacrum) have also been measured. We believe that in combination, these measurements at least form a reasonable proxy for evaluation of 3-dimensional correction. Although we chose the T5eT12 and T2eT12 levels for measurement of thoracic kyphosis to allow standardization for our prospective scoliosis database, these may not have been the angle between the maximally tilted thoracic vertebrae. We acknowledge that the mean coronal Cobb angle and fulcrum flexibility indicates that the curves were not too severe or stiff; this is because we have included a consecutive series of patients who underwent the Ponte procedure at our center with 2-year follow-up regardless of curve severity or stiffness, to avoid selection bias.
204
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Conclusions In this case series of major thoracic AIS curves treated with segmental pedicle screw instrumentation and Ponte osteotomies, there was improvement in the coronal and sagittal radiological parameters. A prospective controlled study is needed to determine whether pedicle screw instrumentation and Ponte osteotomies influence outcomes and complications. References [1] Hibbs RA. A report of fifty-nine cases of scoliosis treated by the fusion operation. J Bone Joint Surg Am 1924;6:3e34. [2] Moe JH. A critical analysis of methods of fusion for scoliosis: an evaluation in two hundred and sixty-six patients. J Bone Joint Surg Am 1958;40:529e54. [3] Smith-Petersen MN, Larson CB, Aufranc OE. Osteotomy of the spine for correction of flexion deformity in rheumatoid arthritis. J Bone Joint Surg Am 1945;27:1e11. [4] Lagrone MO, Bradford DS, Moe JH, et al. Treatment of symptomatic flatback after spinal fusion. J Bone Joint Surg Am 1988;70: 569e80. [5] Voos K, Boachie-Adjei O, Rawlins BA. Multiple vertebral osteotomies in the treatment of rigid adult spine deformities. Spine (Phila Pa 1976) 2001;26:526e33. [6] McMaster MJ. A technique for lumbar spinal osteotomy in ankylosing spondylitis. J Bone Joint Surg Br 1985;67:204e10. [7] Kostuik JP, Maurais GR, Richardson WJ, et al. Combined single stage anterior and posterior osteotomy for correction of iatrogenic lumbar kyphosis. Spine (Phila Pa 1976) 1988;13:257e66. [8] Cho KJ, Bridwell KH, Lenke LG, et al. Comparison of SmithPetersen versus pedicle subtraction osteotomy for the correction of fixed sagittal imbalance. Spine (Phila Pa 1976) 2005;30:2030e7. [9] Ponte A, Vero B, Siccardi GL. Surgical treatment of Scheuermann’s hyperkyphosis. In: Winter RB, editor. Kyphosis (Progress in Spinal Pathology). Bologna, Italy: Aulo Gaggi; 1984. p. 75e80. [10] Ponte A. Posterior column shortening for Scheuermann’s kyphosis: an innovative one-stage technique. In: Haher T, Merola A, editors. Surgical Techniques for the Spine. 1st ed. New York, NY: Thieme Verlag; 2003. p. 107e13. [11] Shufflebarger HL, Clark CE. Effect of wide posterior release on correction in adolescent idiopathic scoliosis. J Pediatr Orthop B 1998;7:117e23. [12] Shufflebarger HL, Geck MJ, Clark CE. The posterior approach for lumbar and thoracolumbar adolescent idiopathic scoliosis: posterior shortening and pedicle screws. Spine (Phila Pa 1976) 2004;29:269e76. [13] Geck MJ, Macagno A, Ponte A, et al. The Ponte procedure: posterior only treatment of Scheuermann’s kyphosis using segmental posterior shortening and pedicle screw instrumentation. J Spinal Disord Tech 2007;20:586e93.
[14] Kim YJ, Lenke LG, Kim J, et al. Comparative analysis of pedicle screw versus hybrid instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2006;31: 291e8. [15] Liljenqvist U, Lepsien U, Hackenberg L, et al. Comparative analysis of pedicle screw and hook instrumentation in posterior correction and fusion of idiopathic thoracic scoliosis. Eur Spine J 2002;11: 336e43. [16] Lowenstein JE, Matsumoto H, Vitale MG, et al. Coronal and sagittal plane correction in adolescent idiopathic scoliosis: a comparison between all pedicle screw versus hybrid thoracic hook lumbar screw constructs. Spine (Phila Pa 1976) 2007;32:448e52. [17] Imrie M, Yaszay B, Bastrom TP, et al. Adolescent idiopathic scoliosis: should 100% correction be the goal? J Pediatr Orthop 2011;31(1 Suppl):S9e13. [18] Newton PO, Yaszay B, Upasani VV, et al. Preservation of thoracic kyphosis is critical to maintain lumbar lordosis in the surgical treatment of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2010;35:1365e70. [19] Halanski MA, Cassidy JA. Do multilevel Ponte osteotomies in thoracic idiopathic scoliosis surgery improve curve correction and restore thoracic kyphosis? J Spinal Disord Tech 2011 Dec 21 [Epub ahead of print]. [20] Pizones J, Izquierdo E, Sanchez-Mariscal F, et al. Does wide posterior multiple level release improve the correction of adolescent idiopathic scoliosis curves? J Spinal Disord Tech 2010;23:e24e30. [21] Kim YJ, Lenke LG, Bridwell KH, et al. Free hand pedicle screw placement in the thoracic spine: is it safe? Spine (Phila Pa 1976) 2004;29:333e42. [22] Shah SA. Derotation of the spine. Neurosurg Clin N Am 2007;18: 339e45. [23] Luk KD, Cheung KM, Lu DS, et al. Assessment of scoliosis correction in relation to flexibility using the fulcrum bending correction index. Spine (Phila Pa 1976) 1998;23:2303e7. [24] Diab MG, Franzone JM, Vitale MG. The role of posterior spinal osteotomies in pediatric spinal deformity surgery: indications and operative technique. J Pediatr Orthop 2011;31:S88e98. [25] Sucato DJ, Agrawal S, O’Brien MF, et al. Restoration of thoracic kyphosis after operative treatment of adolescent idiopathic scoliosis: a multicenter comparison of three surgical approaches. Spine (Phila Pa 1976) 2008;33:2630e6. [26] Lehman Jr RA, Lenke LG, Keeler KA, et al. Operative treatment of adolescent idiopathic scoliosis with posterior pedicle screw-only constructs: minimum three-year follow-up of one hundred fourteen cases. Spine (Phila Pa 1976) 2008;33:1598e604. [27] Suk SI, Lee SM, Chung ER, et al. Selective thoracic fusion with segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis: more than 5-year follow-up. Spine (Phila Pa 1976) 2005;30:1602e9. [28] Coe JD, Arlet V, Donaldson W, et al. Complications in spinal fusion for adolescent idiopathic scoliosis in the new millennium: a report of the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976) 2006;31:345e9.
