Warning: mkdir(): Permission denied in /home/virtual/lib/view_data.php on line 87 Warning: chmod() expects exactly 2 parameters, 3 given in /home/virtual/lib/view_data.php on line 88 Warning: fopen(/home/virtual/e-kjs/journal/upload/ip_log/ip_log_2025-04.txt): failed to open stream: No such file or directory in /home/virtual/lib/view_data.php on line 95 Warning: fwrite() expects parameter 1 to be resource, boolean given in /home/virtual/lib/view_data.php on line 96 Prevalence and Risk Factors for Postoperative Neurological Complications in Spinal Deformity Surgery: A Systematic Review and Proportional Meta-Analysis
  • Advanced Search
Neurospine Search

CLOSE
Man, Lui, Lam, Cheung, and Cheung: Prevalence and Risk Factors for Postoperative Neurological Complications in Spinal Deformity Surgery: A Systematic Review and Proportional Meta-Analysis

Original Article

Neurospine 2025;22(1):243-263.

Published online: March 31, 2025

DOI: https://doi.org/10.14245/ns.2449364.682

Prevalence and Risk Factors for Postoperative Neurological Complications in Spinal Deformity Surgery: A Systematic Review and Proportional Meta-Analysis

Yam Wa Man, Jedidiah Yui Shing Lui, Chor Yin Lam, Jason Pui Yin Cheung, Prudence Wing Hang Cheung

Department of Orthopaedics and Traumatology, The University of Hong Kong, Hong Kong SAR, China

Corresponding Author Prudence Wing Hang Cheung Department of Orthopaedics and Traumatology, The University of Hong Kong, 5/F, Professorial Block, Queen Mary Hospital, 102 Pokfulam Road, Pokfulam, Hong Kong SAR, China Email: gnuehcp6@hku.hk

Received December 14, 2024       Revised January 28, 2025       Accepted February 2, 2025

Copyright © 2025 by the Korean Spinal Neurosurgery Society

This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Objective

To investigate the incidence of postoperative neurological complications among patients who underwent spinal deformity surgery and to determine the significant risk factors for postoperative neurological complications.

Methods

Six databases PubMed, Web of Science, Scopus, MEDLINE, Embase, and Cochrane Library have been searched to identify observational studies from inception until January 2025. Inclusion criteria were patients aged ≥10 years with postoperative neurological complications after spinal deformity surgery. Stata/MP18.0 was used to conduct the meta-analysis in this review. The summary incidence estimates, proportion with 95% confidence intervals (CIs) and weights were pooled by the random-effects restricted maximum likelihood model.

Results

The search strategy identified 53 articles with 40,958 patients for final review. Overall incidence of postoperative neurological complications was 7% (95% CI, 5.0%–9.0%; p < 0.001; I2 = 98.34%) in which incidence estimates for patients with adult spinal deformity and underwent 3-column spinal osteotomies were 12% (95% CI, 9%–16%; p < 0.001; I2 = 93.17%) and 18% (95% CI, 8%–31%; p < 0.001; I2 = 94.68%) respectively. Preoperative neurological deficit was the risk factor with highest overall odds ratio (OR, 2.86; 95% CI, 1.85–4.41; p = 0.01; I2 = 76.20%), followed by the presence of kyphosis (OR, 1.13; 95% CI, 0.75–1.70; p = 0.02; I2 = 81.80%) and age at surgery (OR, 1.04; 95% CI, 1.01–1.08; p = 0.04; I2 = 68.80%).

Conclusion

Preoperative neurological deficit, the presence of kyphosis and age at surgery were significant risk factors for postoperative neurological complications. Therefore, comprehensive preoperative assessment and surgical planning are crucial to minimize the risk of developing postoperative neurological complications or the deterioration of pre-existing neurologic deficits.

Keywords: Spinal deformity surgery, Neurological complications, Postoperative complications, Spinal deformity

INTRODUCTION

Iatrogenic spinal cord injury is one of the clinical issues in spinal deformity surgery in terms of the clinical and socioeconomic burden. The incidence of iatrogenic spinal cord injury ranged from 0.24% to 12.50% in Norway and the United States [1-3]. It is also suggested the overall cost for patients with postoperative neurological complications in the first 2 years was around $112,970.54 in the United States [4]. Moreover, nerve root motor deficits and spinal cord injuries have been suggested as the most common postoperative neurologic deficits in spinal deformity surgeries [5,6]. Apart from the socioeconomic concerns, the devastating consequences including paraplegia and tetraplegia associated with the neurologic deficits are challenges of spinal deformity surgery despite the advancements in surgical techniques and intraoperative neurophysiological monitoring (IONM) techniques.
Common postoperative neurological complications including radiculopathy, motor or sensory deficits, and spinal cord injury, can result from neural tissue damage caused by direct contusion, compression, spinal cord ischemia or the misplacement of screws and rods, as well as insufficient removal of intervertebral discs and osteophytes during surgical procedures. Additionally, an increased number of vertebrae with instrumentation and higher preoperative pelvic tilt have been demonstrated as risk factors for neurologic deficits, while revision surgery with technical difficulties in anatomical considerations and surgical procedures contributed to a higher incidence of neurological complications [7,8].
Although the consequences and risks of postoperative neurologic deficits in spinal deformity surgery have been reported in the literature, there is a paucity of studies investigating the significance of the identified risk factors by statistical analysis. Therefore, a comprehensive review of the significant risk factors for postoperative neurological complications is necessary and this review aims to provide an overview of risk factors and the prevalence of postoperative neurologic deficits in spinal deformity surgery.

MATERIALS AND METHODS

This proportional meta‐analysis complied with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) 2020 [9], MOOSE (Meta‐analysis of Observational Studies in Epidemiology) reporting guidelines and SPIDER framework [10,11]. The review protocol was registered in the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY Number: INPLASY202450103).

1. Search Strategy

The electronic databases of PubMed, Web of Science, Scopus, MEDLINE, Embase, and Cochrane Library were searched from inception until 9th January 2025. The following queries were used to search the records: (‘neurological complication’ OR ‘neurologic complication’ OR ‘neurological deficit’ OR ‘neurologic deficit’ OR ‘neurological impairment’ OR ‘spinal cord injury’ OR ‘spinal cord damage’ OR ‘spinal cord deficit’ OR ‘spinal cord infarction’ OR ‘spinal cord ischemia’ OR ‘spinal cord dysfunction’ OR ‘spinal epidural hematoma’ OR ‘radiculopathy’ OR ‘myelopathy’ OR ‘motor deficit’ OR ‘sensory deficit’ OR ‘motor impairment’ OR ‘sensory impairment’ OR ‘cauda equina syndrome’ OR ‘nerve root injury’ OR ‘nerve root deficit’ OR ‘paraplegia’ OR ‘tetraplegia’ OR ‘hemiplegia’) AND (‘spinal deformity’ OR ‘scoliosis’ OR ‘kyphosis’ OR ‘lordosis’) AND (‘spinal deformity surgery’ OR ‘spinal correction surgery’) AND (‘postoperative’ OR ‘postsurgery’ OR ‘after surgery’ OR ‘following surgery’), followed by manual searching to identify additional relevant studies.

2. Inclusion Criteria and Exclusion Criteria

We selected retrospective and observational articles, including case-control and cohort studies that met the inclusion criteria. The inclusion criteria were as follows: (1) publications in a peer‐reviewed journal in English, (2) reported cases of postoperative neurological complications or deficits including spinal cord injury, motor deficit, sensory deficit, nerve root deficit or myelopathy in spinal deformity surgery, (3) reported sample size and age range or mean age of study population, (4) patients with diagnosis of spinal deformity. Two reviewers (YWM and JYSL) independently screened and reviewed the titles and abstracts of all searching records.
After the preliminary screening of the titles and abstracts, eligible full-text articles will be downloaded for close reading and evaluation of methodological quality using 9 items from the Joanna Briggs Institute (JBI) critical appraisal checklist for prevalence studies. Studies were excluded due to sampling bias, insufficient sample sizes, inappropriate statistical analysis, inconsistent measurements, poor reporting and low response rates without bias mitigation. The study was considered low methodological quality if one of the items was not fulfilled (graded with no, unclear, or not applicable). Meanwhile, searching results including review articles, conference abstracts, case reports, letters to the editor, reported cases of transient or traumatic neurological complications and studies with observations of pediatric patients aged below 10 years only were excluded from the study selection process. Any disagreements between the 2 reviewers were discussed to reach consensus.

3. Data Extraction

Required data were retrieved independently by 2 reviewers (YWM and JYSL) with a designated data extraction sheet: article titles, name of authors, year of publication, time horizon, study region or country, study design, population size, mean age or age range of population, types of spinal deformity (adolescent idiopathic scoliosis [AIS], adult spinal deformity [ASD] or studies including more than 1 type of spinal deformities), number of cases of neurological complications or neurologic deficits, surgical procedures or techniques, type of neurological complications, risk factors along with odds ratio (OR) and 95% confidence interval (CI) if presented.

