Wh e n idiopathic adolescent scoliosis involves 2

J Neurosurg Spine 10:000 000, 10:214 219, 2009 Shoulder balance after surgery in patients with Lenke Type 2 scoliosis corrected with the segmental pedicle screw technique Clinical article *Mi n g Li, M.D., 1 Su x i Gu, M.D., 1 Ji a n q i a n g Ni, M.D., 2 Xi u t o n g Fa n g, M.D., 3 Xi a o d o n g Zh u, M.D., 1 a n d Zh i y u Zh a n g, M.D. 1 1 Department of Orthopedics, Scoliosis Research Center, Changhai Hospital, Second Military Medical University, Shanghai; 2 Department of Orthopedics, Yuhuangding Hospital of Tsingtao Medical University, Yantai, Shangdong; and 3 Department of Orthopedics, Beijing Shijitan Hospital, Ninth Clinical Medical College of Peking University, Beijing, China Object. The authors evaluated the effectiveness of Lenke Type 2 criteria in scoliosis correction with the segmental pedicle screw (PS) technique, with emphasis on shoulder balance. Methods. Twenty-five consecutive patients with Lenke Type 2 scoliosis (structural double thoracic curves, sidebending Cobb angle > 25, or T2 5 kyphosis > 20 ) who underwent segmental PS instrumentation were included in this study. At surgery, the patients were an average of 14.1 years of age, and the average duration of follow-up was 2.9 years. For radiological evaluation of the patients, preoperative, postoperative, and the latest available follow-up radiographs were used. The difference between right and left shoulder heights was determined to assess shoulder balance. All patients were treated with fusion of both the proximal and distal curves. Results. The mean preoperative proximal thoracic curve of 43 was corrected to 21 postoperatively, a 51.2% correction. The preoperative lower thoracic curve of 61 was corrected to 23, for a 62.3% correction. The preoperative shoulder height difference of 5.92 ± 12.52 mm (range: 31 to +14 mm, negative designating a lower left shoulder) was improved to 1.52 ± 8.12 mm. Postoperatively, no patient had significant or moderate shoulder imbalance, 4 patients had minimal shoulder imbalance, and 21 patients had balanced shoulders. Conclusions. Although Lenke Type 2 criteria were developed wth Cotrel-Dubousset instrumentation, they are successfully applied to determining thoracic fusion when segmental PS instrumentation is used. (DOI: 10.3171/2008.11.SPINE08524) Ke y Wo r d s idiopathic double structural thoracic scoliosis segmental pedicle screw instrumentation shoulder balance Wh e n idiopathic adolescent scoliosis involves 2 thoracic curves, an important preoperative decision is whether fusion and instrumentation of both curves is necessary, or whether, after fusion and instrumentation of the main curve only, the upper, proximal curve will spontaneously correct itself in compensation for the surgical straightening of the lower curve. The authors of several studies have addressed this question and devised guidelines to use in deciding on the appropriate protocol. 5,7 9,16,18,19 Surgical techniques used to treat idiopathic adolescent scoliosis have changed and improved over the Abbreviation used in this paper: PS = pedicle screw. * Drs. Li and Gu contributed equally to this study. years, with each change increasing the extent of curvature correction possible. This has lessened the predictive power of the older guidelines because the upper thoracic curve involving fewer vertebrae and attached to the rigid rib cage is stiffer than the lower curve, and the flexibility that was adequate to compensate for the amount of straightening achievable in the lower curve with older methods cannot fully cope with the increased straightening power of the newer methods. In a series of studies, Lenke et al. 10 14 devised a newer scoliosis classification system suitable for the newer hook and rod (Cotrel-Dubousset) system that replaced the older rod (Harrington) system on which the original classification was based. This classification system is now being applied to the segmental PS fixation system, the newest system currently in use. Pedicle screw fixation attaches 214 J. Neurosurg.: Spine / Volume 10 / March 2009

