Ilgenfritz, Ryan M. MD*; Yaszay, Burt MD†; Bastrom, Tracey P. MA†; Newton, Peter O. MD†; Harms Study Group
In 2001, Lenke et al1 published a new surgical classification system for adolescent idiopathic scoliosis (AIS). This classification defines 6 distinct curve patterns based on the location, magnitude, and flexibility of the primary curve(s) of the deformity. It is further classified on the basis of the apical deviation of the lumbar curve and sagittal profile of the thoracic spine. This system is now widely used by surgeons as a surgical guide to differentiate between primary and compensatory curves to determine the extent of fusion needed in various AIS spinal deformities.2
After a selective fusion for Lenke 1C and 5C AIS curve types, there is concern that the uninstrumented compensatory curve will continue to progress over time. Previously, many studies have been developed to study the risk of progression of uninstrumented compensatory curves after selective fusion for AIS.2–23 Although the results of the majority of these studies have been favorable, most have involved the retrospective application of this classification system to a series of patients treated surgically for AIS in an attempt to validate the ability to predict the long-term outcome of spinal deformities that are fused selectively. To date, there have been no studies using prospectively collected data beyond 2 years postoperatively to determine the natural history of uninstrumented compensatory curves after selective fusion for AIS spinal deformity.
In this study, a prospectively collected multicenter database was used to analyze the natural history of uninstrumented compensatory curves during a 5-year postoperative period after selective fusion for Lenke 1C and 5C AIS spinal deformity.
MATERIALS AND METHODS
After institutional review board approval, a prospectively collected multicenter database was used to identify patients treated with selective fusion for AIS. The patients included were treated with selective main thoracic (MT) or thoracolumbar/lumbar (TL/L) fusion with a minimum 5-year follow-up. Selective MT fusion was defined as a fusion in which the lowest instrumented vertebra was L1 or above for spinal deformity patterns classified as Lenke 1. Selective TL/L fusion was defined as a fusion in which the most cranial instrumented vertebra was at T9 or below for spinal deformity patterns classified as Lenke 5. All selected cases were classified with a Lenke lumbar modifier C.
Radiographical measurements were collected from preoperative, first-erect, 1-year, 2-year, and 5-year postoperative images. These measurements included MT and TL/L Cobb angles, coronal balance as measured by C7-CSVL, thoracic kyphosis Cobb angles, and lumbar lordosis Cobb angles. Additionally, the degree of axial rotation of the compensatory curve was approximated using the Perdriolle torsionmeter method.
Preoperative, first erect, 1-year, 2-year, and 5-year postoperative coronal, sagittal, and axial (Perdriolle) radiographical outcomes were compared using repeated measures analysis of variance with Bonferroni post hoc comparisons (P < 0.05).
Twenty-four selectively fused Lenke 1C curves and 21 selectively fused Lenke 5C curves with a minimum of 5 years of follow-up data were identified. All patients were in the Risser grade of 4 or 5 at 5 years of postoperative follow-up, indicating relative skeletal maturity.
Each group included patients treated with either an anterior or a posterior surgical approach and instrumentation. In the Lenke 1C group, 17 (70.8%) of the 24 patients were treated with an anterior approach (Figure 1) and 7 (29.2%) of the 24 patients were treated with a posterior approach (Figure 2). In the Lenke 5C group, 15 (71.4%) of the 21 were treated with an anterior approach (Figure 3), and 6 (28.6%) of the 21 were treated with a posterior approach.
In the Lenke 1C group, the preoperative primary curve coronal Cobb angle was at an average 49° ± 9°. Postoperatively, the primary MT curve corrected to 21° ± 7° at the first erect (P = 0.007), which demonstrated an average of 2° to 3° increase each year until the 5-year visit (Figure 4). The first erect MT curve percent correction averaged 58%, which showed a nonsignificant decline between subsequent visits compared with the 5-year postoperative period (Table 1). However, the overall change in percent correction from first erect and 5 years was significantly different (P ≤ 0.001). The preoperative compensatory lumbar curve coronal Cobb angle was an average 40° ± 6°. Postoperatively, these curves corrected to 28° ± 9° at first erect (P < 0.001), which further corrected another 5° on average at 1 year. The compensatory curve then showed an average increase of 2° between 2 and 5 years and remained stable between the 2- and 5-year periods (Figure 4). The percent correction of the compensatory curve was averaged 32% at first erect, which significantly increased to 44% at 1 year (P = 0.02) and remained relatively stable throughout the follow-up period (Table 1). The change in percent correction from first erect up to 5 years was not significantly different (P = 0.10).
