Rigid dorsal kyphosis associated with endplate irregularities, Schmorl's nodes, and wedging of apical vertebrae was first described by Scheuermann in 1921.1 Sorensen's radiographic criterion of a minimum of 3 adjacent vertebrae wedged at least 5° was proposed in 1964.2 Prevalence of Scheuermann's kyphosis has been reported to be between 0.4% and 10%.2,3 A familial tendency, but not a clear gender predilection, has been observed by a number of authors.3–7 Patients may present with midthoracic or thoracolumbar disease associated with the apex of deformity.9,10
Operative management has been advocated for adolescents with progressive kyphosis over 70°, for those who have had progression despite bracing, for patients with intractable back pain, and finally for unacceptable cosmetic deformity.11,12 One of the early reported series by Bradford et al of surgical management by a posterior approach with Harrington compression instrumentation, in the majority of patients, was characterized by significant loss of correction at follow-up.13 The authors of that study advocated combined anterior-posterior surgery to minimize loss of correction, which did improve in their next reported series.14 Surgical management with Luque instrumentation was associated with a rate of junctional kyphosis of 68%.15 The problem of junctional kyphosis occurring adjacent to a corrective arthrodesis for Scheuermann's kyphosis was further illustrated by Reinhardt and Bassett and has been subsequently reported in nearly all series.16–20
There has been a paucity of reports on the results of operative management of Scheuermann's kyphosis using current techniques and implant systems. These studies have been characterized by relatively small cohorts (21–46 patients) and incomplete radiographic analysis.17–21 Variable recommendations for combined anterior-posterior or posterior-alone interventions have been made.17,19,21
The main aims of this study were to analyze the following issues relating to the operative management of Scheuermann's kyphosis in a relatively large cohort of patients:
- To compare outcomes of anterior-posterior to posterior arthrodesis alone, including radiographic analysis and complications.
- To evaluate loss of correction and the occurrence of proximal and distal junctional kyphosis and to identify factors related to same.
- To explore the role of pelvic incidence in treatment planning and evaluation of the patient with Scheuermann's kyphosis.
Materials and Methods
A retrospective multicenter study involving ten institutions was conducted by members of the Harms Study Group. Inclusion criteria for the study were as follows:
- Diagnosis of Scheuermann's kyphosis (wedging of ≥5° of 3 successive vertebrae, with or without endplate irregularities and Schmorl's nodes)
- Surgical correction of kyphosis with current generation multisegmental instrumentation
- Minimum 2 years clinical and radiographic follow-up
Institutional Review Board or equivalent (for Germany site) approval was obtained by each site before data collection and analysis. Patient information was entered onto a secure, HIPAA-compliant, Web-based data bank after appropriate deidentification. Analysis was done centrally after all data were received.
Clinical data retrieved from medical records included age at surgery, gender, surgeon, duration of surgery, estimated blood loss, blood products given, levels instrumented, instrumentation type, anchor type and levels, performance of anterior release (open or thoracoscopic), structural graft, nonstructural bone graft, posterior osteotomies, complications, and implant failure.
Radiographic data were entered for the preoperative, first postoperative erect, 1 year, 2 year, and final follow-up visit. Radiographic parameters measured have been previously described and illustrated22 and included thoracic kyphosis (greatest Cobb, T2–T12, T5–T12, T10–L2), lumbar lordosis (T12–S1, greatest Cobb), C7–S1 sagittal translation, proximal junctional kyphosis (PJK), distal junctional kyphosis (DJK), pelvic incidence, first lordotic disc, sagittal stable vertebra, lateral kyphosis, apical translation.
