Adding Thoracic Fusion Levels in Lenke 5 Curves: Risks and Benefits
Lark, Robert K. MD*; Yaszay, Burt MD†; Bastrom, Tracey P. MA†; Newton, Peter O. MD†; Harms Study Group
*Department of Orthopaedic Surgery, Duke University Medical Center, Durham, NC; and
†Rady Children's Hospital and Health Center, San Diego, CA.
Address correspondence and reprint requests to Burt Yaszay, MD, 3030 Children's Way, Suite 410, San Diego, CA 92123; E-mail: email@example.com
Acknowledgment date: November 19, 2011. First revision date: May 30, 2012. Acceptance date: June 5, 2012.
The manuscript submitted does not contain information about medical device(s)/drug(s).
Harms Study Group from DePuy Spine Inc. and Children's Specialists Foundation Orthopedic Research and Education Fund grants were received to support 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: for example, honoraria, gifts, consultancies, royalties, stocks, stock options, decision-making position.
Study Design. Multicenter; review of prospectively collected data.
Objective. To determine the risks and potential benefits of nonselective versus selective fusion in a matched set of patients with Lenke 5 curves.
Summary of Background Data. The Lenke classification suggests a limited thoracolumbar/lumbar fusion for type 5 curves, although many experienced adolescent idiopathic scoliosis surgeons, at times, include a fusion of the thoracic curve.
Methods. Prospectively collected cases from a multicenter database were analyzed. Patients with Lenke type 5 scoliosis curves were divided into 2 groups: 109 selective or short (only thoracolumbar/lumbar curve fused), and 41 nonselective or long (both thoracolumbar/lumbar and thoracic curves fused). Patients were then matched on the basis of the preoperative radiographical and clinical measures. Two-year postoperative radiographical and clinical outcomes were compared, using analysis of variance, with Bonferroni correction (P < 0.008).
Results. Twenty-nine matched pairs (58 patients) with Lenke 5 curves were identified. There were no preoperative differences between groups in age, thoracic or lumbar Cobb angle, curve flexibility, thoracic kyphosis, clinical trunk flexibility, or Scoliosis Research Society outcomes questionnaire scores. Postoperatively, patients in the nonselective group exhibited greater coronal correction for thoracic (residual Cobb; 22° vs. 12°) and lumbar curves (residual Cobb; 19° vs. 13°). However, the longer fusions had significantly less thoracic kyphosis (27° vs. 18°), truncal side bending (14 vs. 10 cm), and rotational flexibility (53° vs. 42°). There was no difference in clinical balance or Scoliosis Research Society questionnaire, version 22, scores.
Conclusion. Adolescent idiopathic scoliosis surgeons attempt to achieve balanced correction with the fewest motion segments fused. Our data suggest that fusion of the thoracic curve in primary thoracolumbar scoliosis may improve coronal correction, but at the cost of decreased thoracic kyphosis and clinical flexibility 2 years postoperatively.
Multiple classification schemes have been proposed over the years in an effort to better understand curve types and fusion levels in adolescent idiopathic scoliosis.1–3 Generally, it is recommended to treat the primary or structural curve and leave flexible compensatory curves unfused. Significant attention has been given to the King Moe type II or Lenke type 1C curves with regard to fusion levels to preserve motion segments in the lumbar spine.4–12 This may not always be the case with regard to the thoracolumbar/lumbar (TL/L) curve. Some surgeons may be less concerned about fusing the compensatory thoracic curve above the structural thoracolumbar curve.
The relatively small amount of motion in the thoracic spine may not be as critical to maintain, compared with the lumbar spine. Unfortunately, little is known about the consequences of losing motion segments in the thoracic spine. The purpose of the present study was to determine the effects of fusing the thoracic and lumbar spine (long fusion) when compared with fusing just the TL/L curve (short fusion) in patients with primary TL/L scoliosis (Lenke type 5).
After institutional review board approval, a retrospective review of a multicenter prospectively collected adolescent idiopathic scoliosis database was conducted to select all patients with a Lenke type 5 curve. The curve pattern defines a major TL/L curve with nonstructural thoracic and upper thoracic region curves (Cobb angle <25° on best bend films; and <20° kyphosis at T2–T5 and T10–L2). Only patients with a minimum of 2 years of postoperative follow-up were included.
Patients were divided into 2 groups: those who underwent a short (selective) fusion of just the TL/L curve (n = 109) and those who underwent a long (nonselective) fusion inclusive of the thoracic curve (n = 41) (Figure 1). A matched pair design was created. A patient who underwent a long fusion was matched to a patient with a short fusion based on a preoperative thoracic Cobb angle within 2°, followed by a lumbar Cobb angle within 2°. Flexibility measures were then matched within 5° on bending thoracic Cobb and 15% when expressed as a percentage of preoperative Cobb angle.
