Proximal Thoracic Alignment Change Influences Cervical Sagittal Alignment After Correction Surgery in Patients With Lenke Type 2 Adolescent Idiopathic Scoliosis
Article information
Abstract
Objective
To examine the factors influencing cervical sagittal alignment (CSA) after posterior correction and fusion surgery (PSF) for patients with Lenke type 2 adolescent idiopathic scoliosis (AIS).
Methods
A total of 102 female patients with Lenke 2 AIS and a minimum 2-year follow-up were included. The upper instrumented vertebra was T2 in all patients. Sagittal and coronal parameters were measured before and 2 years after surgery. Patients were categorized into cervical malalignment (CM) and noncervical malalignment (NCM) groups following Passias’ criteria. Radiographic factors influencing CSA were analyzed.
Results
Preoperatively, 57 patients (55.9%) were assigned to the CM group and 45 patients (44.1%) to the NCM groups. The cervical lordosis (CL) in CM group was more kyphotic (19.3° vs. 3.3°), smaller proximal thoracic kyphosis (PTK; 9.7° vs. 15.4°), and smaller T1 slope (7.1° vs. 14.0°) than those in the NCM group. Main thoracic kyphosis (MTK) did not show significantly difference between the 2 groups (11.3° vs. 14.4°). Two years after surgery, the CM group demonstrated significant improvements in CSA. PTK increased from 9.7° to 13.5°, T1 slope increased from 7.1° to 10.5°, and cervical kyphosis improved from -19.3° to -8.8°, while MTK remained unchanged (11.3° vs. 11.6°).
Conclusion
PSF significantly improved CSA in patients with preoperative CM. Increased PTK, correlated with improved CL, suggests that PSF-induced PTK enhancement, rather than MTK, drives T1 slope and subsequent CSA improvement.
INTRODUCTION
Adolescent idiopathic scoliosis (AIS) is characterized by a three-dimensional deformity involving vertebral rotation and lateral coronal curvature, often accompanied by flattening of sagittal profiles [1]. Since a severely progressed deformity can compromise mental, physical, and pulmonary conditions, correction surgery has been indicated. The principal aim of surgery for patients with AIS is to correct the coronal deformity and achieve ideal sagittal profiles while maintaining or achieving optimal coronal and sagittal balance, which may positively impact the patient’s overall musculoskeletal function and quality of life [2].
Over the past few decades, research on AIS has mainly focused on finding an optimal method to correct the coronal deformity. The introduction of pedicle screw construct has revolutionized correction methods by providing a larger corrective force and enhancing coronal correction and axial de-rotation [3]. However, radiological studies of patients with AIS who underwent posterior correction and fusion surgery (PSF) using a pedicle screw construct revealed loss of normal thoracic kyphosis (TK), which may lead to the development of cervical malalignment (CM) [4,5]. Youn et al. [2] demonstrated significant correlations between the postoperative cervical radiographic parameters and health-related quality of life in patients with AIS. Additionally, a separate study analyzing 191 consecutive patients with cervical spondylotic myelopathy revealed a strong association between cervical kyphosis and severe myelopathy symptoms [6,7]. These findings indicate that restoring satisfactory sagittal alignment may be crucial for patients with AIS to prevent the onset of further degenerative disorders during adulthood [8,9].
More recently, emphasis on understanding the surgical impact of postoperative cervical sagittal alignment (CSA) in AIS has been growing [2,10]. A significant association was reported between decreased cervical lordosis (CL) and decreased TK in patients with Lenke 3C and 6C AIS [11]. Several studies have reported a relationship between postoperative CSA and thoracic sagittal alignment in patients with Lenke types 1, 5, or 6 AIS [2,12-14] who underwent PSF [2,12-14]. However, few studies have exclusively investigated the association between CSA and thoracic sagittal alignment in patients with Lenke type 2 AIS who underwent PSF.
The objective of this study was to evaluate the relationship between changes in CSA and TK, and to determine the factors that influence CSA after PSF for Lenke type 2 AIS.