Shoulder balance in Lenke Type 2 scoliosis treated with PSs the instrumentation more firmly to the vertebrae than hooks or wires, and enables an even greater correction of the major curve to be made. Thus, the Lenke criteria may or may not be completely suitable for PS instrumentation. One of the most important measures of success in scoliosis surgery is the achievement of shoulder balance because of its importance for patient appearance and satisfaction. Our aim in the following study was to report the effects of surgical correction on shoulder balance in a patient population with double thoracic curves, the proximal curve fulfilling Lenke criteria for double fusion, who were treated with double fusions using the segmented PS fixation technique. Methods Patient Selection Twenty-five consecutive patients with Lenke Type 2 double structural thoracic curves, who underwent double fusion with segmental PS fixation and were followed up for > 2 years, were selected and their records reviewed retrospectively. J. Neurosurg.: Spine / Volume 10 / March 2009 Data Collection The main and proximal thoracic curves, shoulder balance, and spinal balance were measured on standing anteroposterior radiographs. The flexibility of each curve was measured using supine side-bending radiographs. Kyphosis was measured for the T2 5, T5 12, and T10 L2 vertebral segments using lateral radiographs. Curve angles were estimated on standing, side-bending, and lateral radiographs using the Cobb technique. Spinal imbalance was measured on the standing anteroposterior radiograph as the deviation from the central sacral line of a plumb line dropped down from T-1. A deviation > 2 cm was considered significant. Shoulder balance was determined from the difference on the standing radiograph between the heights of the lateral ends of the left and right clavicles. A shoulder height difference > 5 mm was considered significant, and was considered positive when the left shoulder was higher than the right, and negative when the left shoulder was lower than the right. Side-bending and T2 5 curvatures were used to select patients with Lenke 2 scoliosis. Patients with Lenke 2 scoliosis are by definition unable to straighten their proximal thoracic curves to < 25 curvature, or have an upper thoracic (T2 5) kyphosis > 20, and thus are recommended for upper thoracic and lower curve fusion surgeries. Spinal balance, lower thoracic (T5 12) kyphosis, thoracolumbar kyphosis (T10 L2), and lumbar (T12 L5) lordosis were measured to determine the Lenke 2 subcategory classifications: lumbar spine modifier (A, B, or C) and thoracic sagittal modifier (, neutral [N], or +). 12 Shoulder balance was the outcome variable, and was recorded as the difference (in millimeters) between left and right shoulder heights. Shoulder balance was also graded as follows: Grade 3, significant imbalance, > 3 cm; Grade 2, moderate imbalance, 2 3 cm; Grade 1, minimal imbalance, 1 2 cm; Grade 0, balanced, 0 1 cm. 12 Postoperative proximal and main thoracic curvatures, shoulder height difference, and spinal balance were determined using the standing radiograph obtained at the last follow-up visit > 2 years after surgery. Surgical Technique All surgeries were performed by the first author (M.L.) using PS fixation. For the fusion of both the proximal and distal curves, instrumentation and fusion were performed from T-1 or T-2 to the distal neutral vertebra of the lower thoracic curve. Monoaxial PSs were inserted segmentally on the correction side (concave sides) of both the upper and lower thoracic curves, and in every second or third vertebra on the support side (convex sides of the curves). After screw insertion, the upper curve was corrected with a 90 clockwise rotation of a concave rod contoured to the normal sagittal profile of the instrumented segment. Rods were then inserted on the convex sides of both curves and connected to the concave rods using rod connectors. In some patients, when the correction of the proximal curve resulted in enough straightening of the curve to allow insertion of a longer rod contoured to the normal sagittal contour of the instrumented segment, a longer rod spanning the entire fusion length was substituted for short rods