Abstract
Objective:
Loss of lumbar lordosis (LL) in degenerative deformity activates spinal compensatory mechanisms to maintain neutral C7 sagittal vertical axis (C7SVA), such as an increase in pelvic tilt (PT) and decreased thoracic kyphosis (TK). We study the extent to which PT increase and TK reduction contribute to the compensation of pelvic incidence (PI)-LL mismatch.
Methods:
A cohort of 43 adult patients with adult degenerative thoracolumbar deformity were included in this retrospective study. Radiographic spinopelvic measurements were obtained before and after corrective surgery. Pearson correlations were calculated.
Results:
Preoperative PI-LL mismatch significantly correlated with an increase in PT and a decrease in TK in the whole cohort r = +0.66 (95% confidence interval [CI] 0.44–0.8) and r = −0.67 (95% CI − 0.81–−0.47), respectively, at a relative rate of 0.37 (standard deviation [SD]: 0.07) and − 0.57 (SD: 0.09), respectively. In patients with low PI, only TK showed a significant correlation with PI-LL mismatch, r = −0.56 (95% CI − 0.8 to − 0.16), at a rate of − 0.57 (SD: 0.19). The high PI subgroup showed a significant correlation with PT, TK, and C7SVA, r = 0.62 (95% CI 0.26–0.82), r = −0.8 (95% CI − 0.9–−0.58), and r = 0.71 (95% CI 0.41–0.87) at rates of 0.48 (SD: 0.11), −0.72 (SD: 0.12), and 0.62 (SD: 1.27).
Conclusions:
Decreased TK represented a more consistent compensatory mechanism in patients with high and low PI when compared to an increase in PT. PI-LL mismatch induced more pronounced changes in TK than did PT in both subgroups. Patients with high PI relied more on increases in PT and a relative decrease in TK to compensate for PI-LL mismatch than patients with low PI.
Keywords: Adult spinal deformity, pelvic incidence, pelvic tilt, sagittal balance, thoracolumbar deformity
INTRODUCTION
Loss of lumbar lordosis (LL) in adult degenerative and adult spinal deformity (ASD) can activate compensatory mechanisms to maintain a neutral C7 sagittal vertical axis (C7SVA).[1,2] In the context of ASD, positive sagittal imbalance – a parameter strongly correlated with a disability, pain, and negative health outcomes[3] – may be offset by decreased thoracic kyphosis (TK) or increased pelvic tilt (PT).[4,5] However, these compensatory changes are energy-dependent and may ultimately cause fatigue and decreased health status.[6]
In the case of ASD, loss of LL is insidious and driven primarily by multisegmental disc degeneration with the potential for significant disability if substantial pelvic incidence (PI)-LL mismatch develops.[7] The primary goal of surgical correction is to restore age-appropriate spinopelvic alignment balance[8] as restoration of sagittal balance and PI-LL mismatch are closely linked to improved pain and functional status postoperatively.[7] Failure to achieve age-appropriate alignment goals postoperatively with persisting PI-LL mismatch has been linked to an increased risk of adjacent segment disease, junctional failure and junctional kyphosis, and overall disability.[9] Risk of reoperation has been shown to be 10 times higher in those with unresolved PI-LL mismatch when compared to those with the restoration of PI-LL postoperatively.[9]
Postoperative “normalization” of activated compensatory mechanisms impacts spinal alignment and should be considered when planning surgical intervention. In the present study, we investigate if the activation of compensatory mechanisms is uniform or whether individual pelvic parameters have an influence. Further, we explore the extent to which compensatory alignment changes regress after corrective surgery to better predict long-term spinal alignment after corrective surgery.