Preoperative and 2-year postoperative coronal and sagittal radiographical measures were compared. Clinical assessment of trunk motion was measured in the coronal, sagittal, and axial (rotation) planes by a trained research coordinator (Figure 2). The Scoliosis Research Society (version 22) questionnaire was administered preoperatively and 2 or more years postoperatively.
Statistical Package for Social Sciences (SPSS, Chicago, IL) was used for all analyses in this study. The data were checked for normality and equal variances. Multivariate analysis of variance was used to compare radiographical measures, trunk motion data, and Scoliosis Research Society questionnaire scores between the short and long fusion groups. A P value of less than 0.05 was selected as statistically significant, and adjusted by using Bonferroni correction for multiple statistical tests to P < 0.008.
A total of 150 patients with Lenke 5 curves were identified. One hundred and nine (73%) of those patients underwent short fusion, whereas 41 patients (27%) had a long fusion. Preoperatively, the groups were significantly different in thoracic Cobb angle magnitude and thoracic translation (apex to C7 plumb) (P ≤ 0.001) (Table 1).
The 29 matched pairs were similar in all preoperative variables measured (Table 2). The average age at the time of surgery was 15 ± 2 years in both groups (P > 0.05). The thoracic Cobb angle magnitude was 34 ± 8° in both fusion groups (P > 0.05) and thoracic flexibility was similar (54 ± 19% in the long fusion group; 60 ± 18% in the short fusion group; P > 0.05). The number of levels fused was significantly different, with more levels fused in the long fusion group (12 ± 1) compared with the short fusion group (4 ± 1; P < 0.001). Operative time trended toward being greater in the long fusion group and estimated blood loss was significantly greater in the long fusion group (Table 3).
Postoperative radiographs revealed a significant coronal correction of the lumbar curve in both groups with a significantly smaller residual Cobb angle in those who underwent a long fusion (13 ± 6° vs. 19 ± 6°; P = 0.001) (Figure 3). The thoracic curve also improved significantly in both groups with less postoperative deformity in the long fusion group (11 ± 6° vs. 22 ± 8°; P = 0.000) (Figure 4). In the sagittal plane, there was significantly less postoperative thoracic kyphosis (T5–T12) after the long fusion compared with the short (18 ± 6° vs. 27 ± 12°; P = 0.002) (Table 4).
Measures of trunk motion were also found to be significantly different between the 2 groups. Patients who underwent a long fusion were significantly less flexible in forward flexion, lateral flexion, and axial rotation (Table 5). Other clinical measures including thoracic rib hump and lumbar angle of trunk rotation as measured by scoliometer were no different between the groups, postoperatively.
There were no statistically significant differences in cosmesis measures such as trunk shape and coronal decompensation, pre- or postoperatively between the groups (Table 6). Scoliosis Research Society questionnaire domains, including pain, self-image, general function, and activity also demonstrated no difference (Table 7).
The typically recommended treatment for Lenke 5 curves is fusion only of the primary TL/L curve.3,13 However, nearly 27% of fusions for these curves were treated nonselectively by experienced adolescent idiopathic scoliosis surgeons. This is in contrast with 17% of patients with primary thoracic scoliosis undergoing nonselective fusion by surgeons with similar experience as reported by Newton et al.11 This suggests that protecting thoracic motion segments may not be approached with the same vigilance as preserving lumbar motion, or that the Lenke classification criteria for a structural minor curve are interpreted differently for the thoracic and lumbar regions.
One of the primary purposes in performing a selective thoracic fusion is to minimize the loss of lumbar motion and theoretically the risk of lumbar degeneration or pain. Multiple studies have demonstrated the negative long-term risks of extending a scoliosis spinal fusion into the lower lumbar spine.14–19 Reducing the number of fused levels maximizes spinal flexibility and distributes stress across more distal lumbar motion segments.20 Theoretically, this may diminish the long-term risk of disc degeneration at adjacent distal levels. Therefore, it is clear why many studies have focused on the topic of selective thoracic fusion.4–12
Some attention has been given to the concept of selective TL/L fusion. In their multicenter study, Sanders et al 21 attempted to determine when the associated thoracic curve could be left unfused in a major TL/L curve. They determined that patients who had a TL/L to thoracic Cobb angle magnitude ratio of 1.25 or greater, a flexible thoracic curve (bending to 20° or less), and closed triradiate cartilage had good results and did not require further surgery. Another study retrospectively evaluated whether the Lenke classification could predict a successful selective TL/L fusion based on the criteria for a Lenke 5 curve.13 In their patients, the radiographical and clinical outcomes supported the suggestions for a selective fusion of a Lenke 5 curve. The authors, however, did report on 1 patient (of 21) who required a revision operation for a progressive thoracic curve.