to increase the stability of the construct. After locking the rods, the longitudinal members were cross-linked by transfixators. Clinical Evaluation Interpretation of the shoulder height difference in all cases was confirmed clinically by the treating surgeon, using the grading system described above. Shoulder imbalance was categorized as significant (> 3 cm), moderate (2 3 cm), minimal (1 2 cm), or absent (< 1 cm side-toside difference). Results Demographics and Preoperative Data The 25 patients (including 8 male and 17 female patients) all had Lenke Type 2 scoliosis (that is, both proximal and distal curves were classified as structural) and all underwent fusion of both curves with the segmental PS fixation technique. The mean preoperative proximal thoracic curvature was 43. During side bending it straightened 25 40, a 42% decrease in curvature. The mean preoperative main thoracic curvature was 61, and straightened 15 during side bending to 46, a 25% decrease in curvature. In 6 patients the left shoulder was elevated > 5 mm above the right; 8 patients had level shoulders (< 5 mm difference); and in 11 patients the right shoulder was elevated > 5 mm above the left. The complete data set for all patients is shown in Table 1. Curve Correction, Trunk Balance, and Shoulder Height Surgery achieved a 51.2% correction of the proximal curve, from the 43 preoperative value to 21 postoperatively. In the lower thoracic curve, surgery achieved a 215

M. Li et al. TABLE 1: Double thoracic spinal fusions in patients with Lenke Type 2 scoliosis* Case No. Age at Op (yrs) Length of FU (yrs) PT PT SB PT IMPO PT PO 1 12.3 2.1 44 35 37 42 56 33 36 37 2BN 6 9 8 0 2 12.4 3.6 32 27 17 22 88 45 41 43 2CN 14 16 5 4 3 14.2 2.2 38 28 20 23 56 40 23 33 2AN 10 12 6 0 4 12.1 2.5 32 31 22 26 52 45 36 35 2AN 6 9 7 4 5 15.6 3.1 34 26 22 27 62 53 22 30 2AN 11 14 14 7 6 13.4 2.4 45 25 24 27 79 51 33 36 2AN 7 9 12 5 7 16.0 2.5 40 25 15 21 54 22 19 24 2AN 0 0 8 0 8 16.9 3.0 38 33 21 19 60 33 24 24 2AN 0 0 0 2 9 16.7 4.0 56 31 35 38 47 21 25 30 2AN 3 0 10 3 10 13.3 2.6 56 35 33 30 52 26 23 27 2BN 0 0 8 3 11 12.6 3.3 45 40 25 28 94 54 26 29 2CN 0 9 13 8 12 13.2 3.1 53 26 18 19 66 23 10 10 2BN 0 0 0 5 13 14.0 2.2 55 32 16 20 49 20 15 16 2AN 2 8 12 5 14 13.7 4.0 40 25 18 20 51 28 18 19 2AN 0 7 7 0 15 13.5 3.5 46 26 16 14 46 21 15 17 2AN 13 8 10 7 16 13.0 2.7 32 26 14 16 42 22 17 20 2AN 22 7 0 8 17 13.8 2.3 36 27 15 17 53 21 20 24 2BN 21 8 22 0 18 17.8 3.5 37 24 13 15 41 20 15 17 2AN 31 14 14 7 19 12.7 2.5 44 28 20 21 63 38 27 31 2BN 14 6 17 10 20 14.3 2.0 45 30 17 18 58 27 12 15 2AN 25 9 5 0 21 15.0 3.7 50 40 20 22 88 37 24 26 2CN 17 7 15 5 22 12.7 3.8 60 42 40 43 72 59 45 47 2CN 19 6 5 0 23 14.5 4.1 35 25 20 18 79 35 26 24 2CN 17 3 25 21 24 14.0 3.7 37 29 10 13 58 21 18 20 2AN 7 4 27 0 25 14.5 2.4 43 29 25 28 66 30 16 18 2CN 15 3 24 7 * FU = follow-up; = main thoracic upright Cobb; IMPO = main thoracic immediate postoperative Cobb; PO = main thoracic final postoperative Cobb; SB = main thoracic side-bending Cobb; PT = proximal thoracic upright Cobb measurement; PT IMPO = proximal thoracic immediate postoperative Cobb; PT PO = proximal thoracic final postoperative Cobb; PT SB = proximal thoracic side-bending Cobb; SHD = shoulder height difference; TS = trunk shift. According to the Lenke Classification, A, B, and C are lumbar spine modifiers and N indicates normal sagittal thoracic alignment. Negative values designate low left shoulder. Negative values designate a shift to the right. SB IMPO PO Lenke Type Preop SHD Final SHD Preop TS Final TS 62.3% decrease, from the preoperative value of 61 to a postoperative value of 23. Three patients had deranged trunk balance before surgery. Postoperatively 2 patients showed improved balance and 1 remained unchanged. There was no new postsurgical trunk imbalance in patients who had had balanced trunks before surgery. After surgery, left shoulder elevation was increased in all patients. This increase would, of course, increase shoulder balance in those with initial right shoulder elevation, and decrease shoulder balance in those with initial normal shoulders or left shoulder elevation. Table 1 shows the 