MATERIALS AND METHODS
Between 2007 and 2021, our institution performed surgery in 113 patients diagnosed with Lenke type 2 AIS. The inclusion criteria were as follows: diagnosis of Lenke type 2 AIS; age younger than 20 years at the time of surgery; no trauma, congenital vertebral deformity, neuromuscular disorders, or other pathological conditions; no previous spine surgery; with a minimum of 2 years of follow-up after surgery. Among these patients, 11 were excluded (1 was male; 9 had follow-up periods <2 years, 1 patient had unsuitable preoperative image). Finally, 102 patients with Lenke type 2 AIS who underwent PSF with pedicle screw construct were included in the analysis. The demographic and clinical characteristics of the patients are presented in Table 1. All demographic information was retrospectively obtained by reviewing patient records and radiological images.
Demographic and radiographical characteristics of patients (n=102)
CM was defined if a patient met at least one of the following criteria as defined by Passias et al. [15]: T1 slope minus C2–7 CL (T1–CL) >20°; C2–7 sagittal vertical axis (SVA) >40 mm; C2–7 CL <10°. Based on the CSA before surgery and at the 2-year postoperative follow-up, all patients were divided into CM and non-CM (NCM) groups, and statistically relevant factors were analyzed.
This study was approved by the Committee on Ethics and Institutional Review Board of Keio University Hospital (approval number: 20090042), and all subjects gave informed consent for inclusion before treatment.
1. Surgical Technique
All patients underwent posterior surgery utilizing using similar techniques for correcting deformities [16,17]. In all patients, the upper instrumented vertebra (UIV) was T2, and the lower instrumented vertebra (LIV) was the last touching vertebra [18]. Pedicle screws were placed segmentally on both sides using a freehand technique with the ball-tip method [19]. On the concave side of the proximal thoracic curve, where diameter of pedicles were narrowed, typically at T3 and T4, sublaminar tapes were used instead of pedicle screws [20]. Ponte osteotomies were performed at the proximal thoracic curve to achieve maximum correction and restoration of the proximal TK (PTK), and a resection of inferior articular process was performed at every level in the fusion area. Corrective maneuvers involved rotating the rod on the concave side of the main thoracic (MT) region, followed by in situ rod bending. Furthermore, a differential rod-contouring technique (hypokyphotic-bent rod) was implemented using a cantilever technique on the convex side to assist derotation. Segmental compression and distraction techniques were performed to horizontalize UIV and LIV.
2. Measurements
All patients in this study had long-cassette standing whole-spine radiographs preoperatively and at 2-year postoperative follow-up. To obtain a standard lateral view, the patients stood looking straight ahead, feet shoulder-width apart, elbows bent, knuckles in the supraclavicular fossa bilaterally, and knees locked [12,21]. We measured the Cobb angles of the proximal thoracic (PT Cobb) and MT curves. CL is typically measured using the Cobb angles from C2 to C7, with lordosis was defined as a positive value and kyphosis defined as a negative value. The C7 SVA was measured as the horizontal distance between the C7 plumb line and the superior-posterior corner of the S1 vertebra and was defined as a positive value when the C7 plumb line was located anteriorly to the superior-posterior corner of the S1 vertebra, and vice versa. The C2–7 SVA was measured as the distance between the C2 and C7 plumb lines. The C2 slope, T1 slope, C1–2 angle, PTK angle for the T1–5 angle (PTK [T1–5]), main TK (MTK) for the T5–12 angle (MTK [T5–12]), T10–L2 kyphosis, lumbar lordosis (LL) for the T12–S1 angle, sacral slope (SS), pelvic tilt (PT), and pelvic incidence (PI) were measured, and positive values define angles that open posteriorly, whereas negative values define angles that open anteriorly in the sagittal plane. Measurements for all patients were conducted by one of the investigators with 10 years of experience in the field of spine.
3. Clinical Outcome
The Scoliosis Research Society (SRS)-22 questionnaire [22] was used to evaluate clinical outcomes between the NCM and CM group at the final follow-up.
4. Statistical Analysis
Means±standard deviations were used to describe continuous variables and frequencies, and percentages were used to summarize categorical variables. Continuous variables with a normal distribution were compared between groups using an independent samples t-test and paired t-test in the same group. For variables that did not meet the assumption of normality, the Mann-Whitney U-test was employed. Categorical variables were summarized as frequencies (n) and analyzed using the chi-square test or Fisher exact test as appropriate. Effect sizes were calculated using Cohen d. Following Cohen (1988, statistical power analysis for the behavioral sciences, 2nd ed), effect sizes were categorized as follows: negligible (d<0.2), small (0.2≤d<0.5), moderate (0.5≤d<0.8), and large (d≥0.8).