The sagittal stable vertebra is defined as the most proximal thoracolumbar or lumbar vertebra intersected (not bisected) by the posterior sacral vertical line, which is the vertical line constructed from the posterior corner of the superior sacral endplate.22 The first lordotic disc is defined as the most proximal thoracolumbar or lumbar disc with ≥5° of anterior opening.22 PJK is defined as kyphosis measured from 1 segment cephalad to the upper end instrumented vertebrae (EIV) to the proximal instrumented vertebrae with an abnormal value defined as ≥10°.22 DJK is defined as kyphosis measured from 1 segment caudal to the EIV to the EIV with an abnormal value again defined as ≥10° of kyphosis.22 C7 sagittal plumbline is defined as the vertical line dropped from the center of the C7 vertebrae on the lateral radiograph with the measurement taken from the posterior superior corner of the sacrum perpendicular to the vertical line. A negative value indicates that the plumbline falls posterior to the posterior corner of the sacrum and positive values denote an anterior position of the vertical line.22 Pelvic incidence is defined as an angle subtended by a line drawn from the midpoint of a line drawn through the center of both femoral heads to the midpoint of the superior sacral endplate and a line drawn perpendicular to the midpoint of the sacral endplate on the lateral radiograph (Figure 1).22 Lateral kyphosis apical translation is defined as the horizontal distance from the center of the kyphosis apical vertebrae to a vertical line drawn through the posterior superior corner of the sacrum.22
Independent samples t test was used to determine statistical differences, if any, between Group 1 (anterior-posterior surgery) and Group 2 (posterior surgery alone). In addition, Pearson correlation analysis, as well as the nonparametric Mann-Whitney U test, was performed to determine any correlation between the various variables under study.
Seventy-eight patients met the inclusion criteria. There were 25 female and 53 male patients with a mean age of 16.7 years (range, 9–27 years). Mean follow-up was 2.9 years (range, 2–6 years). Forty-two patients underwent combined anterior-posterior surgery (Group 1) and the remainder underwent posterior arthrodesis alone (Group 2). The mean preoperative greatest Cobb kyphosis was 78.8° for the whole group and 82.6° and 74.4°, respectively, for Groups 1 and 2. Information on coronal plane curvature was available for 65 patients. There were 25 thoracolumbar or lumbar curves, mean 13.8° (range, 10°–23°) and 19 thoracic curves, mean 16° (range, 10°–35°) in 35 of the 65 patients. Demographic data comparing the whole cohort and the groups with or without anterior release are contained in Table 1.
Operative data for the entire cohort and Groups 1 and 2 is contained in Table 2. Of note, significantly more patients in Group 2 (58.3%) had all pedicle screw constructs compared with Group 1 (4.8%). In Group 1, anterior release was done by standard thoracotomy in 38% and by thoracoscopy in 62%.
Thirteen complications in ten patients occurred in Group 1 (23.8%) and 2 complications in 22 patients occurred in Group 2 (5.5%) (P = 0.03). Complications are listed in Table 3. A total of 5 reoperations were required: 3 in Group 1 and 2 in Group 2. Three of the reoperations were for revision of patients with junctional kyphosis: 2 distal and 1 proximal. Two of these were in Group 2. One patient in Group 1 had a revision for rod breakage and pseudarthrosis. A final patient in Group 1 required irrigation and debridement for a wound infection.
Greatest Cobb kyphosis and lordosis (T12–S1) preoperatively and postoperatively, and at follow-up. Pelvic incidence are shown in Table 4 and Figures 2 and 3. Loss of greatest Cobb kyphosis correction in Group 1 was 3.3° (P = 0.10), whereas a loss of correction of 6.4° (P = 0.00) occurred in Group 2 patients. In both groups, a significant increase in lordosis occurred between the first postoperative visit and final follow-up. Pelvic incidence was 39.9° (range, 7°–59°) in Group 1 and 40.8° (range, 10°–66 °) in Group 2 (P = 0.78). C7 sagittal plumbline data are contained in Table 5. In both groups, the plumbline fell more posteriorly (became more negative) following surgery.
There was a direct linear correlation of lordosis to pelvic incidence both before surgery (P = 0.001) and at follow-up (P = 0.000); i.e., the greater the pelvic incidence, the larger the lordosis value. Lordosis also correlated with kyphosis at final follow-up (P = 0.02); however, there was no correlation between lordosis and kyphosis before surgery (P = 0.23). Kyphosis did not correlate with pelvic incidence at any time frame (P = 0.58).