In Lenke 5C curves, the preoperative primary curve Cobb angle averaged 46° ± 7°. Postoperatively, the primary TL/L curves corrected to a mean of 11° ± 5° at first erect (P = 0.007), with an average of 5° loss of correction at 1 year and then remained stable at 5 years postoperatively (Figure 5). The first erect primary lumbar correction averaged 77%, which significantly declined at 1 year (P ≤ 0.001) and then remained stable at 5 years (P > 0.10, Table 2). The overall change from first erect (77%) up to 5 years (66%) was significantly different (P ≤ 0.001). The preoperative thoracic compensatory curve Cobb angle averaged 25° ± 9°. Postoperatively, these curves corrected to a mean of 15° ± 8° at first erect (P < 0.001) and remained stable throughout the follow-up period (Figure 5). The first erect percent correction of the compensatory curve averaged 37%, which did not significantly change throughout the 5-year postoperative period (P > 0.10, Table 2). The overall change from first erect up to 5 years was not significantly different (P = 0.17).
In the Lenke 1C group, there was no significant change in sagittal profile in the instrumented curve (thoracic kyphosis) between the preoperative and first erect visits (P > 0.10) or any of the subsequent postoperative visits (P > 0.10, Table 3). There was also no significant change in lumbar lordosis after instrumentation of the MT curve at first erect or any subsequent follow-up visits (P > 0.10, Table 3). In Lenke 5C curves, there was an initial significant decrease in sagittal profile in the instrumented curve (lumbar lordosis) at the first erect visit (P = 0.004). The changes during the remaining postoperative period were not significant (P > 0.10). The sagittal profile of the uninstrumented curve (thoracic kyphosis) did not significantly change between preoperative and first erect and remained stable throughout the postoperative period (P > 0.10).
In Lenke 1C curves, the preoperative radiographical lumbar axial rotation (Perdriolle) averaged 18° ± 6.4°, which showed a nonsignificant decrease at first erect (P = 0.06) and then remained stable for the postoperative period (P > 0.10) (Table 4). In Lenke 5C curves, the preoperative radiographical thoracic axial rotation (Perdriolle) averaged 10° ± 5°, which showed no significant change after fusion of the lumbar curve (P > 0.10) or during the postoperative follow-up period (P > 0.10).
Coronal balance (C7-CSVL) decreased slightly from an average of 1.9 cm to the left preoperatively in patients with Lenke 1C to 1.3 cm at first erect (P = 0.11). The coronal balance then remained relatively unchanged for the remaining postoperative period, with no significant differences (P > 0.10, Table 5). Coronal balance (C7-CSVL) was unchanged in patients with Lenke 5C from preoperative (−2.1) to first erect (−2.0) (P = 0.88). However, there was a significant decrease (shift toward neutral) between the first erect and 1-year visit (P = 0.02) and even further between the 1-year and 2-year visit (P = 0.02, Table 5). A slight, but nonsignificant increase in imbalance was observed between 2 and 5 years (average change of 0.1 cm, P = 0.08).
Although the majority of the patients with Lenke 1C were treated anteriorly, current findings indicate no significant differences in change over time between the 2 approaches for coronal, sagittal, and Perdriolle measures (P > 0.10). The average percent compensatory lumbar coronal correction remained stable between the approaches for the Lenke 1C curves at final follow up (38% ± 16% anterior, 40% ± 27% posterior). Similarly, the uninstrumented sagittal lordosis seemed, on average, similar (53° ± 15° anterior, 61° ± 11° posterior) with both groups showing a modest decrease from preoperative (2° anterior, 5° posterior). The Perdriolle measure was similarly stable between both approaches, with a similar final average rotation (17° ± 6° anterior, 17° ± 9° posterior).
Similar findings were observed in the Lenke 5C curves. The percent compensatory thoracic coronal correction was on average slightly better in the posterior group (39% ± 20%) than the anterior (27% ± 34%); however, there is significant overlap in the variability between the groups as indicated by the standard deviation. Uninstrumented thoracic sagittal kyphosis remained stable and similar between approaches at final follow up (25° ± 13° anterior, 32° ± 6° posterior) with both groups showing a slight increase from preoperative (5° anterior, 3° posterior). The Perdriolle measure remained stable from preoperative in both groups with a similar final average (9° ± 6° anterior, 10° ± 4° posterior).
The concept of selective fusion in spinal deformity surgery was first introduced by Moe19 more than 50 years ago. The goals of selective spinal fusion are to obtain excellent, long-lasting correction of all components of the deformity (including coronal and sagittal deformities, as well as axial rotation) while maintaining as many cranial and caudal motion segments as possible.1,2,11,19–21,24 The spinal deformity surgeon must carefully weigh the benefits of selective fusion versus the risk of compensatory curve progression in each case.
