Abstract
-
Objective
To compare perioperative and complication outcomes, focusing on the prevention of sagittal translation (ST), between a novel “prioritized correction with multiple-rod construct” (PC-MRC) technique and traditional multiple-rod constructs (M-RC) in adult spinal deformity (ASD) patients undergoing 3-column osteotomies (3-COs).
-
Methods
In this retrospective study, 101 ASD patients with a minimum 2-year follow-up after 3-COs were divided into 2 groups: PC-MRC (n=65) and M-RC (n=36). The PC-MRC technique involved initial osteotomy closure with short rods followed by global alignment correction with long rods. Radiographic and clinical parameters were assessed preoperatively, postoperatively, and at final follow-up.
-
Results
The PC-MRC group had significantly shorter operation time and lower estimated blood loss (p=0.045 and p=0.007, respectively). Major coronal and kyphotic deformity correction rates were similar between groups. No significant correction loss occurred at the final follow-up. Crucially, the incidence of ST was significantly lower in the PC-MRC group (1.5% vs. 25.0%, p<0.001). Correspondingly, the overall rate of neurological injury was lower in the PC-MRC group (7.7% vs. 22.2%, p=0.037).
-
Conclusion
The PC-MRC technique offers a versatile and rigid fixation for 3-COs in ASD patients, facilitating significant correction of global deformity. This straightforward technique effectively prevents massive blood loss and ST caused by 3-COs, while minimizing the risk of neurological complication.
-
Keywords: Adult spinal deformity, Complications, Osteotomy, Pedicle subtraction osteotomy, Vertebral column resection
INTRODUCTION
Surgical intervention for adult spinal deformity (ASD) patients is indicated to alleviate chronic back pain and improve quality of life [
1,
2]. To achieve a sufficient amount of correction, 3-column osteotomies (3-COs), such as pedicle subtraction osteotomy (PSO) and vertebral column resection (VCR), are essential for correcting rigid deformities and restoring global alignment [
3]. Despite their high correction efficacy, 3-COs remain technically demanding due to subsequent instability of the osteotomy site, massive bleeding, and the risk of visceral or spinal cord injuries [
4,
5]. The incidence of intraoperative neurophysiological monitoring (IONM) alerts related to 3-COs has been reported to range from 2% to 27% [
6-
8], while postoperative neurological deficits occur in 5% to 35% of cases [
9-
13].
Sagittal translation (ST) at the osteotomy site is an unintended complication following the 3-column release of the spine, resulting in spinal cord distraction and neurological compromise [
14-
16]. Since its introduction, the cantilever technique has become the predominant measure employed in rod insertion [
17]. Meanwhile, Qian et al. [
16] found the cantilever approach to insert the rod from proximal to distal could potentially precipitate the occurrence of ST. Moreover, Qiao et al. [
18] analyzed 171 ASD patients who underwent 3-COs using the traditional cantilever technique and discovered an ST incidence rate of 10.3%, in which 27.8% had neurological injury. Reportedly, early temporary rods were used to stabilize the osteotomy site. However, the procedure of gap closure and main rod implantation often involves repeated removals or adjustments of temporary rods. These steps can be cumbersome and time-consuming, adversely affecting surgical efficiency and prolonging operative duration. Moreover, the sudden vertebral subluxation during the application of cantilever technique can hardly be avoided, given the temporary rod is removed.
To the best of our knowledge, although multiple studies have reported the occurrence and risk factors of ST following 3-COs, an effective technical strategy to avoid the occurrence of ST remains lacking. In this study, we proposed a novel technique—prioritized correction with multiple rod construct (PC-MRC)—for severe and rigid ASD patients who underwent 3-COs. This technique is achieved by 1 or 2 short rods insertion first to close the gap on the convex side and restore regional spinal stability. The purpose of this study was to investigate the feasibility and perioperative and complication outcomes of PC-MRC in ASD patients undergoing 3-COs.
MATERIALS AND METHODS
1. Patient Population
This study was performed according to the Helsinki Declaration and approved by the institutional review board (IRB) of the Affiliated Hospital of Nanjing University Medical School (IRB No. 2021-398-01). A retrospective analysis was performed on 101 consecutive patients (30 males and 71 females, aged 18 years or older) with kyphoscoliosis who had undergone 3-CO (PSO or VCR) using multiple rods in Affiliated Hospital of Nanjing University Medical School from June 2019 to April 2022. The inclusion criteria were fusion levels over 6 segments and the spinal radiograph were accessible for a minimum of 2 years postoperatively at the time of data extraction. Patients with history of spinal surgery or staged operations were excluded from the cohort (
Fig. 1). These patients underwent 2 distinct surgical procedures and were accordingly assigned to 2 separate groups: PC-MRC group and traditional multiple-rod constructs (M-RC) group. In June 2019, our team pioneered the first application of the PC-MRC technique. During the data collection period, both surgical techniques (PC-MRC and M-RC) were concurrently employed and consistently performed by the same experienced surgical team. The etiologies were congenital kyphoscoliosis (n=48), degenerative kyphoscoliosis (n=40), neuromuscular kyphoscoliosis (n=13). Indications for surgery were as follows: rigid spinal deformity causing coronal and sagittal plane malalignment; severe dysfunction of lower limbs; persistent back pain; and/or neurological deficit.
Patient demographics were reviewed and recorded: age at surgery, gender, length of follow-up, diagnoses, fusion levels, 3-CO type (PSO or VCR), osteotomy level, thoracic tri-cortical pedicle screws (3C-PS) at upper instrumented vertebra (UIV), and fusion to sacrum/pelvic. Primary concerns were the surgical details and complication data. So, we recorded operative time and estimated blood loss (EBL). The development of ST was defined as the distance between the lower posterior edge of the cranial vertebra and the upper posterior edge of the caudal vertebra was more than 5 mm at the osteotomy level according to Chang’s study [
19]. ST can be diagnosed if such abnormalities were identified on either intraoperative fluoroscopy or 2-week postoperative x-rays. Other Complications were recorded as intraoperative events (Dural tears/cerebrospinal fluid [CSF] leak, and IONM events), immediate postoperative complications (including deep wound infection, wound hematomas, and nerve root deficit), and long-term complications (implant failure, pseudarthrosis, proximal junctional kyphosis/proximal junctional failure (PJK/PJF), biplanar global malalignment, etc.) noted at follow-up. In addition, the Scoliosis Research Society-22 (SRS-22) questionnaire was employed to evaluate health-related quality of life, and the scores were recorded both preoperatively and at the final follow-up.