25 patients divided into 3 classes, according to preoperative shoulder balance status: those with left shoulder elevation (Cases 1 6), those with normal shoulder balance (Cases 7 15), and those with right shoulder elevation (Cases 16 25). Although surgery raised the left shoulder relative to the right in all patients, this effect was minimal in the 2 groups in which it would worsen the outcome, amounting to 2.5 mm in the preoperatively left-elevated group and 3 mm in the preoperatively normal shoulder group. In contrast, the effect was substantial (a difference of 13.3 mm) in the group with preoperative right shoulder elevation, decreasing the imbalance to < 10 mm in all but 1 patient in this group. Before surgery, 1 patient had significant shoulder imbalance (Grade 3), 3 had moderate (Grade 2), 9 had minimal (Grade 1), and 12 patients (48%) had balanced shoulders (Grade 0). After surgery, no patients had significant or moderate shoulder imbalance, only 4 patients had minimal shoulder imbalance, and 21 patients (84%) had balanced shoulders. 216 J. Neurosurg.: Spine / Volume 10 / March 2009

Shoulder balance in Lenke Type 2 scoliosis treated with PSs Surgical Complications There were no neurological or vascular complications related to PS placement or instrumentation in any patient, and no patient experienced postoperative worsening of shoulder and trunk imbalance. There were no cases of pseudarthrosis or implant failure. One patient had a spontaneous pneumothorax that was treated by chest tube insertion (Fig. 1). Discussion Our results demonstrate that fusing both the main and proximal thoracic curves with PS fixation in patients who fulfill the Lenke criteria for double fusions (patients with Lenke Type 2 scoliosis) gives good results for shoulder balance. Of our 25 patients, 21 (84%) ended up with normal shoulder balance, and the other 4 (16%) had only minimal imbalance. The Lenke Type 2 classification criteria include the presence of a double thoracic curve, corresponding to King Type V.21 In young patients King found that although an upper curve was present, fusion of the lower curve only caused deterioration of shoulder balance. He thought in these cases that overcorrection of the lower curve overwhelmed the ability of the upper curve to compensate, and recommended fusion of both curves for patients with King Type V curves.6 Pedicle screws, which provide a strong corrective force for the lower curve, are becoming popular worldwide. This strong correction makes more crucial the ability to determine whether the upper curve should be fused. Lenke et al.11 argued that if the preoperative bending Cobb angle was > 25 and/or the T2 5 kyphosis was > 20, the upper curve should be surgically fused. In their later study using the Lenke criteria to decide whether to fuse the upper curve, Cil and colleagues3 found that the postoperative T-1 tilt, the clavicular angle, and the height of the coracoid process in patients with fused and unfused upper curves were similar. This result suggests that the Lenke criteria may be the proper standard to use for upper curve fusion. The most recent classification system by the Scoliosis Research Society,1,15 does not shed light on this matter because it is designed specifically for adult spinal deformities. In the surgeries we performed in the present study, monoaxial PSs were used without the assistance of extension holders, and the anatomic placement technique. The PS technique provided a stronger correction force in both the sagittal and coronal planes. In the PS technique, the correction force is applied in the concave side. This is the left side in most cases; thus the left shoulder is usually elevated. Intraoperative radiography was used to ensure achievement of shoulder balance prior to leaving the operating room. No vertebral derotation was performed in this cohort, and it is controversial whether screws or hooks have better correction power. Kim and colleagues5 believed that segmental PS instrumentation offered a significantly better major curve correction, less perioperative blood loss, and improved pulmonary function values than hook constructs, and did not generate