METHODS
Approval for this study was obtained by the local institutional review board. An institutional database of patients with ASD, who underwent surgical treatment at a single institution has been searched for eligible patients. Inclusion criteria were pathologic PI-LL at baseline, minimum of three-level fusion, normal postoperative sagittal alignment parameters (C7SVA as well as PI-LL), and postoperative follow-up whole spine X-ray within 6 weeks of surgery. Exclusion criteria were incomplete pre- or postoperative whole spine standing X-rays. After screening of the database, a cohort of 43 adult patients with adult degenerative thoracolumbar deformity were included in this retrospective study, and the following radiographic measurements were obtained before and after corrective surgery: C7SVA, TK, T10L2 sagittal Cobb angle, LL, PT, and PI. Parameters were measured using Surgimap Spine software (Nemaris INC., New York, NY, USA) and then extracted into a spreadsheet. Patients were divided into low PI (40.4°, standard deviation [SD]: 6.2) and high PI (57.2°, SD: 6.9) subgroups. For the whole cohort as well as the low and high PI subgroups, Pearson correlations were calculated for preoperative PI-LL mismatch and preoperative PT, TK, T10L2, and C7SVA using GraphPad (Graphpad Holdings, LLC; San Diego, USA). Further correlations were calculated for perioperative change in PI-LL and perioperative change in PT, TK, T10L2 as well as C7SVA. The results were compared between low and high PI subgroups.
RESULTS
A total of 43 patients were identified in this retrospective study. Preoperative whole spine standing X-rays were used to determine C7SVA, TK, T10L2 sagittal Cobb angle, PT, PI, and PI-LL. Patients then have been dichotomized into low and high PI subgroups, which is shown in Table 1.
Table 1.
Baseline radiographic parameters
| Low PI group | High PI group | |
|---|---|---|
| n | 21 | 22 |
| PI | 40.4±6.2 (21-48) | 57.2±6.9 (49-77) |
| PI-LL | 11±11.7 (−20-31) | 22.8±19.1 (−18-63) |
| T10-L2 | 3±11.4 (−24-27) | 3.2±13.4 (−21-30) |
| PT | 16.3±8.2 (−4-29) | 25.7±8.4 (11-54) |
| TK | 40.9±11.7 (18-60) | 31.1±17.2 (5-77) |
| C7SVA | 79.6±64.9 (−43-254) | 82.2±62.2 (−9-208) |
Values are given as mean±SD (range). Patients have been dichotomized into low or high PI groups. Radiographic patient parameters at baseline in individuals with low and high PI. C7SVA - C7 sagittal vertical axis (mm), PI - Pelvic incidence (°), LL - Lumbar lordosis, PI-LL - PI-LL mismatch (°), PT - Pelvic tilt (°), T10-L2 - Sagittal Cobb angle T10-L2 (°), TK - Thoracic kyphosis (°)
Preoperative PI-LL mismatch significantly correlated with an increase in PT and a decrease in TK in the whole cohort r = +0.66 (95% confidence interval [CI] 0.44–0.8), P < 0.0001 and r = −0.67 (95% CI − 0.81–−0.47), P < 0.0001, respectively, at a relative rate of 0.37 (SD 0.07) and −0.52 (SD 0.09), respectively, [Table 2 and Figure 1]. PI-LL mismatch correlated with T10-12 angle with r = 0.42 (95% CI 0.14–0.64), P = 0.0043 [Table 2].
Table 2.
Preoperative radiographic measurements
| PI-LL versus T10-L2 | PI-LL versus PT | PI-LL versus TK | PI-LL versus C7SVA | |
|---|---|---|---|---|
|
| ||||
| PI-LL preoperative, whole cohort | ||||
| Pearson r | ||||
| R | 0.4218 | 0.6553 | −0.6747 | 0.2502 |
| 95% CI | 0.1428-0.6387 | 0.442-0.7985 | −0.8095-−0.4725 | −0.05038-0.5093 |
| R2 | 0.1779 | 0.4294 | 0.4553 | 0.06262 |
| P | ||||
| P (two-tailed) | 0.0043 | <0.0001 | <0.0001 | 0.1013 |
| Slope | 0.2597±0.08614 | 0.3677±0.06619 | −0.519±0.08761 | 0.1884±0.1125 |
|
| ||||
| PI-LL preoperative, low PI subgroup | ||||
|
| ||||
| Pearson r | ||||
| R | 0.5957 | 0.2753 | −0.5555 | 0.1116 |
| 95% CI | 0.2208-0.8172 | −0.1774-0.6319 | −0.7962-−0.1628 | −0.3363-0.5183 |
| R2 | 0.3549 | 0.07581 | 0.3085 | 0.01246 |
| P | ||||
| P (two-tailed) | 0.0044 | 0.2270 | 0.0089 | 0.6301 |
| Slope | 0.4194±0.1297 | 0.2687±0.2153 | −0.5575±0.1915 | 0.6216±1.27 |
|
| ||||
| PI-LL preoperative, high PI subgroup | ||||
|
| ||||
| Pearson r | ||||
| R | 0.6808 | 0.6162 | −0.804 | 0.7118 |
| 95% CI | 0.3636-0.8566 | 0.2629-0.8238 | −0.9153-−0.5785 | 0.4146-0.8718 |
| R2 | 0.4635 | 0.3797 | 0.6464 | 0.5067 |
| P | ||||
| P (two-tailed) | 0.0005 | 0.0023 | <0.0001 | 0.0002 |
| Slope | 0.27±0.07715 | 0.4773±0.1148 | −0.7215±0.1193 | 0.6216±1.27 |
Preoperative radiographic measurements in the whole patient cohort, as well as low and high PI subgroups. Correlations between sagittal parameters and PI-LL mismatch are given. CI - Confidence interval, C7SVA - C7 sagittal vertical axis, PI - Pelvic incidence, PI-LL - PI-lumbar lordosis mismatch, PT - Pelvic tilt, T10-L2 - Sagittal Cobb angle T10-L2, TK - Thoracic kyphosis
Figure 1.
