Secondary Logo

Journal Logo

Preservation of Thoracic Kyphosis Is Critical to Maintain Lumbar Lordosis in the Surgical Treatment of Adolescent Idiopathic Scoliosis

Newton, Peter O., MD*†; Yaszay, Burt, MD*†; Upasani, Vidyadhar V., MD*; Pawelek, Jeff B., BA; Bastrom, Tracey P., MA; Lenke, Lawrence G., MD; Lowe, Thomas, MD§; Crawford, Alvin, MD; Betz, Randal, MD; Lonner, Baron, MD** Harms Study Group

doi: 10.1097/BRS.0b013e3181dccd63

Study Design. Retrospective analysis of prospectively collected multicenter series.

Objective. To evaluate the sagittal profile of surgically treated adolescent idiopathic scoliosis (AIS) patients.

Summary of Background Data. With the increasing popularity of segmental pedicle screw spinal instrumentation, thoracic kyphosis (TK) is often sacrificed to achieve coronal and axial plane correction.

Methods. Radiographs of AIS patients with a Lenke type 1 deformity and minimum 2-year follow-up after selective thoracic fusion (lowest instrumented vertebra of T11, T12, or L1) were evaluated. Changes in TK were correlated with changes in lumbar lordosis (LL). Patients were divided according to approach (open/thoracoscopic anterior vs. posterior). Analysis of variance was used to compare pre and postoperative radiographic measures.

Results. Two hundred fifty-one patients (age: 14 ± 2 years) were included. Sixty seven percentages of the patients had anterior surgery (97 open anterior, 71 thoracoscopic) and 33% (83 patients) had posterior spinal fusion. A decrease in postoperative TK was significantly correlated (P ≤ 0.001) with a decrease in LL at first erect (r = 0.3), 1 year (r = 0.4) and 2 years (r = 0.4), independent of surgical approach. LL decreased significantly at the first erect regardless of approach (P = 0.003); however, at 2-year postoperative TK and LL were significantly decreased after a posterior approach (P ≤ 0.001) when compared with an anterior approach that added kyphosis. The decrease in LL (5.6° ± 9.7°) was nearly twice the decrease in TK (2.8° ± 11.4°) in the posterior group at 2-years.

Conclusion. Given that thoracic AIS is often associated with a preexisting reduction in TK, ideal surgical correction should address this deformity. Procedures which further reduce TK also reduce LL. It is unclear if the loss of LL from thoracic scoliosis correction will compound the loss of LL that occurs with age and lead to further decline in sagittal balance. With this concern, we recommend a posterior column lengthening and/or an anterior column shortening to achieve restoration of normal TK and maximal LL.

Analysis of changes in the sagittal profile in surgically treated patients with adolescent idiopathic scoliosis showed that a decrease in postoperative thoracic kyphosis results in a decrease in lumbar lordosis. A premature, iatrogenic loss of lumbar lordosis may predispose patients to develop flat back deformity and unbalanced forward sagittal alignment.

From the *Department of Orthopedic Surgery, University of California, San Diego; †Department of Orthopedics, Rady Children's Hospital-San Diego, CA; ‡Department of Orthopedics, WA University, St. Louis, MO; §Department of Orthopedics, Woodridge Spine Orthopedics, Wheat Ridge, CO; ¶Department of Orthopedics, Cincinnati Children's Hospital, Cincinnati, OH; ∥Department of Orthopedics, Shriners Hospital for Children, Philadelphia, PA; and **Scoliosis Associates, New York, NY.

Acknowledgment date: March 20, 2009. First revision date: June 18, 2009. Second revision date: July 24, 2009. Acceptance date: July 27, 2009.

The device(s)/drug(s) that is/are the subject of this manuscript is/are not FDA-approved for this indication and is/are commercially available in the United States (off-label use in pediatrics).

Corporate/Industry funds were received in support of this work. 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: e.g., honoraria, gifts, consultancies, royalties, stocks, stock options, decision making position.

This study was conducted at Rady Children's Hospital and Health Center San Diego, CA.

Supported by a grant to the Harms Study Group from DePuy Spine Inc.

Thomas Lowe, MD, is deceased.