Ponte Osteotomies With Pedicle Screw Instrumentation in the Treatment of Adolescent Idiopathic Scoliosis Suken A. Shah, MDa,*, Arjun A. Dhawale, MDa, Jon E. Oda, MDb, Petya Yorgova, MSa, Geraldine I. Neiss, PhDa, Laurens Holmes, Jr., PhD, DrPHa, Peter G. Gabos, MDa a
Department of Orthopaedic Surgery, Nemours/Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, USA b Department of Orthopaedic Surgery, Central California Children’s Hospital, 9300 Children’s Blvd, Madera, CA 93636, USA Received 24 May 2012; revised 12 March 2013; accepted 15 March 2013
Abstract Study design: Review of prospective database. Objectives: To report the results of Ponte osteotomy with pedicle screw instrumentation for major thoracic adolescent idiopathic (AIS) curves. Summary of background data: Ponte osteotomy for achieving coronal and sagittal correction of major thoracic curves in AIS with pedicle screw instrumentation is a widespread technique, but results have not been well described. Methods: Review of 87 consecutive AIS patients with Lenke 1e4 curves who underwent Ponte osteotomies and pedicle screw instrumentation by 2 surgeons at a single institution. Surgical details, blood loss, and complications were recorded. We evaluated coronal and sagittal radiological measurements and Scoliosis Research Societye22 (SRS-22) questionnaire scores over 2-year follow-up. Results: The mean preoperative thoracic coronal Cobb angle was 57 9.7 , fulcrum flexibility was 47.2%, and lateral Cobb angle was 17.8 4 . The mean estimated blood loss (EBL), expressed as percent estimated blood volume, was 35.8 20.5 mL. There was significant improvement in coronal thoracic Cobb angle, percent correction, and apical vertebral translation over 2-year follow-up (p ! .05). In hypokyphotic curves, there was a significant increase in lateral thoracic T5eT12 kyphosis from 8.1 to 18.3 (p ! .001). In hyperkyphotic curves, mean lateral thoracic T5eT12 kyphosis improved from 45 to 26 (p ! .001). Median SRS-22 domains were higher after treatment (p ! .05). Complications included significant hypotension (1), EBL greater than 75% estimated blood volume (2), and wound infection needing drainage (2). There were neuromonitoring signal changes in 7 patients but no significant neurological complications. Conclusions: In this case series of major thoracic AIS curves treated with segmental pedicle screw instrumentation and Ponte osteotomies, there was an improvement in the coronal and sagittal radiological parameters. A prospective controlled study is needed to determine whether pedicle screw instrumentation and Ponte osteotomies influence outcomes and complications. Ó 2013 Scoliosis Research Society. Keywords: Thoracic; Spinal deformity; Sagittal; Coronal; Curve correction
Author disclosures: SAS (grant from the Setting Scoliosis Straight Foundation; provision of writing assistance, medicines, equipment, or administrative support from Harms Study Group; board membership with Scoliosis Research Society and Setting Scoliosis Straight Foundation; patents from DePuy Synthes Spine; AAD (none); JEO (none); PY (none); GIN (none); LH (none); PGG (payment for lecture from DePuy Synthes Spine). Corporate or industry funds were received to support this work. Although one or more of the author(s) has/have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this manuscript, benefits will be directed solely to a research fund, foundation, educational institution, or other nonprofit organization that the author(s) has/have been associated with. *Corresponding author. Department of Orthopaedic Surgery, Nemours/ Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, USA. Tel.: (302) 651-5904; fax: (302) 651-5951. E-mail address: [email protected] (S.A. Shah). 2212-134X/$ - see front matter Ó 2013 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2013.03.002
Introduction The technique of partial facetectomy for the purpose of obtaining a solid posterior spinal fusion in scoliosis surgery was first described by Hibbs [1] in 1924 and later by Moe [2]. In 1945, Smith-Petersen et al. [3] described a singlelevel posterior column osteotomy for correction of fixed lumbar flexion deformity in the fused spine resulting from rheumatoid arthritis and/or ankylosing spondylitis. Indications for performing the Smith-Petersen osteotomy were expanded for correcting kyphotic deformities with other etiologies in the thoracic and lumbar spine [4-8]. In 1984, Alberto Ponte et al. [9,10] advocated a similar procedure consisting of wide posterior releases, complete facet joint
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
excisions, and posterior column shortening at multiple levels for correcting sagittal-plane deformity in the unfused spine in Scheurmann’s kyphosis. Shufflebarger and Clark [11] and Shufflebarger et al. [12] first described the use of this procedure for the correction of lumbar and thoracolumbar adolescent idiopathic scoliosis (AIS) and showed better curve correction with use of both hooks and pedicle screws as anchors. Geck et al. [13] reported the use of all thoracic pedicle screw instrumentation with the Ponte procedure for Scheuermann’s kyphosis, with good results. Although the use of all posterior pedicle screw constructs in AIS has provided better correction of the coronal plane deformity than hook and hybrid instrumentation [14,15], results with respect to sagittal plane deformity correction have been more variable; some authors