Radiographic measurements were performed by Surgimap Spine Software, ver. 2.2.9 (Surgimap, USA). Anteroposterior and lateral spinal radiographs were obtained preoperatively, at 2 weeks postoperatively, and at final follow-up. On coronal films, we measured major Cobb angle and global coronal malalignment (GCM, defined as distance between C7 plumb line and center sacral vertical line). Meanwhile, we have incorporated the Nanjing Classification for Coronal Plane Balance. This classification divides coronal plane balance into 3 types: type A, GCM< 3 cm; type B, GCM> 3 cm and C7 plumb line (C7PL) shifted to the concave side of the curve; type C, GCM> 3 cm and C7PL shifted to the convex side. Type A is further divided into 3 subtypes: type A
a, GCM≤1 cm; type A
b, GCM > 1 cm and C7PL shifted to the concave side of the curve; and type A
c, CBD> 1 cm and C7PL shifted to the convex side. Patients with type A
c and type C coronal malalignment are reportedly at greater risk of postoperative coronal imbalance following posterior osteotomy [
20,
21]; We also assessed the following sagittal parameters: (1) global kyphosis (GK): the angle between the superior end plate of the maximally tilted upper end vertebra to the inferior end plate of the maximally tilted lower end vertebra; (2) pelvic incidence (PI): the angle between the line that connected the hip joint axis with the center point of the superior margin of the S1 vertebra and the line that formed a right angle with the superior margin of the S1 vertebra; (3) pelvic tilt: the angle between the line that connected the center point of the superior margin of the S1 vertebra with the hip joint axis and the vertical line; (4) sacral slope (SS): the angle between the superior margin of the sacrum and the horizontal line; (5) lumbar lordosis (LL): the angle between the upper endplate of the L1 vertebra and the lower endplate of the L5 vertebra between the upper endplate of the L1 vertebra and the lower endplate; (6) sagittal vertical axis (SVA): the vertical distance between the plumb line from C7 and posterior-superior edge of the sacral endplate. To minimize measurement bias, the same surgeon (observer 1) performed duplicate measurements of all parameters and assessment of ST, with an interval of at least 3 days between the 2 sessions. The mean values were used for statistical analysis. To evaluate interobserver consistency, a second orthopedic surgeon (observer 2) independently repeated the ST assessments after observer 1 completed the initial measurements. Additionally, 15 patients were randomly selected using a random number generator, and their radiographic parameters were remeasured by observer 2. The reproducibility of these measurements was assessed by comparing them with the original results from observer 1. They were only provided with the anonymized imaging studies.
Before surgery, the surgical plan—including the selection of UIV/lower instrumented vertebra (LIV), osteotomy type, and location—was determined through a multidisciplinary team discussion following comprehensive integration of each patient’s clinical and imaging data. Special consideration was given to the indications for pelvic fusion, which were evaluated based on multiple clinical, radiographic, and biomechanical factors, to achieve adequate deformity correction while preserving distal motion segments whenever possible. The key factors considered include: (1) coronal plane considerations: The LIV is ideally selected at the most cephalad vertebra intersected by the coronal sacral vertical line, corresponding to the stable or neutral vertebra. Optimal coronal balance, as measured by GCM, should be restored within the normal range (<3 cm). Particularly in patients with type A
c and type C deformities, greater emphasis is placed on horizontalization of the lumbosacral curve, often leading to pelvic fusion [
20,
21]; (2) sagittal plane considerations: LIV selection must align with overall sagittal alignment goals, particularly the harmonization of PI minus LL (PI–LL). When maximal LL is required, fusion may extend to the sacrum or pelvis. This decision is also influenced by the planned osteotomy level and the extent of fusion. Pelvic fixation is often required to counteract significant lever arm forces and prevent distal junctional failure; (3) distal degenerative changes and disc status: If the first disc distal to the LIV exhibits severe degeneration, spondylolisthesis, or spinal stenosis, extension of the fusion construct may be warranted; (4) patient-specific factors: Older patients and those with osteoporosis are more likely to undergo pelvic fixation to enhance construct stability [
22].
Both groups underwent identical preoperative preparations. All surgeries were performed in a single stage via a posterior approach under general anesthesia. Somatosensory evoked potential and transcranial motor evoked potential monitoring of the spinal cord was initiated prior to surgical intervention. The spine was exposed to transverse processes according to the predetermined fusion strategy. Pedicle screws were placed as planned, and internal fixation was performed using the Solera spinal internal fixation system (Medtronic, Inc., USA). If necessary, S2 alar iliac (S2AI) screws and/or thoracic tri-cortical pedicle screws were inserted under O-arm navigation (Stealth Station S7 Navigation System, Medtronic, Inc.). Then asymmetrical 3-CO (PSO or VCR) was performed on the presumed osteotomy level. The concave side should be conducted first to complete the transposition and decompression of the spinal cord. A short rod was placed across osteotomy site to prevent the subluxation.
Here shows the different steps of PC-MRC technique (
Fig. 2): After completing the osteotomy on the convex side, the osteotomy site is directly closed by inserting short rods from the convex side, which simultaneously achieves correction of the regional deformity around the osteotomy area. Then the coronal and sagittal balance could be restored by 2 long correction rods. Specifically, for patients with Nanjing-type A
c and C coronal plane deformities, local correction was followed by targeted correction of lumbosacral functional curve. Lumbosacral fusion was performed to establish a stable foundation, thereby achieving coronal plane balance. Finally, the implants would constitute M-RC using cross-link or connectors with 2 tulips. In contrast, in the traditional M-RC group, all corrective maneuvers are performed on 2 long rods, while the additional rods serve solely as supplemental fixation to enhance the biomechanical stability of the construct.
All analyses were generated using IBM SPSS Statistics ver. 26.0 (IBM Co., USA) software package. Statistical results are presented as mean±standard deviation, median, interquartile range, or odds ratio (OR) with corresponding confidence intervals (CI), as appropriate. Demographics, intraoperative conditions, and complication data were under descriptive statistics, what followed was the analysis to compare the radiological parameters between the time points by paired t-tests and chi-square test. Binary logistic regression analysis was conducted to identify risk factors for ST, with age, PC-MRC technique, osteotomy type (PSO vs. VCR), osteotomy level, pelvic fusion, preoperative sagittal deformity severity, and intraoperative blood loss included as covariates. A p-value<0.05 was considered statistically significant.
RESULTS
1. Patients’ Demographics and Surgical Details
A total of 101 consecutive patients (30 males and 71 females) were analyzed and the mean follow-up time was 24.9±2.3 months. There were 65 cases in the PC-MRC group and 36 cases in the M-RC group. There was a growing preference for PC-MRC, and 90% of the M-RC patients had surgery before April 2021. The PC-MRC group had a mean age of 45.4 (range, 23–65) years, with a median of 41 and an interquartile range of 37 to 57. The M-RC group had a mean age of 40 (range, 25–65) years, with a median of 44 and an interquartile range of 30 to 53. A chi-square test for age category distribution indicated no statistically significant difference between the 2 groups (χ
2=0.613, p=0.736), suggesting comparable age distributions (
Table 1).