the neurological problems associated with the latter instrumention. However, Vora et al.20 argued that the PS does not give a better correction of Lenke Type 2 adolescent idiopathic scoliosis in the coronal plane than the hybrid construct or the universal system with a segmental collar button intraspinous wire construct. Vertebral derotation and proximal thoracic curve convex compression/concave distraction are certainly helpful in controlling postoperational shoulder imbalance. However, in selective lower curve fusion, the correction power of the lower curve is hard to predict precisely, and uncertainty as to the outcome is inevitable. Double fusion is therefore a safer option in Lenke Type 2 deformities. We believe that PS instrumentation achieves true 3-column fixation and enhanced mechanical control of Fig. 1. Radiographs obtained in a 17-year-old girl. Preoperatively, the patient had an upper thoracic curve of 53 and lower thoracic curve of 66 (A and B). Her shoulders were level and there were no abnormalities in the sagittal plane. The patient underwent fusion of both the upper and lower thoracic curves from T2 L1 (C). Image obtained 2 years postoperatively demonstrates that the upper curve is 19 and the lower curve is 10 (D). The shoulders remained level. J. Neurosurg.: Spine / Volume 10 / March 2009 217

M. Li et al. instrumented vertebral segments compared to hook and wire construct. Pedicle screw constructs have demonstrated improved stabilization in the lumbar spine and improved thoracic deformity correction in adolescent idiopathic scoliosis compared to traditional hook-based constructs. 2 The correction of rib cage deformity by instrumentation alone is possible, but reports of its efficacy vary widely. Harvey et al. 4 reported that the success rate for Cotrel-Dubousset instrumentation alone was 82% in 83 patients. However Shufflebarger and King 17 showed the efficacy rate for derotation by Cotrel-Dubousset instrumentation was merely 37%. None of our patients had significant rib cage deformities, so we have no data on the use of instrumentation for this condition. Results of Other Studies There are no other reports describing correction of shoulder balance in a cohort exactly like ours, that is, consisting only of patients with Lenke Type 2 scoliosis, who underwent double fusion with PS fixation. In the following discussion we compare our data with the most similar literature available. A report by Suk et al., 18 whose primary purpose was to ascertain when double thoracic fusions were needed and when single fusions would suffice, contains pre- and postoperative shoulder height data in 12 patients (out of 40) who, like our patients, had Lenke Type 2 scoliosis and underwent double fusions with PS fixation. When data from these 12 patients are analyzed separately, Suk and colleagues results are similar to ours. The degree of surgical correction of the main thoracic curve in these patients was identical to ours (60%). After surgery, 75% of their patients had balanced shoulders and 25% had minimal imbalance. This is a slightly worse outcome than the one we report (84 and 16%), however this difference in outcome is probably due to the somewhat higher incidence of preoperative moderate or severe imbalance in their patients (33%) compared to ours (16%). Kuklo et al. 7 measured shoulder balance in their study comparing outcomes after anterior and posterior fusion of the main thoracic curve. Of the 10 patients with Lenke Type 2 scoliosis in their 85-patient group, 6 had moderate and 3 minimal shoulder imbalances preoperatively. Surgery reduced the group with shoulder imbalance to 1 patient with moderate and 3 with minimal imbalance. The type of instrumentation used in these patients was not recorded, but the correction of the main thoracic curve (52%) achieved was similar to that seen with the older types of instrumentation. These postoperative results for single fusion in patients with Lenke Type 2 scoliosis (40% with some imbalance) are worse than our results (16% with some imbalance) using double fusion. However, Kuklo et al. s patients, like the patients in the study by Suk and associates, had