reported loss of thoracic kyphosis with pedicle screw constructs [14,16,17]. Preservation and restoration of thoracic kyphosis are important for maintaining lumbar lordosis and preventing flatback and loss of sagittal balance with aging [18]. Ponte osteotomy for achieving coronal, rotational, and sagittal correction of major thoracic curves in AIS with pedicle screw instrumentation is a widespread technique, but results have not been well described in the literature. Halanski and Cassidy [19] recently reported the use of Ponte osteotomies in 18 patients with thoracic idiopathic scoliosis (Lenke 1 and 2) and compared the results with 19 patients who underwent inferior facetectomy. They concluded that the Ponte osteotomy group had increased blood loss and operative time with no significant differences in the coronal and sagittal correction. The limitations of this retrospective series were the inadequate follow-up (only 6-week to 4-month radiographs were used for radiographic evaluation); the small number of patients, which resulted in an underpowered study; the significant baseline differences in Cobb angles with differences in curve flexibility, which made postoperative comparison difficult; and non-randomization with surgeon pre-selection of patients. These major limitations precluded any conclusion. Pizones et al. [20] reported a retrospective study of idiopathic scoliosis patients treated with hybrid instrumentation and sublaminar wires. A total of 21 patients underwent a wide posterior release and 25 patients underwent a standard posterior release. The authors found better coronal curve correction in the wide posterior release group, although there were no significant differences in the kyphosis correction or complications. This series included no patients treated only with pedicle screw constructs, and the number of patients treated with wide posterior release was relatively small. There is an important difference between the posterior column shortening osteotomy described by previous authors for Scheuermann’s kyphosis, major thoracolumbar scoliosis, and lumbar scoliosis [9-13] and the osteotomy performed for achieving restoration of kyphosis in major thoracic scoliosis with hypokyphosis: The latter involves lengthening the posterior column to produce kyphosis.
197
In the current study, we reviewed a prospectively collected, single-center database of consecutive AIS patients with major thoracic curves, who were treated with pedicle screw instrumentation and the modified Ponte osteotomy involving wide posterior release, complete facetectomy, and posterior column lengthening (in hypokyphotic curves). Patients had a minimum postoperative follow-up of 2 years. Our primary hypothesis was that the modified Ponte osteotomy with pedicle screw instrumentation resulted in postoperative restoration of thoracic kyphosis and sagittal balance. We also evaluated the coronal curve correction, blood loss, neuromonitoring signal changes, complications, and Scoliosis Research Society (SRS)-22 questionnaire outcomes. Materials and Methods After we obtaining institutional review board approval, we queried a prospectively collected database of AIS patients with major thoracic curves (Lenke 1e4) who underwent modified Ponte osteotomies and pedicle screw instrumentation consecutively by 2 surgeons at a single institution between November 2005 and August 2009, with minimum 2-year follow-up data available. Patients with Lenke 5 and 6 curves, patients with Lenke 1e4 curves with follow-up less than 2 two years, and patients who also underwent thoracoscopic anterior release were excluded. A total of 87 patients met the inclusion criteria and had adequate radiographs. Broadly, indications for the Ponte osteotomy were preoperative coronal Cobb angle greater than 50 with fulcrum flexibility less than 45%, hypokyphosis (lateral Cobb angle less than 20 ) or hyperkyphosis (lateral Cobb angle greater than 40 ). Technique With the patient prone on the Jackson table, the spine was exposed subperiosteally from the proximal to distal levels of planned fusion with an electrocautery blade and adequate hemostasis was achieved. We exposed The transverse processes at each level. The spinous processes were removed, and the ligamentum flavum was exposed. We osteotomized a portion of the inferior articular process at each level to expose the superior articular process of the lower vertebra. The cartilage of the superior articular process was decorticated. We carefully removed the ligamentum flavum with a Kerrison rongeur across the dorsal extent of the spinal canal (Fig. 1A). Using a narrow, double-action Leksell rongeur or Kerrison rongeur, we removed the superior articular facet, disarticulating the superior vertebrae from the inferior vertebrae posteriorly (Fig. 1B). We used a Penfield dissector to palpate the neural foramen and ensure complete facet resection. Gelfoam was laid across the vertebral interspace to tamponade epidural bleeding (Fig. 1C). The procedure was repeated on the opposite facet and at adjacent levels as planned. We placed freehand pedicle screws [21] at all levels on the concave side
198
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Fig. 1. Technique of performing Ponte osteotomy. (A) Removal of the ligamentum flavum with a Kerrison rongeur. (B) Removal of the facet with a narrow, double-action Leksell rongeur. (C) Completed osteotomy with Gelfoam in vertebral interspace.