After publication of the Lenke classification for AIS in 2001,1 the majority of studies on compensatory curve progression after selective fusion have focused on retrospective review of cases in which the principles of selective fusion were presumed to have been followed correctly. In this circumstance, patients seem to do well clinically and radiographically after selective thoracic1,2,5–8,10–12,23,25 and TL/L2,16,18 fusion. With the exception of 2-year postoperative data, there is little prospectively collected data in the literature regarding selective fusion for AIS. This study suggests that in Lenke 1C and 5C, AIS deformity patterns fused selectively, the uninstrumented compensatory curves do not seem to progress in coronal, sagittal, or axial rotation measures up to 5 years postoperatively.
Previous studies have demonstrated that spontaneous correction of the uninstrumented lumbar curve in selective fusions of primary thoracic deformities matches the correction of the primary thoracic deformity.2,5–7,26 Jansen et al8 found the correction of the MT and compensatory lumbar curves to be significantly correlated. This study demonstrated similar findings, with ultimate spontaneous lumbar correction within 15% of the instrumented thoracic correction.
In the Lenke 1C group in this study, the lumbar compensatory curves seemed to stabilize at 2 years with no significant change from 2 to 5 years postoperatively (Table 1, Figure 4). This supports data from prior retrospective studies evaluating the results of selective fusion in primary MT curve patterns.2,4–12,17,22,23,26 Compensatory curve lumbar lordosis remains stable up to 5 years of postoperative follow-up. There was no significant change noted in the compensatory lumbar curve axial rotation measurements with the use of the Perdriolle torsionmeter method.15,26–28
Although there have been fewer studies looking at the correction of the unfused thoracic curve in selective fusions of primary TL/L curves, several have demonstrated that spontaneous correction of the thoracic curve does indeed occur.2,16,18 Sanders et al16 found that when the ratio between the major TL/L Cobb and the thoracic Cobb angles are 1.25 or greater, and/or the thoracic curve bends to 20° or less on side bending, the resulting Cobb correction is similar between the instrumented TL/L and uninstrumented thoracic curves. This study found that the spontaneous thoracic correction in Lenke 5C curves mirrored that of the instrumented TL/L correction, similar to the pattern observed in the Lenke 1C curves.
Also similar to the Lenke 1C (primary thoracic) curves, the Lenke 5C curve patterns in our study demonstrated stable thoracic compensatory curve correction after first erect imaging, with no significant change to 5 years postoperatively (Table 2, Figure 5). This also supports data from prior retrospective studies focused on the results of selective fusion in primary TL/L/lumbar curve patterns.2,16,18 Compensatory curve thoracic kyphosis remains stable from first erect to 5-year postoperative imaging. There was no significant change noted in compensatory thoracic curve axial rotation measurements.15,26–28
Perhaps the most striking finding of our data was how closely the magnitude of the uninstrumented compensatory curves seemed to mirror the magnitude of the instrumented primary curves (Tables 1, 2; Figures 4, 5). This finding was evident in both Lenke 1C and 5C deformity patterns investigated in this study. This suggests that these curves responded as true compensatory curves in this cohort of patients. This supports prior retrospective data from the literature.22
A potential limitation of this study is the low number of patients with Lenke 1C and 5C that were available for review from the prospective database. This can be expected to improve over time as the database continues to mature. Additionally, this study was not able to control for the surgical technique used by the surgeons. Planned correction rates and absolute criteria for performing a selective fusion were determined by the individual surgeon. Ideally, the benefit of evaluating data from a multicenter database is to average the effect of any specific surgeon.
Another potential limitation of the generalizability of the results of this study to the current AIS surgical population is the relatively high number of patients in the study that underwent surgical correction through an anterior approach and instrumentation. We recognize that currently the majority of patients undergo posterior instrumentation and fusion for these spinal deformity patterns. Depending on the study, there is some debate whether an equivalent deformity correction can be achieved from either approach or if one is better.4,10,12 This may be a reflection of the type of instrumentation used posteriorly with greater correction generally being attained with segmental pedicle screw fixation. In 1999, Betz et al4 compared 2 cohorts of patients undergoing either anterior or posterior instrumentation and fusion for major thoracic AIS spinal deformities for 2 years postoperatively. In their study there was no statistical difference in spontaneous lumbar curve correction between anterior and posterior instrumentation and fusion, reporting correction of 51% in both groups. Also in 1999, Lenke et al10 retrospectively reviewed 123 cases of patients with primary thoracic-compensatory lumbar AIS who were treated with selective thoracic fusion by either an anterior or a posterior instrumentation and fusion. They noted a major thoracic curve correction of 58% in the anterior group and 38% in the posterior group and a spontaneous compensatory lumbar curve correction (SLCC) of 56% and 37%, respectively. In 2008, Patel et al12 reviewed prospectively collected data from 132 patients with AIS who underwent selective thoracic fusion and compared SLCC for anterior and posterior approaches. The average SLCC in the anterior approach cohort was 44% ± 19%, and the average SLCC in the posterior instrumentation group was 49% ± 19%.