worse preoperative imbalance than our patients. Cil and colleagues 3 investigated whether proximal curves that fulfilled the King rules for double fusion, but were too mild or flexible to fulfill the newer Lenke rules, really needed double fusions. The main curve correction was ~ 56% whether the proximal curve was included in the surgery. Cil et al. did not measure shoulder height, but their data for clavicle angle, another measure of shoulder imbalance, showed improvement from 3.3 to 0.4 when the proximal curve was included in the fusion, and from 1.5 to 1.1 in patients who only underwent fusion of the main curve. The authors concluded that these patients, who would have undergone double fusion surgery under the King but not the Lenke rules, did not need proximal curve fusion. The postoperative increase in clavicle angle is similar in patients with single and double fusions, so as far as what can be told about shoulder imbalance from clavicle angle, these patients falling just outside the Lenke rules for double fusion do not seem to have needed fusion of the upper curve. In a much older study, Lee and associates 9 reported on the results of double thoracic fusion on shoulder balance in 138 patients with an average proximal thoracic curvature of 37. Before surgery, 48% had balanced shoulders (as in our study), and 63% had balanced shoulders postoperatively. Among our patients, 84% had balanced shoulders after surgery. The major difference between their study and ours was that theirs was performed before PS fixation had become a common practice, and using the older techniques, they achieved only 38% correction of the main thoracic curve, compared to our 60% correction of this curve. Looking at the 4 studies as a group, we can draw 2 tentative conclusions. First, in patients with Lenke Type 2 scoliosis who undergo double fusions, success in restoring shoulder balance may be related to the degree of correction of the main thoracic curve; the 2 studies in which PS surgery was used reported better outcomes than the 2 studies in which the older methods were used. Second, the results of Cil et al. suggest that patients with curvatures not quite severe enough to fall within the Lenke 2 classification seem to do equally well regardless of whether the proximal curve is fused or left unfused. Effect of Including Kyphosis Under the Lenke classification, either of 2 defects in the proximal thoracic curve could place a patient into the Lenke 2 category too much rigidity or too much kyphosis. None of the other classification systems use kyphosis to categorize the proximal curve. Proximal kyphosis patients will obviously benefit from double fusions, but including them in the double fusion category along with patients with overly rigid upper curves will, by itself, influence the shoulder balance data. How much this inclusion affects the data is not known. Shortcomings and Strengths of the Present Study Shortcomings of this study are its retrospective nature and lack of a control group. However, our primary intention was to provide insight into the problems encountered in day-to-day practice, using examples from our own practice. Another shortcoming is that no data on patient satisfaction are included. A strength of the study, however, is that uniform inclusion criteria were used, and all patients underwent treatment with the same surgical technique. 218 J. Neurosurg.: Spine / Volume 10 / March 2009

Shoulder balance in Lenke Type 2 scoliosis treated with PSs Conclusions Our results demonstrate that the segmental PS technique applied to double-fusion treatment of Lenke Type 2 scoliosis can yield good results. We believe that the Lenke criteria are suitable for use in deciding which curves require fusion and which do not that is, a proximal thoracic bending curve that does not straighten to < 25 should be treated with double fusion. References 1. Berven SH, Lowe T: The Scoliosis Research Society classification for adult spinal deformity. Neurosurg Clin N Am 18: 207 213, 2007 2. Bess RS, Lenke LG, Bridwell KH, Cheh G, Mandel S, Sides B: Comparison of thoracic pedicle screw to hook instrumentation for the treatment of adult