of the major thoracic curve and at most levels on the convex side (implant density was typically greater than 85%). The Ponte osteotomies were performed before the screws were placed, to allow access laterally through the foramen. We used a burr, gearshift probe, ball-tipped feeler, and tap to prepare the pedicle screw path. To confirm position, we stimulated the screws directly using electromyographic
thresholds after fluoroscopic examination. Differential rod contouring (more kyphosis on concave sided rod versus less kyphosis on convex sided rod) was performed [22]. In hypokyphotic deformities, posterior column distraction was performed, and in hyperkyphotic deformities, posterior column compression was performed. Segmental and/or en bloc derotation [22] were performed. All cases used the Expedium Spine System (DePuy Synthes Spine, Inc., a Johnson & Johnson Company, Rayhnam, MA) posterior segmental instrumentation. We paid meticulous attention to decortication of the laminae and transverse processes at each vertebral level to obtain a good posterior fusion. We used local autograft, b-tricalcium sulfate, or allograft bone for fusion. At the levels of Ponte osteotomy, we placed longitudinal bone slivers across the superior and inferior laminae and transverse processes. Precautions were taken to place no small, morselized pieces of graft posteriorly at the levels of the osteotomies, because these could encroach on the spinal canal. All cases were performed under somatosensory evoked potentials (SSEP), transcranial motor evoked potentials (TcMEP), and direct pedicle screw electromyographic monitoring. Surgical details including operating time, instrumentation, levels of osteotomy, blood loss, neuromonitoring signal changes, hospital stay duration, and complications were reviewed. We reviewed clinical outpatient visits for rib hump scoliometer measurements and any complications. A single observer who was not associated with the surgeries evaluated standard preoperative and postoperative coronal and sagittal radiological measurements that were recorded on standing radiographs at first-erect (the first postoperative visit, usually 3e4 weeks after surgery), 1-, and 2-year follow-up. Figure 2A-D presents a radiographic case example. Coronal measurements included main thoracic and thoracolumbar/lumbar Cobb angle, fulcrum flexibility rate, apical vertebral translation, percent correction, and fulcrum bending correction index [23]. Sagittal measurements included Cobb angles at levels T2eT12, T5eT12, T10eL2, T2eT5, and T12eS1; junctional kyphosis; and global sagittal balance. In this study, as defined by the SRS, the groups were classified on the lateral T5e12 Cobb angle as follows: thoracic hypokyphosis less than 20 , normal kyphosis 20 to 40 , and hyperkyphosis grater than 40 . Criteria for radiological evaluation of pseudarthrosis were loss of Cobb angle correction of greater than 10 and/or instrumentation failure. We evaluated patient self-reported SRS-22 individual domain and total scores recorded at preoperative and follow-up visits when available. Statistical analysis We tested data for normality and performed descriptive and comparative statistics as applicable using parametric and nonparametric tests. We used repeated measures analysis of variance to compare preoperative and postoperative
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
199
Fig. 2. Case illustration of a 14-year-old girl with a Lenke 2B curve, who underwent 5-level Ponte osteotomy with pedicle screw instrumentation. (A) Preoperative posteroanterior (PA) radiograph: main thoracic curve, 74 ; fulcrum bending, 47 ; fulcrum flexibility, 36%; and lumbar curve, 35 . (B) Preoperative lateral radiograph: lateral thoracic T2eT12 kyphosis, 17 , and T5eT12 kyphosis, 4 . (C) Postoperative 2-year follow-up PA radiograph: main thoracic curve, 20 (73% correction), and lumbar curve, 10 . (D) Postoperative 2-year follow-up lateral radiograph: lateral thoracic T2eT12 kyphosis, 25 , and T5eT12 kyphosis, 15 .
radiological measurements at defined intervals. We used median-based chi-square test to compare normalized selfreported SRS scores. All tests were 2-tailed with significance set at p ! .05. We used SPSS version 17.0 (Chicago, IL) and STATA version 11.0 (College Station, TX) for statistical analyses. Results There were 23 males and 64 females. The mean age at surgery was 14.7 2.3 years (range, 10e19 years). Preoperative Risser grades were as follows: grade 0, 12 patients; grade 1, 7; grade 2, 10; grade 3, 17; grade 4, 29; and grade 5, 12 patients. Lenke classification of the curves was as follows: Lenke 1, 50 patients; Lenke 2, 29; Lenke 3, 5; and Lenke 4, 3 patients. Lumbar modifier was A in 39 cases, B in 25, and C in 23 cases. Thoracic kyphosis according to the Lenke classification was negative (less than 10 kyphosis) in 25 patients, neutral (10 e40 ) in 53 patients, and positive (greater than 40 kyphosis) in 9 patients. According to the study definition, 51 of 87 patients (59%) had preoperative thoracic hypokyphosis (less than 20 ), 27 (31%) had normal kyphosis (20 e40 ), and 9 (10%) had hyperkyphosis (greater than 40 ). The mean preoperative thoracic coronal Cobb angle was 57 9.7 (range, 40 e81 ), fulcrum flexibility was 47.2%, and lateral Cobb angle was 17.8 4 (range, 19 e49 ).
The mean number of Ponte osteotomies performed was 4 1 (range, 1e7) per patient, and the mean number of levels fused was 11 1.7 (range, 6e15). The mean operating time was 321 63 minutes (range, 210e475 minutes). Operating time was less than 300 minutes in 34 patients, 300e399 minutes in 42 patients, and greater than 400 minutes in 11 patients. Estimated blood loss (EBL) was 1,508 874 mL (range, 412e4,810 mL). The mean EBL, expressed as percent estimated blood volume (EBV), was 35.8 20.5 mL (range, 9.4e129.6 mL) and was less than 25 mL in 31 patients, 26e50 mL in 41, 51e75 mL in 13, and greater than 75 mL in 2. The mean volume of cell saver transfusion was 476 375 mL (range, 0e2,125 mL), and the mean volume of other blood products transfused was 315 346 mL (range, 0e1,717 mL). The mean length of hospital stay was 5.2 0.7 days (range, 4e9 days), including the day of surgery. Rod material used was stainless steel in 76 patients, cobalt chrome in 9, and titanium in 2. Rod diameter was 5.5 mm in 85 patients and 6.35 mm in 2 patients. We recorded details of derotation maneuvers [22] in 85 patients and included segmental derotation in 54 patients, en bloc maneuvers in 11, and both in 20 patients. There was significant improvement in coronal main thoracic Cobb angle, percent correction, fulcrum bending correction index, and apical vertebral translation over 2-year follow-up (p ! .05) (Table 1). The mean coronal thoracic