4. Methodological Quality Assessment, Risk of Bias and Quality of the Evidence

The JBI critical appraisal checklist for prevalence studies was used to evaluate the methodological quality of the included articles reporting prevalence data [12]. The checklist has addressed 9 questions regarding the methodological quality and study bias of study designs, including sample size and data analysis, with respective instructions (Supplementary Table 1) [7,13-64]. The evaluation of each question was defined as yes, no, unclear, or not applicable. The study would be included when the information and data had been clearly stated in the article.
Meanwhile, the quality of nonrandomized studies was assessed using the Newcastle-Ottawa Scale (NOS), which evaluates 3 domains: selection of study groups, comparability of groups and outcome ascertainment in 8 items [65]. It considers factors such as sample representativeness, response rate, measurement of exposures or risk factors, control of confounding variables and statistical methods. The study would be regarded as good or high quality if it was rated more than 6 out of 9 points in total. The 2 reviewers performed the assessment for each included study independently, while a discussion with the third reviewer, who served as the final adjudicator, was arranged when there was an inconsistent response.
The quality of evidence of relevant outcomes (risk factors) in studies was categorized based on the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) guidance as high, moderate, low or very low. Four out of 5 domains, including risk of bias, inconsistency, indirectness and imprecision were assessed, whereas the evaluation of publication bias was not available among the observational studies. The rating would be downgraded from 1 to 3 levels when the risks were presented [66,67]. Further discussion for the confirmation of certainty of evidence among studies was conducted by 2 reviewers.

5. Data Analysis

Stata/MP18.0 (StataCorp LLC, College Station, TX, USA) was used to conduct the meta-analysis in this review. The summary incidence estimates, proportion with 95% CIs and weights were pooled by the random-effects restricted maximum likelihood model. Forest plots were performed by the meta suite in the Stata, which allows the evaluation of the effect size of included articles by using the Freeman-Tukey Double Arcsine Transformation. The heterogeneity of the meta-analysis was presented by the I-squared statistic (I2).
Pooled analysis followed by sensitivity analysis was conducted to further investigate the incidence of postoperative neurological complications in different types of spinal deformities and surgical procedures. To further investigate the significance of risk factors for postoperative neurological complications in spinal deformity surgery, subgroup analysis of common risk factors was conducted by evaluating the OR and 95% CIs. Studies involving complex spinal deformities, severe or rigid spinal deformities and with reported median age or age range have been selected for subgroup analyses. Patients aged 10 to 18 years were classified as the AIS group while patients aged above 18 years old were classified as the ASD group. Other studies that included both adult and pediatric patients were excluded from the subgroup analysis. Meta-regression analysis was then conducted to explore potential sources of heterogeneity and a leave-one-out analysis was performed to evaluate the change in overall effect size when one included study was excluded from the pooled analysis.
The evaluation of publication bias in the proportional meta-analysis is difficult as the proportional data is not adequately adjusted for the related tests such as Egger’s test and Begg’s test [68]. Meanwhile, the use of funnel plots would lead to erroneous or misleading results when pooling studies reporting the incidence or proportional data [69]. Nevertheless, non-parametric trim and fill analysis was conducted to evaluate the publication bias of this study. This method investigates the number of studies potentially missing from the analysis due to publication bias and adjusts the overall effect size estimate.

RESULTS

The study selection process was summarized in Fig. 1. There were 45,879 results identified in the initial search. After removing duplicate results (n= 15,514), titles and abstracts of 30,365 articles were screened. There were 263 full-text articles being reviewed (Fig. 1), of which 210 articles were excluded due to studies without reported incidence of postoperative neurological complications (n= 115), age range or mean age (n= 13) and articles with observations of pediatric patients only (n= 61), syndromic scoliosis or spinal deformities secondary to other diseases only (n= 21). As a result, 53 articles with 40,958 patients were included in the meta-analysis.

1. Methodological Quality Assessment, Risk of Bias and Quality of the Evidence

All studies included in the meta-analysis had a relatively low risk of bias in terms of methodological quality and study design. Responses to the assessment by the 2 reviewers were detailed in Supplementary Table 1. Meanwhile, all included studies scored greater than 6 out of 9 in the NOS assessment (Table 1), indicating the good quality of the observational studies. The details of GRADE assessment for risk factors, including patient’s age at surgery, body mass index (BMI), diabetes mellitus, the presence of kyphosis, blood loss and preoperative neurologic deficits, were also presented in Supplementary Table 2. The quality of evidence from cohort and case-control studies was upgraded to moderate quality due to the large sample size.

2. Study Characteristics and Publication Bias

All 53 included studies were published from 1994 to 2024, the main characteristics of each article were listed in Table 1. All included articles were retrospective studies, including 46 cohort studies and 7 case-control studies. They were all designed to investigate patients with postoperative neurologic deficits after undergoing spinal deformity correction surgery, of which 33 studies included the observations of perioperative neurological complications or nonneurological surgical complications. Surgical procedures such as spinal fusion and instrumentation (n = 13), sublaminar wiring and subtransverse process wiring (n= 1), 3-column spinal osteotomies (3-CO) (n= 5) including pedicle subtraction osteotomies (PSOs) (n = 1) and vertebral column resections (VCRs) (n= 7), posterior column osteotomies (PCOs) (n= 3) including Smith-Petersen osteotomy (SPO) and Ponte osteotomy were included, while more than one type of surgical techniques had been discussed in the remaining included studies (n= 23). Sample size ranged from 13 to 8,870. Patients with ASD, AIS or studies including more than one type of spinal deformities in the study cohort were recruited from all included studies. Risk factors for postoperative neurological complications with OR and 95% CIs had been reported in 9 studies.
The publication bias was evaluated by using the funnel plot and non-parametric trim and fill analysis. A symmetrical distribution of included studies was illustrated in the Supplementary Fig. 1, suggesting that the publication bias might not be present in this study (p= 0.546).

3. Incidence of Postoperative Neurological Complications

The incidence of postoperative neurological complications ranged from 0.15% to 38.46% among all included articles. Possible reasons for the large range of incidence were follow-up periods or population characteristics, and sample sizes varied among studies. The overall incidence of postoperative neurological complications was 7% (95% CI, 5.0%–9.0%; p < 0.001; I2= 98.34%) (Table 2), indicating the high heterogeneity among all included studies (Supplementary Fig. 2).
There were 47 and 28 studies eligible for the subgroup analyses based on the type of spinal deformities and surgical techniques. The pooled incidence estimates of AIS, ASD and studies including more than one type of spinal deformities were 0.3% (95% CI, 0.2%–0.5%; p= 0.0429; I2= 40.57%), 12% (95% CI, 9%–16%; p< 0.001; I2= 93.17%), and 4% (95% CI, 2%–7%; p < 0.001; I2= 97.56%), respectively (Fig. 2), while the overall pooled incidence estimate based on the type of spinal deformity was 7% (95% CI, 5%–9%; p< 0.001; I2= 98.56%). The difference among different types of spinal deformity subgroups was statistically significant.
The pooled incidence estimates of spinal fusion and instrumentation, PVCR or VCR, 3-CO, PCO, or Ponte osteotomy were 4% (95% CI, 2%–6%; p< 0.001; I2= 97.16%), 6% (95% CI, 4%–10%; p = 0.024; I2= 58.28%), 18% (95% CI, 8%–31%; p < 0.001; I2= 94.68%), 4% (95% CI, 0%–17%; p< 0.001; I2= 93.43%), respectively (Fig. 3). A wide 95% CIs of 3-CO subgroup indicated that the pooled incidence estimate was not precise, due to small sample size or insufficient outcomes of interest of those included studies. The overall pooled incidence estimate based on surgical procedures was 6% (95% CI, 4%–9%; p< 0.001; I2= 97.33%), indicating the difference between different surgical procedures was statistically significant.

4. Risk Factors for Postoperative Neurological Complications

The risk factors for postoperative neurological complications in spinal deformity surgery from the included studies were shown in Tables 1, 2 and Supplementary Table 3, with OR and 95% CIs. Among all included studies, risk factors including the male gender, PSO, coronal curve size, osteotomy index, diabetes mellitus, fusion to the sacrum, IONM alerts, the presence of kyphosis, blood loss, Cobb angle ≥ 70°, age at surgery, preoperative neurologic deficits, history of spine surgery, interbody fusion and BMI are potential risk factors for the incidence of postoperative neurological complications.
The overall ORs for patient’s age at surgery, BMI, diabetes mellitus, the presence of kyphosis, blood loss and preoperative neurologic deficits ranged from 1.00 to 2.86 (Supplementary Figs. 3–8). Notably, the highest overall OR was found in the preoperative neurologic deficit subgroup (OR, 2.86; 95% CI, 1.85– 4.41; p= 0.006; I2= 76.20%), suggesting that a higher incidence of postoperative neurological complications was observed in patients with pre-existing neurologic deficits. The presence of kyphosis (OR, 1.13; 95% CI, 0.75–1.70; p= 0.019; I2= 81.80%) should be taken into consideration during surgical planning, as well as patient’s age at surgery (OR, 1.04; 95% CI, 1.01–1.08; p= 0.041; I2= 68.80%). Blood loss was a potential risk factor (OR, 1.00; 95% CI, 1.00–1.01; p< 0.001; I2= 91.90%), whereas BMI (OR, 1.10; 95% CI, 1.04–1.16; p= 0.692; I2= 0%) and diabetes mellitus (OR, 1.82; 95% CI, 1.15–2.88; p = 0.286; I2= 20.10%) were not statistically significant.