To identify predictors of postoperative transition from the CM group to the NCM group (CM-NCM), univariate analyses were performed. Variables with p-values <0.15 in univariate analyses were subsequently entered into a multivariate analysis. The Pearson coefficients for correlations between the CL and other sagittal alignment parameters were calculated. All statistical analyses were performed using IBM SPSS Statistics ver. 22.0 (IBM Co., Armonk, NY, USA). A p-value of <0.05 was considered statistically significant.
RESULTS
1. Proportion of CM
Before surgery, 57 patients (55.9%) were CM and 45 patients (44.1%) were NCM. At postoperative 2 years, number of CM patients significantly decreased to 40 patients (39.2%).
2. Comparison of the Radiographic Parameters Between CM and NCM Groups
We compared the preoperative and 2-year postoperative radiographic parameters between the CM and NCM groups (Table 2). The coronal parameters, including the mean PT Cobb angle and MT Cobb angle, were similar between the 2 groups both preoperatively and the 2-year postoperative follow-up.
Comparison of preoperative and 2-year postoperative radiographic parameters between the preoperative CM and NCM groups
The mean preoperative cervical alignment in the CM group was significantly kyphotic than that in the NCM group (CL; -19.3°±8.7° vs. 3.3°±8.1°, p<0.001). The mean C2 slope was significantly larger (23.8°±8.8° vs. 12.2°±5.5°, p<0.001), the mean C1–2 angle was smaller (-36.1°±7.8° vs. -31.0°±8.2°, p=0.002), and the mean C7 SVA was smaller (-18.2±23.0 mm vs. -8.1±21.0 mm, p=0.024) in the CM group than that in the NCM group. At the 2-year postoperatively, although the mean C2 slope and C2–7 lordosis significantly improved in the CM group, the mean values remained significantly different between the 2 groups. Meanwhile, the difference in mean C7 SVA was no longer significant.
Regarding the preoperative thoracic sagittal parameters, the mean preoperative T1 slope and PTK in the CM group were significantly smaller (T1 slope: 7.1°±8.0° vs. 14.0°±6.3°, p<0.001; PTK: 9.7°±6.8° vs. 15.4°±6.6°, p<0.001) compared to those in the NCM group. Meanwhile, the mean values for MTK were similar between the 2 groups. At 2 years postoperatively, the mean T1 slope and PTK in the CM group significantly increased (T1 slope: preoperative 7.1°±8.0°, 2-year postoperative 10.5°±6.7°, p<0.001; PTK: preoperative 9.7°±6.8°, 2-year postoperative 13.5°±6.9°, p<0.001). In contrast, no significant changes were observed between preoperative and postoperative measurements in the NCM group. Although the mean T1 slope in the CM group remained significantly smaller than that of the NCM group, the difference in mean PTK was no longer significant.
Regarding lumbopelvic parameters, no significant differences were observed between the CM group and the NCM group both preoperatively and 2-years postoperatively. Additionally, no significant changes were observed between the preoperative and 2-years postoperative measurements in any of the parameters within each group.
3. Factors that Influence Postoperative Remaining and Improving Cervical Alignment
At 2-year postoperatively, among 57 patients in the CM group, 31 patients (54.4%) improved to NCM (CM-NCM group), whereas 26 patients had remained as CM (CM-CM group). To identify the factors that influence the improvement of cervical alignment after surgery, we compared the preoperative sagittal parameters between the CM-CM and CM-NCM groups.
Regarding the coronal parameter, no statistically significant differences were observed between the CM-CM and CM-NCM groups (Table 3).