200
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Table 1 Coronal radiological measurements. Variable Coronal main thoracic Cobb angle Preoperatively First erect 1 year 2 years Thoracic Cobb % correction First erect 1 year 2 years Thoracic fulcrum bend correction index First erect 1 year 2 years Thoracic apicaleC7 plumbline translation Preoperatively First erect 1 year 2 years Thoracic apicaleCSVL translation Preoperatively First erect 1 year 2 years Thoracolumbar/lumbar Cobb angle Preoperatively First erect 1 year 2 years Thoracolumbar/lumbar Cobb % correction First erect 1 year 2 years Thoracolumbar/lumbar fulcrum bend correction index First erect 1 year 2 years Thoracolumbar/lumbar apical translation Preoperatively First erect 1 year 2 years
Mean
Standard deviation
p !.001
57.1 15.7 18.3 20.2
9.7 5.1 4.9 5.3
71.5 69.6 63.9
9.8 12.1 11.1
183.5 179 165
122.5 122.7 113.5
4.6 1.0 1.5 1.5
2.6 1.0 0.9 1.0
4.4 0.3 0.8 1.0
3.3 1.5 1.3 1.4
35.1 11.4 11.7 12.3
11.3 6.7 5.7 6.5
67.1 68.3 65.1
17.5 15.4 16.3
!.001
!.001
!.001
Complications
!.001
!.001
.16
!.001 105.8 214.7 21.4
105.6 84.2 28.0
1.2 0.9 0.8 0.6
1.5 1.3 1.1 1.2
T5eT12 Cobb angle increased significantly from 8.1 7.5 to 18.3 6.0 (p ! .001) (Table 2), which was 2.5 kyphosis improvement per osteotomy level. The mean sagittal balance improved from 0.6 3.8 cm (range, 10.8 to 9.2 cm) preoperatively to 0.1 2.8 cm (range, 7.5 to 8.8 cm) at 2-year follow-up (p 5 .01) (Table 2). Radiological data are summarized in Figures 3A and 3B. Mean thoracic rib hump scoliometer measurements reduced from 15 4.4 preoperatively to 7.4 3.6 at 1-year follow-up and to 6.9 3.9 at 2-year follow-up. Mean lumbar scoliometer measurements decreased from 8 4.6 preoperatively to 3.7 2.7 at 1-year follow-up and to 4.2 3.2 at 2-year follow-up. Median SRS-22 individual domain and total scores improved after treatment, as shown in Table 3 and Figure 4.
!.001
First erect, first standing follow-up visit; CSVL, central sacral vertical line.
Cobb angle improved from 57.1 9.7 (range, 40 e81 ) to 20.2 5.3 (range, 6 e38 ). The thoracic apical eC7 plumbline translation improved from 4.6 2.6 cm ( 6.6 to 8.8 cm) to 1.5 1.0 cm (range, 1.8 to 3.8 cm). The mean lateral T2eT12 Cobb angle measurement increased from 27 14.5 (range 9 to 59 ) to 30.5 10.5 (range, 6 e62 ) (p5.03) (Table 2). The mean lateral thoracic T5eT12 Cobb angle measurement increased from 17.8 4.0 (range, 19 to 49 ) to 19.5 7.4 (range, 2 e41 ) (p 5 .3) (Table 2). In hyperkyphotic curves (greater than 40 ), the mean T5eT12 kyphosis improved from 44.6 2.8 to 25.8 8.6 (p ! .001) (Table 2). In hypokyphotic curves (less than 20 ), the mean lateral thoracic
Intraoperative complications included a significant episode of hypotension in 1 patient and EBL greater than 75% EBV in 2 patients (2.3%). We observed intraoperative neuromonitoring signal changes in 7 patients (8%). These included TcMEP changes in 3 patients (3.4%), SSEP changes in 2 (2.3%), and both TcMEP and SSEP changes in 2 patients (2.3%). The intraoperative monitoring signal changes occurred during exposure in 1 patient, during or after pedicle screw insertion in 3, and during or after curve correction maneuvers after instrumentation in 3s. The signal changes were temporary. Correction maneuvers included pausing surgery for a few minutes, raising mean arterial pressure, changing the position of the extremities, and removing and reinserting screws. None of the surgeries required staging. The cause of the TcMEP changes was hypotension in 2 cases, screw malposition in 1, and corrective maneuvers in 2. There were no significant neurological complications in any patient, except 1 who developed axillary nerve palsy (Parsonage-Turner syndrome, or brachial neuritis). The patient recovered 3 months postoperatively; we believed that this was not related to the Ponte osteotomy. Two patients (2.3%) developed a wound infection needing surgical drainage. One patient developed a superior mesenteric artery syndrome 2 weeks postoperatively that was treated with hospitalization with no food or drink, nasogastric/nasojejunal tubes, and total parenteral nutrition. One patient with a positive family history developed deep vein thrombosis 12 days after surgery and was treated with warfarin and enoxaparin. There was 1 instrumentationrelated complication: a distal pedicle screw loosened in the vertebra. We noted no cases of obvious pseudarthrosis at the 2-year follow-up. Discussion Wide facetectomy consists of removing the inferior articular process of the cephalad level, releasing the capsule, and removing articular cartilage. The Ponte
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
201
Table 2 Sagittal radiographic measurements. Variable Lateral T2eT12 Cobb angle Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (hypokyphosis group!20 ) Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (normal kyphosis group 20 e40 ) Preoperatively First erect 1 year 2 years Lateral T5eT12 Cobb angle (hyperkyphosis groupO40 ) Preoperatively First erect 1 year 2 years Lateral T10eL2 Cobb angle Preoperatively First erect 1 year 2 years Lateral T2eT5 Cobb angle Preoperatively First erect 1 year 2 years Proximal junctional kyphosis First erect 1 year 2 years Distal junctional kyphosis First erect 1 year 2 years Lordosis (T12eS1) Preoperatively First erect 1 year 2 years Sagittal balance (C7 plumbline to sacrum) Preoperatively First erect 1 year 2 years
Mean
Standard deviation
p .03
27.0 29.0 30.3 30.5
14.5 9.2 9.7 10.5
17.8 18.5 18.9 19.5
4.0 5.8 6.3 7.4
.3
!.001 8.1 16.9 17.5 18.3 25.7
7.5 4.9 5.4 6.0 4.9
19.4 20 19.6
6.1 7.0 8.3
!.001
!.001 44.6 24.8 23.8 25.8
2.8 5.8 6.3 8.6
3.5 4.3 4.8 4.4
10.8 8.2 8.1 8.0
.47
.08 9.6 11.2 11.3 11.0
6.6 6.9 6.9 6.9
7.2 7.9 8.0
4.7 5.5 5.3
6.6 9.4 9.1
8.2 8.9 9.3
58.3 54.2 58.1 55.9
13.7 12.6 10.8 12.0
.2
!.001
.006
.01 0.6 0.76 0.71 0.14
3.8 4.0 3.1 2.8
osteotomy differs by removing the interspinous ligament and ligamentum flavum (an important tether to posterior lengthening) and resecting the entire facet (superior and
Fig. 3. Radiological data summarized in graphs. (A) Changes in mean coronal mid-thoracic (MT) Cobb angle in 87 patients and changes in mean sagittal T5e12 Cobb angle in 51 patients with preoperative hypokyphosis. (B) Changes in mean thoracic apical-C7 plumbline translation, mean thoracic apicalecentral sacral vertical line (CSVL), and translation and mean sagittal balance (C7 plumblineesacrum) in 87 patients.