PJK occurred in 25 patients, 32.1% of the total cohort. The preoperative magnitude of the PJK measurement for the whole group averaged 1.0°. Of the 25 patients who developed PJK >10°, the mean magnitude of PJK was 21.5° (range, 11°–25°), >20° more than the preoperative mean. One patient in Group 2 was reoperated on and was treated with a 1-level proximal extension of the fusion to the first thoracic vertebrae. A second patient in Group 1 was noted to have a clinical deformity related to PJK but revision surgery was not performed. DJK over 10° occurred in 4 patients (5.1%) with a mean value of 22° (range, 17°–28°). Of the 4 patients, there was 1 patient in each group who had a reoperation to extend the fusion distally (Figure 4).
The following radiographic and operative parameters were found to correlate with the development of proximal junctional kyphosis. There was a direct linear correlation between the magnitude of PJK and follow-up greatest Cobb kyphosis measurements (P = 0.00) and a trend in that direction for preoperative greatest Cobb kyphosis (P = 0.072). In other words, a larger magnitude of kyphosis both before surgery and at final follow-up tended to be associated with the development of PJK. A comparison of those patients who developed PJK with those who did not is contained in Table 6. Those patients that developed PJK had a lower kyphosis percentage correction compared with those patients that did not develop PJK.
Although pelvic incidence did not differ between those patients who developed PJK and those who did not, the magnitude of PJK had a direct linear correlation with magnitude of pelvic incidence (P = 0.00). That is, the greater the pelvic incidence for the individual, the higher the magnitude of PJK (Figure 5). Kyphosis/pelvic incidence ratios at follow-up were analyzed between patients who developed PJK and those who did not. The kyphosis/pelvic incidence ratio was 2.1 and 1.2, respectively (P = 0.02). No significant differences were noted for the lordosis/pelvic incidence ratios.
Patients who had the arthrodesis extended proximally to the level of or cephalad to the Cobb end vertebrae were less likely to develop PJK (P = 0.04) than those who were fused caudal to the proximal end vertebrae as illustrated in Figure 6. The distribution of proximal fusion levels for patients who did and did not develop PJK are illustrated in Figure 7. There was no apparent correlation of proximal anchor type (hook or screw) with the development of PJK (P = 0.36).
Of the 4 patients who developed distal junctional kyphosis, all had the distal extent of arthrodesis to the level of or caudal to the Cobb end vertebra. In 3 of these patients, fusion ended 2 levels cephalad to the sagittal stable vertebra and in 1, 3 levels cephalad; thus, all patients who developed DJK were fused proximal to the sagittal stable vertebrae. Fusion was cephalad to the first lordotic disc in 2 patients and caudad in 2 patients. Statistical significance could not be found for the development of DJK in relation to the level of fusion as compared with Cobb end vertebrae, sagittal stable vertebra, or first lordotic disc between groups due to small sample size. Of the 4 patients who developed DJK, the 2 patients that had combined anterior-posterior surgery both had structural interbody grafts, while the other 2 patients who had posterior surgery only did not have structural grafts.
In addition to standard radiographic analysis, we have begun to explore the role of a fixed pelvic determinant of sagittal alignment, pelvic incidence, in the occurrence of junctional kyphosis, and in surgical decision-making for Scheuermann's kyphosis.