Patient demographics was showed in
Table 1. The mean fusion levels were 12 (PC-MRC: 12.2, M-RC: 10.8). 3-COs included PSO (PC-MRC: 44, M-RC: 20) and VCR (PC-MRC: 21, M-RC: 16) which were mostly carried out at: T9 in 2 cases (PCMRC: 2, M-RC: 0), T11 in 9 cases (PC-MRC: 6, M-RC: 3), T12 in 12 cases (PC-MRC: 7, M-RC: 5), L1 in 21 cases (PC-MRC: 13, M-RC: 8), L2 in 35 cases (PC-MRC: 21, M-RC:14), L3 in 18 cases (PC-MRC: 13, M-RC: 5), L4 in 4 cases (PC-MRC: 3, M-RC: 1). In PC-MRC group, there were 3 cases (4.6%) using 3C-PS at UIV, and 21 cases (32.3%) had their LIV extending to the pelvis with S2AI. While in M-RC group, 8 cases (22.2%) had sacropelvic fixation. The PC-MRC group had less EBL and shorter operation time, which reached statistical significance (p=0.007 and p=0.045, respectively). No significant difference was observed in sex, pathologies, and fusion level (p=0.294, p=0.689, and p=0.115, respectively).
The interobserver and intraobserver consistency tests demonstrated that all intraclass correlation coefficient (ICC) values exceeded 0.75, indicating good measurement reliability (
Table 2). As a representative example, preoperative Cobb angle measurements achieved an interobserver ICC of 0.994 (95% CI, 0.991–0.995) and an intraobserver ICC of 0.990 (95% CI, 0.971–0.997). All of the 101 patients had severe kyphoscoliosis, that the mean major Cobb angle was 69.3°± 20.7° (range, 45°–120°) and the mean GK angle was 72.6°± 22.7° (range, 48°–120°) preoperatively. The preoperative prevalence of coronal imbalance was 27 of 65 (41.5%) in the PC-MRC group and 17 of 36 (47.2%) in the M-RC group. With respect to types A
c and C, which carries high risk of postoperative coronal imbalance—the proportion was 33 of 65 (50.8%) in the PC-MRC group versus 16 of 36 (44.4%) in the M-RC group, showing no significant difference was observed in the overall distribution of coronal alignment between the 2 groups, indicating comparable baseline coronal balance (χ
2=11.105, p=0.893). There was no significant difference in other radiographical parameters before operation (
Table 3).
3-COs ensured adequate correction in both groups: the mean correction rate for Cobb angle was 22.2%±10.6% and 35.0%±12.2% in PC-MRC and M-RC group, and the mean correction rate for GK angle was 54.7%±17.0% and 58.3%±18.5%. There was no significant loss of correction at final follow-up (
Tables 4 and
5).
Regarding to coronal and sagittal balance, both groups could maintain well global alignment at follow-up (
Tables 3–
5;
Fig. 3). In the coronal plane, GCM was improved from 24.8±20.1 mm to 16.9±12.5 mm and maintained 18.6±11.4 mm at final follow-up in PC-MRC group; In M-RC group, GCM was 6.7±24.5 mm postoperatively and maintained 13.0±16.1 mm at final follow-up. In the sagittal plane, the improvement of sagittal parameters was observed in both groups. In PC-MRC group, |PI–LL| improved from 35.6° ± 21.9° to 14.8° ± 9.5° and SVA improved from 55.0±51.5 mm to 35.7±16.9 mm; In M-RC group, |PI–LL| improved from 26.4°± 23.8° to 17.1°± 15.2° and SVA improved from 47.6±42.9 mm to 23.5±17.4 mm.
1) Intraoperative events
The incidence of ST was significantly lower in the PC-MRC group compared to the M-RC group, with 1 of 65 cases (1.5%) versus 9 of 36 cases (25%), respectively (χ
2=14.296, p<0.001) (
Table 6). Among the 9 ST cases in M-RC, 5 occurred IONM events and developed new neurologic deficit (
Fig. 4). Meanwhile, the results of binary logistic regression indicated that the use of PC-MRC technology was an independent protective factor against the occurrence of ST (p=0.009; 95% CI, 0.006–0.484) (
Table 7). Regarding overall nerve injury, there were 5 (7.7%) versus 8 cases (22.2%) between PC-MRC and M-RC groups, respectively, which was statistically significant (χ
2=4.362, p<0.037) (
Table 6). No significant differences were observed between the 2 groups in the incidence of IONM events or dural tears/CSF leaks (χ
2=0.997 and χ
2=0.092, p=0.318 and p=0.762, respectively).
2) Follow-up complications
There was no significant difference in follow-up complications between the 2 groups. In the PC-MRC group, one patient had deep wound infection 1 month after surgery, and the secretion culture showed Pseudomonas aeruginosa infection. However, conservative treatment for 2 weeks was ineffective, and the wound was healed after debridement and drainage. At final follow-up, although PJK developed in 15 (23.1%) and 8 cases (22.2%) in PC-MRC and M-RC groups respectively, which showed no statistical significance (χ
2=0.010, p=0.922), these patients did not require revision surgery since the symptoms were often vague and no global malalignment or PJF was raised secondary to PJK. The rod or screw broken was not observed in both groups. In the PC-MRC group, 3 patients met coronal imbalance according to Nanjing classification [
20,
21] (GCM>3 cm) and 3 patients met sagittal imbalance (SVA>5 cm); however, both GCM and SVA showed substantial improvement compared to preoperative values.
Due to variations in patient compliance during actual clinical management, 28 completed SRS-22 questionnaires have been collected from the PC-MRC group and 12 from the M-RC group to date (
Table 8). Based on the currently available data, baseline characteristics were comparable between the 2 groups prior to surgery. At the final follow-up, patients in the PC-MRC group demonstrated higher scores in the pain domain and achieved a significantly higher total SRS-22 score (p=0.005 and p=0.003, respectively), indicating that the PC-MRC technique is associated with a significant improvement in health-related quality of life.
DISCUSSION
Our study introduces a novel and effective technique—PC-MRC—designed to reduce the incidence of ST, minimize intraoperative blood loss, and lower the risk of neurological complications. Satisfactory deformity correction was achieved and maintained at final follow-up. We emphasize that PC-MRC is not merely an addition of implants but represents a strategic refinement of surgical workflow compared to conventional M-RC. The core principle of this approach involves assigning distinct biomechanical roles to individual rods: short rods are utilized for osteotomy site closure, whereas long rods are reserved for global coronal and sagittal realignment. By reorganizing the surgical sequence and strategically modulating biomechanical forces, we established a stepwise correction strategy characterized by a “local-first, global-second” sequence. This constitutes a key conceptual and methodological advancement at the strategic level. Importantly, the technique directly targets the underlying mechanism of ST, highlighting its clinical value in enhancing surgical safety rather than prioritizing maximal correction magnitude.