inferior articular process) and ventral portion of the facet capsule and pars out to the foramen. The Ponte osteotomy was originally described for Scheuermann’s kyphosis [9,10], and Shufflebarger et al. [11,12] demonstrated its role in lumbar and thoracolumbar scoliosis. Indications for this procedure in thoracic scoliosis have not been well defined. Although a recent review article [24] mentioned that scoliosis greater than 70 to 75 that does not bend down to less than 40 is an indication in current practice, no reference or data are provided. Some surgeons have argued for the routine use of this osteotomy in AIS; consequently, this study provides valuable evidence-based information for spinal deformity surgeons for decision making. Because there is considerable variability in obtaining bending radiographs, we have been using the fulcrum bending film to assess the curve flexibility [23]. In thoracic hypokyphosis, normal kyphosis could also be restored effectively with an anterior release [25]; however, this would entail a combined anterior and posterior procedure. Imrie et al. [17] found that in Lenke 1 curves where high coronal plane correction (greater than 80%) was achieved, patients had a loss of kyphosis compared with the low coronal plane correction (less than 40%) group, in which there was a gain in kyphosis. In hyperkyphosis, there was a significant reduction of almost 20 in the mean kyphosis, and in
202
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Table 3 Scoliosis Research Society (SRS)e22 patient outcomes. SRS domain Pain Preoperatively 1 year 2 years Self-image Preoperatively 1 year 2 years General function Preoperatively 1 year 2 years Mental health Preoperatively 1 year 2 years Satisfaction Preoperatively 1 year 2 years Total Preoperatively 1 year 2 years
Median
Interquartile range
4.2 4.6 4.6
0.6 0.7 0.8
3.4 4.6 4.6
0.8 0.6 0.6
4.8 4.8 5.0
0.8 0.5 0.5
4.2 4.6 4.4
0.8 0.8 1.0
3.5 5.0 5.0
1.5 0.5 0.5
4.1 4.6 4.5
0.6 0.4 0.4
p !.001
!.001
.13 Fig. 4. Median SRS outcomes preoperatively and at 1 and 2 years. !.001
!.01
!.001
hypokyphosis we observed a mean increase in thoracic kyphosis of 10 . The Ponte osteotomy allows better segmental vertebral rotational correction, although we were unable to objectively assess this in our series because placement of thoracic pedicle screws prevented comparison of vertebral rotation by the Nash-Moe method on radiographs. A radiographic method of assessment of pedicle screw rotation would not allow comparison of the preoperative and postoperative vertebral rotation because the Nash-Moe and pedicle screw grading systems are different. Computed tomography scans were not routinely performed in our series, but there was an improvement in clinical scoliometer measurements. We have sought to improve thoracic restoration of kyphosis (alleviation of lordosis) in these patients, not just for cosmesis, but more importantly because an improvement in thoracic kyphosis results in an improvement in lumbar lordosis for the long term in these patients. Cho et al. [8] compared the results of Smith-Petersen versus pedicle subtraction osteotomy for the correction of fixed sagittal imbalance in adults with kyphotic spinal deformities, and reported that a Smith-Petersen osteotomy could achieve 10.7 of correction at each level. In our series, we found a significantly lower kyphosis correction rate of 2.5 per osteotomy level in the hypokyphotic curves. One reason could be the different patient populations. The series of Cho et al. included adult patients with a heterogeneous group of lumbar kyphotic deformities, who underwent primary or revision surgeries. Our series included AIS patients who underwent primary surgeries;
more than half had hypokyphosis. Hypokyphosis correction needs to be more controlled because excessive posterior distraction could result in spinal cord deficit and might increase the risk of pseudarthrosis. In the absence of a control group, it is difficult to draw definite conclusions. We reviewed our AIS patients who did not undergo Ponte osteotomies, but we found that there were significant baseline differences and lower preoperative mean Cobb angles, which made comparison subject to potential inaccuracies and bias. Lehman et al. [26] reported a single-center series of 114 consecutive AIS patients treated with posterior pedicle screw constructs with a mean age of 14.9 years, mean main thoracic coronal Cobb angle of 59.2 , mean preoperative lateral T5eT12 Cobb angle of 25.8 , and mean of 10 levels fused. The mean main thoracic coronal Cobb angle decreased from 59.2 to 16.8 postoperatively, the mean lateral Cobb angle decreased from 26.2 to 14.5 postoperatively, and the C7 sagittal plumbline decreased from 7.1 to 15.6 mm postoperatively [26]. Lehman et al. acknowledged the decrease in thoracic kyphosis, which they attributed to the use of the 5.5-mm stainless-steel rod, the use of monoaxial screws, and the absence of posterior soft tissue release in the lordotic segment. Suk et al. [27] reported improved thoracic kyphosis with the use of a stiffer, 7-mm, stainless-steel rod. Although we used the 5.5 rods in 85 of 87 patients, we were able to achieve kyphosis correction, presumably because of the Ponte osteotomies, pre-contouring of kyphosis in an ultra-high-strength, stainless-steel rod, and distraction of the concave apex. We cannot comment on the co-relation of the improved sagittal contour and the rod diameter based on the study data, because we used 6.35-mm rods in only 2 patients, whereas 85 patients had 5.5-mm rods. However Suk et al. showed better restoration of the sagittal balance with use of the 7-mm stainless-steel rods. The procedure of pedicle screw instrumentation and Ponte osteotomies may be associated with a relatively longer operating time and increased blood loss; however, we have no way to separate out the variables in the surgical technique that may affect the outcome without a true control group. Our institution’s practice of calculating EBL
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