5. Meta Regression and Leave-One-Out Analysis

Meta-regression analysis revealed that the sample size (p<0.001) was the significant factor associated with the high heterogeneity (Supplementary Fig. 9). The result of leave-one-out analysis indicated that the omitting of studies might not cause a substantial change in the overall pooled incidence of postoperative neurological complications (Supplementary Fig. 10) and the overall pooled incidence of postoperative neurological complications in spinal deformity surgery was 7%.

DISCUSSION

This systematic review and meta-analysis is the first comprehensive review to provide empirical evidence of significant risk factors and the incidence of postoperative neurological complications after spinal deformity surgery.

1. Incidence of Postoperative Neurological Complications in Subgroups of the Type of Spinal Deformity and Surgical Procedure

The estimated overall pooled incidence of postoperative neurological complications in spinal deformity surgery among the included studies was 7%. This is consistent with the findings in published studies (from 4.0% to 21.2%) conducted by Chen et al. [56] and Kelly et al. [49]. According to the types of spinal deformity, the overall incidence of postoperative neurological complications of ASD (12.0%) is higher than that in AIS (0.3%), or studies including more than one type of spinal deformities in the study cohort (4.0%). ASD had a higher risk of postoperative neurologic deficits due to the complex conditions associated with degenerative changes. Spinal deformities accompanied by degenerative disc disease, spinal stenosis or osteoporosis usually require long-segment spinal fusions, osteotomies or combined procedures for the surgical intervention. This will increase not only the difficulty of surgical procedures but also the risk of surgery-related postoperative neurologic deficits.
Our results also demonstrated that the overall incidence of postoperative neurological complications was higher when performing 3-CO (18.0%) than other surgical procedures, including spinal fusion and instrumentation (4.0%), PVCR or VCR (6.0%), PCO or Ponte osteotomy (4.0%). Higher incidence occurred for 3-CO used for the correction of complex or severe spinal deformities, this suggests that partial or complete removal of vertebral bodies for the correction of spinal alignment will increase the risk of postoperative neurological complications associated with the spinal cord compression or stretch-associated injury.

2. Postoperative Neurological Complications Associated With Preoperative Neurologic Deficits, the Presence of Kyphosis, Age at Surgery and Blood Loss

Patients with pre-existing neurologic deficits in which the spinal cord or nerve roots are compromised usually have a higher risk of developing postoperative neurological complications or deteriorating neurological conditions after surgical interventions. The spinal cord can be at risk during surgical procedures, particularly in cases of pre-existing spinal cord compression, dysfunction associated with spinal deformity, or ligamentous ossifications [56,70]. Spinal instrumentation may also cause encroachment or damage of the spinal cord [71,72]. Moreover, the impingement of bony structures into the spinal cord or stretching of the anterior spinal artery may lead to vascular injury while the excessive shortening of the spinal column may reduce blood flow to the spinal cord [73-75]. Apart from the potential risks of damage to the spinal cord, nerve root foraminal compression may also result from a shortening of the posterior column during osteotomies such as PSO and SPO [42].
In addition to preoperative neurologic deficits, the presence of kyphosis and the patient’s age at surgery are important risk factors contributing to postoperative neurological complications. It is known that the abnormal forward curvature of the spine in severe kyphosis would further increase the compression or tension to the neural structures including the spinal cord and nerve roots during the surgical procedures. This also increases the surgical complexity and leads to a higher risk for neurologic deficits. Furthermore, patients with the location of apex in the distal thoracic region are a high-risk population for postoperative neurological deterioration in the correction of kyphotic deformity [76]. Meanwhile, age-related postoperative neurological complications could be attributed to the senescence-induced stress asymmetry in the spine, ossification of the facet joints, decreased flexibility in spinal muscles and ligaments and reduced capacity of the spinal cord for stress tolerance and repair [56]. Possible causes of postoperative neurologic deficits also include intraoperative excessive blood loss with impaired blood circulation and insufficient oxygen supply to the spinal cord and obesity with free fatty acids-induced lipotoxicity and inflammation triggered by oxidative stress may lead to neurological dysfunction or neurodegeneration [19]. Our results of subgroup analysis also corroborated the association between the identified risk factors and the incidence of postoperative neurological complications, which are consistent across studies [21,47,51-53,55].
Preoperative assessment with comprehensive neurological assessment and tailored surgical strategies are crucial to reduce the risk of postoperative neurological complications and deterioration of pre-existing neurologic deficits by ensuring the optimal conditions for the surgery, especially for the high-risk population including elderly patients with degenerative scoliosis, osteoporosis or spondylolisthesis, as well as patients with preoperative neurologic deficits. Preoperative cardiovascular risk assessment and continuous glucose monitoring are necessary for patients with comorbidities, including cardiovascular disease and diabetes mellitus, to reduce the risk of perioperative adverse outcomes. The assessment of the preoperative neurologic deficits, anti-inflammatory therapy and infection control also allow for the improvement of patients’ preoperative neurological functions.
Meanwhile, preoperative neurologic deficits, spinal tumors or inflammatory lesions that can cause compression of the spinal cord or nerve roots should be identified with the assistance of magnetic resonance imaging (MRI) scans, whereas motor and sensory functions should be evaluated through neurological examination or diagnostic tests such as electromyography and nerve conduction test. IONM, including motor-evoked potential and somatosensory-evoked potential monitoring, can also be used to monitor intraoperative neurological status and evaluate the functional integrity of neural structures. Furthermore, minimizing kyphosis correction involves detailed surgical planning with age-adjusted alignment goals or gradual deformity correction, which could minimize the risk of overcorrection and neurological injuries and optimize clinical outcomes. Intraoperative measures, including the use of tranexamic acid for preventing excessive blood loss, reducing the blood transfusion rate, and improving the neurological outcomes are mandatory [77,78].

3. Sensitivity Analysis for High Heterogeneity Across Studies

The included studies had varying sample sizes (from 13 to 8,870 cases or patients). Despite the I2 value is related to a number of selected studies and pooled estimates of the meta-analysis, the high I2 value might not indicate the high heterogeneity across the studies. High heterogeneity is a common phenomenon in the meta-analysis of observational studies [79,80]. Therefore, the subgroup analysis and meta-regression analysis have been conducted to identify the possible factors contributing to the high heterogeneity.

4. Strengths and Limitations

This review has several limitations. Firstly, observational studies without sample randomization may contribute to selection bias and difficulty in establishing the cause-and-effect relationship between risk factors and postoperative neurological complications although the majority of included studies adopting the retrospective cohort study design. Meanwhile, retrospective studies were susceptible to publication bias which may lead to the overestimation of effects when negative findings had not been shown in the results. Establishing the associations between risk factors and study outcomes might also be limited by the recall bias and cofounding of the retrospective study designs.
Secondly, the reported postoperative neurologic deficits or neurological complications varied between studies with different study objectives, in which some studies included the observations of postoperative motor deficits only. The high heterogeneity across studies indicated a significant variation in study outcomes, while context-specific results increased the uncertainty and reduced the reliability of the pooled estimates. Further study with additional stratifications by study setting or surgeon expertise could help to address the sources of variability leading to the high heterogeneity. Adopting standardized protocols for reporting methodologies and results could also enhance the comparability across different studies. Nevertheless, some studies suggested that high heterogeneity is common in proportional meta-analysis due to the characteristics of proportional data, in which small variance is observed in studies with small sample sizes. Therefore, the high heterogeneity in the proportional meta-analysis does not imply inconsistent data exists or that study effects are dispersed across a wide range [68,81].
Thirdly, the use of random-effects models could lead to wider CIs. Therefore, the results should be interpreted together with subgroup analysis or meta-regression analysis for a more precise evaluation of the characteristics of the included studies [82]. A relatively large number of studies (n= 53) with a low risk of bias was included to enhance the authenticity of the study. Inclusion of articles, the assessment of study methodological quality and data extraction from the included articles were conducted independently by the 2 reviewers to optimize the objectivity of this study. The methodology applied in this proportional meta-analysis is another merit of the study design.

5. Strategies for Addressing High Heterogeneity, Confounding, Selection Bias in Meta-analysis

Although the high heterogeneity in the proportional meta-analysis is sometimes inevitable, conducting subgroup analysis or meta-regression using study-level characteristics is necessary to study the causes of the heterogeneity. Moreover, individual participant data meta-analysis is another approach that allows more comprehensive subgroup analyses according to the patient characteristics to improve the data quality and generalizability when the raw data are available [83]. This also provides stronger evidence to identify the risk factors that have a causal effect on the study outcomes. Apart from the study methodology of meta-analysis, propensity score matching in observational studies demonstrated similar effects as randomization can overcome the imbalances in patient-level covariates, control the confounding effects and reduce the selection bias [84,85].