Correlations between preoperative PI-LL and perioperative change in PI-LL and sagittal spinopelvic parameters in the whole patient cohort as well as low and high PI subgroups. PI-LL - Pelvic incidence-lumbar lordosis, (a) PI-LL preoperative measures for whole cohort, (b) Change in lumbar lordosis for whole cohort, (c) PI-LL preoperative measures for low PI subgroup, (d) Change in lumbar lordosis preoperative measures for low PI subgroup, (e) PI-LL preoperative measures for high PI subgroup, (f) Change in lumbar lordosis preoperative measures for high PI subgroup
In patients with low PI, only TK showed a significant correlation with PI-LL, r = −0.56 (95% CI − 0.8 to −0.16), at a rate of − 0.57 (SD 0.19). The high PI subgroup showed a significant correlation with PT, TK, and C7SVA, r = 0.62 (95% CI 0.26–0.82), r = −0.8 (95% CI − 0.9–−0.58), and r = 0.71 (95% CI 0.41–0.87) at rates of 0.48 (SD: 0.11), 0.72 (SD: 0.12), and 0.62 (SD: 1.27) [Table 3].
Table 3.
Postoperative change in radiographic measures of lumbar lordosis in whole cohort, low pelvic incidence and high pelvic incidence subgroups
| Delta LL versus delta T10-L2 | Delta PI-LL versus delta PT | Delta PI-LL versus delta TK | Delta PI-LL versus delta C7SVA | |
|---|---|---|---|---|
|
| ||||
| ∆LL, whole cohort | ||||
| Pearson r | ||||
| R | 0.138 | 0.04679 | −0.6334 | |
| 95% CI | −0.4209-0.1693 | −0.3653-0.2815 | 0.3179-0.8227 | |
| R2 | 0.01905 | 0.002189 | 0.4012 | |
| P | ||||
| P (two-tailed) | 0.3774 | 0.7833 | 0.0007 | |
| Slope | −0.1018±0.1141 | −0.03929±0.1418 | 0.8061±0.2053 | |
|
| ||||
| ∆LL, low PI subgroup | ||||
|
| ||||
| Pearson r | ||||
| R | 0.06492 | 0.4821 | −0.4877 | 0.3166 |
| 95% CI | −0.4933-0.3888 | −0.7746-−0.01964 | −0.05785-0.809 | −0.7253-0.2572 |
| R2 | 0.004214 | 0.2324 | 0.2378 | 0.1002 |
| P | ||||
| P (two-tailed) | 0.7857 | 0.0428 | 0.0769 | 0.2701 |
| Slope | −0.04801±0.174 | −0.4455±0.2024 | 0.728±0.3762 | −1.394±1.206 |
|
| ||||
| ∆LL, high PI subgroup | ||||
|
| ||||
| Pearson r | ||||
| R | 0.05297 | 0.1505 | −0.7686 | −0.5404 |
| 95% CI | −0.4841-0.3989 | −0.3559-0.5885 | 0.313-0.9366 | −0.8611-0.08795 |
| R2 | 0.002806 | 0.02264 | 0.5907 | 0.2921 |
| P | ||||
| P (two-tailed) | 0.8245 | 0.5643 | 0.0057 | 0.0861 |
| Slope | −0.04358±0.1936 | 0.09308±0.1579 | 0.8101±0.2248 | −2.283±1.185 |
Correlations between sagittal parameters and∆LL are given. CI - Confidence interval, C7SVA - C7 sagittal vertical axis, PI - Pelvic incidence, LL - Lumbar lordosis, PI-LL - PI-LL mismatch, PT - Pelvic tilt, T10-L2 - Sagittal Cobb angle T10-L2, TK - Thoracic kyphosis