Address correspondence and reprint requests to Peter O. Newton, MD, 3030 Children's Way, Ste 410, San Diego, CA 92123; E-mail:

The current objective in the surgical management of adolescent idiopathic scoliosis (AIS) is to maximize deformity correction, achieve coronal and sagittal balance, and retain spinal flexibility.1,2 Traditionally, the focus of scoliosis surgery has been to correct the coronal deformity of the main curves. However, early surgical intervention with posterior distraction devices resulted in improvement of coronal deformity at the expense of the sagittal profile.3,4 Long-term follow-up of these patients has then demonstrated the consequences of this “flatback syndrome” including back pain, degenerative disc disease, and inability to stand erect.5 Therefore, optimal surgical management of AIS requires the maintenance of sagittal balance, including lumbar lordosis (LL).

Patients with a primary thoracic scoliosis are typically hypokyphotic relative to nonscoliosis patients. Recent studies have evaluated the influence of a spinal fusion on thoracic kyphosis (TK) and overall sagittal alignment. Depending on instrumentation and correction techniques used, postoperative TK has been shown to either increase or decrease.6–10 The effect of a selective thoracic fusion on LL has not been evaluated.

In the coronal plane, correction of the thoracic scoliosis results in a spontaneous lumbar curve correction (SLCC). The amount of SLCC is dependent on multiple variables including surgical approach, instrument type, curve flexibility, lowest instrumented vertebrae, and thoracic curve correction achieved.11–14 We would anticipate that a selective thoracic fusion and resulting kyphosis would have a similar influence in the sagittal plane. The purpose of this study was to evaluate the sagittal profile of surgically treated Lenke type I patients and determine the variables that effect postoperative LL.

Back to Top | Article Outline

Materials and Methods

A multicenter prospective database of patients receiving surgical correction for AIS was used for this study. Patients with a Lenke type I deformity corrected by a selective thoracic fusion (lowest instrumented vertebra of T11, T12, or L1) and a minimum 2-year follow-up were evaluated. Sagittal alignment was assessed for each patient using radiographs collected before surgery and at first-erect, 1-year, and 2-year postoperative visits.

TK and LL were measured on all radiographs in an attempt to identify a relationship between the 2 variables. In this study, TK was defined as the angle between T5 and T12, whereas LL was the angle measurement between T12 and the sacrum. During the 2-year follow-up interval, the progression of TK and LL was tracked for each patient.

Sagittal alignment measurements were analyzed using Statistical Package for the Social Sciences (SPSS, Inc., Chicago, IL). Descriptive data were calculated for the radiographic data. Changes in TK were correlated with changes in LL at each visit using Pearson's correlation analysis (P < 0.05). Patients were subsequently subdivided according to approach (open/thoracoscopic anterior vs. posterior). Analysis of variance was used to compare preoperative and postoperative radiographic measures (P < 0.05) to determine the effect of surgical approach on sagittal alignment after surgery.

Back to Top | Article Outline


Results from the AIS database indicated that 251 patients met the inclusion criteria and were included in the analysis. The average age of the cohort was 14.5 ± 2.1 year. Anterior surgery was performed on 168 (67%) of included individuals with 97 (39% of total surgeries) involving an open anterior approach and 71 (28% of total surgeries) completed thoracoscopically; a posterior spinal fusion was performed on 83 patients (33%). For this cohort, the average proximal thoracic curve was 25.0° ± 7.6°, the mean thoracic curve was 51.1° ± 8.2°, and the lumbar curve averaged 32.1° ± 9.1°. Additional demographic data classified according to the surgical approach is presented in Table 1.

Table 1

Table 1

Radiographic measurements analyzed with Pearson's correlation revealed a linear relationship between TK and LL. A decrease in postoperative TK was significantly correlated (P ≤ 0.001) with a decrease in LL at the first erect (r = 0.3), 1-year follow-up (r = 0.4), and 2-year follow-up (r = 0.4) visits, independent of surgical approach (Figures 1–3). Focusing only on the first erect, 1-year, and 2-year visits, Figures 4 and 5 further illustrate the correlation between TK and LL because they both increased with each visit.

