Clinically Significant Differences Exist Between Curves in Operative Idiopathic Early-Onset Scoliosis and Adolescent Idiopathic Scoliosis

McElroy, Mark J. MS*; Sponseller, Paul D. MD*; Fuhrhop, Sara K. BS*; Russell, Colin J.*; Newton, Peter O. MD†,‡; Marks, Michelle C. MA, PT; Sanders, James O. MD§; Yazici, Muharrem MD; Pawelek, Jeff B. BS; Akbarnia, Behrooz A. MD‡,‖; the Harms Study Group, and the Growing Spine Study Group

doi: 10.1097/BRS.0b013e318294eb37

Study Design. Retrospective analysis.

Objective. To determine if statistically significant differences exist between operative idiopathic early-onset scoliosis (IEOS) and adolescent idiopathic scoliosis (AIS) in primary curve characteristics, stable vertebra, kyphosis, or lumbar deformity and, if so, to identify the clinical significance of these differences.

Summary of Background Data. To our knowledge, no study has statistically compared radiographical measures of operative IEOS and AIS.

Methods. We identified operative patients (60, IEOS; 1537, AIS) in 2 multicenter databases and measured preoperative radiographical parameters of interest. The measurements were compared using the Student t test and other appropriate statistical methods (significance, P= 0.05).

Results. The IEOS and AIS groups were significantly different in primary curve magnitude (70°± 20°vs. 54°± 13°, respectively; P< 0.001), stable vertebra location (L3.0 ± 1.4 vs. L2.1 ± 2.2, respectively; P= 0.001), and T2 to T12 kyphosis (40°± 15°vs. 31°± 13°, respectively; P< 0.001). Distribution of major curve apex was unimodal centered close to the thoracolumbar junction in IEOS versus bimodal in AIS. Primary curve type was thoracic in 83% and 79% and thoracolumbar/lumbar in 17% and 21% of patients with IEOS and AIS, respectively. When the overall cohorts were separated into curve types and these subsets were compared, statistically significant differences were found between IEOS and AIS in thoracic primary curves (apex, magnitude, direction, proximal and distal Cobb vertebrae, and lumbar deformity) and thoracolumbar/lumbar primary curves (magnitude).

Conclusion. Significant radiographical differences exist between operative IEOS and AIS curves. IEOS curves are greater in magnitude, more kyphotic, less well compensated, and have a more caudal apex and stable vertebra. These findings suggest that younger patients may require more distal instrumentation and that proximal fixation techniques should consider the additional pullout forces created by the greater kyphosis.

Level of Evidence: 3

We statistically compared radiographical characteristics between operative idiopathic early-onset scoliosis (60 patients) and adolescent idiopathic scoliosis (1537 patients). Younger patients showed a larger curve, more severe kyphosis, less compensation, and a more caudal apex and stable vertebra. They may therefore require more distal instrumentation and carefully planned proximal anchors.

*Department of Orthopaedic Surgery, The Johns Hopkins Medical Institutions, Baltimore, MD

Department of Orthopaedic Surgery, Rady Children's Hospital and Health Center, San Diego, CA

Department of Orthopaedic Surgery, University of California, San Diego, CA

§Department of Orthopaedic Surgery, University of Rochester, Rochester, New York, NY

Hacettepe University School of Medicine, Ankara, Turkey; and the

San Diego Center for Spinal Disorders, La Jolla, CA.

Address correspondence and reprint requests to Paul D. Sponseller, MD, c/o Elaine P. Henze, BJ, ELS, Medical Editor and Director, Editorial Services, Department of Orthopaedic Surgery, The Johns Hopkins University/Johns Hopkins Bayview Medical Center, 4940 Eastern Ave, No. A665, Baltimore, MD; E-mail:

Acknowledgment date: January 10, 2013. Revision date: March 21, 2013. Acceptance date: April 1, 2013.