Perioperative change in LL showed significant correlations with TK in the entire cohort, r = +0.63 (95% CI 0.31–0.82) at a rate of + 0.80 (SD: 0.20), with PT and TK (trend toward significance) in the low PI group, r = −0.48 (95% CI −0.77–0.02) and r = +0.49 (95% CI − 0.06–0.8) at a rate of −0.44 (SD 0.20) and + 0.72 (SD 0.38) [Table 4]. In the high PI group, significant correlations were found for TK and C7SVA (trend toward significance), r = 0.76 (95% CI 0.31–0.93) and r = −0.54 (95% CI −0.86–0.09) at rates of 0.81 (SD: 0.22) and −2.3 (SD: 1.2). The ratio of PT: TK change preoperatively was 0.46 (SD: 0.28) in the low PI group and 1.47 (SD: 2.16) in the high PI group.
Table 4.
Correlations between sagittal parameters and preoperative PI-LL mismatch and change in LL postoperatively across groups
| T10-L2 | PT | TK | C7SVA | PT/TK ratio | |
|---|---|---|---|---|---|
| Preoperative PI-LL, whole cohort | +0.26 (±0.09) | +0.37 (±0.06) | −0.52 (±0.09) | - | - |
| Perioperative∆LL, whole cohort | - | - | +0.81 (±0.21) | - | - |
| Preoperative PI-LL, low PI subgroup | +0.42 (±0.13) | 0.27 (±0.22) | −0.56 (±0.19) | - | 0.46 (±0.28) |
| Perioperative∆LL, low PI subgroup | - | −0.45 (±0.2) | +0.73 (±0.38) | - | - |
| Preoperative PI-LL, high PI subgroup | +0.27 (±0.08) | 0.48 (±0.11) | −0.72 (±0.12) | +0.62 (±1.27) | 1.5 (±2.2) |
| Perioperative∆LL, high PI subgroup | - | +0.09 (±0.16) | +0.81 (±0.22) | −2.28 (±1.90) | - |
Values are given as Person correlation coefficients. Blank cells indicate that no statistically significant correlations were found. C7SVA - C7 sagittal vertical axis (mm), PI - Pelvic incidence (°), LL - Lumbar lordosis, PI-LL - PI-lumbar lordosis mismatch (°), PT - Pelvic tilt (°), T10-L2 - Sagittal Cobb angle T10-L2 (°), TK - Thoracic kyphosis (°)
DISCUSSION
The results of the present study demonstrate that the degree of PI may influence the compensation patterns in patients with ASD and significant PI-LL mismatch. It was further shown that the overall reduction of TK was more consistently normalized after surgical deformity correction, while PT was not. Anticipating perioperative changes in PT and TK are important to consider in surgical planning.
Distinct spinal deformities can activate distinct compensation patterns. Lumbar hypolordosis is associated with an increase in PT and a decrease in TK.[2,5] Like the results of our study, Diebo et al. found a linear correlation between increases in PI-LL mismatch and both PT increase and TK decrease.[10] While reduction of TK is more frequently observed in younger patients (often under the age of 45) with a mobile, nonrigid spine, it requires contraction of the erector spinae muscle, as shown in Figure 2,[5] which has been linked to increased thoracic muscle pain over the course of the day due to muscle fatigue.[11]
Figure 2.