Figure 1

Figure 1

Figure 2

Figure 2

Figure 3

Figure 3

Figure 4

Figure 4

Figure 5

Figure 5

Differences in anterior and posterior surgical approaches also influenced the sagittal alignment of patients. Between the preoperative and first erect visits, patients who underwent an anterior surgery experienced an increase in TK, whereas posterior patients exhibited an overall decrease in kyphosis (Figures 4, 6). During the course of 2 years, the anterior procedures lead to a steady increase in kyphosis from 19.9° ± 11.2° before surgery to 27.8° ± 9.9° at 2-year postoperative. On the other hand, the posterior approach decreased kyphosis from 23.0° ± 12.0° before surgery to 17.8° ± 7.6° at first erect (Figure 4). In this same posterior group, despite gradual increases in kyphosis at the 1- and 2-year postsurgical visits, the average kyphosis measurement at 2 years postoperative was only 20.2° ± 8.4°.

Figure 6

Figure 6

Table 2

Table 2

LL decreased significantly from the preoperative to the first-erect visit regardless of approach (P = 0.003) (Figures 5, 6). In the anterior surgical group, lordosis decreased from −59.0° ± 11.2° before surgery to −54.1° ± 11.6° at first erect. Similar results were noted in the posterior surgical group with average values for lordosis decreasing from −63.2° ± 11.8° before surgery to −54.2° ± 12.0° at first erect (Figure 5). At 2 years, there was slight increase in lordosis from the first erect in both the anterior and posterior approach (−62.0° ± 11.4° vs. −57.6° ± 11.5°, respectively). Furthermore, the overall decrease in LL (5.6° ± 9.7°) was nearly twice the decrease in TK (2.8° ± 11.4°) in the posterior group at 2 years after surgery (Table 2, Figure 6). Collectively, 2-years postoperative TK and LL were significantly decreased after a posterior approach (P ≤ 0.001) when compared with an anterior approach (Table 2, Figure 6).

Back to Top | Article Outline


Better understanding of spinal sagittal balance has increased the importance of this surgical objective among scoliosis surgeons.1 Patients who have a poor sagittal profile after scoliosis surgery are presenting with the consequences of what is now known as “flatback syndrome.”5 Symptoms associated with an iatrogenic loss of LL resulting in a positive sagittal balance include an inability to stand erect and disabling back pain. The surgical treatment of this problem involves corrective osteotomies that can have significant perioperative complications.3–5,15,16

Prevention of “flatback syndrome” has generally focused on spinal fusions that extend into the lumbar spine. Recommendations to prevent an iatrogenic loss of LL include avoidance of distraction instrumentation in the lumbar spine, contouring the rod with LL, and intraoperative positioning that enhances physiologic lordosis.3,4 Little discussion has been given to the influence of a thoracic fusion on the sagittal profile of an unfused lumbar spine. The results of this study suggest that preservation of TK is also critical in preventing an iatrogenic loss of LL. A decrease in postoperative TK in Lenke type 1 curves significantly correlated with a decrease in LL.

Control of the thoracic sagittal profile has been shown to be influenced by instrument type. Kim et al10 demonstrated a significant difference in the 2-year postoperative kyphosis between the use of all pedicle screws compared with all hooks (17° vs. 26°, respectively). A similar difference has been shown between all screw and hybrid instrumentation.8,9 Lowenstein et al7 found a 10° loss of kyphosis in patients that had pedicle screw instrumentation compared with only 3° in those that had a hybrid construct. The cause for this difference in TK may be the difference in deformity correction attained by the instrument types. The typical main thoracic curve is hypokyphotic compared with control patients.6 This is thought to result from an overgrowth of the anterior spine that is responsible for the curve formation seen in AIS.17–19 Greater correction would return this increased length anteriorly. Without increasing length posteriorly, this would lordose the spinal column. Pedicle screws have been suggested to provide better coronal and rotational correction than hybrid or hook instrumentation.7,8,10,20,21 This may be responsible for the lost kyphosis seen in all pedicle screw constructs.

The surgical approach has also been shown to affect the sagittal profile with an anterior spinal instrumentation being more kyphogenic than a posterior approach.13,14,22,23 Discectomy and compression along the rod lead to shortening of the anterior column when performing an anterior fusion. A posterior approach relies on lengthening the posterior column to gain kyphosis. Betz et al13 demonstrated that posterior fusions failed to improve on thoracic hypokyphosis in 60% of cases, whereas 81% of anterior cases had normal postoperative kyphosis. Interestingly, 40% of patients with normal preoperative kyphosis were hyperkyphotic after surgery following an anterior spinal instrumentation. Potter et al14 and Rhee et al23 reported an increase in TK with anterior instrumentation compared with a loss in posterior cases. The study by Rhee et al included fusion of all curve types. Therefore, the effect on LL could not be evaluated. Potter et al evaluated 40 Lenke type 1 patients but had no analysis of postoperative LL.