The concept of minimizing fusion levels in TL/L curves is not novel in the literature. Many of the proponents for anterior fusion have suggested that an anterior approach can save fusion levels compared with a posterior approach.22–24 On average, the number of saved levels range from 1 to 1.6. The majority of these studies have focused on the caudal extent of the fusion. Other authors have suggested that the combination of wide posterior releases combined with segmental instrumentation can result in similar or improved results, without some of the problems associated with an anterior approach.25
The present study is the first to evaluate the effects of extending the fusion into the thoracic spine in a matched set of patients. Those fused into the thoracic spine had significantly better postoperative thoracic Cobb angle measurements. This allowed for greater lumbar curve correction in the long fusion group. In both groups, the resulting thoracic and lumbar curve corrections had a similar effect on the coronal alignment.
In the sagittal plane, fusion of the thoracic spine caused a decrease in thoracic kyphosis compared with a gain in kyphosis in the short fusion group. Overall, there was a significant difference in the thoracic kyphosis between the 2 groups. A great deal of attention has been given to sagittal profile after fusion of a main thoracic curve, especially with the use of modern segmental instrumentation. Multiple studies have demonstrated the influence of surgical approach and instrumentation type.26–31 The specific concerns discussed are the risk of junctional kyphosis and reduced lumbar lordosis associated with postoperative thoracic hypokyphosis.32,33 It is unclear what the relative effects of decreasing thoracic kyphosis are, when fusing into the lumbar spine in patients with long fusion.
Clinically, the most significant finding was the relative decrease in flexibility associated with the long fusion. Patients who underwent a short fusion were relatively more flexible than those who had their fusion extended into the thoracic spine. As already discussed, decreased lumbar mobility associated with a lumbar fusion may increase the risk of degenerative changes and low back pain.14–19 It is unclear if the increased immobility of a longer thoracic and lumbar fusion will also increase the risk of degeneration or low back pain compared with a patient with a short lumbar fusion. Long-term studies will be needed to address this question. In the more immediate future, the importance of flexibility in the patient's postoperative activity (sport, dance, etc.) should be considered when selecting fusion levels for a Lenke 5 curve.
One of the limitations of this study is the number of anterior and posterior approaches used to treat the patients. Our intent was to obtain a very close match of patients with similar preoperative values. As such, we did mix anterior and posterior approaches to obtain the closest match possible. Some may consider that an anterior-only approach may result in better flexibility, because there is no dissection through the posterior paraspinous musculature. We did analyze the anterior versus posterior selective fusions separately, but were underpowered to report the results (n = 6, posterior selective fusions). The data did suggest, however, that the flexibility measures were influenced most by the long fusion as opposed to the surgical approach.
It is unclear why 27% of the patients in our cohort had their thoracic spine included in the fusion. Although the patients matched similarly, specific clinical factors or patient concerns (skeletal maturity, shoulder balance, rib prominence, etc.) may have influenced the specific treatment plan. Interestingly, at short-term follow-up, whether the patient had a short or long fusion did not influence patient outcome as measured by the Scoliosis Research Society questionnaire. Overall, the relative importance of minor differences in coronal and sagittal correction as well as truncal flexibility need to be balanced with individual surgeon and patient criteria that determine fusion levels for Lenke 5 curves. Only long-term follow-up will help us to understand the risk and benefit of preserving thoracic motion segments in primary TL/L curves.
Fusion of the thoracic curve in a Lenke type 5 curve improves coronal correction.
Fusion of the thoracic curve in a Lenke type 5 curve reduces thoracic kyphosis and clinical flexibility.
At 2 years postoperatively, Scoliosis Research Society questionnaire scores and clinical balance are not significantly different between a matched set of patients that had either a selective or nonselective fusion of their Lenke 5 curve.
1. 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.
2. Coonrad RW, Murrell GA, Motley G, et al. A logical coronal pattern classification of 2000 consecutive idiopathic scoliosis cases based on the scoliosis research society–defined apical vertebra. Spine (Phila Pa 1976) 1998;23:1380–91.
3. 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-A:1169–81.
4. 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 (Phila Pa 1976) 2008;33:1068–73.
5. Jansen RC, van Rhijn LW, Duinkerke E, et al. Predictability of the spontaneous lumbar curve correction after selective thoracic fusion in idiopathic scoliosis. Eur Spine J 2007;16:1335–42.
6. Edwards CC II, Lenke LG, Peelle M, et al. Selective thoracic fusion for adolescent idiopathic scoliosis with C modifier lumbar curves: 2- to 16-year radiographic and clinical results. Review. Spine (Phila Pa 1976) 2004;29:536–46.