spinal deformity. Spine 32:555 561, 2007 3. Cil A, Pekmezci M, Yazici M, Alanay A, Acaroglu RE, Deviren V, et al: The validity of Lenke criteria for defining structural proximal curves in patients with adolescent idiopathic scoliosis. Spine 30:2550 2555, 2005 4. Harvey CJ, Betz RR, Clements DH, Huss GK, Clancy M: Are there indications for partial rib resection in adolescent idiopathic scoliosis patients treated with Contrel-Dubousset instrumentation? Spine 18:1593 1598, 1993 5. Kim YJ, Lenke LG, Kim J, Bridwell KH, Cho SK, Cheh G, et al: A comparative analysis of pedicle screw versus hybrid instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine 31:291 298, 2006 6. King HA, Moe JH, Bradford DS, Winter RB: Selection of fusion levels in thoracic idiopathic scoliosis. J Bone Joint Surg Am 65:1302 1313, 1983 7. Kuklo TR, Lenke LG, Graham EJ, Won DS, Sweet FA, Blanke KM, et al: Correlation of radiographic, clinical, and patient assessment of shoulder balance following fusion versus nonfusion of the proximal thoracic curve in adolescent idiopathic scoliosis. Spine 27:2013 2020, 2002 8. Kuklo TR, Lenke LG, Won DS, Graham EJ, Sweet FA, Belz RR, et al: Spontaneous proximal thoracic curve correction after isolated fusion of the main thoracic curve in adolescent idiopathic scoliosis. Spine 26:1966 1975, 2001 9. Lee CK, Denis F, Winter RB, Lonstein JE: Analysis of the upper thoracic curve in surgically treated idiopathic scoliosis. A new concept of the double thoracic curve pattern. Spine 18:1599 1608, 1993 10. Lenke LG, Betz RR, Bridwell KH, Clementts DH, Harms J, Lowe TO, et al: Intraobserver and interobserver reliability of the classification of thoracic adolescent idiopathic scoliosis. J Bone Joint Surg Am 80:1097 1106, 1998 11. Lenke LG, Betz RR, Haher TR, Lapp MA, Merola AA, Harms J, et al: Multisurgeon assessment of surgical decision-making in adolescent idiopathic scoliosis. Curve classification, operative approach, and fusion levels. Spine 26:2347 2353, 2001 12. Lenke LG, Betz RR, Harms J: Adolescent idiopathic scoliosis. A new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am 83:1169 1181, 2001 13. Lenke LG, Bridwell KH, Baldus C, Blanke K: Preventing decompensation in King Type II curves treated with Cotrel- Dubousset instrumentation: strict guidelines for selective thoracic fusion. Spine 17:S274 S281, 1992 14. Lenke LG, Bridwell KH, O Brien MF, Baldus C, Blanke K: Recognition and treatment of the proximal thoracic curve in adolescent idiopathic scoliosis treated with Cotrel Dubousset instrumentation. Spine 19:1589 1597, 1994 15. Lowe T, Berven SH, Schwab FJ, Bridwell KH: The SRS classification for adult spinal deformity: building on the King/Moe and Lenke classification systems. Spine 31 (19 Suppl):S19 S125, 2006 16. Margulies JY, Floman Y, Robin GC, Neuwirth MG, Kuflik P, Weidenbaum M, et al: An algorithm for selection of instrumentation levels in scoliosis. Eur Spine J 7:88 94, 1998 17. Shufflebarger HL, King WF: Composite measurement of scoliosis-a new method of analysis of the deformity. Spine 12: 228 232, 1987 18. Suk SI, Kim JH, Kim WJ, Lee SM, Liu Y, Lee CS, et al: Treatment of King type V adolescent idiopathic scoliosis by separate curve derotation technique with segmental pedicle screw fixation. J Korean Soc Spine Surg 5:224 230, 1998 19. Suk SI, Kim WJ, Lee CS, Kim JH, Chung ER, Lee JH: Indications of proximal thoracic curve fusion in thoracic adolescent idiopathic scoliosis. Spine 25:2342 2349, 2000 20. Vora V, Orth MS, Crawford A, Babekhir N, Boachie-Adjei O, Lenke L, et al: A pedicle screw construct gives an enhanced posterior correction of adolescent idiopathic scoliosis when compared with other constructs. Spine 32:1869 1874, 2007 21. Winter RB, Denis F: The King type V curve pattern. Its analysis and surgical treatment. Orthop Clin North Am 25:353 362, 1994 Manuscript submitted December 19, 2007. Accepted November 25, 2008. Address correspondence to: Xiutong Fang, M.D., Department of Orthopedics, Beijing Shijitan Hospital, The Ninth Clinical Medical College of Peking University, Beijing, 100038, China. email: fangxiutong@163.com. J. Neurosurg.: Spine / Volume 10 / March 2009 219