Correction of kyphosis in this series was similar to what has been reported in other recent studies.17–20 The mean preoperative kyphosis was greater for those patients undergoing anterior release (82.6°) versus those who had posterior surgery alone (74.4°). This represents the tendency of the surgeons in the study group to perform anterior release for larger magnitude curvature. What cannot be ascertained from our data are whether posterior surgery alone might have resulted in similar curve correction as with the anterior-posterior approach. Hosman et al did not note any difference in curve correction with or without the addition of anterior release. These authors, however, aimed for correction of kyphosis in the “high normal” range.17 The desired end result of kyphosis surgery is to achieve a curve within a normal range; therefore, absolute curve correction or percent correction figures can be misleading as a smaller preoperative kyphosis requires less correction than a curve of larger magnitude, which is why our analysis focused on absolute endpoints and ranges rather than percentage correction.23 Kyphosis was corrected within desirable ranges in both groups in our study. Curvatures in Groups 1 and 2 at final follow-up were 52.0° and 41.8° (T2–T12), respectively, and 39.7° and 28.9° (T5–T12), respectively. Ponte and Siccardi advocate a posterior technique for curves of all magnitudes with wide posterior resections and posterior shortening using compressive instrumentation.21 They reported correction of kyphosis from 79° to 34° with minimal loss of correction at final follow-up.21 Recently, Johnston et al reported equivalent corrections using the Ponte technique and a threaded posterior compression construct between patients having posterior surgery alone and those undergoing anterior release and fusion.24 No differences in maintenance of correction were noted. In our series, a loss of correction of 6.4° (T2–T12) was noted in Group 2 patients and loss of correction was statistically insignificant in those patients who had anterior release. Perhaps this is related to a more rapid onset of fusion of apical segments, preserving correction more effectively. The tendency for better preservation of correction with the addition of anterior release and fusion has been pointed out by a number of authors.14,18,19 Although the differences in loss of correction were small, they were statistically significant. Interobserver variability was not evaluated as part of this study but is potentially a flaw in any multicenter study. Loss of correction using current surgical techniques has been reported to be 3° to 4° and 6° for combined surgery and posterior surgery, respectively.18,19 Loss of correction of approximately 15° has been associated with implant removal.20 The differences noted between the groups in our series should not be related to the posterior anchors used as more patients in Group 2 than in Group 1 had all screw constructs, which one might expect to result in improved correction and maintenance of correction. A more in-depth analysis of anchor choice (i.e., screws vs. hooks) is needed to assess the role of screw fixation in kyphosis surgery and that construct's ability to preserve correction compared with all hooks or hybrid hook and screw constructs.
Lordosis was statistically similar between Groups 1 and 2, before surgery, despite a significantly greater kyphosis noted in Group 1. However, lordosis was significantly less in Group 2 both after surgery and at final follow-up. Lordosis diminished as a result of kyphosis correction in both groups and then rose several degrees at final follow-up. This gradual balancing of kyphosis and lordosis over time has been reported previously.20 The fact that lordosis was similar for the 2 groups before surgery is possibly reflective of the fact that pelvic incidence was similar for the 2 groups. Pelvic incidence directly correlated with lordosis, before surgery, in this study as has been shown by a number of authors.25–27 This would suggest that lordosis below an unfused thoracic spine is influenced directly by pelvic incidence and perhaps only secondarily by kyphosis as suggested by our analysis. The interplay of kyphosis, lordosis, and pelvic incidence is aptly summarized by Jackson who states, “The lumbar lordosis [has] to complement the pelvic lordosis [reflected by pelvic incidence] and be concordant with the thoracic kyphosis for there to be congruent sagittal spinopelvic alignment.”26 Legaye et al similarly found a greater correlation of lordosis to pelvic incidence than lordosis to kyphosis.25
Lordosis after arthrodesis of the thoracic spine appeared to be influenced by the magnitude of the postoperative kyphosis. It is likely that the compensatory mechanism of the spine is both responsive to intrinsic pelvic morphology as well as changes in alignment and rigidity imparted by a spinal arthrodesis. Preoperative kyphosis did not correlate with pelvic incidence. This suggests that fixed pelvic morphometry as assessed by pelvic incidence is not a causative factor in the development of Scheuermann's kyphosis; rather, its etiology is more likely intrinsic to the thoracic spine.
Sagittal balance as assessed by C7 sagittal plumbline shifted posteriorly behind the sacrum in both groups after surgery indicating a restoration of more physiologic alignment. This has also been reported in other series.17,19,20 The final sagittal balance appears to be less posterior than reported for normal adolescents in 1 study.28 In that study by Vedantam et al,28 the C7 sagittal plumbline was measured against the anterior superior corner of the sacrum as opposed to the posterior corner as used in our study. Nevertheless, the mean sagittal vertical axis in their study was −5.6 cm in comparison to −0.4 and −1.1 cm in Groups 1 and 2, respectively, in our study, likely, a significant difference.