3-COs are widely used in the correction of severe fixed spinal deformities with sagittal and/or coronal malalignment [
3]. Although sufficient correction could be provided by 3-COs, the high procedural risks often hinder the surgeon from performing aggressive osteotomies and achieving optimal correction. The 3-COs, particularly the wedge PSO, entail substantial technical challenges, primarily in the following domains: (1) neurological injury: During the procedure, the dural sac and nerve roots are at considerable risk of injury within the confined spinal canal. This risk is especially pronounced during closure of the osteotomy site, where posterior vertebral wall fragments may herniate into the canal or the intervertebral foramen may undergo acute narrowing, potentially resulting in neural compression [
5]; (2) spinal instability during rod installation: following osteotomy, the high risk procedure occurs in the gap closure and rod installation process, where long instrumentation generates a strong leverage force that concentrated on the osteotomy site and causing dislocation; (3) hemorrhage and vascular injury: Spinal cord ischemia may arise from intraoperative hypotension, direct vascular injury, or compromised regional perfusion following deformity correction, all of which pose a risk for irreversible neurological deficits [
12].
Specifically speaking, first of all, a major concern is the elevated risk of neurological complications, that greatly hamper patients’ quality of life and even cause permanent deficits. Zhang et al. [
10] observed significantly higher risk of neurological complication in VCR group (4 of 12, 33.3%) than in multiple levels asymmetrical Ponte osteotomy group (1 of 26, 3.8%). Khashan et al. [
12] measured the lower-extremity motor scores (LEMSs) in 199 ASD patients managed with 3-COs, and the analysis showed that 10% of the patients had LEMSs decline immediately postoperatively and 5% still suffered from lower-extremity motor deficits at 1-year follow-up. According to Li et al. [
23] retrospective study on 65 cases who developed neurologic deficit after corrective surgery for scoliosis, mechanical injury to spinal cord is the primary cause of postoperative neurological complications in 3-CO surgery. Therefore, it is essential to ensure complete resection of the posterior spinal structures and perform spinal cord transposition to fully decompress the spinal canal space [
24,
25].
However, preventing ST and its associated neurological injury remains challenging [
14,
15]. Of the 35 severe pediatric deformity patients, Lenke et al. [
26] reported 1 case occurred IONM events along with vertebral subluxation upon closing osteotomy site. In patients with ankylosing spondylitis, Qian et al. [
16] reported that the occurrence of ST was associated with inappropriate manual manipulation of osteotomy, gap closure, and rod insertion. Traditionally, the entitled corrective strength was fulfilled by 2 long rods, the surgeon needed to stabilize the osteotomy site and simultaneously balance the global alignment. In this process, the correction maneuver is difficult, and spinal instability and instantaneous vertebral subluxation are likely to occur during the correction. Especially when applying cantilever technique, due to the force exerted by a longer lever arm and the effect of trunk gravity, the ends of the unstable osteotomy gap are prone to shift in opposite directions, ventrally and dorsally, leading to ST. Although temporary rods are commonly used to provide interim stability, they are frequently removed and reinstalled during the procedure, limiting their effectiveness. In our M-RC group, ST occurred in 25% of cases, indicating that temporary fixation alone is insufficient to prevent this complication, while also adding time and complexity to the operation.
While in PC-MRC technique, we separate the complex correction surgery into discrete, manageable steps. A short rod is used directly to accomplish bone-to-bone gap closure from the convex side. The key innovation is that the short rods contribute to local stabilization and partial correction, allowing long rods to focus exclusively on the reconstruction of global coronal and sagittal balance without worry of vertebral subluxation. In the current study, the postoperative x-rays showed only 1 case (1.5%) reached ST in PC-MRC group that was a markedly lower incidence comparing to M-RC group (9 cases, 25%). Among these 9 cases in M-RC group, 5 of them occurred IONM alert and developed new neurologic deficit. Although these new developed neurologic deficits had relieved or disappeared at final follow-up, the overall incidence of nerve injury was significantly lower in the PC-MRC group, underscoring the neuroprotective advantage of this technique. Furthermore, binary logistic regression analysis identified that the use of PC-MRC was an independent protective factor against ST (
Table 7). Due to the limited number of ST events in our cohort, strict adherence to the “rule of ten,” which recommends at least 10 outcome events per predictor variable to ensure model stability, was not feasible. The underlying overfitting effects are likely associated with the very large OR and wide CI. Consequently, our results should be validated in larger, prospective, multicenter studies to confirm the protective effect of PC-MRC against ST.
According to Khashan et al. [
12], vascular causes also play an important role in postoperative neurological deficits. Blood loss in 3-CO surgery is mainly from the osteotomy site and intraspinal venous plexus bleeding, which could result in spinal cord hypoperfusion [
27,
28]. Meanwhile, the incidence of postoperative epidural hematoma would be greatly increased due to inadequate hemostasis. In our study, the PC-MRC group had significantly less EBL than M-RC group, likely due to early gap closure, and no cases of epidural hematoma were observed. These findings suggest that prompt, stable bone-to-bone closure of the osteotomy site effectively controls bleeding after 3-COs.
Another strength of the PC-MRC is the strong fixation achieved by the 2 short rods around the osteotomy and/or lumbosacral region, with 2 long rods securing the global balance. Lower rates of rod fracture, reoperation rates, and pseudoarthrosis were reported in numerous studies [
29-
33]. Hyun et al. [
29] compared radiographical outcomes of 2-RC construct versus M-RC construct across 3-CO sites and the 2-RC group had 17% (11 of 66) cases of rod breakage at osteotomy site with 5 of them required revision while there was no implant failure in the M-RC group during the 2-year follow-up. Luca et al. [
30] conducted a finite element analysis and concluded that the M-RC could significantly achieve stable biomechanical effects around 3-CO region. Guevara-Villazón et al. [
31] observed 18% (6 of 33) and 3% (1 of 33) pseudoarthrosis in 2-RC group and M-RC group, respectively. In our series, no rod fractures or pseudoarthrosis were observed at the 3-CO site in the PC-MRC group during a minimum 2-year follow-up, further supporting the structural robustness of this construct.
It is worth noting that construct rigidity may increase the incidence of PJK due to the distinct density between proximal normal bone substance and fixation area [
34]. Meanwhile, several studies reported that no difference in the rate of PJK was found between M-RC and 2-RC constructs [
29,
30,
33]. However, this observation may be skewed due to a lack of uniform baseline characteristics among the study populations, noting that these investigations did not adequately account for variations in preoperative parameters. In the present study, a similar rate of PJK (23.1% vs. 22.2%) was observed between the 2 groups during our follow-up, which may be attributed to similar sagittal parameters postoperatively. So, whether the solid construct could be associated with the development of PJK requires further investigation and long-term follow-up.