CONCLUSION

This systematic review and meta-analysis reveals that preoperative neurologic deficit, the presence of kyphosis and the patient’s age are significant risk factors for postoperative neurological complications or neurological deterioration in spinal deformity surgery. This evidence can aid in predicting the risk of postoperative neurological complications in patients. Moreover, clinical guidelines involve comprehensive preoperative assessments and patient-specific surgical strategies are crucial for managing pre-existing neurologic deficits and comorbidities, particularly in high-risk populations such as elderly patients with degenerative spine conditions. Preoperative optimization includes anti-inflammatory therapy, infection control to improve the preoperative conditions of patients and MRI scans to allow a precise understanding of the lesions. Intraoperative measures such as the application of IONM, gradual kyphosis correction and the use of tranexamic acid could minimize the risk of postoperative neurological complications and improve the clinical outcomes in spinal deformity surgery.

Supplementary Materials

Supplementary Table 1.
Critical appraisal results for included studies using JBI critical appraisal checklist for prevalence studies
ns-2449364-682-Supplementary-Table-1.pdf
Supplementary Table 2.
The certainty assessment of evidences based on GRADE (Grading of Recommendations Assessment, Development, and Evaluation)
ns-2449364-682-Supplementary-Table-2.pdf
Supplementary Table 3.
Risk factors with reported odds ratios in included studies
ns-2449364-682-Supplementary-Table-3.pdf
Supplementary Fig. 1.
Funnel plot for publication bias. CI, confidence interval; REML, restricted maximum likelihood.
ns-2449364-682-Supplementary-Fig-1.pdf
Supplementary Fig. 2.
Forest plot of the overall pooled incidence estimates of postoperative neurologic complications in spinal deformity surgery. CI, confidence interval; REML, restricted maximum likelihood.
ns-2449364-682-Supplementary-Fig-2.pdf
Supplementary Fig. 3.
Forest plot representing odds ratios of postoperative neurologic complications associated with patient’s age at surgery. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-3.pdf
Supplementary Fig. 4.
Forest plot representing odds ratios of postoperative neurologic complications associated with body mass index. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-4.pdf
Supplementary Fig. 5.
Forest plot representing odds ratios of postoperative neurologic complications associated with diabetes mellitus. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-5.pdf
Supplementary Fig. 6.
Forest plot representing odds ratios of postoperative neurologic complications associated with the presence of kyphosis. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-6.pdf
Supplementary Fig. 7.
Forest plot representing odds ratios of postoperative neurologic complications associated with blood loss. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-7.pdf
Supplementary Fig. 8.
Forest plot representing odds ratios of postoperative neurologic complications associated with preoperative neurologic deficit. OR, odds ratio; CI, confidence interval.
ns-2449364-682-Supplementary-Fig-8.pdf
Supplementary Fig. 9.
Bubble plot for sample size of study. CI, confidence interval.
ns-2449364-682-Supplementary-Fig-9.pdf
Supplementary Fig. 10.
Leave-one-out analysis. CI, confidence interval; REML, restricted maximum likelihood.
ns-2449364-682-Supplementary-Fig-10.pdf

NOTES

Conflict of Interest

The authors have nothing to disclose.

Funding/Support

This work was supported by Department Seed Fund (#200011025), the University of Hong Kong under PWHC.

Author Contribution

Conceptualization: CYL, JPYC, PWHC; Data curation: YWM, JYSL; Formal analysis: YWM; Funding acquisition: PWHC; Methodology: YWM, JYSL, CYL, JPYC, PWHC; Project administration: PWHC; Visualization: YWM, JYL, CYL, JPYC, PWHC; Writing – original draft: YWM, JYSL, CYL, JPYC, PWHC; Writing – review & editing: YWM, JYSL, CYL, JPYC, PWHC.