Scoliosis was noted in 35 patients (53.8%) for whom these data were available. The curves were of mild to moderate magnitude (10°–35°) and would not in themselves require surgical treatment. This parallels the series by Murray et al in which 72% of patients had scoliosis with a mean curve size of 16°.29
The complication rate was significantly greater in the Group 1 patients (23.8%) than in Group 2 patients (5.5%). The majority of complications in Group 1 were approach-related such as pneumothorax or pulmonary effusion or related to the magnitude of the procedure (acute renal failure and pulmonary embolism). There were 2 patients in Group 1 and 2 in Group 2 who were reoperated on for instrumentation or alignment-related issues. Symptomatic or clinically evident junctional kyphosis was noted in 2 patients in each group, 1 proximally and 1 distally, in each group. Radiographic junctional kyphosis of >10° that was asymptomatic will be discussed further below. The 1 case of pseudarthrosis occurred in the patient who had an anterior release and fusion. Discussion of complications in the recent literature is incomplete. Lowe and Kasten focused on the issue of junctional kyphosis in detail but did not report other maloccurrences.18 Johnston et al indicated 1 case of rod breakage but did not mention any analysis of junctional kyphosis or surgical complications in their series.24 Papagelopoulos et al reported 1 death from SMA syndrome, implant prominence in 3 patients, and fractured rods in 2 patients, but no pseudarthroses in a series of 21 patients.19 No mention of differences in complication rate was made for patients based on whether or not anterior surgery was performed. In a prospective series by Poolman et al,20 of 23 patients all of whom had anterior release, 1 case of aortic injury and chylothorax was reported, 3 cases of PJK, and 9 cases of implant removal for prominence of instrumentation or screw breakage were reported. Finally, Hosman et al found a complication rate of 18.8% in patients undergoing posterior surgery only versus 41% in those having the addition of an anterior procedure in 2 cohorts with statistically similar preoperative curve magnitude.17 The majority of complications in that series were implant- or alignment-related. Based on the results from our series and those of other reports in which complications were reported in detail, it would appear that posterior surgery only should be favored over combined surgery whenever possible for diminishing complication rates.
A high rate of junctional kyphosis was noted in our series. PJK was noted in 25 patients (32.1%) and DJK occurred in 4 patients (5.1%). Despite this high rate of radiographic findings, the problem required reoperation or was clinically problematic in only 4 patients (5.1%). Few authors have written in-depth about this problem and its avoidance. Lowe and Kasten reported a radiographic rate of PJK of 30% ranging from 12° to 49° and of DJK of 28% ranging from 10° to 30°.18 In their study, PJK was related to >50% correction of the curve magnitude in 5 of 10 patients who developed PJK. PJK was also related to fusing short (caudal to) the proximal Cobb end vertebrae by 1 or 2 levels. The authors recommended correcting kyphosis to no less than 40°. Reinhardt and Bassett pointed out the importance of including any vertebrae wedged >5° even if they extended beyond the Cobb end vertebrae in order to minimize this problem.16 In our series, PJK was not found to correlate with the degree or percentage of correction except in an inverse fashion. What was noted was that patients with large curves before surgery and at follow-up were more likely to develop PJK. It is unclear if this is indicative of a greater difficulty of the patient with a larger curve to assume physiologic global balance without inducing local imbalance adjacent to a fusion, or if there are factors at play related to ligamentous integrity adjacent to the fusion and to technical aspects of exposure of the spine and anchor placement or contouring of the rods at the ends of the construct. Hosman et al recommended correcting kyphosis to the high normal range of 40° to 50°.17 These authors point out the comment by Stagnara et al that patients have their own particular “physiognomy” that must be respected in achieving spinal balance.30 We believe this concept is reflected by the parameter of pelvic incidence as discussed below.