The manuscript submitted does not contain information about medical device(s)/drug(s).

The Growing Spine Foundation and Harms Study Group Foundation funds were received in support of this work.

Relevant financial activities outside the submitted work: consultancy, grants, royalties, expert testimony, payment for lecture, patents, stocks, payment for development of educational presentations.

Article Outline

Idiopathic scoliosis can develop at any age in children and is categorized on the basis of when it occurs. Infantile idiopathic scoliosis (IIS) manifests between 0 and 3 years, juvenile idiopathic scoliosis (JIS) manifests between 4 and 9 years, and adolescent idiopathic scoliosis (AIS) manifests between 10 and 18 years.1

The curve patterns of AIS have been well studied. Although many classification systems have been proposed,2–12 only 2 have been widely adopted for use: the King system,13 published in 1986, and the Lenke system,14,15 which superseded the King system when it was published in 2001. Both provide a framework for physicians and researchers to communicate their findings, using shared terminology in the search for the most successful treatment for each curve type. However, the Lenke system requires bending radiographs, which prevents its direct application in many young patients because of the difficulty of performing this imaging technique in a reliable and reproducible manner in these patients. Comparing curves of these younger patients with one another and with older patients therefore relies on individual measurements rather than a comparison of classifications such as Lenke types.

Our goal was to compare patients with operative idiopathic early-onset scoliosis (IEOS), for which we considered patients 5 years and younger, with patients with operative AIS to determine if statistically significant differences exist in primary curve characteristics, stable vertebra location, or kyphosis. Where a difference did exist, we considered its clinical significance.

Back to Top | Article Outline


This study was approved by our institutional review boards. We searched 2 multicenter databases and retrieved preoperative data and radiographs of patients with idiopathic scoliosis who had undergone surgery.

Back to Top | Article Outline

Patient Cohorts

Idiopathic Early-Onset Scoliosis

We searched the Growing Spine Study Group's database for data of children with IEOS (excluding those with neuromuscular, congenital, or syndromic diagnoses) and identified 60 patients (53% female, 47% male). Preoperative radiographs of patients were obtained between 1994 and 2011 at an average age of 3.6 ± 1.4 years, with all patients being 5 years or younger. The primary curve was thoracic in 50 patients and thoracolumbar/lumbar in 10 patients. All patients were diagnosed as idiopathic and had otherwise normal clinical examinations and normal magnetic resonance imaging studies when indicated.

Back to Top | Article Outline

Adolescent Idiopathic Scoliosis

We searched the Harms Study Group's database for children with AIS and identified 1537 patients (79% female, 21% male). These patients had preoperative radiographs taken between 1995 and 2011 at an average age of 12.3 ± 2.5 years, and all were between 10 and 18 years of age. The primary curve was thoracic in 1219 patients and thoracolumbar/lumbar in 318 patients.

Back to Top | Article Outline

Radiographical Measurements

All measurements were performed by 2 members of the study group. Differences were resolved via discussion between the 2 members, with a third member being consulted if disagreement persisted.

Back to Top | Article Outline

Individual Parameters

The following measurements were made on the preoperative radiographs: in the coronal plane, highest stable vertebra, primary curve direction, Cobb angle, proximal Cobb end vertebra, apical vertebra, and distal Cobb end vertebra; in the sagittal plane, T2 to T12 kyphosis.

Back to Top | Article Outline

Curve Classification

The primary curve was defined as the curve with the largest Cobb angle and was classified as either thoracic (primary curve apex at the T11 to T12 disc space and above) or thoracolumbar/lumbar (primary curve apex at T12 and below).

Back to Top | Article Outline

Lumbar Deformity

Lumbar spinal deformity was categorized using the methodology for the lumbar modifier in the Lenke classification system.15 Specifically, we examined the relationship between the lumbar vertebral pedicles and the central sacral vertebral line (CSVL) (a vertical line drawn through the center of S1 on a coronal radiograph) to stratify patients into 1 of the following 3 descriptions: CSVL between pedicles, CSVL touching apical body, and CSVL completely medial to apical body.