Comparison of radiographic parameters: CM-CM vs. CM-NCM
Regarding thoracic sagittal parameters, although the preoperative mean T1 slope, PTK, and MTK were similar between the CM-CM and CM-NCM groups, the mean T1 slope and PTK were significantly larger in the CM-NCM group at the 2-year postoperative follow-up. The mean PTK increased from 8.8°± 5.8° to 15.3°±5.6° (p<0.001), and the T1 slope increased from 8.1°±8.3° to 12.4°±6.2° (p<0.001). The mean C2–7 lordosis improved from -17.7°±7.9° to -0.3°±5.5° (p<0.001); correspondingly, the C2 slope decreased from 23.8°±8.6° to 13.7°±7.9° (p<0.001), the C1–2 angle increased from -36.1°±7.4° to -33.3°± 6.6° (p=0.027), and the C2–7 SVA decreased from 16.7±11.1 mm to 9.8±16.9 mm (p=0.027). However, there was no significant differences between the 2 groups in MTK (12.4°±6.5° vs. 12.1°±4.8°, p=0.791).
Univariate analyses were performed to identify potential predictors of postoperative transition from the CM group to the NCM group (Table 4). Variables with p-values <0.15 in univariate analyses were subsequently entered into a multivariate analysis. The multivariate analysis revealed that 2-year postoperative C2 slope was the only independent predictor of persistent CM following surgery. Preoperative C2–7 lordosis, 2-year postoperative C2–7 SVA, C7SVA, T1 slope, and PTK were not found to be significant predictors in the multivariate model.
Multivariate analysis of predictors for preoperative CM to postoperative NCM
4. Parameters Relating to Change in CL
To investigate the factors influencing changes in CL, we used univariate correlation analysis (Table 5). Changes in CL was significantly correlated with changes in C2 slope, T1 slope, PTK, MTK, and C7 SVA. Among them C2 slope, T1 slope, and PTK were strongly correlated (PTK, r=0.655; T1 slope, r=0.570; C2 slope, r=-0.691).
Correlation of radiographic parameters with 2-year postoperative changes of PT Cobb angle, MT Cobb angle, or CL
DISCUSSION
1. Key Findings and Implications for Lenke Type 2 AIS Surgery
This study evaluated the factors influencing CSA after correction surgery for patients with Lenke type 2 AIS, and high incidence of CM was observed (59.3%). We found that the PTK and T1 slopes were significantly correlated with CL. Patients with CM initially present with a more kyphotic curve, lower T1 slope, and reduced PTK. However, surgical correction not only increased the T1 slope and PTK but also that these enhancements, particularly PTK, significantly contributed to CL improvement. Our findings provide valuable insights for optimizing sagittal correction strategies for Lenke type 2 AIS.
2. T1 Slope and Its Role in CSA
Previous research has established a strong link between T1 slope and overall spinal sagittal balance, including CSA [23-25]. Studies by Lee et al. [5] and Akbar et al. [26] demonstrated this connection, with findings suggesting that T1 slope correction can influence CL and postoperative TK. Our study supports these observations, as patients with preoperative CM exhibited a smaller T1 slope. This finding reinforces the role of T1 slope in cervical alignment and suggests that surgical correction can lead to postural adjustments in the CL, ultimately improving overall sagittal balance.
3. PTK, but not MTK, Improves T1 Slope and CSA
Elevated TK is associated with a corresponding increase in T1 slope [27]. This heightened T1 slope directly influences cervical curvature, leading to a compensatory response characterized by augmentation of CL [28]. In a retrospective study conducted by Berthonnaud et al. [29] 160 lateral radiographs of asymptomatic adults were analyzed, which revealed that changes in CL correlated with variations in TK to maintain an upright posture. In AIS, proximal (T1–5) PTK and distal (T5–12) MTK play distinct roles in shaping the global thoracic sagittal curvature and influencing the occurrence of hypokyphosis. Notably, only the proximal segment demonstrated a significant relationship with the cervical spine [30]. We conducted univariate correlation analysis to explore how PTK and MTK affected CL changes. The findings revealed a substantial correlation between changes in CL and PTK as well as a modest correlation with MTK. Clement et al. [31] concluded that the correlations between PTK and distal CL indicated geometrical equivalence. Specifically, a 60% increase in PTK was transferred to a gain in distal CL. In this study, a 2-year postoperative assessment of the CM group revealed significant improvements in PTK from 9.7° preoperatively to 13.5° postoperatively. Additionally, the T1 slope exhibited a notable increase from 7.1° preoperatively to 10.5° postoperatively. Consequently, the CL angle showed a significant enhancement, improving from -19.3° preoperatively to -8.8° postoperatively. The observed reciprocal lordotic change in the cervical spine in the preoperative CM group was regarded as an adaptive response to the increased T1 slope, which is attributed to the elevation of the PTK but not MTK, aiming to preserve a horizontal gaze.
