Abstract
Study Design:
Multi-center, prospective, observational cohort.
Objective:
To compare myelopathic vs. non-myelopathic ambulatory patients in short- and long-term neurologic function, operative treatment, and patient-reported outcomes.
Methods:
Pediatric deformity patients from 16 centers were enrolled with the following inclusion criteria: aged 10-21 years-old, a Cobb angle ≥100° in either the coronal or sagittal plane or any sized deformity with a planned 3-column osteotomy, and community ambulators. Patients were dichotomized into 2 groups: myelopathic (abnormal preoperative neurologic exam with signs/symptoms of myelopathy) and non-myelopathic (no clinical signs/symptoms of myelopathy).
Results:
Of 311 patients with an average age of 14.7 ± 2.8 years, 29 (9.3%) were myelopathic and 282 (90.7%) were non-myelopathic. There was no difference in age (P = 0.18), gender (P = 0.09), and Risser Stage (P = 0.06), while more patients in the non-myelopathic group had previous surgery (16.1% vs. 3.9%; P = 0.03). Mean lower extremity motor score (LEMS) in myelopathic patients increased significantly compared to baseline at every postoperative visit: Baseline: 40.7 ± 9.9; Immediate postop: 46.0 ± 7.1, P = 0.02; 1-year: 48.2 ± 3.7, P < 0.001; 2-year: 48.2 ± 7.7, P < 0.001). The non-myelopathic group had significantly higher LEMS immediately postoperative (P = 0.0007), but by 1-year postoperative, there was no difference in LEMS between groups (non-myelopathic: 49.3 ± 3.6, myelopathic: 48.2 ± 3.7, P = 0.10) and was maintained at 2-years postoperative (non-myelopathic: 49.2 ± 3.3, myelopathic: 48.2 ± 5.7, P = 0.09). Both groups improved significantly in all SRS domains compared to preoperative, with no difference in scores in the domains for pain (P = 0.12), self-image (P = 0.08), and satisfaction (P = 0.83) at latest follow-up.
Conclusion:
In severe spinal deformity pediatric patients presenting with preoperative myelopathy undergoing spinal reconstructive surgery, myelopathic patients can expect significant improvement in neurologic function postoperatively. At 1-year and 2-year postoperative, neurologic function was no different between groups. While non-myelopathic patients had significantly higher postoperative outcomes in SRS mental-health, function, and total-score, both groups had significantly improved outcomes in every SRS domain compared to preoperative.
Keywords: severe pediatric spinal deformity, pediatric spine surgery, scoliosis, myelopathy, neurologic function, neurologic deficit
Introduction
Patients with severe pediatric spinal deformity often require complex spinal reconstruction to maintain neurologic function and prevent significant functional deterioration.1-5 Preoperative neurologic deficits are seen in as many as 15%-23% of pediatric patients with complex deformities.3,6,7 Due to the severe nature of these deformities, many patients have undergone prior spinal surgery and present with compromised pulmonary function, and/or neurologic deficits, elevating the risk of intraoperative or postoperative complications.1-3,6 Patient characteristics such as larger coronal and sagittal Cobb angles, rigid deformities, high deformity angular ratios (DAR), and abnormal spinal cord morphology are risk factors for intraoperative neuromonitoring data changes and/or loss.2,3,6,8-10 In addition to these risk factors, the presence of preoperative neurologic deficits can significantly increase the odds of postoperative complications.3,6
Though many pediatric patients with severe deformity requiring spinal reconstruction have a neuromuscular condition and utilize a wheelchair for mobility, ambulators represent a starkly different cohort. Myelopathic, ambulatory patients with severe spinal deformity represent a population at high-risk for further loss of neurologic function.2,3,6 Furthermore, multiple studies have identified preoperative neurologic deficits as a significant independent risk factor for intraoperative and postoperative complications. Learning more about this high-risk population undergoing major spinal reconstruction surgery may benefit patients and surgeons in both the preoperative counseling and planning stages to adequately manage patient expectations. Thus, in ambulatory patients undergoing major pediatric spinal deformity reconstructive surgery, we sought to compare the short- and long-term perioperative outcomes, neurologic function, and patient-reported outcomes in myelopathic versus non-myelopathic patients.
Methods
Study Design
The Fox Pediatric Spinal Deformity Study was a prospective, observational, multi-center study to determine both radiographic and clinical outcomes of complex spinal deformity surgery in pediatric patients. