Coronal imbalance is one of the major findings in patients with adult spinal deformity (ASD),1 it plays an important role in decreased health related quality of life.2–4 Depending on the definition used, the prevalence of coronal imbalance in ASD varied from 19.3% to 34.8%.5,6 Although coronal imbalance in ASD is common and has become one of the focuses of important health problems, the mechanism of coronal imbalance is still not clear. It was supposed that coronal imbalance might be caused by postural compensation from back pain associated with spinal stenosis, degeneration of discs and facet joints, and atrophy of paraspinal muscles.6 Nevertheless, there is no solid evidence to support this hypothesis. Recently, it has been reported that the ability to level the coronal tilt of L4 and L5 had great impact on the ability to achieve coronal balance,7 but the relationship between global coronal alignment and coronal tilt of L4 or L5 is not clear.
On the contrary, although coronal imbalance has been paid more and more attention to, there is only one widely used global parameter to reflect coronal alignment, which is the measurement of distance between C7 plumb line and central sacral vertical line (CSVL) despite a wide range of names have been used for this parameter such as coronal balance difference,6 coronal balance,7 coronal malalignment,8 global coronal malalignment.9 we name it as C7 migration. As a distance parameter, C7 migration has its inherent defects, such as being easily influenced by patient's size, which may affect the accuracy of this parameter to reflect global coronal alignment. For example, a small-sized patient may have a small range of normality of this distance parameter. To overcome the defects of distance parameter, we used an angular index—coronal T1 pelvic tilt (CTPT)—to reflect global coronal alignment. CTPT is defined as the angle between a vertical line and the line connecting the middle point of S1 endplate to T1 centroid (Figure 1A). The purpose of this study is to verify the validity of this new index for investigating the relationship between global coronal alignment and regional coronal parameters such as coronal tilt of L4 or L5.
MATERIALS AND METHODS
Approval by the Clinical Research Ethics Committee of our hospital was obtained before this study. We retrospectively collected data of ASD patients with preoperative standing full-spine x-rays in one single institution between January 2016 and May 2019. Inclusion criteria included: patient of either sex, age >18 years. Exclusion criteria included: previous spinal surgery, previous hip or knee surgery, congenital deformity, post-traumatic deformity, neuromuscular disease, spinal tumor, Pott deformity, pelvic deformity, absolute discrepancy of leg length >20 mm. Eventually, 146 patients were involved in this study. The mean age of patients was 53.6 years (range: 19–78 years), with 28 males and 118 females.
Full-spine standing coronal view of radiographs was analyzed. Each parameter of coronal radiographic measurements was defined as below (Figure 1A, B): C7 migration, defined as the horizontal distance between the C7 centroid and CSVL, C7 centroid on the left side of CSVL was recorded as negative (−), right side as positive (+);CTPT, defined as the angle between a vertical line and a line connecting middle point of S1 endplate and T1 centroid. Left angle was recorded as negative (−), right angle as positive (+);major Cobb angle, defined as the angle between the superior endplate of the most tilted vertebra cranially and the inferior endplate of the most tilted vertebra caudally (left scoliosis was recorded as negative [−], right scoliosis as positive [+]);lumbosacral fractional curve, defined as the angle between the superior endplate of L4 and the line formed by pedicles of S1 (left fractional curve was recorded as negative [−], right fractional curve as positive [+]);L4 coronal tilt, defined as the angle between superior endplate of L4 and the horizontal line9 (left tilt was recorded as negative [−], right tilt as positive [+]);L5 coronal tilt, defined as the angle between superior endplate of L5 and the horizontal line9 (left tilt was recorded as negative [−], right tilt as positive [+]).