4. Preoperative Alignment and Surgical Techniques Influence PTK
Our study revealed that the effect of surgery on PTK depends on the preoperative spinal alignment. Patients with preoperative CM exhibited postoperative increases in both PTK and MTK, while those in the preoperative NCM group experienced reductions in both parameters. This aligns with the findings of Garg and Mehta [32] who observed similar variations in TK response to surgery based on preoperative alignment. The observed differences might be due to the baseline PTK in the NCM group being greater than that in the CM group. Additionally, techniques like partial facetectomy, posterior column elongation [33], and proper rod contouring can all play a role.
5. C2 Slope and Its Role in CSA
The positioning of the upper cervical area, encompassing factors such as the C1–2 angle and C2 slope, is important for maintaining the patient’s horizontal gaze [23]. The study of Lamas et al. [34] on spinal deformities underlines a marked increase in C1–2 lordosis when compared with symptom-free individuals, which aligns with our research findings. Patients in the CM group exhibited greater C1–2 lordosis and higher C2 slopes compared to the NCM group. Theoretically, following correction surgery, an improvement in C2–7 lordosis would be expected, leading to a concomitant decrease in C2 slope and C1–2 lordosis. However, achieving optimal CL restoration after surgery is still a complex process. Despite employing a standardized surgical technique, complete correction of cervical kyphosis was not observed in all patients.
A previous study on adult deformity reported larger cervical kyphosis as an independent risk factor for the development of adult cervical kyphosis [9]. Our findings demonstrated a similar trend, although this association did not reach statistical significance likely due to the relatively small sample size of this study. Multivariate analysis identified postoperative C2 slope as an independent predictor of maintaining the compensatory mechanism. This finding suggests that correction of CL may not solely depend on achieving overall sagittal balance but also significantly involve the flexibility of the lower cervical spine. As preoperative lower cervical spine flexibility was not routinely assessed in AIS patients within this study, its impact on postoperative CL could not be fully evaluated. Further investigation is warranted to explore the role of lower cervical spine flexibility in determining the success of CL correction following surgery in AIS.
6. Limited Impact of LL on CSA
Akbar et al. [26] demonstrated that the LL does not impact the CL because there is sufficient thoracic compensation that eliminates the need for compensation from the cervical spine. Similarly, Hiyama et al. [35] compared 42 patients with AIS and 24 normal adolescents, and found that LL was not correlated with CL. These findings are consistent with our results. Specifically, in this study, no differences were observed in lumbar sagittal parameters between the CM and NCM groups before or after the 2-year follow-up period.
7. Limitations and Future Directions
This study has limitations inherent to its retrospective design. Firstly, we were only able to assess the SRS scores between the NCM and CM group at the final follow-up. This limits our ability to track changes over time and identify potential trends. Secondly, the retrospective design makes it difficult to establish a direct causal link between the CSA and the patients’ clinical outcomes. Nevertheless, despite these limitations, our study is the first to explore the factors influencing CSA after correction surgery for patients with Lenke type 2 AIS. Additionally, we assessed the relationship between the changes in PTK, MTK and CL to provide valuable insights into this domain.
CONCLUSION
This study exclusively investigated the relationship between CSA and thoracic sagittal alignment in Lenke type 2 AIS patients undergoing PSF. Our findings revealed significant improvement in CSA among patients with preoperative CM following PSF. Improves in CL were significantly correlated with increases in T1 slope and PTK. This suggests that PSF, rather than focusing on increasing MTK, might indirectly improve CSA through PTK-mediated increases in T1 slope. These findings highlight the intricate interplay between PSF correction strategies and their influence on both thoracic and cervical spinal alignments, warranting further investigation for optimizing postoperative outcomes in Lenke type 2 AIS patients.
Notes
Conflict of Interest
The authors have nothing to disclose.
Funding/Support
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Author Contribution
Conceptualization: KW; Data curation: XL; Formal analysis: XL, TI; Methodology: KW; Project administration: SS, KT, TI, TO, MO, OT, NN, MM, KW; Visualization: SS, KT, MM, MN; Writing – original draft: XL; Writing – review & editing: TO, MO, OT, NN, MN, KW.