Back to Top | Article Outline

Statistical Methods

All the parameters were compared between patients with IEOS and patients with AIS in 3 ways: (1) complete patient cohorts, (2) subset of patients with thoracic primary curves, and (3) subset of patients with thoracolumbar/lumbar primary curves. Curve direction was compared using Fisher exact test, vertebral levels (stable vertebra, apex, proximal and distal end vertebrae) were compared using the Wilcoxon rank-sum test, and Cobb angle and kyphosis were compared using the Student t test. The level of significance was set at P= 0.05.

Back to Top | Article Outline


Comparison of All Curves

The primary curve in the IEOS cohort was thoracic in 50 patients and thoracolumbar/lumbar in 10 patients. In the AIS cohort, the primary curve was thoracic in 1219 patients and thoracolumbar/lumbar in 318 patients. The curve apex had a bimodal distribution in AIS versus a unimodal distribution in IEOS, with the IEOS distribution being closer to the thoracolumbar/lumbar junction (Figure 1). Patients with IEOS had a greater curve magnitude (P< 0.001), more caudal highest stable vertebra (P= 0.001), and more severe T2 to T12 kyphosis (P< 0.001) than patients with AIS (Table 1) (Figure 2A–D).

Back to Top | Article Outline

Comparison of Thoracic Curves

For patients whose primary curve was thoracic (i.e., those whose primary curve apex was at the T11 to T12 disc space and above), those with IEOS had a significantly larger curve magnitude (P< 0.001), more caudal apical vertebra (P< 0.001), more caudal proximal Cobb end vertebra (P< 0.001), and more caudal distal Cobb end vertebra (P< 0.001) than those with AIS. Curve direction was more often to the left in IEOS (69%) and to the right in AIS (95%) (Table 2). The CSVL was more likely to stay between the lumbar pedicles in patients with IEOS than in patients with AIS (P< 0.001) (Figure 3).

Back to Top | Article Outline

Comparison of Thoracolumbar/Lumbar Curves

For patients whose primary curve was thoracolumbar/lumbar (i.e., those whose primary curve apex was at T12 and below), those with IEOS had a significantly larger magnitude (P= 0.03) than those with AIS. The 2 groups did not differ in apical vertebra (P = 0.41), proximal Cobb end vertebra (P= 0.34), distal Cobb end vertebra (P= 0.89), or direction (P= 0.70) (Table 3).

Back to Top | Article Outline


The works of King, Coonrad, Lenke, and subsequent researchers has been instrumental in creating an understanding of curve patterns and developing treatment methods for AIS.13–16 The differences between IEOS and AIS, however, have not been thoroughly investigated or statistically analyzed in a manner that might begin to elucidate variation in causes and preferred treatment methods. One of the barriers to such analysis has been the inability to apply the Lenke system to younger patients, who often do not have bending radiographs. Mishiro et al17 proposed a modified Lenke system for IIS and JIS using only upright coronal and lateral radiographs, but this new system has not been widely adopted as of the time of submission. Because of this limitation, we analyzed and compared parameters measurable on standard coronal and sagittal radiographs. We limited the comparison to the latest preoperative radiographs to ensure that the groups were being compared at points of greatest equivalence, thus minimizing analysis at differing stages of curve progression.

Overall, compared with patients with AIS, those with IEOS exhibited larger curves, a more caudal stable vertebra, larger kyphosis, and a unimodal pattern of primary curve apex distribution centered closer to the thoracolumbar/lumbar junction (vs. the bimodal distribution seen in AIS). When we limited the comparison to patients with thoracic primary curves, we found that those with IEOS had a more caudal apical vertebra, more caudal proximal and distal Cobb end vertebra, larger curve magnitude, a higher percentage of left curves, and a higher percentage of lumbar deformity in which the CSVL stayed between the lumbar pedicles. When we limited the comparison to patients who had thoracolumbar/lumbar primary curves, those with IEOS also had a larger curve magnitude. The larger curve magnitude in IEOS may partially represent the desire to delay surgery as long as possible in these skeletally immature patients.