The radiographic measurements were done using Surgimap (version 2.2.15; Spine Software, New York, NY) by two independent researchers. Each parameter was measured twice by each spine surgeon, and the intraclass correlation coefficients were analyzed using the mean values measured by each spine surgeon. Intra- and inter-rater reliabilities were excellent with kappa values ranging from 0.807 to 0.916. Pearson correlation coefficient (r) was calculated to test for associations between C7 migration, CTPT and regional coronal parameters. After the initial correlation analysis, variables with a value of P < 0.05 were further investigated with stepwise multiple regression analysis and the coefficient of determination (R2) was calculated. The statistical analysis was performed using SPSS computer software (version 24; SPSS, Chicago, IL). A value of P < 0.05 was considered as statistically significant.
The mean CTPT angle was 0.43° ± 3.4° (from −8.7° to 12.5°), the mean C7 migration was 3.3 ± 24.6 mm (from −61.0 to 70.2 mm), and the mean L4 coronal tilt was −3.4° ± 13.5° (from −27.8° to 38.5°). The mean L5 coronal tilt was −2.6° ± 9.0° (from −21.9° to 32.5°). The means of other parameters are shown in Table 1.
Relationship Between CTPT, C7 Migration, L4, and Other Parameters
Pearson correlation analysis showed that CTPT was significantly correlated with C7 migration (r = 0.993, P < 0.001). For each regional parameter, CTPT was significantly correlated with fractional curve (r = −0.344, P < 0.001), L4 coronal tilt (r = 0.339, P < 0.001), and L5 coronal tilt (r = 0.234, P = 0.004), but not with major Cobb angle (r = 0.092, P = 0.267). Similarly, C7 migration significantly correlated with fractional curve (r = −0.346, P < 0.001), L4 coronal tilt (r = 0.342, P < 0.001), and L5 coronal tilt (r = 0.235, P = 0.004), but not with major Cobb angle (r = 0.090, P = 0.281) (Table 2).
Additionally, lumbosacral fractional curve was significantly associated with L4 coronal tilt (r = −0.994, P < 0.001), L5 coronal tilt (r = −0.877, P < 0.001) and major Cobb angle (r = −0.702, P < 0.001) (Table 2).
The results of the correlation analysis showed that L4 coronal tilt, L5 coronal tilt, fractional curve and CTPT values had significantly associations with the C7 migration value (Table 2; P < 0.05). Entry of these four variables into stepwise multiple regression analysis revealed that only CTPT was significant independent predictor of C7 migration. The coefficient of multiple determination (R2) of CTPT was 0.985, which means 98.5% of the variance in the C7 migration that CTPT can explain; similarly, four variables including C7 migration, L4 coronal tilt, L5 coronal tilt, fractional curve shown in Table 2 were significantly associated with CTPT, only C7 migration was independent predictor of CTPT after stepwise multiple regression analysis (R2 = 0.985), and the following linear regression equation was obtained (Figure 2):
On the basis of this regression equation, the coronal imbalance threshold for CTPT was found to be 2.8° or 4.2° when 20 or 30 mm of C7 migration was set as threshold of coronal imbalance, respectively.
Additionally, entry of fractional curve-associated three variables including L4 coronal tilt, L5 coronal tilt, and major Cobb angle into stepwise multiple regression analysis revealed that only L4 coronal tilt was an independent predictor of lumbosacral fractional curve (R2 = 0.988).
This study showed that CTPT was highly correlated with C7 migration, and had a similar relationship with coronal tilt of L4 or L5, fractional curve, and major Cobb angle. Moreover, neither L4 coronal tilt nor L5 coronal tilt was an independent predictor of C7 migration.