is to add the weight of sponges rinsed in saline and squeezed for cell salvage to the cell salvage blood obtained, plus any estimates of blood on the drapes, confirmed by the anesthesiologist. There may be variation in the calculation of blood loss over multiple centers, and this may reflect institutional practices and not just surgeon or technical factors. We saw intraoperative neuromonitoring changes in 8% of our patients, although there were no significant or permanent neurological complications. The cause of the TcMEP changes was hypotension in 2 cases, screw malposition in 1, and corrective maneuvers (translation and distraction) in 2. All changes were reversed and signals returned to baseline with appropriate corrective maneuvers. Our monitoring team is well qualified and experienced, and prides itself on a spectacular track record of sensitivity and specificity of neurological alerts. Without doing a complete facetectomy on the concave side, it is not possible to obtain further sagittal and coronal plane correction. We believe that the contracted facet and soft tissues on the concave side are an important structural tether to correction of scoliosis and restoration of kyphosis. However, there will be a gap after correction. One danger of releasing the concave side of a thoracic curve is that the spinal cord hugs the concave pedicle at the apex, and if the curve apex is hypokyphotic or lordotic, the cord is posterior along the dorsal lamina and articular process. The surgeon must exercise extreme caution in this area when positioning the rongeur and resecting bone or soft tissue as part of the release. Our infection rate was 2.3% in this series of patients. The study population was a consecutive series of patients during a time when we had a cluster of infections. Since that time, our infection rate has returned to near 0 for AIS, even with the routine use of the Ponte osteotomy. Increased operative time and bleeding may be implicated as factors, but this has not been our experience in subsequent years. Although we observed no cases of pseudarthrosis over the 2-year followup period, this is a potential complication that will need to be re-evaluated on further follow-up. Because facetectomy is a critical component of obtaining a posterior fusion, removal of the facets along with posterior distraction of the laminae could theoretically impede fusion. Due attention should be paid to decortication of the laminae and transverse processes and use of bridging autograft along with allograft to obtain a posterolateral fusion. Excessive distraction should never be performed because of the risks of spinal cord deficit and pseudarthrosis. The use of TcMEP with this technique is important to improve detection of impending neurological deficit. Coe et al. [28] reported neurological complications in 14 patients (0.32%), wound infections in 59 patients (1.35%), deep vein thrombosis in 2 patients (0.05%), fatal blood loss in 1 patient (0.02%), and implant-related problems in 28 patients (0.64%), in a total of 4,369 patients who underwent posterior spinal instrumentation and fusion. Pseudarthrosis rates and neuromonitoring changes were not reported. Lehman et al. [26] reported 5 complications in 114 patients, including 3
203
deep wound infections and 2 cases of adding on extension to the primary fusion; there were no short-term or longterm neurological complications. Blood loss and neuromonitoring changes were not reported.
Limitations Although this is the first large, consecutive, single-center case series of Ponte osteotomies with pedicle screw instrumentation performed for thoracic AIS, this study has limitations. First, in the absence of a control group, we can make no conclusions about the increased operative time, blood loss, or neuromonitoring changes, because there is no way to separate out the variables in the surgical technique that may affect the outcome. Historical controls are not sufficient to make conclusions. Second, our follow-up of 2 years is not sufficient to evaluate pseudarthrosis, which is a potential risk with any spinal osteotomy. Third, SRS-22 questionnaire results were not available for all patients. When they were available, they showed that patients were satisfied with their appearance postoperatively. Fourth, we do not have the ability to perform body surface contour assessments, and suggest that these do not always correlate with Cobb angle measurement or degree of postoperative correction. Threedimensional analysis would probably enable better understanding; however, we did not have this information. We are currently collecting prospective data on 3-dimensional alignment changes preoperatively and postoperatively from EOS images (EOS Imaging, Paris, France), but this will require longer follow-up. The contour of the thoracic cage has been measured with the scoliometer and captured with digital photography. Scoliometer measurements have been reported for assessment of the horizontal plane and thoracic cage contour. Pree and postecomputed tomography scans of these patients were not available, because this practice exposes patients to a tremendous amount of radiation and is not our current clinical practice. We have presented detailed radiological data using universally accepted standard deformity measurements on lateral and coronal projections. Apart from Cobb angle, thoracic apicalC7 plumbline translation, thoracic apicalecentral sacral vertical line translation, and sagittal balance (C7 plumbline to sacrum) have also been measured. We believe that in combination, these measurements at least form a reasonable proxy for evaluation of 3-dimensional correction. Although we chose the T5eT12 and T2eT12 levels for measurement of thoracic kyphosis to allow standardization for our prospective scoliosis database, these may not have been the angle between the maximally tilted thoracic vertebrae. We acknowledge that the mean coronal Cobb angle and fulcrum flexibility indicates that the curves were not too severe or stiff; this is because we have included a consecutive series of patients who underwent the Ponte procedure at our center with 2-year follow-up regardless of curve severity or stiffness, to avoid selection bias.