The recurring patterns of a lower apical vertebra and a lower stable vertebra in IEOS may suggest that children with IEOS require more distal fusion than those with AIS. Such action comes with multiple drawbacks, including limited flexibility and potential stress concentration. In addition, the greater kyphosis seen in IEOS increases pullout stress on proximal anchors (Figure 4A and B). Although several studies have shown casting to be effective in treating less severe IEOS curves,18–21 these results have not been shown to extend to more severe curves such as those in this study. In these severe curves and in more moderate curves that fail nonoperative treatment, multiple established and experimental surgical treatments exist.22

Radiographical curve characteristics in idiopathic scoliosis have been well examined,23–34 and many of our findings reflect reports in the literature that have characterized primary curve sidedness, Lenke type, thoracic kyphosis, and apical vertebra in idiopathic scoliosis. Primary curves are predominantly left-sided in patients with IIS23 and early-onset scoliosis,24 but right-sided patterns predominate in JIS25,28,33 and AIS.28 Our study illustrates these findings: primary thoracic curves were primarily left sided in IEOS (66%) and right sided in AIS (95%). Current literature indicates that patients with primary thoracic curves in EOS and IIS have more pronounced kyphotic tendencies than in patients with JIS and AIS, who tend to exhibit thoracic kyphosis that is less than or equal to normal.10,26,27,30,33,35,36 In our study, the magnitude of T2 to T12 kyphosis was significantly higher in patients with IEOS (40°± 15°) than in those with AIS (31°± 13°). Our study is somewhat less reflective of existing literature on the modal apex of the primary thoracic curve. In our IEOS group, the modal apex of the primary thoracic curve was T10, followed by T9. Two previous studies reported T9 as the modal apex of the primary thoracic curve in IIS29 and JIS,25 although these studies included patients treated both nonoperatively and surgically. Modal apex in our AIS group was T9 followed by T8, which agrees with the T8/T9 average of 1 study32 but differs slightly from the T8 mode in another study.31 Reports of stable vertebra could not be found for comparison.

The primary limitation of our study was the size of the IEOS group compared with the size of the AIS group. AIS is a much more common disorder. As a result, the IEOS group was significantly smaller despite sourcing data from a large, multicenter study group. However, we think the similarities with existing literature show our IEOS group to be representative and suitable for comparison.

A second limitation was that patients were retrospectively identified in the study groups' records based on whether or not they had surgery. Consequently, rather than comparing “operative” IEOS and “operative” AIS as defined by a homogenously used standard of “operative,” we actually compared IEOS and AIS in the children considered as operable by the array of surgeons. This methodology has the potential to introduce bias because many factors play a role in a surgeon's decision. For instance, it is possible that the greater curve magnitude in IEOS results from a desire to postpone surgery in all but the most severe deformities. To address such possibilities fully, we would need to also examine the deformity in patients in each database who have not been considered for surgical intervention.

Back to Top | Article Outline


Our study shows significant differences in radiographical characteristics of IEOS and AIS. Deformity in the younger population exhibits greater kyphosis, larger curve magnitude, and a more distal curve apex, all of which present treatment challenges that are less prevalent in adolescent patients.

Back to Top | Article Outline

Key Points

* To our knowledge, no previous studies have statistically compared the radiographical characteristics of large cohorts of operative IEOS and operative AIS.

* Patients with IEOS showed larger curve magnitude, more severe kyphosis, less curve compensation, more caudal curve apex, and a more caudal highest stable vertebra.

* The clinical significance of these findings include the need for more distal instrumentation and carefully planned proximal anchors in IEOS secondary to the more caudal and kyphotic curve characteristics.