As a widely used distance parameter to reflect global coronal alignment, C7 migration has its inherent defects: calibration is needed before the measurement, which not only makes this parameter become more complicated to measure, but also might be a potential source of measurement error; C7 migration might be affected by a patient's body size. To overcome these defects, we introduced a novel global coronal tilt angle—CTPT, and focused on CTPT as a simple and practical index for evaluating global coronal alignment. CTPT is an angle formed by a vertical line and the line connecting T1 centroid and middle point of S1. we recommended CSVL as the vertical line due to several reasons: CSVL is recommended by Scoliosis Research Society 3-dimensional classification committee to use in two-dimensional radiographs, although CSVL is not as reproducible as central hip vertical axis in 3-dimensional evaluation of spinal deformity10; CSVL is one of the lines that spine surgeons are most familiar with and widely practice in spinal deformity; CSVL is the same line used in the measurement of C7 migration; CSVL is a vertical line drawn upward from the middle point of S1, which is just one of two points that makes up the other line. Those reasons mentioned above make this angle very convenient and efficient to measure.
As an angular parameter, CTPT had several advantages when compared to C7 migration. First, calibration is not needed before the measurement, which not only makes the measurement simpler but also removes a potential source of measurement error. Second, CTPT is not affected by a patient's body size. Lastly, CTPT is not affected by C7 segment, which could more accurately reflect the truncal coronal alignment. In a normal healthy standing person, the ideal value of CTPT would be 0°, which is the same ideal value as C7 migration (0 mm), these features make CTPT easier to use.
In the present study, we determined the relationship between CTPT, C7 migration, and regional coronal parameters. According to measurements of coronal parameters, strong correlation between CTPT and C7 migration was demonstrated. In addition, the linear regression equation between CTPT and C7 migration was established: CTPT (°) = −0.02 + 0.14∗ C7 migration (mm). Based on this regression equation, the threshold of coronal imbalance for CTPT could be obtained once the threshold for C7 migration value is determined. Different values of C7 migration were set as threshold for coronal imbalance such as 20 mm,11 30 mm,6 or 40 mm.5 When >20 mm of C7 migration was set as coronal imbalance, the value of coronal imbalance for CTPT would be >2.8°. Accordingly, the threshold for CTPT would be 4.2° or 5.6° when 30 or 40 mm of C7 migration was set as threshold of coronal imbalance, respectively. We used 20 mm of C7 migration as strict threshold of coronal imbalance (CTPT 2.8°). For the convenience's purpose, 3° of CTPT might be used as threshold of coronal imbalance. But further clinical research is needed to validate it.
Our present study also showed C7 migration had significant correlation with L4 coronal tilt, L5 coronal tilt, or fractional curve, but not with major Cobb angle. CTPT had a similar relationship with L4 coronal tilt, fractional curve, L5 coronal tilt and major Cobb angle. Therefore, we believe CTPT could be used as a simple and practical means of evaluating global coronal alignment.
The mechanism of coronal imbalance is still not clear. Fractional curve was thought to be critical in the correction of coronal imbalance. Campbell et al12 even believed that the patient is very unlikely to achieve appropriate coronal balance if a present fractional curve was not addressed at the time of index surgery. However, regarding its relationship with global coronal alignment, the present study showed fractional curve was significantly associated with C7 migration or CTPT (Table 2, P < 0.05), but not an independent predictor of C7 migration or CTPT after multiple regression analysis. In other words, correction of fractional curve might be helpful but not determinant to achieve coronal balance.
Lewis et al7 suggested horizontalization of L4 and L5 vertebrae is critical to achieve coronal balance in ASD patients fused to the pelvis. However, the present study showed C7 migration had significant correlation with L4 coronal tilt or L5 coronal tilt, but either L4 coronal tilt or L5 coronal tilt was not an independent predictor for C7 migration after multiple regression analysis. Thus, neither L4 coronal tilt nor L5 coronal tilt was an independent predictor for C7 migration. In other words, correction of L4 coronal tilt would provide a horizontal base in the caudal part of lumbar spine and help correct the coronal imbalance, but not assure the correction of coronal imbalance.