This study supports the greater kyphogenic effect of anterior spinal instrumentation during a posterior procedure. At 2-years postoperative, an anterior approach added TK resulting in a gain in LL. A posterior approach decreased TK causing a loss in LL. Interestingly, there was not a direct linear relationship between the changes in kyphosis with the changes in LL. There was a 5.6° loss of lordosis associated with a 2° loss of kyphosis in the posterior cases. On the other hand, there was only a 3° gain in LL with an 8° gain in TK after an anterior procedure. It is unclear why the posterior approach would have a greater effect on LL than an anterior approach. This may reflect differences in the influence of posterior column lengthening compared with anterior shortening. Overall, there was no difference in the total sagittal balance between the 2 groups.

Posterior spinal fusion has become the treatment of choice for Lenke type I curves. There is also increasing use of thoracic pedicle screws to maximize deformity correction. This study suggests that there is a risk for a loss in LL with this procedure. Techniques to lengthen the posterior column must be used and they include: avoiding compression, using distraction for deformity correction, and contouring kyphosis into the rod. These techniques are still limited by the tension created by the soft tissue and bone anatomy of the posterior spine. We attempt to minimize this effect by performing releases similar to a Ponte osteotomy described for Scheuermanns kyphosis.24 Removing the facets and soft tissues (supraspinous and interspinous ligaments, ligamentum flavum, and facet joint capsule) between the posterior elements allows for greater distraction abilities along the length of the posterior column. However, further research is needed to determine whether these posterior releases positively effect the sagittal profile of a hypokyphotic thoracic spine.

The influence of a thoracic fusion on the lumbar spine is well documented in the coronal plane. SLCC after a selective thoracic fusion was first described in the 1950s.25 Since, then multiple studies have evaluated the various factors that result in the greatest SLCC. The data are conflicting with some authors suggesting that the surgical approach can affect the amount of SLCC.12,14 Others have demonstrated no difference in SLCC between anterior or posterior approaches.11,13 Our data would suggest that the surgical approach does influence the lumbar spine in the sagittal plane. However, this is not a direct effect of the approach but is more likely a secondary effect of the resulting TK. Patel et al11 had a similar experience in the SLCC when anterior and posterior approaches were matched by percent thoracic curve correction. To maintain coronal and sagittal balance, the lumbar spine needs to respond to the shape of the thoracic spine.

Potential limitations to this study are the inability to evaluate for differences in intraoperative techniques and instrumentation (rod type/stiffness) used at the different centers. The extent of posterior releases or anterior discectomies and the amount of compression or distraction done may result in differences in the postoperative TK. Although the goal is to use techniques that maximize kyphosis in the hypokyphotic spine, the actual influence in the current data could not be determined.

It is also unclear what the effect of this loss in LL after a decrease in TK will have in the patient's long-term outcome. The consequences of “flatback syndrome” following Harrington instrumentation are seen decades after the index procedure. The most severe cases were also the ones that had their instrumentation extend into the lower lumbar spine. All our patients had their instrumentation at L1 or above. However, because there is a loss of lordosis with aging, it is reasonable to use techniques that would minimize a decrease in LL.

Back to Top | Article Outline


The changes in thoracic sagittal alignment associated with a selective thoracic fusion also effect changes in the sagittal alignment of the unfused lumbar spine. The loss of TK associated with many posterior instrumentation methods is also associated with a loss of LL. Given the long-term importance of LL in maintaining sagittal balance, it seems prudent to consider using surgical approaches for the correction of thoracic AIS which maintain, if not increase, TK and thus LL.