* Future research may include a classification system applicable to early-onset scoliosis and comparison of the levels and types of instrumentation used in IEOS and AIS.

Back to Top | Article Outline


The authors thank Elaine P. Henze, BJ, ELS, for her editorial work, and Tracey Bastrom, MA, for her statistical analysis. The authors also thank the Growing Spine Study Group and the Harms Study Group, for contribution of data for this study. The Harms Study Group (HSG) is a collaborative cohort of worldwide distinguished surgeons dedicated to the advancement of treatment for children and adolescents with spine deformity. Through comprehensive, multicenter prospective research studies, questions regarding treatment approach and techniques to achieve desired outcomes are studied with a commitment to be internationally recognized for the highest quality published research on new spinal deformity surgery techniques.

Back to Top | Article Outline


1. Akbarnia BA. Management themes in early onset scoliosis. J Bone Jt Surg Am 2007;89:42–54.
2. Berven SH, Lowe T. The scoliosis research society classification for adult spinal deformity. Neurosurg Clin N Am 2007;18:207–13.
3. Lowe T, Berven SH, Schwab FJ, et al. The SRS classification for adult spinal deformity: building on the King/Moe and Lenke classification systems. Spine (Phila Pa 1976) 2006;31:S119–25.
4. Negrini A, Negrini S. The three-dimensional easy morphological (3-DEMO) classification of scoliosis, part II: repeatability. Scoliosis 2006;1:23.
5. Negrini S, Atanasio S, Fusco C, et al. 3-DEMO classification of scoliosis: a useful understanding of the 3rd dimension of the deformity. Stud Health Technol Inform 2008;135:139–53.
6. Negrini S, Negrini A. The three-dimensional easy morphological (3-DEMO) classification of scoliosis - part III, correlation with clinical classification and parameters. Scoliosis 2007;2:5.
7. Negrini S, Negrini A, Atanasio S, et al. Three-dimensional easy morphological (3-DEMO) classification of scoliosis, part I. Scoliosis 2006;1:20.
8. Poncet P, Dansereau J, Labelle H. Geometric torsion in idiopathic scoliosis: three-dimensional analysis and proposal for a new classification. Spine (Phila Pa 1976) 2001;26:2235–43.
9. Qiu G, Zhang J, Wang Y, et al. A new operative classification of idiopathic scoliosis: a Peking Union Medical College method. Spine (Phila Pa 1976) 2005;30:1419–26.
10. Sangole AP, Aubin CE, Labelle H, et al. Three-dimensional classification of thoracic scoliotic curves. Spine (Phila Pa 1976) 2009;34:91–9.
11. Schwab F, Farcy JP, Bridwell K, et al. A clinical impact classification of scoliosis in the adult. Spine (Phila Pa 1976) 2006;31:2109–14.
12. Stokes IAF, Sangole AP, Aubin CE. Classification of scoliosis deformity three-dimensional spinal shape by cluster analysis. Spine (Phila Pa 1976) 2009;34:584–90.
13. King HA, Moe JH, Bradford DS, et al. The selection of fusion levels in thoracic idiopathic scoliosis. J Bone Joint Surg Am 1983;65:1302–13.
14. Lenke LG, Betz RR, Harms J, et al. Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am 2001;83:1169–81.
15. Lenke LG, Edwards CC, Bridwell KH. The Lenke classification of adolescent idiopathic scoliosis: how it organizes curve patterns as a template to perform selective fusions of the spine. Spine (Phila Pa 1976) 2003;28:S199–S207.
16. Coonrad RW, Murrell GA, Motley G, et al. A logical coronal pattern classification of 2,000 consecutive idiopathic scoliosis cases based on the Scoliosis Research Society-defined apical vertebra. Spine (Phila Pa 1976) 1998;23:1380–91.