Fig. 1.
Flow diagram detailing study selection process. Adapted from Page MJ, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.
ns-2449364-682f1.jpg
Fig. 2.
Forest plot of pooled incidence estimates of postoperative neurologic complications in spinal deformity surgery by type of spinal deformity subgroup. CI, confidence interval; REML, restricted maximum likelihood.
ns-2449364-682f2.jpg
Fig. 3.
Forest plot of pooled incidence estimates of postoperative neurologic complications in spinal deformity surgery by surgical procedure subgroup. CI, confidence interval; 3-CO, 3-column osteotomy; PCO, posterior column osteotomy; PVCR, posterior vertebral column resection; VCR, vertebral column resection; REML, restricted maximum likelihood.
ns-2449364-682f3.jpg
Table 1.
Characteristics of included studies (n=53)
Source/quality score (NOS) Study region Time horizon (follow-up period) Study design and methodology (multicenter/single center) Age (yr), mean/range Sample size (no. of cases of neurological complications) Subtype of spinal deformity/diagnosis/inclusion criteria (surgical techniques) Type of neurologic complications/study scope Risk factors reported
Zhang et al., [13] 2017/(6) China 2008-2013 (≥ 2 yr) Retrospective cohort (single center) 16.3 ± 6.4 62 (3 motor deficit; 4 sensory deficit; 1 paraplegia) Severe congenital spinal deformities (PVCR) Transient neurologic deficits and permanent neurologic deficits Age ≥ 18 yr; EBL 50%; pulmonary dysfunction
Kundnani et al., [14] 2010/(7) Singapore 2004–2008 (unknown) Retrospective cohort (single center) 13.6 ± 2.3 354 (2 motor deficits) AIS (spinal fusion and instrumentation) Postoperative neurologic deficits Intraoperative SSEP and NMEP alerts
Boachie-Adjei et al., [15] 2020/(7) USA 2011–2015 (2 yr) Retrospective cohort (multicenter) 14.6 ± 2.8 286 (27 new neurologic deficit) Complex spinal deformity with curves ≥ 100º (VCR) Postoperative neurologic deficits Complex deformities; osteotomy grade
Eskilsson et al., [16] 2017/(6) Sweden 2007–2012 (≥ 1 yr) Retrospective cohort (single center) 55.0 ± 16.0 104 (10 motor deficit) ASD (PSO) Postoperative complications N/A
Buckland et al., [17] 2018/(8) USA 2006–2016 (2 yr) Retrospective case-control (multicenter) 14.6 ± 2.2 2,210 (7 nerve root or spinal cord injury) AIS (PO) Perioperative neurologic deficits N/A
Bjerke et al., [18] 2017/(6) Africa Unknown (≥ 12 mo) Retrospective cohort (single center) 14 88 (4 motor deficit) Severe spinal deformity (PCO) Postoperative neurologic deficits Osteotomy
Vitale et al., [19] 2010/(6) USA 1995–2005 (unknown) Retrospective cohort (single center) 15.7 162 (2 new neurologic deficit) Idiopathic scoliosis; neuromuscular scoliosis; congenital scoliosis; kyphosis; syndromic scoliosis; spondylolisthesis (spinal fusion and instrumentation) Postoperative complications True electrophysical event; cardiopulmonary comorbidities
Stone et al., [20] 2024/(6) USA Unknown (2 yr) Retrospective cohort (multicenter) 14.4 ± 2.0 61 (2 incomplete spinal cord injury) Severe AIS with curves > 100º (PCO; VCR) Postoperative complications N/A
Boachie-Adjei et al., [21] 2014/(6) Africa Unknown (unknown) Retrospective cohort (single center) 14.3 145 (7 neurologic deficit) Spinal deformities (spinal fusion and instrumentation; SPO/PO; PSO; PVCR) Postoperative complications Osteotomy and resection procedures; kyphosis preoperative neurologic deficits; higher BMI; osteotomy index; coronal curve size
Van Halm-Lutterodt et al., [22] 2023/(6) China 2009–2015 (≥ 2 yr) Retrospective cohort (single center) 23.97 38 (6 neurologic deficit) Severe rigid scoliosis or kyphoscoliosis (SPO/ PO) Postoperative complications PO treatment
Fujioka et al., [23] 1994/(6) Japan Unknown (≥ 3 mo) Retrospective cohort (unknown) 5–57 134 (1 paraplegia; 1 monoplegia) Scoliosis (Harrington rod instrumentation; Dwyer’s operation; Zielke’s operation; Luque’s operation; Cotrel-Dubousset instrumentation) Postoperative neurological sequelae N/A
Li et al., [24] 2020/(6) USA Unknown (≥ 21 mo) Retrospective cohort (single center) 46.30 ± 19.17 124 (8 neurologic deficit) ASD (spinal fusion and instrumentation) Postoperative complications 3-CO
Miller et al., [25] 2019/(6) USA 2008–2015 (unknown) Retrospective cohort (single center) 5–79 125 (18 neurological deficit with 15 motor deficit) Thoracolumbar scoliosis (spinal fusion and instrumentation) Postoperative neurological deficits N/A
Qiao et al., [26] 2018(c)/(7) China 2000–2014 (≥ 2 yr) Retrospective cohort (single center) 46.2 ± 18.5 171 (1 nerve root injury; 4 spinal cord injury) ASD (3-CO) Postoperative neurological complications Osteotomy
Huang et al., [27] 2019/(6) China 2010–2015 (≥ 2 yr) Retrospective cohort (single center) 22.6 82 (1 incomplete paraplegia; 2 nerve root injury) Severe thoracic deformities (PVCR) Intraoperative alerts and postoperative neurological outcomes Osteotomy; osteotomy gap closure during PVCR
Bartley et al., [28] 2017/(8) USA 1995–2014 (≥ 2 yr) Retrospective cohort (multicenter) 14.8 ± 2.2 3,582 (9 nerve root injury; 5 spinal cord injury; 8 weakness/IONM alerts requiring intervention) AIS (spinal fusion and instrumentation) Perioperative and delayed complications N/A
Zhang et al., [29] 2018/(7) China 2012–2016 (≥ 2 yr) Retrospective cohort (single center) 61.07 ± 5.14 131 (1 nerve root injury; 4 sensory deficit; 7 radiculopathy; 3 peripheral nerve palsy) ADS (posterior lumbar interbody fusion combined with pedicle screw fixation) Postoperative complications N/A
Zhang et al., [30] 2024/(6) China 2020–2023 (≥ 1 yr) Retrospective case-control (single center) 36.53 ± 11.98 40 (8 postoperative neurologic deficit) Severe and complex ASD (spinal fusion and instrumentation; 3-CO) Postoperative complications N/A
Smith et al., [31] 2017(a)/(6) USA 2013–2015 (≥ 90 day) Retrospective cohort (multicenter) 62.3 23 (3 nerve root motor deficit/motor deficit; 1 nerve root sensory deficit; 1 radiculopathy; 2 mental status change; 1 other) ACD (3-CO) Postoperative complications N/A
Smith et al., [32] 2016/(7) USA 2008–2012 (≥ 2 yr) Retrospective cohort (multicenter) 56.2 291 (81 neurological complications) ASD (3-CO; decompression) Postoperative complications N/A
Qiao et al., [33] 2018(b)/(6) China 2002–2014 (unknown) Retrospective cohort (single center) 21–78 5,377 (8 neurologic deficit) Degenerative spinal deformity; idiopathic scoliosis; congenital scoliosis; kyphosis; neuromuscular scoliosis; syndromic scoliosis; ankylosing spondylitis; Scheuermann disease (PSO; VCR; SPO; instrumentation) Delayed postoperative neurologic deficits Age; osteotomy procedures
Papadopoulos et al., [34] 2015/(6) Africa 2002–2009 (≥ 2 mo) Retrospective cohort (single center) 14 45 (1 complete spinal cord injury; 1 nerve root injury) Congenital deformity; secondary to tuberculosis of the spine (PVCR) Intraoperative and Postoperative complications N/A
Williamson et al., [35] 2024/(6) USA and Canada 2018–2022 (6 wk) Retrospective case-control (multicenter) 61.0 ± 14.6 249 (51 neurological complications) Complex ASD (3-CO) Perioperative complications N/A
Coe et al., [36] 2006/(6) USA 2001–2003 (unknown) Retrospective cohort (unknown) 10–17 6,334 (31 neurological complications) AIS (spinal fusion and instrumentation) Perioperative complications N/A
Hariharan et al., [37] 2022/(6) USA Unknown (≥ 10 yr) Retrospective cohort (multicenter) 14.6 ± 2.1 284 (3 neurologic complications) AIS (spinal fusion and instrumentation) Perioperative complications N/A
Charosky et al., [7] 2012/(7) France 2002–2007 (≥ 1 yr) Retrospective cohort (multicenter) 63 306 (11 motor deficit; 7 sensory deficit; 1 cauda equina syndrome; 6 other) ASD (spinal fusion and instrumentation; PSO; SPO) Perioperative complications No. of instrumented vertebra fusion to the sacrum; PSO; high preoperative pelvic tilt ≥ 26°
Smith et al., [38] 2017/(6) USA Unknown (2 yr) Retrospective cohort (multicenter) 60.7 82 (8 radiculopathy; 8 motor deficit; 3 nerve root injury; 2 sensory deficit; 2 mental status change; 1 stroke) ASD (3-CO) Postoperative complications N/A
Akmes et al., [39] 2013/(6) Turkey Unknown (≥ 25 mo) Retrospective cohort (unknown) 9–34 64 (4 neurologic deficits) Idiopathic thoracic scoliosis (sublaminar wiring/subtransverse process wiring) Postoperative complications Sublaminar wiring technique
Boachie-Adjei et al., [40] 2021/(6) Africa Unknown (≥ 2 yr) Retrospective cohort (single center) 17.8 13 (5 motor deficit) Gamma deformity (halo gravity traction; VCR) Postoperative lower extremity motor deficits Myelopathic patients with thoracic deformities
Sacramento-Domínguez et al., [41] 2015/(6) USA 2002–2013 (unknown) Retrospective cohort (single center) 14 ± 6.5 98 (7 spinal cord injury) Severe kyphoscoliosis; hyperkyphosis (PVCR) Perioperative complications The level of apex; BMI
Kato et al., [42] 2018/(7) North America, Europe, and Asia 2011–2012 (2 yr) Retrospective cohort (multicenter) 56.8 265 (61 motor deficit) ASD (revision surgery; 3-CO) Perioperative lower extremity motor scores N/A
Wang et al., [43] 2024/(7) China 2010–2018 (≥ 1 yr) Retrospective cohort (single center) 20.3 ± 15.0 120 (1 nerve root injury; 1 spinal cord injury) Severe spinal deformities (PVCR) Intraoperative alerts and postoperative neurological complications N/A
Yamato et al., [44] 2017/(6) Japan 2011–2013 (≥ 3 mo) Retrospective cohort (multicenter) 57.7 1,192 (67 neurological disorder) ASD (spinal fusion and instrumentation) Perioperative neurological deficits N/A
Qiao et al., [45] 2018(a)/(6) China 2005–2014 (≥ 2 yr) Retrospective cohort (single center) 36.5 ± 19.7 109 (11 neurological complications) ASD (PSO/VCR) Perioperative complications N/A
Kara et al., [46] 2023/(6) Turkey 2017–2021 (2 yr) Retrospective cohort (unknown) 32 83 (5 neurologic deficit) Severe spinal deformity (PVCR; PSO; SPO) Perioperative neurologic deficits N/A
Li et al., [47] 2022/(7) China Unknown (≥ 2 yr) Retrospective cohort (single center) 25 8,870 (8 complete paraplegia; 57 incomplete paraplegia) AIS; congenital scoliosis; spinal deformity in NF-1; neuromuscular scoliosis (revision surgery; osteotomy; spinal fusion and instrumentation) Perioperative neurological deficits Preoperative neurological deficits; age; Cobb angle ≥ 70°
Mehta et al., [48] 2021/(8) USA 1995–2017 (≥ 2 yr) Retrospective cohort (multicenter) 10–23 3,464 (6 nerve root deficit; 3 spinal cord injury) AIS (spinal fusion and instrumentation) Perioperative complications N/A
Kelly et al., [49] 2014/(7) USA 2009–2011 (≥ 2 yr) Retrospective case-control (multicenter) 18–80 207 (18 postoperative sensory/motor deficit) Primary scoliosis, kyphosis or kyphoscoliosis with major Cobb angle ≥ 80° (spinal fusion and instrumentation; 3-CO) Perioperative complications N/A
Kim et al., [50] 2017/(7) USA 2008–2014 (≥ 6 wk) Retrospective case-control (multicenter) 57 564 (1 bowel/bladder dysfunction; 1 epidural hematoma; 2 femoral neuralgia; 14 mental status changes; 25 motor deficit; 2 myelopathy; 10 nerve root injury; 1 peroneal nerve palsy; 35 radiculopathy; 14 sensory deficit; 3 stroke; 8 other) ASD (decompression; osteotomy; interbody fusion; anterior thoracotomy; costotransversectomy; revision surgery) Perioperative neurologic complications Revision surgery; diabetes mellitus; male
Yagi et al., [51] 2019/(6) Japan Unknown (2 yr) Retrospective cohort (multicenter) 54.7 ± 18.6 285 (21 radiculopathy; 7 limb numbness; 1 hypoesthesia) ASD (PSO; revision surgery; spinal fusion and instrumentation) Postoperative neurological complications Age; Schwab-SRS type-type L; Schwab-SRS sagittal modifier-pelvis tilt “++”; diabetes mellitus
Mohanty et al., [52] 2023/(7) North America 2008–2021 (6 wk) Retrospective cohort (multicenter) 61.5 ± 1.12 205 (32 motor deficit) Adult congenital/degenerative/ idiopathic scoliosis; iatrogenic spinal deformity (PCO; PSO; VCR; decompression) Postoperative LEMS decline Age; higher BMI; higher frailty score; longer operating room times; greater estimated blood loss; IONM alerts; PSO
Kim et al., [53] 2012/(7) Korea 1997–2006 (≥ 2 yr) Retrospective cohort (single center) 33.5 233 (6 permanent neurologic deficit) Spinal deformities (PVCR; osteotomy) Perioperative complications More than 5 levels of fusion; preoperative neurologic deficits; blood loss; kyphosis; interbody fusion; history of spine surgery
Leong et al., [54] 2016/(6) UK 2006–2012 (unknown) Retrospective cohort (single center) 14.7 2,291 (6 permanent neurologic deficit) Spinal deformities (growing rod procedure; operations for hemivertebra) Perioperative neurologic complications Posterior corrections for kyphosis; scoliosis associated with a syndrome
Xie et al., [55] 2014/(6) China 2004–2011 (≥ 2 yr) Retrospective cohort (single center) 17.5 76 (6 postoperative neurologic deficit) Severe and rigid spinal deformities (PVCR) Perioperative neurologic complications Pre-existing neurologic dysfunction associated with intraspinal and brain stem anomalies; thoracic hyperkyphosis; levels of vertebral column resected
Chen et al., [56] 2020/(6) China 2006–2017 (≥ 2 yr) Retrospective cohort (unknown) 21 177 (22 neurological complications) Severe and rigid spinal deformities including scoliosis, kyphosis and kyphoscoliosis (gravity cranial ring traction; SPO; PSO; PVCR; BOBO; MVCR) Postoperative neurological complications and spinal cord function Age; etiology; spinal cord functional classification;severity of deformity and S-DAR; intraoperative kyphosis correction rate; osteotomy site and grade
Ushirozako et al., [57] 2018/(6) Japan 2010–2016 (≥ 1 yr) Retrospective cohort (unknown) 18–84 295 (1 motor deficit; 6 nerve root injury) ASD (3CO; lumbar lateral interbody fusion) Postoperative neurological complications Change in pelvic tilt of > 17.5°
Glassman et al., [58] 2018/(7) USA Unknown (≥ 2 yr) Retrospective cohort (multicenter) 40.2–78.5 138 (1 cauda equina; 1 spinal cord deficit; 7 motor deficit; 4 sensory deficit; 2 radiculopathy) Degenerative scoliosis (unknown) Perioperative complications N/A
Yoo et al., [59] 2024/(6) Korea 2018–2020 (> 2 yr) Retrospective case-control (single center) 68.8 ± 8.1 90 (9 myelopathy) ASD (spinal fusion and instrumentation) Delayed-onset neurological deficits N/A
Bianco et al., [60] 2014/(7) USA; France 1999–2012 (6 wk) Retrospective cohort (multicenter) 55.850 ± 12.842 423 (11 spinal cord deficit; 6 nerve root injury; 2 cauda equina syndrome; 51 motor deficit or paralysis) ASD (3-CO) Intraoperative and Postoperative complications Age (> 60 yr); 2 osteotomies; one thoracic osteotomy; major blood loss
Pateder and Kostuik [61] 2005/(7) USA 1992–2000 (≥ 3 yr) Retrospective cohort (unknown) 20–86 407 (12 nerve root injury) ASD (spinal fusion and instrumentation) Postoperative lumbar nerve root palsy Complex spinal deformities; staged and revision procedures
Riley et al., [62] 2018/(6) USA 2009–2011 (≥ 2 yr) Retrospective cohort (single center) 50.7 ± 16.0 74 (6 nerve root injury and 4 spinal cord injury) ASD (VCR; PSO; PCO; spinal fusion and instrumentation) Changes in health-related quality of life measures after surgery N/A
Pastorelli et al., [63] 2011/(6) Italy 2005–2009 (unknown) Retrospective case-control (single center) 18–70 172 (1 paraparesis) Idiopathic/congenital/syndromic scoliosis; kyphoscoliosis (spinal fusion and instrumentation) Transient and permanent postoperative neurologic deficits Severe scoliosis combined with kyphosis; pre-existing neurological deficits
Zhang et al., [64] 2018(b)/(7) China 2012–2017 (≥ 2 yr) Retrospective cohort (single center) 61.16 ± 5.15 153 (2 nerve root injury; 4 sensory deficit/ neuropathy; 7 radiculopathy; 4 peripheral nerve palsy; 2 delirium) Adult degenerative scoliosis (spinal fusion and instrumentation) Postoperative complications BMI; diabetes; smoking; hypertension; cardiac comorbidity