Back to Top | Article Outline

Key Points

  • A postoperative loss in kyphosis after a thoracic fusion results in a loss in lordosis of the uninstrumented lumbar spine.
  • Anterior thoracic fusion added thoracic kyphosis and lumbar lordosis at 2-year follow-up.
  • Posterior thoracic fusion decreased thoracic kyphosis and lumbar lordosis at 2-year follow-up.
Back to Top | Article Outline


1.Majdouline Y, Aubin CE, Robitaille M, et al. Scoliosis correction objectives in adolescent idiopathic scoliosis. J Pediatr Orthop 2007;27:775–81.
2.Bridwell KH. Surgical treatment of idiopathic adolescent scoliosis. Spine 1999;24:2607–16.
3.Lagrone MO, Bradford DS, Moe JH, et al. Treatment of symptomatic flatback after spinal fusion. J Bone Joint Surg Am 1988;70:569–80.
4.Potter BK, Lenke LG, Kuklo TR. Prevention and management of iatrogenic flatback deformity. J Bone Joint Surg Am 2004;86–A:1793–808.
5.La Grone MO. Loss of lumbar lordosis. A complication of spinal fusion for scoliosis. Orthop Clin North Am 1988;19:383–93. Jonge T, Dubousset JF, Illes T. Sagittal plane correction in idiopathic scoliosis. Spine 2002;27:754–60.
7.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 2007;32:448–52.
8.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 2006;31:291–8.
9.Vora V, Crawford A, Babekhir N, et al. A pedicle screw construct gives an enhanced posterior correction of adolescent idiopathic scoliosis when compared with other constructs: myth or reality. Spine 2007;32:1869–74.
10.Kim YJ, Lenke LG, Cho SK, et al. Comparative analysis of pedicle screw versus hook instrumentation in posterior spinal fusion of adolescent idiopathic scoliosis. Spine 2004;29:2040–8.
11.Patel PN, Upasani VV, Bastrom TP, et al. Spontaneous lumbar curve correction in selective thoracic fusions of idiopathic scoliosis: a comparison of anterior and posterior approaches. Spine 2008;33:1068–73.
12.Lenke LG, Betz RR, Bridwell KH, et al. Spontaneous lumbar curve coronal correction after selective anterior or posterior thoracic fusion in adolescent idiopathic scoliosis. Spine 1999;24:1663–71; discussion 72.
13.Betz RR, Harms J, Clements DH III, et al. Comparison of anterior and posterior instrumentation for correction of adolescent thoracic idiopathic scoliosis. Spine 1999;24:225–39.
14.Potter BK, Kuklo TR, Lenke LG. Radiographic outcomes of anterior spinal fusion versus posterior spinal fusion with thoracic pedicle screws for treatment of Lenke Type I adolescent idiopathic scoliosis curves. Spine 2005;30:1859–66.
15.Farcy JP, Schwab FJ. Management of flatback and related kyphotic decompensation syndromes. Spine 1997;22:2452–7.
16.Bradford DS, Tribus CB. Vertebral column resection for the treatment of rigid coronal decompensation. Spine 1997;22:1590–9.
17.Deacon P, Archer IA, Dickson RA. The anatomy of spinal deformity: a biomechanical analysis. Orthopedics 1987;10:897–903.
18.Dickson RA. The aetiology of spinal deformities. Lancet 1988;1:1151–5.
19.Siu King Cheung C, Tak Keung Lee W, Kit Tse Y, et al. Abnormal peri-pubertal anthropometric measurements and growth pattern in adolescent idiopathic scoliosis: a study of 598 patients. Spine 2003;28:2152–7.
20.Suk SI, Lee CK, Kim WJ, et al. Segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis. Spine 1995;20:1399–405.
21.Watanabe K, Lenke LG, Bridwell KH, et al. Comparison of radiographic outcomes for the treatment of scoliotic curves greater than 100 degrees: wires versus hooks versus screws. Spine 2008;33:1084–92.
22.Muschik MT, Kimmich H, Demmel T. Comparison of anterior and posterior double-rod instrumentation for thoracic idiopathic scoliosis: results of 141 patients. Eur Spine J 2006;15:1128–38.
23.Rhee JM, Bridwell KH, Won DS, et al. Sagittal plane analysis of adolescent idiopathic scoliosis: the effect of anterior versus posterior instrumentation. Spine 2002;27:2350–6.
24.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;20:586–93.
25.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-A:529–54 passim.

scoliosis; sagittal alignment; adolescent idiopathic scoliosis; postoperative

© 2010 Lippincott Williams & Wilkins, Inc.