Posterior constructs have reportedly been less successful than anterior in restoring sagittal alignment in both primary thoracic and primary thoracolumbar curves.4,25,29 It has been demonstrated that reduction of thoracic kyphosis in selective fusion of primary thoracic curves is correlated with cervical decompensation30 and loss of lumbar lordosis.31 We were unable to compare the incidence of sagittal decompensation above and below the fusion between posterior and anterior constructs critically because of our sample size. This is a limitation of this study and an important area of future research.
The focus of this study was on the 5-year natural history of the uninstrumented compensatory curves in 2 different curve types. Although we did not have the number to compare the 2 approaches adequately, we think that the trends in our 5-year data are comparable with previous research comparing spontaneous correction between the 2 approaches at 2 years postoperatively. This cohort of selective fusion patients with 5-year follow-up provides evidence to the stability of fusions during this postoperative period regardless of construct type, and provides benchmark data for selectively fused patients with AIS at their 5-year visits. Ultimately, further follow-up will be needed to determine the long-term decompensation risk or incidence of degenerative disc disease in selectively fused patients with AIS. Currently, enrollment and follow-up are ongoing. Although there is potential for later progression of the compensatory curves, all patients included in this study had reached relative skeletal maturity by the time of final radiographical follow-up (Risser 4 or 5). On the basis of prior natural history studies of nonoperatively treated patients with AIS, it would be expected that these compensatory curves would not continue to progress beyond skeletal maturity given the mean curve magnitude at 5-year postoperative follow-up.32–35 This is supported by a recent report on 20-year follow-up of selective thoracic fusions, which demonstrated stable coronal and sagittal correction, no revision surgeries, and good to excellent functional scores in these patients.36 To quantify the success of selective thoracic fusion because it relates to possibly preventing degenerative changes in the uninstrumented segment secondary to saving motion segments, weighed against the risk of decompensation during the lifespan, much longer follow-up of these patients and an equivalent long fusion comparison group will be required.
In Lenke 1C and 5C AIS deformity patterns fused selectively, the uninstrumented compensatory curves adjust to match the instrumented primary curves and do not seem to progress between 1 and 5 years postoperatively. The sagittal and axial measures of the compensatory curves remain stable during the postoperative period. Longer follow-up on a larger number of patients will be necessary to evaluate concern for progression of uninstrumented compensatory curves beyond 5 years postoperatively.
* The magnitude of the uninstrumented compensatory curves adjust to match the magnitude of the instrumented primary curves, suggesting that these curves responded as true compensatory curves in this cohort of patients.
* In the Lenke 1C group, the lumbar compensatory curves seemed to stabilize at 2 years with no significant change from 2 to 5 years postoperatively.
* In Lenke 5C curve patterns in our study, the thoracic compensatory curves seemed to remain stable after first erect imaging with no significant change up to 5 years postoperatively.
* In both curve types, the sagittal and axial measures of the compensatory curves remain stable during the postoperative period.
1. 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.
2. Lenke LG, Edwards CC 2nd, 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.
3. Abel MF, Herndon SK, Sauer LD, et al. Selective versus nonselective fusion for idiopathic scoliosis: does lumbosacral takeoff angle change? Spine (Phila Pa 1976) 2011;36:1103–2.
4. Betz RR, Harms J, Clements DH, et al. Comparison of anterior and posterior instrumentation for correction of adolescent thoracic idiopathic scoliosis. Spine (Phila Pa 1976) 1999;24:225–39.
5. Chang KW, Chang KI, Wu CM. Enhanced capacity for spontaneous correction of lumbar curve in the treatment of major thoracic-compensatory C modifier lumbar curve pattern in idiopathic scoliosis. Spine (Phila Pa 1976) 2007;32:3020–9.
6. 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.
7. Edwards CC 2nd, 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. Spine (Phila Pa 1976) 2004;29:536–46.
8. 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.