17. Mishiro T, Lenke LG, Koester LA, et al. Modified Lenke classification system for infantile and juvenile idiopathic scoliosis. Poster presented at the Scoliosis Research Society's 2009 International Meeting on Advanced Spine Techniques (Vienna, Austria), September 23–26, 2009. Abstract available at Accessed on April 2, 2012.
18. D'Astous JL, Sanders JO. Casting and traction treatment methods for scoliosis. Orthop Clin North Am 2007;38:477–84, v.
19. Mehta MH. The rib-vertebra angle in the early diagnosis between resolving and progressive infantile scoliosis. J Bone Joint Surg Br 1972;54:230–43.
20. Garg S, Mehta S, Dormans JP. Modern surgical treatment of primary aneurysmal bone cyst of the spine in children and adolescents. J Pediatr Orthop 2005;25:387–92.
21. Risser JC. Scoliosis treated by cast correction and spine fusion: a long term follow-up study. Clin Orthop Relat Res 1976;86–94.
22. Tis JE, Karlin LI, Akbarnia BA, et al. Early onset scoliosis: modern treatment and results. J Pediatr Orthop 2012;32:647–57.
23. Ceballos T, Ferrer-Torrelles M, Castillo F, et al. Prognosis in infantile idiopathic scoliosis. J Bone Joint Surg Am 1980;62:863–75.
24. Fernandes P, Weinstein SL. Natural history of early onset scoliosis. J Bone Joint Surg Am 2007;89:21–33.
25. Figueiredo UM, James JIP. Juvenile idiopathic scoliosis. J Bone Joint Surg Br 1981;63:61–6.
26. Hayashi K, Upasani VV, Pawelek JB, et al. Three-dimensional analysis of thoracic apical sagittal alignment in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2009;34:792–7.
27. Lenke LG, Betz RR, Clements D, et al. Curve prevalence of a new classification of operative adolescent idiopathic scoliosis: does classification correlate with treatment? Spine (Phila Pa 1976) 2002;27:604–11.
28. Lenke LG, Dobbs MB. Management of juvenile idiopathic scoliosis. J Bone Joint Surg Am 2007;89:55–63.
29. Lincoln TL. Infantile idiopathic scoliosis. Am J Orthop (Belle Mead NJ) 2007;36:586–90.
30. Mac-Thiong JM, Labelle H, Charlebois M, et al. Sagittal plane analysis of the spine and pelvis in adolescent idiopathic scoliosis according to the coronal curve type. Spine (Phila Pa 1976) 2003;28:1404–9.
31. Potter BK, Rosner MK, Lehman RA, et al. Reliability of end, neutral, and stable vertebrae identification in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2005;30:1658–63.
32. Richards BS, Sucato DJ, Johnston CE, et al. Right thoracic curves in presumed adolescent idiopathic scoliosis: which clinical and radiographic findings correlate with a preoperative abnormal magnetic resonance image? Spine (Phila Pa 1976) 2010;35:1855–60.
33. Robinson CM, McMaster MJ. Juvenile idiopathic scoliosis. Curve patterns and prognosis in one hundred and nine patients. J Bone Joint Surg Am 1996;78:1140–8.
34. Sponseller PD, Flynn JM, Newton PO, et al. The association of patient characteristics and spinal curve parameters with Lenke classification types. Spine (Phila Pa 1976) 2012;37:1138–41.
35. Upasani VV, Tis J, Bastrom T, et al. Analysis of sagittal alignment in thoracic and thoracolumbar curves in adolescent idiopathic scoliosis: how do these two curve types differ? Spine (Phila Pa 1976) 2007;32:1355–9.
36. Wang W, Zhu Z, Zhu F, et al. Different curve pattern and other radiographical characteristics in male and female patients with adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2012;37:1586–92.

idiopathic; scoliosis; early-onset; infantile; juvenile; adolescent; Lenke

© 2013 by Lippincott Williams & Wilkins