NOS, Newcastle-Ottawa Scale; PVCR, posterior vertebral column resection; EBL, estimated blood loss; AIS, adolescent idiopathic scoliosis; SSEP, somatosensory-evoked potentials; NMEP, neurogenic motor-evoked potentials; VCR, vertebral column resection; PSO, pedicle subtraction osteotomy; ASD, adult spinal deformity; PO, Ponte osteotomy; PCO, posterior column osteotomy; N/A, not applicable; SPO, Smith-Petersen osteotomy; 3-CO, 3-column osteotomy; ADS, adult degenerative scoliosis; ACD, adult cervical deformity; NF-1, neurofibromatosis type 1; LEMS, lower extremity motor score; IONM, intraoperative neurophysiological monitoring; MVCR, multisegmental vertebral column resection; BOBO, bone-disc bone osteotomy; S-DAR, sagittal deformity angular ratio; BMI, body mass index.

Table 2.
Results summary of pooled estimates and sensitivity analysis for subgroups
Variable Overall incidence estimate (%) 95% CI Heterogeneity (I2 values) (%) p-value
Pooled analysis 7 5–9 98.34 < 0.001
Sensitivity analysis
 Type of spinal deformities
  AIS 0.3 0.2–0.5 40.57 0.043
  ASD 12 9–16 93.17 < 0.001
  > 1 types 4 2–7 97.56 < 0.001
  Overall 7 5–9 98.56 < 0.001
 Surgical techniques
  Spinal fusion and instrumentation 4 2–6 97.16 < 0.001
  3-CO 18 8–31 94.68 < 0.001
  PCO/Ponte osteotomy 4 0–17 93.43 < 0.001
  PVCR/VCR 6 4–10 58.28 0.024
  Overall 6 4–9 97.33 < 0.001
Overall estimates in OR
Risk factors
 Preoperative neurologic deficits 2.86 1.85–4.41 76.20 0.006
 The presence of kyphosis 1.13 0.75–1.70 81.80 0.019
 Age 1.04 1.01–1.08 68.80 0.041
 Blood loss 1.00 1.00–1.01 91.90 < 0.001
 BMI 1.10 1.04–1.16 0.00 0.692
 Diabetes mellitus 1.82 1.15–2.88 20.10 0.286

OR, odds ratio; CI, confidence interval; AIS, adolescent idiopathic scoliosis; ASD, adult spinal deformity; 3-CO, 3-column osteotomy; PCO, posterior column osteotomy; PVCR, posterior vertebral column resection; VCR, vertebral column resection; BMI, body mass index.