Pelvic incidence has been described as a fixed anatomic parameter that directly determines lumbar lordosis.25–27 It is a factor that directly defines an individuals standing posture. It does not appear to directly define kyphosis as borne out by our study. Changes in alignment of regions of the spine as a result of disc disease, spondylolisthesis, or surgical fusion, for example, require the flexible remaining segments of the spine to compensate to maintain overall sagittal balance. This compensatory mechanism is impacted by the fixed pelvic incidence below.17,25–28 We think that pelvic incidence is an important determinant of what the appropriate amount of correction for an individual might be particularly in a multilevel spinal arthrodesis in which the number of segments available for compensation is relatively limited. Our data showed that among those patients who developed junctional kyphosis, increasing pelvic incidence correlated with increasing magnitude of PJK. As the kyphotic spine is straightened, compensation of the spine to create or maintain sagittal balance occurs in the unfused lumbar spine and proximal thoracic spine and appears to be influenced by pelvic incidence. It is likely that the proximal thoracic region is more prone to this problem as it is kyphotic to begin with. Importantly, in our series, the preoperative PJK in those patients who developed >10° PJK at follow-up averaged only 1°. The mean PJK at follow-up in this group was 21.5° (range, 11°–25°) representing what we believe could potentially be a clinically significant problem perhaps resulting in early degenerative changes and pain over time. The normal range of intersegmental kyphosis in the proximal thoracic spine including segments between T1 and T6 has been shown to be between 1° and 5°.23 These levels represent the distribution of the proximal EIV in our series. Further study will be required to determine whether kyphosis within the instrumented spine should be corrected to within a defined ratio to pelvic incidence to minimize PJK. Perhaps assessing normative data of kyphosis, lordosis, and pelvic incidence in unaffected subjects would be of value to create a formula for physiologic correction. No doubt, other factors are at play, including the number of levels within the arthrodesis and the magnitude of the curvature as well as the handling of soft tissue including ligamentous restraints at surgery.
Distal junctional kyphosis in our series occurred in only 4 patients and required revision surgery in 2 patients. The lumbar spine appears to be relatively protected from junctional kyphosis as the unfused lumbar segments below the fusion are lordotic with the gravity line falling posteriorly. Indeed, the C7 sagittal plumbline fell posteriorly in the majority of patients after surgery. In all patients with DJK, arthrodesis was ended proximal to the sagittal stable vertebrae by 2 or 3 vertebrae. Fusion was ended cephalad to the first lordotic disc in 2 patients and caudad in 2 patients. Based on these few cases, the authors recommend including the sagittal stable vertebrae in the arthrodesis in the majority of cases. One may be able to “save a level” in cases in which there is a highly flexible curvature with midthoracic apex as long as the fusion is extended beyond the first lordotic disc.
Good radiographic outcomes have been found in a relatively large series of patients undergoing either combined anterior-posterior or posterior only surgery for Scheuermann's kyphosis. The combined procedure resulted in slightly better maintenance of kyphosis correction. It also resulted in a higher rate of complications. Radiographic proximal junctional kyphosis was relatively common and appears to be associated with large preoperative and postoperative curve magnitude and correlates with pelvic incidence. The authors intend to explore these issues further in prospective fashion.
- Reliable correction of kyphosis is achieved by either a combined anteroposterior or posterior only approach.
- A higher rate of complications is associated with a combined procedure.
- Proximal junctional kyphosis occurred in 32.1% of patients undergoing Scheuermann's kyphosis surgery and is related to arthrodesis caudal to the proximal end vertebra and is influenced by pelvic incidence.
- Distal junctional kyphosis occurred in 5.1% of patients undergoing kyphosis surgery and was always associated with arthrodesis cephalad to the sagittal stable vertebra.
The authors thank Georgia Panagopoulos, PhD, for her assistance in statistical analysis for this study.
1. Scheuermann H. Kyfosis dorsalis juvenilis. Ugeskr Laeger
2. Sorensen KH. Scheuermann's Juvenile Kyphosis
: Clinical Appearances, Radiography, Aetiology, and Prognosis
. Copenhagen: Munksgaard; 1964.