204
S.A. Shah et al. / Spine Deformity 1 (2013) 196e204
Conclusions In this case series of major thoracic AIS curves treated with segmental pedicle screw instrumentation and Ponte osteotomies, there was improvement in the coronal and sagittal radiological parameters. A prospective controlled study is needed to determine whether pedicle screw instrumentation and Ponte osteotomies influence outcomes and complications. References [1] Hibbs RA. A report of fifty-nine cases of scoliosis treated by the fusion operation. J Bone Joint Surg Am 1924;6:3e34. [2] Moe JH. A critical analysis of methods of fusion for scoliosis: an evaluation in two hundred and sixty-six patients. J Bone Joint Surg Am 1958;40:529e54. [3] Smith-Petersen MN, Larson CB, Aufranc OE. Osteotomy of the spine for correction of flexion deformity in rheumatoid arthritis. J Bone Joint Surg Am 1945;27:1e11. [4] Lagrone MO, Bradford DS, Moe JH, et al. Treatment of symptomatic flatback after spinal fusion. J Bone Joint Surg Am 1988;70: 569e80. [5] Voos K, Boachie-Adjei O, Rawlins BA. Multiple vertebral osteotomies in the treatment of rigid adult spine deformities. Spine (Phila Pa 1976) 2001;26:526e33. [6] McMaster MJ. A technique for lumbar spinal osteotomy in ankylosing spondylitis. J Bone Joint Surg Br 1985;67:204e10. [7] Kostuik JP, Maurais GR, Richardson WJ, et al. Combined single stage anterior and posterior osteotomy for correction of iatrogenic lumbar kyphosis. Spine (Phila Pa 1976) 1988;13:257e66. [8] Cho KJ, Bridwell KH, Lenke LG, et al. Comparison of SmithPetersen versus pedicle subtraction osteotomy for the correction of fixed sagittal imbalance. Spine (Phila Pa 1976) 2005;30:2030e7. [9] Ponte A, Vero B, Siccardi GL. Surgical treatment of Scheuermann’s hyperkyphosis. In: Winter RB, editor. Kyphosis (Progress in Spinal Pathology). Bologna, Italy: Aulo Gaggi; 1984. p. 75e80. [10] Ponte A. Posterior column shortening for Scheuermann’s kyphosis: an innovative one-stage technique. In: Haher T, Merola A, editors. Surgical Techniques for the Spine. 1st ed. New York, NY: Thieme Verlag; 2003. p. 107e13. [11] Shufflebarger HL, Clark CE. Effect of wide posterior release on correction in adolescent idiopathic scoliosis. J Pediatr Orthop B 1998;7:117e23. [12] Shufflebarger HL, Geck MJ, Clark CE. The posterior approach for lumbar and thoracolumbar adolescent idiopathic scoliosis: posterior shortening and pedicle screws. Spine (Phila Pa 1976) 2004;29:269e76. [13] Geck MJ, Macagno A, Ponte A, et al. The Ponte procedure: posterior only treatment of Scheuermann’s kyphosis using segmental posterior shortening and pedicle screw instrumentation. J Spinal Disord Tech 2007;20:586e93.
[14] Kim YJ, Lenke LG, Kim J, et al. Comparative analysis of pedicle screw versus hybrid instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2006;31: 291e8. [15] Liljenqvist U, Lepsien U, Hackenberg L, et al. Comparative analysis of pedicle screw and hook instrumentation in posterior correction and fusion of idiopathic thoracic scoliosis. Eur Spine J 2002;11: 336e43. [16] Lowenstein JE, Matsumoto H, Vitale MG, et al. Coronal and sagittal plane correction in adolescent idiopathic scoliosis: a comparison between all pedicle screw versus hybrid thoracic hook lumbar screw constructs. Spine (Phila Pa 1976) 2007;32:448e52. [17] Imrie M, Yaszay B, Bastrom TP, et al. Adolescent idiopathic scoliosis: should 100% correction be the goal? J Pediatr Orthop 2011;31(1 Suppl):S9e13. [18] Newton PO, Yaszay B, Upasani VV, et al. Preservation of thoracic kyphosis is critical to maintain lumbar lordosis in the surgical treatment of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2010;35:1365e70. [19] Halanski MA, Cassidy JA. Do multilevel Ponte osteotomies in thoracic idiopathic scoliosis surgery improve curve correction and restore thoracic kyphosis? J Spinal Disord Tech 2011 Dec 21 [Epub ahead of print]. [20] Pizones J, Izquierdo E, Sanchez-Mariscal F, et al. Does wide posterior multiple level release improve the correction of adolescent idiopathic scoliosis curves? J Spinal Disord Tech 2010;23:e24e30. [21] Kim YJ, Lenke LG, Bridwell KH, et al. Free hand pedicle screw placement in the thoracic spine: is it safe? Spine (Phila Pa 1976) 2004;29:333e42. [22] Shah SA. Derotation of the spine. Neurosurg Clin N Am 2007;18: 339e45. [23] Luk KD, Cheung KM, Lu DS, et al. Assessment of scoliosis correction in relation to flexibility using the fulcrum bending correction index. Spine (Phila Pa 1976) 1998;23:2303e7. [24] Diab MG, Franzone JM, Vitale MG. The role of posterior spinal osteotomies in pediatric spinal deformity surgery: indications and operative technique. J Pediatr Orthop 2011;31:S88e98. [25] Sucato DJ, Agrawal S, O’Brien MF, et al. Restoration of thoracic kyphosis after operative treatment of adolescent idiopathic scoliosis: a multicenter comparison of three surgical approaches. Spine (Phila Pa 1976) 2008;33:2630e6. [26] Lehman Jr RA, Lenke LG, Keeler KA, et al. Operative treatment of adolescent idiopathic scoliosis with posterior pedicle screw-only constructs: minimum three-year follow-up of one hundred fourteen cases. Spine (Phila Pa 1976) 2008;33:1598e604. [27] Suk SI, Lee SM, Chung ER, et al. Selective thoracic fusion with segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis: more than 5-year follow-up. Spine (Phila Pa 1976) 2005;30:1602e9. [28] Coe JD, Arlet V, Donaldson W, et al. Complications in spinal fusion for adolescent idiopathic scoliosis in the new millennium: a report of the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976) 2006;31:345e9.