9. Kamimura M, Ebara S, Kinoshita T, et al. Anterior surgery with short fusion using the Zielke procedure for thoracic scoliosis: focus on the correction of compensatory curves. J Spinal Disord 1999;12:451–60.
10. 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; discussion 1672.
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. 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.
13. Puno RM, Johnson JR, Ostermann PA, et al. Analysis of the primary and compensatory curvatures following Zielke instrumentation for idiopathic scoliosis. Spine (Phila Pa 1976) 1989;14:738–43.
14. Richards BS. Lumbar curve response in type II idiopathic scoliosis after posterior instrumentation of the thoracic curve. Spine (Phila Pa 1976) 1992;17(suppl):S282–6.
15. Ritzman TF, Upasani VV, Bastrom TP, et al. Comparison of compensatory curve spontaneous derotation after selective thoracic or lumbar fusions in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2008;33:2643–7.
16. 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; discussion 714.
17. Takahashi J, Newton PO, Ugrinow VL, et al. Selective thoracic fusion in adolescent idiopathic scoliosis: factors influencing the selection of the optimal lowest instrumented vertebra. Spine (Phila Pa 1976) 2011;36:1131–41.
18. Wang T, Zeng B, Xu J, et al. Radiographic evaluation of selective anterior thoracolumbar or lumbar fusion for adolescent idiopathic scoliosis. Eur Spine J 2008;17:1012–8.
19. 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.
20. 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.
21. King HA. Selection of fusion levels for posterior instrumentation and fusion in idiopathic scoliosis. Orthop Clin North Am 1988;19:247–55.
22. Kalen V, Conklin M. The behavior of the unfused lumbar curve following selective thoracic fusion for idiopathic scoliosis. Spine (Phila Pa 1976) 1990;15:271–4.
23. Chang MS, Bridwell KH, Lenke LG, et al. Predicting the outcome of selective thoracic fusion in false double major lumbar “C” cases with five- to twenty-four-year follow-up. Spine (Phila Pa 1976) 2010;35:2128–33.
24. Cochran T, Irstam L, Nachemson A. Long-term anatomic and functional changes in patients with adolescent idiopathic scoliosis treated by Harrington rod fusion. Spine (Phila Pa 1976) 1983;8:576–84.
25. 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 (Phila Pa 1976) 2005;30:1859–66.
26. Hosman AJ, Slot GH, van Limbeek J, et al. Rip hump correction and rotation of the lumbar spine after selective thoracic fusion. Eur Spine J 1996;5:394–9.
27. Richards BS. Measurement error in assessment of vertebral rotation using the Perdriolle torsionmeter. Spine (Phila Pa 1976) 1992;17:513–7.
28. Kuklo TR, Potter BK, Lenke LG. Vertebral rotation and thoracic torsion in adolescent idiopathic scoliosis: what is the best radiographic correlate? J Spinal Disord Tech 2005;18:139–47.
29. Tao F, Wang Z, Li M, et al. A comparison of anterior and posterior instrumentation for restoring and retaining sagittal balance in patients with idiopathic adolescent scoliosis. J Spinal Disord Tech 2012;25:303–8.
30. Hwang SW, Samdani AF, Tantorski M, et al. Cervical sagittal plane decompensation after surgery for adolescent idiopathic scoliosis: an effect imparted by postoperative thoracic hypokyphosis. J Neurosurg Spine 2011;15:491–6.
31. Newton PO, Yaszay B, Upasani VV, et al. Preservation of thoracic kyphosis is critical to maintain lumbar lordosis in the surgical treatment of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2010;35:1365–70.
32. Weinstein SL, Zavala DC, Ponseti IV. Idiopathic scoliosis: long-term follow-up and prognosis in untreated patients. J Bone Joint Surg Am 1981;63:702–12.
33. Weinstein SL, Ponseti IV. Curve progression in idiopathic scoliosis. J Bone Joint Surg Am 1983;65:447–55.
34. Weinstein SL. Idiopathic scoliosis. Natural history. Spine (Phila Pa 1976) 1986;11:780–3.
35. Weinstein SL, Dolan LA, Spratt KF, et al. Health and function of patients with untreated idiopathic scoliosis: a 50-year natural history study. JAMA 2003;289:559–67.
36. Larson AN, Fletcher ND, Daniel C, et al. Lumbar curve is stable after selective thoracic fusion for adolescent idiopathic scoliosis: a 20-year follow-up. Spine (Phila Pa 1976) 2012;37:833–9.
adolescent idiopathic scoliosis; natural history; uninstrumented compensatory curves; selective fusion
© 2013 Lippincott Williams & Wilkins, Inc.