The PC-MRC technique not only contributes to the correction of sagittal plane alignment, but also plays a significant role in the reconstruction of coronal plane balance. Preoperatively, 35.4% (23 patients) in the PC-MRC group were classified as type A
c or type C deformities—morphologies associated with a higher risk of postoperative convex-side tilt [
20,
21]. At final follow-up, only 3 patients exhibited persistent radiographic C-type imbalance, and none required revision surgery. This favorable outcome is attributed to the achievement of stable local correction, which allows surgeons greater intraoperative flexibility in fine-tuning coronal alignment. Evidence suggests that the lumbosacral junction serves as the biomechanical foundation of the spine, and restoration of lumbosacral horizontalization is essential for optimal main curve correction and effective coronal compensation [
35]. Following completion of local deformity correction, we typically perform lumbosacral fusion in patients with type A
c and type C deformities to correct functional lumbosacral curves, stabilize the foundation, and subsequently employ long-segment instrumentation to modulate overall spinal alignment. Several studies support a 2-stage approach—initial anterior lateral lumbar interbody fusion (LLIF) for adequate coronal release, followed by posterior spinal correction—to improve coronal balance restoration and may even restore the disc height [
36,
37]. Recently, Hiyama et al. [
38] evaluated radiological outcomes following staged LLIF-based surgery in ASD patients, reporting that patients with Obeid type 2 coronal malalignment (corresponding to Nanjing classification type A
c and type C) were less likely to achieve satisfactory coronal balance compared to those with type 1, highlighting the limitations of current techniques for specific coronal deformity patterns. In contrast, our findings indicate that combining PC-MRC with pelvic fixation can eliminate the need for staged procedures. This integrated strategy can enable safe and efficient osteotomies while maximizing the likelihood of achieving and maintaining ideal postoperative coronal alignment in patients with Obeid type 2 (Nanjing types A
c and C) coronal deformity. Although this observation was not a primary endpoint of the present study, it provides a compelling rationale for future investigations comparing the clinical efficacy and durability of coronal balance between this single-stage approach and conventional staged surgical protocols.
Other limitations of this study include retrospective design, small sample size, rather short term of follow-up. Due to the relatively small sample size, this study was not adequately powered to perform a robust propensity score matching analysis. Variability in underlying pathologies and surgical approaches further limits the generalizability of the findings. Additionally, only a subset of patient-reported outcome questionnaires was available for analysis. A more comprehensive dataset is required to robustly assess and confirm these relationships. Surgeon experience and shifting preference toward PC-MRC may have confounded case selection and outcomes. A learning curve analysis and prospective studies are needed. Also, the complications of implant failure, such as rod fracture, often occur between 2 and 5 years postoperatively [
39,
40], which may cause the low incidence of complications of our series. In addition, the small sample size leads to multivariable model instability, and discrepancies between univariable and multivariable findings should be interpreted with caution. Currently, the PC-MRC technique is indicated specifically for patients with severe sagittal kyphotic deformity coexisting with coronal plane scoliosis, which entails regional 3CO for apex deformity correction and long instrumentation for tuning the global balance. Moreover, the relative effectiveness of PC-MRC in the cervical and upper thoracic vertebrae, as well as highly unstable VCR has not yet been definitively established. Future studies should refine patient selection, clarify technical limitations, and systematically assess long-term implant outcomes.
CONCLUSION
PC-MRC can produce satisfactory correction in adult kyphoscoliosis who underwent 3-COs, and there is no significant loss of correction during a minimum 2 years of follow-up. The stepwise procedure including osteotomy closing at the convex side, short rods correction and fixation followed by long rods global correction which can effectively avoid the ST and massive blood loss and thereby lowering the risk of neurological complication. Additionally, the M-RC could enhance fixation strength across the osteotomy site and minimize the risk of rod fracture in patients with severe deformity.
NOTES
-
Conflict of Interest
The authors have nothing to disclose.
-
Funding/Support
National Natural Science Foundation of China (Grant 82272545) National Natural Science Foundation of China (Grant 82502981).
-
Author Contribution
Conceptualization: CL, ZH, ZZ, ZL; Data curation: CL, JL, AK, XQ; Formal analysis: CL, JL, ZT, ZL; Funding acquisition: ZL; Methodology: CL, JL, ZH, YX, AK, BS, ZZ, YQ, ZL; Project administration: ZL; Visualization: YX, HX, YQ, ZL; Writing – original draft: CL; Writing – review & editing: JL, ZL.
Fig. 1.Flow diagram for patient selection. ASD, adult spinal deformity; 3-CO, 3-column osteotomy; GK, global kyphosis; PSO, pedicle subtraction osteotomy; VCR, vertebral column resection.
Fig. 2.The steps of the “prioritized correction with multiple-rod construct” technique. (A) Step 1: All screws are placed, and the 3-CO (PSO or VCR) is initially performed on the concave side to achieve spinal cord transposition and decompression. (B) Step 2: Short rods are placed across the osteotomy site to prevent subluxation. (C) Step 3: Following completion of the osteotomy on the convex side, the osteotomy gap is directly closed using short rods to attain regional correction and stability. (D) Step 4: Two long corrective rods are subsequently utilized to restore global spinal alignment. 3-CO, 3-column osteotomy; PSO, pedicle subtraction osteotomy; VCR, vertebral column resection.
Fig. 3.PC-MRC case of a patient with congenital scoliosis. (A) Preoperative anterior-posterior x-ray showed a right thoracolumbar curve of 103°. (B) Preoperative lateral x-ray showed a significant kyphosis measuring 97°. (C and D) We performed asymmetric PSO at T12. Two satellite rods were used to complete gap closure and 2 long correction rods were used to regulate global alignment from T4 to L4. Following surgery, the main Cobb angle was corrected to 68°and GK was corrected to 48°. (E and F) At 2-year follow-up, there was no significant loss of correction, maintaining well balance in both sagittal and coronal planes. PCMRC, prioritized correction with multiple-rod construct; PSO, pedicle subtraction osteotomy; GK, global kyphosis.
Fig. 4.M-RC case of a patient with congenital scoliosis who developed sagittal translation (ST). (A and B) Preoperative x-ray demonstrated a hemivertebra deformity at L1 resulted in local kyphoscoliosis. (C and D) We performed VCR at the hemivertebra and got the spine fused from T10 to L4. Though the main Cobb angle and GK had adequate correction from 70° to 32° and from 60° to 18°, respectively, we observed ST (G) and Monitoring events in this patient leading to temporary sensory deficits of the left thigh. (E and F) At 2-year follow-up, there was no significant loss of correction and the symptoms had recovered. M-RC, traditional multiple-rod constructs; VCR, vertebral column resection GK, global kyphosis.