3. Bradford DS, Moe JH, Montalvo FJ, et al. Scheuermann's kyphosis
and roundback deformity: results of Milwaukee brace treatment. J Bone Joint Surg Am
4. Halal F, Gledhill RB, Fraser FC. Dominant inheritance of Scheuermann's juvenile kyphosis
. Am J Dis Child
5. Findlay A, Conner AN, Conner JM. Dominant inheritance of Scheuermann's juvenile kyphosis
. J Med Genet
6. McKenzie L, Silence D. Familial Scheuermann's disease: a genetic and linkage study. J Med Genet
7. Bjersand AJ. Juvenile kyphosis
in identical twins. AJR Am J Roentgenol
8. Van Linthoudt D, Revel M. Similar radiological lesions of localized Scheuermann's disease of the lumbar spine in twin sisters. Spine
9. Blumenthal Sl, Roach J, Herring TA. Lumbar Scheuermann's: a clinical series and classification. Spine
10. Greene TL, Hensinger RN, Hunter LY. Back pain and vertebral changes simulating Scheuermann's disease. J Pediatr Orthop
11. Lowe, TG. Current concepts review, Scheuermann disease. J Bone Joint Surg Am
12. Albanese SH, Albanese SL. Management of Scheuermann's kyphosis
. Semin Spine Surg
13. Bradford DS, Moe JH, Montalvo FJ, et al. Scheuermann's kyphosis
: results of surgical treatment by posterior spine arthrodesis in twenty-two patients. J Bone Joint Surg Am
14. Bradford DS, Ahmed KB, Moe JH, et al. The surgical management of patients with Scheuermann's disease: a review of twenty-four cases managed by combined anterior and posterior spine fusion. J Bone Joint Surg Am
15. Coscia MF, Bradford DS, Ogilvie JW. Scheuermann's kyphosis
: results in 19 cases treated by spinal arthrodesis and L-rod instrumentation. Orthop Trans
16. Reinhardt P, Bassett GS. Short segmental kyphosis
following fusion for Scheuermann's disease. J Spinal Disord
17. Hosman AJ, de Kleuver M, Anderson PG, et al. Analysis of the sagittal plane after surgical management for Scheuermann's disease: a view on overcorrection and the use of an anterior release. Spine
18. Lowe T, Kasten MD. An analysis of sagittal curves and balance after Cotrel-Dubousset instrumentation for kyphosis
secondary to Scheuermann's disease: a review of 32 patients. Spine
19. Papagelopoulos PJ, Peterson HA, Dekutoski MB. Surgical treatment of Scheuermann's disease with segmental compression instrumentation. Clin Orthop
20. Poolman RW, Been HD, Ubags LH. Clinical outcome and radiographic results after operative treatment of Scheuermann's disease. Eur Spine J
21. Ponte A, Siccardi GL. The biomechanical advantage of an innovative posterior technique for correction of Scheuermann's kyphosis
. J Bone Joint Surg Br
22. O'Brien MF, Kuklo TR, Blanke KM, et al. Spinal Deformity Study Group Radiographic Measurement Manual
. Memphis, TN: Medtronic Sofamor Danek; 2004.
23. Bernhardt M, Bridwell K. Segmental analysis of the sagittal plane alignment of the normal thoracic and lumbar spines and thoracolumbar junction. Spine
24. Johnston CE, Elerson E, Dagher G. Correction of adolescent hyperkyphosis with posterior-only threaded rod compression instrumentation: is anterior spinal fusion still necessary? Spine
25. Legaye J, Duval-Beaupere G, Hecquet J, et al. Pelvic incidence: a fundamental pelvic parameter for three-dimensional regulation of spinal sagittal curves. Eur Spine J
26. Jackson RP, McManus AC. Pelvic lordosis and pelvic incidence: the relationship of pelvic parameters to sagittal spinal profile. Curr Opin Orthop
27. Labelle H, Roussouly P, Berthonnaud E, et al. Spondylolisthesis, pelvic incidence, and spinopelvic balance. Spine
28. Vedanatam R, Lenke LG, Keeney JA, et al. Comparison of standing sagittal spinal alignment in asymptomatic adolescents and adults. Spine
29. Murray PM, Weinstein SL, Spratt KF. The natural history and long-term follow-up of Scheuermann's kyphosis
. J Bone Joint Surg Br
30. Stagnara P, Mavroy JC, Dran G, et al. Reciprocal angulation of vertebral bodies in a sagittal plane: approach to references for the evaluation of kyphosis
and lordosis. Spine