Mechanisms involved in reduction of TK (inlet a): Erector spinae activation, and increase in PT (inlet b): Rectus abdominis and gluteal muscle activation. PT - Pelvic tilt, TK - Thoracic kyphosis
Previous studies have shown that compensatory adjustments in PT and TK maintain a neutral C7SVA up to a PI-LL mismatch of 30°.[10,12] Since PI-LL mismatch in our study cohort was between 10° and 22.8°, we expected to observe PT and TK modifications in response to ASD. As evident in Figure 2, abdominal, gluteal, and ischiocrural muscles primarily accomplish PT increase. Eventually, an increased PT maintains the sagittal vertical axis centered in the context of increasing PI-LL mismatch. However, it also correlated with clinical disability scores since it requires muscle activation.[13] As has been shown in previous studies, a compensatory increase in PT can normalize after surgical correction of spinal deformity.[12] Our results are congruent in showing a decrease in PT in the high PI group and suggested a more reliable increase of TK across high and low PI subgroups. Since the axis of rotation for PT regulations is the hip joint, osteoarthritis or dynamic hip pathology may confound findings and limit the body's capability to control PT.[14] Further, PT is influenced by hip flexors and back extensors, as well as abdominal wall muscles and gluteal muscles. Indeed, atrophy of the thoracolumbar extensor muscles produces global kyphosis.[15]
While PT did not change postoperatively, there was marked variation in the range from −26° to 28°. Failure of perioperative PT to consistently correlate with change in PI-LL might be explained by the more complex mechanism of PT regulation, which is potentially influenced by hip joint pathologies. In our study, those patients with low PI relied more on TK reduction than changes in PT. Similarly, in a different study of 34 patients with ADS who underwent correction of lumbar deformity, a compensatory increase in PT on average only reduced by 8° while TK increased by 13°, while no change was observed in 11 of 34 of these patients.[16]
Patients with low PI have reduced capabilities of decreasing sacral slope and pelvic retroversion as seen in those with high PI.[17] Differences in PI compensation patterns between low and high PI subgroups in this study may underlie preoperative PI values. Baseline characteristics may be evaluated to predict when patients may present with a disability, and if earlier intervention may be warranted for those with high PI and exaggerated compensation of TK and PT. This subpopulation may benefit from a greater correction in both parameters postoperatively.
A limitation of this study is the lack of a standardized postoperative physical therapy regimen, which might have had an impact on PT regulation after surgery.
Optimizing the outcome of surgical correction in ASD might require an in-depth understanding of compensation mechanisms and to predict the future changes of these mechanisms. Individual pelvic morphology as well as hip joint assessment should be considered when investigating the benefits of spinal deformity correction. Future studies may examine whether these radiographic and morphologic changes relate to meaningful clinical differences in function and pain scores.
CONCLUSIONS
A substantial decrease in TK represented a more consistent compensatory mechanism in patients with high and low PI when compared to an increase in PT. PI-LL mismatch induced more pronounced changes in TK than did PT in all groups. Postoperatively, the reversal of compensatory changes was more consistent for TK than for PT. Patients with high PI relied more on increases in PT and a relative decrease in TK to compensate for PI-LL mismatch than patients with low PI. Predicting postoperative changes of TK and PT might be important to determine mid- and long-term spinal alignment after spinal deformity correction.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Dreischarf M, Albiol L, Rohlmann A, Pries E, Bashkuev M, Zander T, et al. Age-related loss of lumbar spinal lordosis and mobility - A study of 323 asymptomatic volunteers. PLoS One. 2014;9:e116186. doi: 10.1371/journal.pone.0116186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barrey C, Roussouly P, Perrin G, Le Huec JC. Sagittal balance disorders in severe degenerative spine. Can we identify the compensatory mechanisms? Eur Spine J. 2011;20(Suppl 5):626–33. doi: 10.1007/s00586-011-1930-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Glassman SD, Berven S, Bridwell K, Horton W, Dimar JR. Correlation of radiographic parameters and clinical symptoms in adult scoliosis. Spine (Phila Pa 1976) 2005;30:682–8. doi: 10.1097/01.brs.0000155425.04536.f7. [DOI] [PubMed] [Google Scholar]