Table 1.Comparison of demographics and operative factors between the PC-MRC versus M-RC groups
Table 1.
|
Variable |
PC-MRC (N = 65) |
M-RC (N = 36) |
t/χ2
|
p-value |
|
Age (yr) |
45.4 ± 12.7 |
40.0 ± 13.0 |
1.333 |
0.191 |
|
18–29 |
9 |
7 |
0.613 |
0.736 |
|
30–59 |
43 |
23 |
|
|
|
≥ 60 |
13 |
6 |
|
|
|
Sex, male:female |
17:48 |
13:23 |
1.100 |
0.294 |
|
Pathologies |
|
|
|
|
|
CS |
32 |
16 |
0.745 |
0.689 |
|
DS |
26 |
14 |
|
|
|
NMS |
7 |
6 |
|
|
|
Fusion levels |
12.2 ± 2.7 |
10.8 ± 2.4 |
1.618 |
0.115 |
|
PSO/VCR |
44/21 |
20/16 |
1.470 |
0.225 |
|
T9 |
2 |
0 |
2.346 |
0.885 |
|
T11 |
6 |
3 |
|
|
|
T12 |
7 |
5 |
|
|
|
L1 |
13 |
8 |
|
|
|
L2 |
21 |
14 |
|
|
|
L3 |
13 |
5 |
|
|
|
L4 |
3 |
1 |
|
|
|
3C-PS screws at UIV |
3 |
0 |
1.712 |
0.191 |
|
S2AI or IS at LIV |
21 |
8 |
1.151 |
0.283 |
|
OP time (min) |
372.2 ± 61.0 |
419.7 ± 75.1 |
-2.083 |
0.045 |
|
Blood loss (mL) |
2,322 ± 476 |
2,900 ± 703 |
-2.888 |
0.007 |
Table 2.Interobserver and intraobserver reliability of radiographic measurement preoperatively
Table 2.
|
Variable |
Interobserver (n = 101)
|
Intraobserver (n = 15)
|
|
First measurement |
Second measurement |
ICC (95% CI) |
Observer 1 |
Observer 2 |
ICC (95% CI) |
|
Preoperative Cobb (°) |
69.5 ± 20.8 |
69.5 ± 20.7 |
0.994 (0.991–0.995) |
71.24 ± 20.9 |
70.6 ± 20.4 |
0.990 (0.971–0.997) |
|
Preoperative GK (°) |
72.7 ± 24.3 |
72.5 ± 24.1 |
0.996 (0.994–0.997) |
78.2 ± 26.4 |
77.8 ± 26.7 |
0.996 (0.987–0.999) |
|
Preoperative LL (°) |
38.0 ± 40.1 |
37.9 ± 40.0 |
0.998 (0.997–0.999) |
39.7 ± 46.7 |
39.8 ± 47.2 |
0.998 (0.995–0.999) |
|
Preoperative PI (°) |
39.9 ± 12.5 |
40.0 ± 12.4 |
0.977 (0.967–0.985) |
37.7 ± 13.0 |
36.7 ± 13.2 |
0.982 (0.948–0.994) |
|
Preoperative PT (°) |
16.6 ± 14.9 |
16.5 ± 15.0 |
0.994 (0.991–0.996) |
14.1 ± 15.9 |
13.7 ± 15.5 |
0.992 (0.977–0.997) |
|
Preoperative SS (°) |
23.4 ± 18.5 |
23.5 ± 18.4 |
0.987 (0.981–0.991) |
23.6 ± 13.4 |
23.0 ± 13.6 |
0.992 (0.977–0.997) |
|
Preoperative SVA (mm) |
52.6 ± 48.8 |
52.7 ± 48.3 |
0.998 (0.997–0.999) |
42.7 ± 40.1 |
42.1 ± 40.6 |
0.999 (0.996–0.999) |
Table 3.Comparison of imaging parameters between the PC-MRC versus M-RC groups preoperatively
Table 3.
|
Variable |
PC-MRC (N = 65) |
M-RC (N = 36) |
t/χ2
|
p-value |
|
Cobb (°) |
69.7 ± 20.5 |
68.9 ± 21.5 |
0.125 |
0.902 |
|
GK (°) |
72.6 ± 27.1 |
72.6 ± 18.1 |
-0.003 |
0.998 |
|
LL (°) |
33.8 ± 42.0 |
45.5 ± 36.7 |
-0.894 |
0.378 |
|
PI (°) |
40.8 ± 13.5 |
38.4 ± 10.3 |
0.591 |
0.558 |
|
|PI–LL| (°) |
35.6 ± 21.9 |
26.4 ± 23.8 |
1.203 |
0.237 |
|
PT (°) |
19.2 ± 16.2 |
11.6 ± 11.1 |
1.633 |
0.112 |
|
SS (°) |
21.6 ± 14.1 |
26.6 ± 13.0 |
-1.117 |
0.272 |
|
SVA (mm) |
55.0 ± 51.5 |
47.6 ± 42.9 |
1.559 |
0.129 |
|
GCM (mm) |
24.8 ± 20.1 |
6.3 ± 17.6 |
2.926 |
0.006 |
|
Nanjing classification for coronal plane balance |
|
|
|
|
|
Type Aa
|
15 (23.1) |
6 (16.7) |
11.105 |
0.893 |
|
Type Ab
|
11 (16.9) |
5 (13.9) |
|
|
|
Type Ac
|
12 (18.5) |
8 (22.2) |
|
|
|
Type B |
16 (24.6) |
9 (25.0) |
|
|
|
Type C |
11 (16.9) |
8 (22.2) |
|
|
Table 4.Comparison of imaging parameters between the PC-MRC versus M-RC groups postoperatively
Table 4.
|
Variable |
PC-MRC (N = 65) |
M-RC (N = 36) |
t/χ2
|
p-value |
|
Cobb (°) |
45.6 ± 13.8 |
45.3 ± 17.7 |
0.074 |
0.942 |
|
ΔCobb/Cobb (%) |
34.2 ± 10.6 |
35.0 ± 12.2 |
-0.219 |
0.828 |
|
GCM (mm) |
16.9 ± 12.5 |
6.7 ± 24.5 |
1.580 |
0.123 |
|
GK (°) |
32.7 ± 15.3 |
31.2 ± 17.7 |
0.264 |
0.794 |
|
ΔGK/GK (%) |
54.7 ± 17.0 |
58.3 ± 18.5 |
-0.610 |
0.546 |
|
LL (°) |
48.5 ± 18.7 |
53.0 ± 17.8 |
-0.716 |
0.479 |
|
PI (°) |
41.5 ± 12.4 |
38.0 ± 11.1 |
0.890 |
0.380 |
|
|PI–LL| (°) |
14.8 ± 9.5 |
17.1 ± 15.2 |
-0.558 |
0.580 |
|
PT (°) |
9.7 ± 14.6 |
3.0 ± 10.5 |
1.565 |
0.127 |
|
SS (°) |
31.8 ± 14.2 |
35.0 ± 8.5 |
-0.807 |
0.425 |
|
SVA (mm) |
35.7 ± 16.9 |
23.5 ± 17.4 |
2.026 |
0.051 |
Table 5.Comparison of imaging parameters between the PC-MRC versus M-RC groups at final follow-up
Table 5.