REFERENCES

1. Romero-Muñoz LM, Segura-Fragoso A, Talavera-Díaz F, et al. Neurological injury as a complication of spinal surgery: incidence, risk factors, and prognosis. Spinal Cord 2020;58:318-23.
crossref pmid pdf
2. Daniels AH, Hart RA, Hilibrand AS, et al. Iatrogenic spinal cord injury resulting from cervical spine surgery. Global Spine J 2017;7(1 Suppl):84S-90S.
crossref pdf
3. Æsøy MS, Solvang SH, Grønning M, et al. Epidemiology of persistent iatrogenic spinal cord injuries in Western Norway. Brain Behav 2016;6:e00522.
pmid pmc
4. Williamson TK, Owusu-Sarpong S, Imbo B, et al. An economic analysis of early and late complications after adult spinal deformity correction. Global Spine J 2024;14:789-95.
crossref pmid pdf
5. Iorio JA, Reid P, Kim HJ. Neurological complications in adult spinal deformity surgery. Curr Rev Musculoskelet Med 2016;9:290-8.
crossref pmid pmc pdf
6. Kim HJ, Yao YC, Shaffrey CI, et al. Neurological complications and recovery rates of patients with adult cervical deformity surgeries. Global Spine J 2022;12:1091-7.
crossref pmid pdf
7. Charosky S, Guigui P, Blamoutier A, et al. Complications and risk factors of primary adult scoliosis surgery: a multicenter study of 306 patients. Spine (Phila Pa 1976) 2012;37:693-700.
pmid
8. Qiu Y, Wang S, Wang B, et al. Incidence and risk factors of neurological deficits of surgical correction for scoliosis: analysis of 1373 cases at one Chinese institution. Spine (Phila Pa 1976) 2008;33:519-26.
pmid
9. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.
crossref pmid pmc
10. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008-12.
crossref pmid
11. Cooke A, Smith D, Booth A. Beyond PICO: the SPIDER tool for qualitative evidence synthesis. Qual Health Res 2012;22:1435-43.
crossref pmid pdf
12. Munn Z, Moola S, Lisy K, et al. Methodological guidance for systematic reviews of observational epidemiological studies reporting prevalence and cumulative incidence data. Int J Evid Based Healthc 2015;13:147-53.
crossref pmid
13. Zhang BB, Zhang T, Tao HR, et al. Neurological complications of thoracic posterior vertebral column resection for severe congenital spinal deformities. Eur Spine J 2017;26:1871-7.
crossref pmid pdf
14. Kundnani VK, Zhu L, Tak H, et al. Multimodal intraoperative neuromonitoring in corrective surgery for adolescent idiopathic scoliosis: evaluation of 354 consecutive cases. Indian J Orthop 2010;44:64-72.
crossref pmid pmc pdf
15. Boachie-Adjei O, Duah HO, Yankey KP, et al. New neurologic deficit and recovery rates in the treatment of complex pediatric spine deformities exceeding 100 degrees or treated by vertebral column resection (VCR). Spine Deform 2021;9:427-33.
crossref pmid pdf
16. 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.
crossref pmid
17. Buckland AJ, Moon JY, Betz RR, et al. Ponte osteotomies increase the risk of neuromonitoring alerts in adolescent idiopathic scoliosis correction surgery. Spine (Phila Pa 1976) 2019;44:E175-80.
crossref pmid
18. Bjerke BT, Zuchelli DM, Nemani VM, et al. Prognosis of significant intraoperative neurophysiologic monitoring events in severe spinal deformity surgery. Spine Deform 2017;5:117-23.
crossref pmid
19. Vitale MG, Moore DW, Matsumoto H, et al. Risk factors for spinal cord injury during surgery for spinal deformity. J Bone Joint Surg Am 2010;92:64-71.
crossref pmid
20. Stone LE, Newton PO, Catanzano AA Jr, et al. Severe (>100 degrees) thoracic adolescent idiopathic scoliosis - a comparison of surgical approaches. Global Spine J 2024 Jun 21:21925682241264768doi: 10.1177/21925682241264768. [Epub].
crossref pdf
21. Boachie-Adjei O, Yagi M, Sacramento-Dominguez C, et al. Surgical risk stratification based on preoperative risk factors in severe pediatric spinal deformity surgery. Spine Deform 2014;2:340-9.
crossref pmid
22. Van Halm-Lutterodt NI, Pan A, Al-Saidi NN, et al. Postoperative complications following Schwab-grade-I versus Schwabgrade-II PCO in treating severe rigid kyphoscoliosis patients: comparative matched-group outcomes with minimum 2-year follow-up. Spine J 2023;23:1908-19.
crossref pmid
23. Fujioka H, Shimoji K, Tomita M, et al. Spinal cord potential recordings from the extradural space during scoliosis surgery. Br J Anaesth 1994;73:350-6.
crossref pmid
24. Li XJ, Lenke LG, Jin L, et al. Surgeon-specific risk stratification model for early complications after complex adult spinal deformity surgery. Spine Deform 2020;8:97-104.
crossref pmid pdf
25. Miller SM, Donegan SW, Voigt N, et al. Transcranial motorevoked potentials for prediction of postoperative neurologic and motor deficit following surgery for thoracolumbar scoliosis. Orthop Rev (Pavia) 2019;11:7757.
crossref pmid pmc pdf
26. 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.
crossref pmid pdf
27. Huang ZF, Chen L, Yang JF, et al. Multimodality intraoperative neuromonitoring in severe thoracic deformity posterior vertebral column resection correction. World Neurosurg 2019;127:e416-26.
crossref pmid
28. Bartley CE, Yaszay B, Bastrom TP, et al. Perioperative and delayed major complications following surgical treatment of adolescent idiopathic scoliosis. J Bone Joint Surg Am 2017;99:1206-12.
crossref pmid
29. Zhang XN, Sun XY, Meng XL, et al. Risk factors for medical complications after long-level internal fixation in the treatment of adult degenerative scoliosis. Int Orthop 2018;42:2603-12.
crossref pmid pdf
30. Zhang Y, Yang H, Han C, et al. Precise execution of personalized surgical planning using three-dimensional printed guide template in severe and complex adult spinal deformity patients requiring three-column osteotomy: a retrospective, comparative matched-cohort study. J Orthop Surg Res 2024;19:278.
crossref pmid pmc pdf
31. Smith JS, Shaffrey CI, Lafage R, et al. Three-column osteotomy for correction of cervical and cervicothoracic deformities: alignment changes and early complications in a multicenter prospective series of 23 patients. Eur Spine J 2017;26:2128-37.
crossref pmid pdf
32. Smith JS, Klineberg E, Lafage V, et al. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine 2016;25:1-14.
crossref pmid
33. Qiao J, Xiao L, Zhu Z, et al. Delayed postoperative neurologic deficit after spine deformity surgery: analysis of 5377 cases at 1 institution. World Neurosurg 2018;111:e160-4.
crossref pmid
34. Papadopoulos EC, Boachie-Adjei O, Hess WF, et al. Early outcomes and complications of posterior vertebral column resection. Spine J 2015;15:983-91.
crossref pmid
35. Williamson TK, Mir JM, Smith JS, et al. Contemporary utilization of three-column osteotomy techniques in a prospective complex spinal deformity multicenter database: implications on full-body alignment and perioperative course. Spine Deform 2024;12:1793-801.
crossref pmid pdf
36. Coe JD, Arlet V, Donaldson W, et al. Complications in spinal fusion for adolescent idiopathic scoliosis in the new millennium. A report of the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976) 2006;31:345-9.
crossref pmid
37. Hariharan AR, Shah SA, Petfield J, et al. Complications following surgical treatment of adolescent idiopathic scoliosis: a 10-year prospective follow-up study. Spine Deform 2022;10:1097-105.
pmid
38. Smith JS, Shaffrey CI, Klineberg E, et al. Complication rates associated with 3-column osteotomy in 82 adult spinal deformity patients: retrospective review of a prospectively collected multicenter consecutive series with 2-year follow-up. J Neurosurg Spine 2017;27:444-57.
crossref pmid
39. Akmeşe R, Kemal Us A. Comparison of subtransverse process wiring and sublaminar wiring in the treatment of idiopathic thoracic scoliosis. J Spinal Disord Tech 2013;26:79-86.
crossref pmid
40. Boachie-Adjei O, Sacramento-Dominguez C, Ayamga J, et al. Characterization of complex vertebral transposition (gamma deformity)>180 degrees: clinical and radiographic outcomes of halo gravity traction and vertebral column resection (VCR). Spine Deform 2021;9:411-25.
crossref pmid pdf
41. Sacramento-Domínguez C, Yagi M, Ayamga J, et al. Apex of deformity for three-column osteotomy. Does it matter in the occurrence of complications? Spine J 2015;15:2351-9.
crossref pmid
42. Kato S, Fehlings MG, Lewis SJ, et al. An analysis of the incidence and outcomes of major versus minor neurological decline after complex adult spinal deformity surgery: a subanalysis of scoli-RISK-1 study. Spine (Phila Pa 1976) 2018;43:905-12.
pmid
43. Wang S, He F, Guo L, et al. A critical event frequent lead to reversible spinal cord injury during vertebral column resection surgery. Eur Spine J 2024;33:3628-36.
crossref pmid pdf
44. Yamato Y, Matsuyama Y, Hasegawa K, et al. A Japanese nationwide multicenter survey on perioperative complications of corrective fusion for elderly patients with adult spinal deformity. J Orthop Sci 2017;22:237-42.
crossref pmid
45. Qiao J, Xiao L, Sun X, et al. Three column osteotomy for adult spine deformity: comparison of outcomes and complications between kyphosis and kyphoscoliosis. Br J Neurosurg 2018;32:32-6.
crossref pmid
46. Kara GK, Turan K, Çamurcu Y, et al. The neurologic deficit risk of three different kinds of spinal osteotomies and perioperative management. J Turk Spinal Surg 2023;34:49-53.
crossref
47. 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.
pmid
48. Mehta NN, Talwar D, Harms Study Group; et al. Unplanned return to the operating room (UPROR) after surgery for adolescent idiopathic scoliosis. Spine Deform 2021;9:1035-40.
crossref pmid pdf
49. Kelly MP, Lenke LG, Shaffrey CI, et al. Evaluation of complications and neurological deficits with three-column spine reconstructions for complex spinal deformity: a retrospective Scoli-RISK-1 study. Neurosurg Focus 2014;36:E17.
crossref pmc
50. Kim HJ, Iyer S, Zebala LP, et al. Perioperative neurologic complications in adult spinal deformity surgery: incidence and risk factors in 564 patients. Spine (Phila Pa 1976) 2017;42:420-7.
pmid
51. Yagi M, Michikawa T, Hosogane N, et al. Risk, recovery, and clinical impact of neurological complications in adult spinal deformity surgery. Spine (Phila Pa 1976) 2019;44:1364-70.
crossref pmid
52. Mohanty S, Hassan FM, Lenke LG, et al. Patient and procedural risk factors for decline in lower-extremity motor scores following adult spinal deformity surgery. J Neurosurg Spine 2023;39:157-67.
crossref pmid
53. Kim SS, Cho BC, Kim JH, et al. Complications of posterior vertebral resection for spinal deformity. Asian Spine J 2012;6:257-65.
crossref pmid pmc
54. Leong JJH, Curtis M, Carter E, et al. Risk of neurological injuries in spinal deformity surgery. Spine (Phila Pa 1976) 2016;41:1022-27.
crossref pmid
55. Xie JM, Zhang Y, Wang YS, et al. The risk factors of neurologic deficits of one-stage posterior vertebral column resection for patients with severe and rigid spinal deformities. Eur Spine J 2014;23:149-56.
crossref pmid pdf
56. Chen J, Shao XX, Sui WY, et al. Risk factors for neurological complications in severe and rigid spinal deformity correction of 177 cases. BMC Neurol 2020;20:433.
crossref pmid pmc pdf
57. Ushirozako H, Yoshida G, Kobayashi S, et al. Transcranial motor evoked potential monitoring for the detection of nerve root injury during adult spinal deformity surgery. Asian Spine J 2018;12:639-47.
crossref pmid pmc pdf
58. Glassman SD, Bridwell KH, Shaffrey CI, et al. Health-related quality of life scores underestimate the impact of major complications in lumbar degenerative scoliosis surgery. Spine Deform 2018;6:67-71.
crossref pmid pmc
59. Yoo SJ, Jang HJ, Moon BJ, et al. Application of transverse process hooks at distal thoracic vertebrae in uppermost vertebral instrumentation for adult spinal deformity surgery: special focus on delayed-onset neurologic deficits. Neurospine 2024;21:1219-29.
crossref pmid pmc pdf
60. Bianco K, Norton R, Schwab F, et al. Complications and intercenter variability of three-column osteotomies for spinal deformity surgery: a retrospective review of 423 patients. Neurosurg Focus 2014;36:E18.
crossref
61. Pateder DB, Kostuik JP. Lumbar nerve root palsy after adult spinal deformity surgery. Spine (Phila Pa 1976) 2005;30:1632-6.
crossref pmid
62. Riley MS, Bridwell KH, Lenke LG, et al. Health-related quality of life outcomes in complex adult spinal deformity surgery. J Neurosurg Spine 2018;28:194-200.
crossref pmid
63. Pastorelli F, Di Silvestre M, Plasmati R, et al. The prevention of neural complications in the surgical treatment of scoliosis: the role of the neurophysiological intraoperative monitoring. Eur Spine J 2011;20 Suppl 1(Suppl 1):S105-14.
crossref pmid pdf
64. Zhang XN, Sun XY, Hai Y, et al. Incidence and risk factors for multiple medical complications in adult degenerative scoliosis long-level fusion. J Clin Neurosci 2018;54:14-9.
crossref pmid
65. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603-5.
crossref pmid pdf
66. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336:924-6.
crossref pmid pmc
67. Piggott T, Morgan RL, Cuello-Garcia CA, et al. Grading of Recommendations Assessment, Development, and Evaluations (GRADE) notes: extremely serious, GRADE’s terminology for rating down by three levels. J Clin Epidemiol 2020;120:116-20.
crossref pmid
68. Barker TH, Migliavaca CB, Stein C, et al. Conducting proportional meta-analysis in different types of systematic reviews: a guide for synthesisers of evidence. BMC Med Res Methodol 2021;21:189.
crossref pmid pmc pdf
69. Cheema HA, Shahid A, Ehsan M, et al. The misuse of funnel plots in meta-analyses of proportions: are they really useful? Clin Kidney J 2022;15:1209-10.
crossref pmid pmc pdf
70. Ahn H, Fehlings MG. Prevention, identification, and treatment of perioperative spinal cord injury. Neurosurg Focus 2008;25:E15.
crossref
71. Wang S, Qiu Y, Liu W, et al. The potential risk of spinal cord injury from pedicle screw at the apex of adolescent idiopathic thoracic scoliosis: magnetic resonance imaging evaluation. BMC Musculoskelet Disord 2015;16:310.
crossref pmid pmc pdf
72. Chang KW, Cheng CW, Chen HC, et al. Correction hinge in the compromised cord for severe and rigid angular kyphosis with neurologic deficits. Spine (Phila Pa 1976) 2009;34:1040-5.
crossref pmid
73. Kawahara N, Tomita K, Kobayashi T, et al. Influence of acute shortening on the spinal cord: an experimental study. Spine (Phila Pa 1976) 2005;30:613-20.
crossref pmid
74. Ampe N, Meersman P, Ghysen D, et al. Lumbar pedicular stress fracture post-laminectomy: a case report. SN Compr Clin Med 2022;4:221.
crossref pmid pmc pdf
75. Park J, Lee JB, Park JY, et al. Spinal cord infarction after decompressive laminectomy for spontaneous spinal epidural hematoma--case report. Neurol Med Chir (Tokyo) 2007;47:325-7.
crossref pmid
76. Kalidindi KKV, Sath S, Sharma JK, et al. Incidence and risk factors for neurological deterioration in posterior corrective surgeries for severe angular kyphotic deformities: a 10-year institutional retrospective study. Global Spine J 2022;12:1199-207.
crossref pmid pdf
77. Morte D, Lammers D, Bingham J, et al. Tranexamic acid administration following head trauma in a combat setting: does tranexamic acid result in improved neurologic outcomes? J Trauma Acute Care Surg 2019;87:125-9.
crossref pmid
78. Hong P, Liu R, Rai S, et al. Does Tranexamic acid reduce the blood loss in various surgeries? An umbrella review of state-of-the-art meta-analysis. Front Pharmacol 2022;13:887386.
crossref pmid pmc
79. Metelli S, Chaimani A. Challenges in meta-analyses with observational studies. Evid Based Ment Health 2020;23:83-7.
crossref pmid pmc
80. Migliavaca CB, Stein C, Colpani V, et al. Meta-analysis of prevalence: I2 statistic and how to deal with heterogeneity. Res Synth Methods 2022;13:363-7.
pmid
81. IntHout J, Ioannidis JP, Rovers MM, et al. Plea for routinely presenting prediction intervals in meta-analysis. BMJ Open 2016;6:e010247.
crossref pmid pmc
82. Cordero CP, Dans AL. Key concepts in clinical epidemiology: detecting and dealing with heterogeneity in meta-analyses. J Clin Epidemiol 2021;130:149-51.
crossref pmid
83. Costa ALF, Lopes SLPC. Highlighting the benefits and disadvantages of individual participant data meta-analysis in radiology. Radiol Imaging Cancer 2024;6:e240018.
crossref pmid pmc
84. Yang JY, Webster-Clark M, Lund JL, et al. Propensity score methods to control for confounding in observational cohort studies: a statistical primer and application to endoscopy research. Gastrointest Endosc 2019;90:360-9.
crossref pmid pmc
85. Ye Y, Kaskutas LA. Using propensity scores to adjust for selection bias when assessing the effectiveness of alcoholics anonymous in observational studies. Drug Alcohol Depend 2009;104:56-64.
crossref pmid pmc
  • LG
    • TOOLS
    Share :
    Facebook Twitter Linked In Google+
    METRICS Graph View
    • 0 Crossref
    •   Scopus
    • 679 View
    • 27 Download
    Journal Impact Factor 3.8
    SURGERY: Q1
    CLINICAL NEUROLOGY: Q1
    Related articles in NS