- 4.Dubousset J, Wicart P, Pomero V, Barois A, Estournet B. Spinal penetration index: New three-dimensional quantified reference for lordoscoliosis and other spinal deformities. J Orthop Sci. 2003;8:41–9. doi: 10.1007/s007760300007. [DOI] [PubMed] [Google Scholar]
- 5.Lamartina C, Berjano P. Classification of sagittal imbalance based on spinal alignment and compensatory mechanisms. Eur Spine J. 2014;23:1177–89. doi: 10.1007/s00586-014-3227-9. [DOI] [PubMed] [Google Scholar]
- 6.Roussouly P, Pinheiro-Franco JL. Sagittal parameters of the spine: Biomechanical approach. Eur Spine J. 2011;20(Suppl 5):578–85. doi: 10.1007/s00586-011-1924-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S, Schwab F. The impact of positive sagittal balance in adult spinal deformity. Spine (Phila Pa 1976) 2005;30:2024–9. doi: 10.1097/01.brs.0000179086.30449.96. [DOI] [PubMed] [Google Scholar]
- 8.Kuntz C, 4th, Levin LS, Ondra SL, Shaffrey CI, Morgan CJ. Neutral upright sagittal spinal alignment from the occiput to the pelvis in asymptomatic adults: A review and resynthesis of the literature. J Neurosurg Spine. 2007;6:104–12. doi: 10.3171/spi.2007.6.2.104. [DOI] [PubMed] [Google Scholar]
- 9.Rothenfluh DA, Mueller DA, Rothenfluh E, Min K. Pelvic incidence-lumbar lordosis mismatch predisposes to adjacent segment disease after lumbar spinal fusion. Eur Spine J. 2015;24:1251–8. doi: 10.1007/s00586-014-3454-0. [DOI] [PubMed] [Google Scholar]
- 10.Diebo BG, Ferrero E, Lafage R, Challier V, Liabaud B, Liu S, et al. Recruitment of compensatory mechanisms in sagittal spinal malalignment is age and regional deformity dependent: A full-standing axis analysis of key radiographical parameters. Spine (Phila Pa 1976) 2015;40:642–9. doi: 10.1097/BRS.0000000000000844. [DOI] [PubMed] [Google Scholar]
- 11.Yang C, Yang M, Wei X, Shao J, Chen Y, Zhao J, et al. Lumbar Lordosis Minus Thoracic kyphosis: A novel regional predictor for sagittal balance in elderly populations. Spine (Phila Pa 1976) 2016;41:399–403. doi: 10.1097/BRS.0000000000001231. [DOI] [PubMed] [Google Scholar]
- 12.Merrill RK, Kim JS, Leven DM, Kim JH, Cho SK. Beyond pelvic incidence-lumbar lordosis mismatch: The importance of assessing the entire spine to achieve global sagittal alignment. Global Spine J. 2017;7:536–42. doi: 10.1177/2192568217699405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schwab FJ, Blondel B, Bess S, Hostin R, Shaffrey CI, Smith JS, et al. Radiographical spinopelvic parameters and disability in the setting of adult spinal deformity: A prospective multicenter analysis. Spine (Phila Pa 1976) 2013;38:E803–12. doi: 10.1097/BRS.0b013e318292b7b9. [DOI] [PubMed] [Google Scholar]
- 14.Buckland AJ, Steinmetz L, Zhou P, Vasquez-Montes D, Kingery M, Stekas ND, et al. Spinopelvic compensatory mechanisms for reduced hip motion (ROM) in the setting of hip osteoarthritis. Spine Deform. 2019;7:923–8. doi: 10.1016/j.jspd.2019.03.007. [DOI] [PubMed] [Google Scholar]
- 15.Lee SH, Kim KT, Suk KS, Lee JH, Seo EM, Huh DS. Sagittal decompensation after corrective osteotomy for lumbar degenerative kyphosis: Classification and risk factors. Spine (Phila Pa 1976) 2011;36:E538–44. doi: 10.1097/BRS.0b013e3181f45a17. [DOI] [PubMed] [Google Scholar]
- 16.Lafage V, Ames C, Schwab F, Klineberg E, Akbarnia B, Smith J, et al. Changes in thoracic kyphosis negatively impact sagittal alignment after lumbar pedicle subtraction osteotomy: A comprehensive radiographic analysis. Spine (Phila Pa 1976) 2012;37:E180–7. doi: 10.1097/BRS.0b013e318225b926. [DOI] [PubMed] [Google Scholar]
- 17.Roussouly P, Nnadi C. Sagittal plane deformity: An overview of interpretation and management. Eur Spine J. 2010;19:1824–36. doi: 10.1007/s00586-010-1476-9. [DOI] [PMC free article] [PubMed] [Google Scholar]