|
Variable |
PC-MRC (N = 65) |
M-RC (N = 36) |
t/χ2
|
p-value |
|
Cobb (°) |
45.6 ± 14.2 |
46.6 ± 17.0 |
-0.197 |
0.845 |
|
GCM (mm) |
18.6 ± 11.4 |
13.0 ± 16.1 |
1.202 |
0.238 |
|
GK (°) |
34.0 ± 15.4 |
32.0 ± 18.1 |
0.365 |
0.717 |
|
LL (°) |
48.7 ± 19.0 |
53.0 ± 19.2 |
-0.672 |
0.506 |
|
PI (°) |
42.9 ± 12.7 |
38.2 ± 11.1 |
1.167 |
0.251 |
|
|PI–LL| (°) |
17.6 ± 11.4 |
17.4 ± 15.2 |
0.061 |
0.952 |
|
PT (°) |
12.1 ± 12.8 |
4.8 ± 9.6 |
1.954 |
0.060 |
|
SS (°) |
30.7 ± 13.7 |
33.4 ± 10.3 |
-0.673 |
0.506 |
|
SVA (mm) |
22.4 ± 15.9 |
21.7 ± 21.3 |
0.109 |
0.914 |
Table 6.Comparison of surgery and postoperative complications between the PC-MRC versus M-RC groups at final follow-up
Table 6.
|
Variable |
PC-MRC (N = 65) |
M-RC (N = 36) |
t/χ2
|
p-value |
|
Intraoperative events |
|
|
|
|
|
Sagittal translation |
1 (1.5) |
9 (25) |
14.296 |
< 0.001 |
|
IONM events |
5 (7.7) |
5 (13.9) |
0.997 |
0.318 |
|
Nerve injury |
5 (7.7) |
8 (22.2) |
4.362 |
0.037 |
|
Dural tears/CSF leak |
6 (9.2) |
4 (11.1) |
0.092 |
0.762 |
|
Follow-up complications |
|
|
|
|
|
Deep wound infection |
1 |
0 |
0.000 |
1.000 |
|
PJK/PJF |
15 (23.1) |
8 (22.2) |
0.010 |
0.922 |
|
Rod/Screw broken |
0 (0) |
0 (0) |
- |
- |
|
Coronal imbalance |
3 (4.6) |
1 (2.8) |
0.206 |
0.650 |
|
Sagittal imbalance |
3 (4.6) |
0 (0) |
3.243 |
0.072 |
Table 7.Binary logistic regression analysis to identify risk factors for the occurrence of ST
Table 7.
|
Variables |
Univariate analysis
|
Multivariate analysis
|
|
p-value |
OR (95% CI) |
p-value |
OR (95% CI) |
|
Age ≥ 60 yr |
0.0538 |
2.219 (0.264–18.663) |
- |
- |
|
PC-MRC technique |
0.005 |
0.047 (0.006–0.388) |
0.009 |
0.053 (0.006–0.484) |
|
Lower lumbar osteotomy (L3–5) |
0.154 |
4.650 (0.563–38.392) |
- |
- |
|
VCR used |
0.816 |
0.853 (0.225–3.244) |
- |
- |
|
Pelvic fusion |
0.199 |
4.000 (0.483–33.105) |
- |
- |
|
GK (°) |
0.720 |
1.005 (0.978–1.032) |
- |
- |
|
PI (°) |
0.955 |
1.002 (0.950–1.056) |
- |
- |
|
LL (°) |
0.133 |
1.013 (0.996–1.031) |
- |
- |
|
|PI–LL| (°) |
0.663 |
1.008 (0.974–1.043) |
- |
- |
|
SVA (mm) |
0.349 |
1.006 (0.993–1.019) |
- |
- |
|
Blood loss (mL) |
0.041 |
1.001 (1.000–1.002) |
0.725 |
1.000 (0.999–1.001) |
Table 8.Health-related quality of life measured with Scoliosis Research Society-22
Table 8.
|
Variable |
PC-MRC (n = 28)
|
M-RC (n = 12)
|
Baseline
|
Final
|
|
Baseline |
Final |
Baseline |
Final |
t-value |
p-value |
t-value |
p-value |
|
Self-image |
2.63 ± 0.74 |
3.32 ± 0.44 |
2.61 ± 0.63 |
3.48 ± 0.43 |
-0.115 |
0.909 |
1.065 |
0.293 |
|
Mental health |
2.67 ± 0.45 |
3.83 ± 0.40 |
2.58 ± 0.38 |
3.95 ± 0.46 |
-0.631 |
0.532 |
0.814 |
0.421 |
|
Function |
2.93 ± 0.76 |
3.60 ± 0.59 |
2.93 ± 0.50 |
3.86 ± 0.30 |
-0.024 |
0.981 |
1.878 |
0.068 |
|
Pain |
2.73 ± 0.40 |
3.45 ± 0.38 |
2.66 ± 0.37 |
3.85 ± 0.39 |
-0.515 |
0.609 |
2.983 |
0.005 |
|
Satisfaction |
- |
3.79 ± 0.62 |
- |
4.09 ± 2.69 |
- |
- |
1.489 |
0.145 |
|
Total |
2.74 ± 0.38 |
3.57 ± 0.20 |
2.69 ± 0.26 |
3.81 ± 0.24 |
-0.461 |
0.647 |
3.119 |
0.003 |
REFERENCES
- 1. Ames CP, Scheer JK, Lafage V, et al. Adult spinal deformity: epidemiology, health impact, evaluation, and management. Spine Deform 2016;4:310-22.
- 2. Pellisé F, Vila-Casademunt A, Ferrer M, et al. Impact on health related quality of life of adult spinal deformity (ASD) compared with other chronic conditions. Eur Spine J 2015;24:3-11.
- 3. Yilgor C, Kindan P, Yucekul A, et al. Osteotomies for the treatment of adult spinal deformities: a critical analysis review. JBJS Rev 2022;10(5).
- 4. Kim KT, Park KJ, Lee JH. Osteotomy of the spine to correct the spinal deformity. Asian Spine J 2009;3:113-23.
- 5. Enercan M, Ozturk C, Kahraman S, et al. Osteotomies/spinal column resections in adult deformity. Eur Spine J 2013;22 Suppl 2(Suppl 2):S254-64.
- 6. Wang S, Yang Y, Zhang J, et al. Frequent neuromonitoring loss during the completion of vertebral column resections in severe spinal deformity surgery. Spine J 2017;17:76-80.
- 7. Shi B, Shi B, Liu D, et al. Scoliosis Research Society-Schwab grade 6 osteotomy for severe congenital angular kyphoscoliosis: an analysis of 17 cases with a minimum 2-year follow-up. Neurosurgery 2020;87:925-30.
- 8. Liu W, Qiu J, Zhu Z, et al. Intraoperative Neurophysiological Monitoring Alerts During Three-Column Osteotomy: incidence and risk factors. J Clin Neurophysiol 2023;40:641-5.
- 9. Berven SH, Deviren V, Smith JA, et al. Management of fixed sagittal plane deformity: results of the transpedicular wedge resection osteotomy. Spine (Phila Pa 1976) 2001;26:2036-43.