7. Dobbs MB, Lenke LG, Walton T, et al. Can we predict the ultimate lumbar curve in adolescent idiopathic scoliosis patients undergoing a selective fusion with undercorrection of the thoracic curve? Spine (Phila Pa 1976) 2004;29:277–85.
8. Winter RB, Lonstein JE. A meta-analysis of the literature on the issue of selective thoracic fusion for the King Moe type II curve pattern in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2003;28:948–52.
9. 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 (Phila Pa 1976) 1999;24:1663–71.
10. Lenke LG, Bridwell KH, Baldus C, et al. Preventing decompensation in King type II curves treated with Cotrel-Dubousset instrumentation. Strict guidelines for selective thoracic fusion. Spine (Phila Pa 1976) 1992;17:S274–81.
11. Newton PO, Faro FD, Lenke LG, et al. Factors involved in the decision to perform a selective versus nonselective fusion of Lenke 1B and 1C (King Moe II) curves in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2003;28:S217–23.
12. Richards BS. Lumbar curve response in type II idiopathic scoliosis after posterior instrumentation of the thoracic curve. Spine (Phila Pa 1976) 1992;17:S282–6.
13. Lenke LG, Edwards CC II, 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:S199–207.
14. Danielsson AJ, Cederlund CG, Ekholm S, et al. The prevalence of disc aging and back pain after fusion extending into the lower lumbar spine. A matched MR study twenty-five years after surgery for adolescent idiopathic scoliosis. Acta Radiol 2001;42:187–97.
15. Cochran T, Irstam L, Nachemson A. Long-term anatomic and functional changes in patients with adolescent idiopathic scoliosis treated by Harrington rod fusion. Spine 1983;8:576–84.
16. Danielsson AJ, Nachemson AL. Back pain and function 23 years after fusion for adolescent idiopathic scoliosis: a case-control study part II. Spine 2003;28:E373–83.
17. Hayes MA, Tompkins SF, Herndon WA, et al. Clinical and radiological evaluation of lumbosacral motion below fusion levels in idiopathic scoliosis. Spine 1988;13:1161–7.
18. Connolly PJ, Von Schroeder HP, Johnson GE, et al. Adolescent idiopathic scoliosis. Long-term effect of instrumentation extending to the lumbar spine. J Bone Joint Surg Am 1995;77:1210–6.
19. Paonessa KJ, Engler GL. Back pain and disability after Harrington rod fusion to the lumbar spine for scoliosis. Spine 1992;17:S249–53.
20. Wilk B, Karol LA, Johnston CE II, et al. The effect of scoliosis fusion on spinal motion: a comparison of fused and nonfused patients with idiopathic scoliosis. Spine (Phila Pa 1976) 2006;31:309–14.
21. Sanders AE, Baumann R, Brown H, et al. Selective anterior fusion of thoracolumbar/lumbar curves in adolescents: when can the associated thoracic curve be left unfused? Spine (Phila Pa 1976) 2003;28:706–13
22. Wang Y, Fei Q, Qiu G, et al. Anterior spinal fusion versus posterior spinal fusion for moderate lumbar/thoracolumbar adolescent idiopathic scoliosis: a prospective study. Spine (Phila Pa 1976) 2008;33:2166–72.
23. Geck MJ, Rinella A, Hawthorne D, et al. Comparison of surgical treatment in Lenke 5C adolescent idiopathic scoliosis: anterior dual rod versus posterior pedicle fixation surgery: a comparison of two practices. Spine (Phila Pa 1976) 2009;34:1942–51.
24. Li M, Ni J, Fang X, et al. Comparison of selective anterior versus posterior screw instrumentation in Lenke 5C adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2009;34:1162–6.
25. Shufflebarger HL, Clark CE. Effect of wide posterior release on correction in adolescent idiopathic scoliosis. J Pediatr Orthop B 1998;7:117–23. PubMed PMID: 9597586.
26. 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.
27. 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.
28. 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.
29. 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.
30. 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.
31. 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.
32. Newton PO, Yaszay B, Upasani VV, et al.; Harms Study Group. Preservation of thoracic kyphosis is critical to maintain lumbar lordosis in the surgical treatment of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2010;35:1365–70.
33. Sucato DJ, Agrawal S, O'Brien MF, et al. Restoration of thoracic kyphosis after operative treatment of adolescent idiopathic scoliosis: a multicenter comparison of 3 surgical approaches. Spine (Phila Pa 1976) 2008;33:2630–6.
adolescent idiopathic scoliosis; thoracolumbar/lumbar curve; selective fusion
© 2013 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.