Additionally, in terms of addressing lumbosacral fractional curve, Bao et al9 emphasized on the horizontalization of L5 as the foundation of coronal spinal alignment. However, regarding the relationship between fractional curve and L4 coronal tilt, L5 coronal tilt, the results of present study showed fractional curve was significantly associated with L4 coronal tilt, L5 coronal tilt and major Cobb angle, but only L4 coronal tilt was an independent predictor of fractional curve. That is, correction of L4 coronal tilt would predict the correction of fractional curve. Since it is L4 coronal tilt, not L5 coronal tilt that predicts the fraction curve, it is reasonable to coronally level the caudal part of lumbar spine until L4, not only coronally level until L5. But further longitudinal clinical studies are needed to verify it.
Several limitations with this study must be mentioned, one of which is the small sample size. Furthermore, the relationship between CTPT and clinical findings, functional scores such as SRS-22 and ODI was not considered in this study. Future research is needed to establish the relationships between CTPT, clinical symptoms, and functional scores in ASD patients. We believe the results of this study can be used as a foundation for that purpose. Despite the above-mentioned limitations, we demonstrated CTPT could be a simple and practical index for evaluating global coronal alignment. None of the coronal tilt of L4/L5, fractional curve, and major Cobb angle was an independent predictor of global coronal alignment.
CTPT strongly correlated with C7 migration, and had a similar relationship with L4 coronal tilt, L5 coronal tilt, fractional curve and major Cobb angle. Therefore, CTPT could be a simple and practical index of evaluating global coronal alignment. Furthermore, neither L4 coronal tilt nor L5 coronal tilt was an independent predictor of global coronal alignment.
- C7 migration had significant correlation with L4 coronal tilt and L5 coronal tilt, but neither L4 coronal tilt nor L5 coronal tilt was an independent predictor of global coronal alignment.
- Coronal T1 pelvic tilt was highly associated with C7 migration, and had a similar relationship with coronal tilt of L4 or L5, fractional curve and major Cobb angle.
- Coronal T1 pelvic tilt could be a practical index for evaluating global coronal alignment.
1. Bradford DS, Tay BKB, Hu SS. Adult scoliosis: surgical indications, operative management, complications, and outcomes. Spine (Phila Pa 1976)
2. Ploumis A, Liu H, Mehbod AA, et al. A correlation of radiographic and functional measurements in adult degenerative scoliosis. Spine (Phila Pa 1976)
3. Koller H, Pfanz C, Meier O, et al. Factors influencing radiographic and clinical outcomes in adult scoliosis surgery: a study of 448 European patients. Eur Spine J
4. Acaroglu E, Guler UO, Olgun ZD, et al. Multiple regression analysis of factors affecting health related quality of life in adult spinal deformity. Spine Deform
5. Ploumis A, Simpson AK, Cha TD, et al. Coronal spinal balance in adult spine deformity patients with long spinal fusions: a minimum 2- to 5-year follow-up study. J Spinal Disord Tech
6. Bao H, Yan P, Qiu Y, et al. Coronal imbalance
in degenerative lumbar scoliosis: prevalence and influence on surgical decision making for spinal osteotomy. Bone Joint J
7. Lewis SJ, Keshen SG, Kato S, et al. Risk factors for postoperative coronal balance in adult spinal deformity surgery. Global Spine J
8. Kurra S, Metkar U, Yirenkyi H, et al. Assessment of coronal spinal alignment for adult spine deformity cases after intraoperative T square shaped use. Spine Deform
9. Bao H, Liu Z, Zhang Y, et al. sequential correction technique to avoid postoperative global coronal decompensation in rigid adult spinal deformity: a technical note and preliminary results. Eur Spine J
10. Sangole A, Aubin CE, Labelle H, et al. The central hip vertical axis: a reference axis for the scoliosis research society three-dimensional classification of idiopathic scoliosis. Spine (Phila Pa 1976)
11. Ibrahim O, Pedro B, Claudio L, et al. Classification of coronal imbalance
in adult scoliosis and spine deformity: a treatment-oriented guideline. Eur Spine J
12. Campbell PG, Nunley PD. The challenge of the lumbosacral fractional curve
in the setting of adult degenerative scoliosis. Neurosurg Clin N Am