- 10. Zhang Y, Tao L, Hai Y, et al. One-stage posterior multiple-level asymmetrical ponte osteotomies versus single-level posterior vertebral column resection for severe and rigid adult idiopathic scoliosis: a minimum 2-year follow-up comparative study. Spine (Phila Pa 1976) 2019;44:E1196-205.
- 11. Eskilsson K, Sharma D, Johansson C, et al. Pedicle subtraction osteotomy: a comprehensive analysis in 104 patients. Does the cause of deformity influence the outcome? J Neurosurg Spine 2017;27:56-62.
- 12. Khashan M, Raad M, El Dafrawy MH, et al. Postoperative changes in neurological function after 3-column osteotomy: risk factor analysis of 199 patients. J Neurosurg Spine 2019;30:568-73.
- 13. Dorward IG, Lenke LG, Stoker GE, et al. Radiographical and clinical outcomes of posterior column osteotomies in spinal deformity correction. Spine (Phila Pa 1976) 2014;39:870-80.
- 14. Buchowski JM, Bridwell KH, Lenke LG, et al. Neurologic complications of lumbar pedicle subtraction osteotomy: a 10-year assessment. Spine (Phila Pa 1976) 2007;32:2245-52.
- 15. Park JS, Kim J, Joo IH, et al. Analysis of risk factors for sagittal translation after pedicle subtraction osteotomy in patients with ankylosing spondylitis. Spine J 2018;18:1356-62.
- 16. Qian BP, Mao SH, Jiang J, et al. Mechanisms, predisposing factors, and prognosis of intraoperative vertebral subluxation during pedicle subtraction osteotomy in surgical correction of thoracolumbar kyphosis secondary to ankylosing spondylitis. Spine (Phila Pa 1976) 2017;42:E983-90.
- 17. Chang KW. Cantilever bending technique for treatment of large and rigid scoliosis. Spine (Phila Pa 1976) 2003;28:2452-8.
- 18. Qiao J, Xiao L, Sun X, et al. Vertebral subluxation during three-column osteotomy in surgical correction of adult spine deformity: incidence, risk factors, and complications. Eur Spine J 2018;27:630-5.
- 19. Chang KW, Chen HC, Chen YY, et al. Sagittal translation in opening wedge osteotomy for the correction of thoracolumbar kyphotic deformity in ankylosing spondylitis. Spine (Phila Pa 1976) 2006;31:1137-42.
- 20. 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 2016;98-B:1227-33.
- 21. Liu Z, Xu Y, Fan C, et al. Coronal imbalance in degenerative scoliosis with type A coronal alignment: an amendment to the Nanjing coronal imbalance classification. J Neurosurg Spine 2025;43:616-23.
- 22. Cunningham BW, Sefter JC, Hu N, et al. Biomechanical comparison of iliac screws versus interbody femoral ring allograft on lumbosacral kinematics and sacral screw strain. Spine (Phila Pa 1976) 2010;35:E198-205.
- 23. Li J, Hu Z, Qian Z, et al. The prognosis and recovery of major postoperative neurological deficits after corrective surgery for scoliosis: an analysis of 65 cases at a single institution. Bone Joint J 2022;104-B:103-11.
- 24. Chen C, Zhao Z, Li J, et al. Case report: transvertebral transposition of the spinal cord for recovery after paraplegia during kyphoscoliosis surgery. Front Neurol 2022;13:915188.
- 25. Okamoto Y, Wakama H, Matsuyama J, et al. Clinical significance of relative pelvic version measurement as a predictor of low back pain after total hip arthroplasty. Eur Spine J 2023;32:4452-63.
- 26. Lenke LG, O'Leary PT, Bridwell KH, et al. Posterior vertebral column resection for severe pediatric deformity: minimum two-year follow-up of thirty-five consecutive patients. Spine (Phila Pa 1976) 2009;34:2213-21.
- 27. Lin G, Chai X, Wang S, et al. Cross-sectional analysis and trend of vertebral and associated anomalies in Chinese congenital scoliosis population: a retrospective study of one thousand, two hundred and eighty nine surgical cases from 2010 to 2019. Int Orthop 2021;45:2049-59.
- 28. Lin G, Chai X, Wang S, et al. Cross-sectional analysis of associated anomalies and vertebral anomaly location in 1289 surgical congenital scoliosis. Eur Spine J 2021;30:3577-84.
- 29. Hyun SJ, Lenke LG, Kim YC, et al. Comparison of standard 2-rod constructs to multiple-rod constructs for fixation across 3-column spinal osteotomies. Spine (Phila Pa 1976) 2014;39:1899-904.
- 30. Luca A, Ottardi C, Sasso M, et al. Instrumentation failure following pedicle subtraction osteotomy: the role of rod material, diameter, and multi-rod constructs. Eur Spine J 2017;26:764-70.
- 31. Guevara-Villazón F, Boissiere L, Hayashi K, et al. Multiple-rod constructs in adult spinal deformity surgery for pelvic-fixated long instrumentations: an integral matched cohort analysis. Eur Spine J 2020;29:886-95.
- 32. Moniz-Garcia D, Stoloff D, Akinduro O, et al. Two- versus multi-rod constructs for adult spinal deformity: a systematic review and Random-effects and Bayesian meta-analysis. J Clin Neurosci 2023;107:9-15.
- 33. Dinizo M, Passias P, Kebaish K, et al. The approach to pseudarthrosis after adult spinal deformity surgery: is a multiple-rod construct necessary? Global Spine J 2023;13:636-42.
- 34. Han S, Hyun SJ, Kim KJ, et al. Rod stiffness as a risk factor of proximal junctional kyphosis after adult spinal deformity surgery: comparative study between cobalt chrome multiple-rod constructs and titanium alloy two-rod constructs. Spine J 2017;17:962-8.
- 35. Brown KM, Ludwig SC, Gelb DE. Radiographic predictors of outcome after long fusion to L5 in adult scoliosis. J Spinal Disord Tech 2004;17:358-66.
- 36. Hiyama A, Katoh H, Sakai D, et al. Changes in spinal alignment following eXtreme lateral interbody fusion alone in patients with adult spinal deformity using computed tomography. Sci Rep 2019;9:12039.
- 37. Wei XP, Yeh CW, Lin EE, et al. Predicting postoperative imbalance in adult spinal deformity staged surgery using predictive thresholds. Sci Rep 2025;15:32002.
- 38. Hiyama A, Sakai D, Katoh H, et al. Postoperative radiological improvement after staged surgery using lateral lumbar interbody fusion for preoperative coronal malalignment in patients with adult spinal deformity. J Clin Med 2023;12:2389.
- 39. Smith JS, Shaffrey E, Klineberg E, et al. Prospective multicenter assessment of risk factors for rod fracture following surgery for adult spinal deformity. J Neurosurg Spine 2014;21:994-1003.
- 40. Smith JS, Shaffrey CI, Ames CP, et al. Assessment of symptomatic rod fracture after posterior instrumented fusion for adult spinal deformity. Neurosurgery 2012;71:862-7.
, Jie Li2,*