    A Commentary on “Does Vertebral Cement Augmentation Reduce Postoperative Proximal Junction Complications in Spinal Deformity Corrective Surgery: A Systematic Review and Meta-analysis”2025 March;22(1)

    Does Vertebral Cement Augmentation Reduce Postoperative Proximal Junction Complications in Spinal Deformity Corrective Surgery: A Systematic Review and Meta-analysis2025 March;22(1)

    Incidence and Risk Factors of Postoperative Ileus in Oblique Lumbar Interbody Fusion Surgery: A Retrospective Study2025 March;22(1)

    Spinal Robotics in Adult Spinal Deformity Surgery: A Systematic Review2024 March;21(1)

    The Role of Exercise in the Alleviation of Neuropathic Pain Following Traumatic Spinal Cord Injuries: A Systematic Review and Meta-analysis2023 September;20(3)

    ABOUT
    SUBSPECIALTY

    Browse all articles >

    ARTICLE CATEGORY

    Browse all articles >

    BROWSE ARTICLES
    AUTHOR INFORMATION
    Editorial Office
    Department of Neurosurgery, CHA Bundang Medical Center,
    CHA University School of Medicine,
    59 Yatap-ro, Bundang-gu, Seongnam 13496, Korea
    Tel: +82-31-780-1924  Fax: +82-31-780-5269  E-mail: support@e-neurospine.org
    The Korean Spinal Neurosurgery Society
    #407, Dong-A Villate 2 Town, 350 Seocho-daero, Seocho-gu, Seoul 06631, Korea
    Tel: +82-2-585-5455  Fax: +82-2-2-523-6812  E-mail: ksns1987@neurospine.or.kr
    Business License No.: 209-82-62443

    Copyright © The Korean Spinal Neurosurgery Society.

    Developed in M2PI

    Zoom in Close layer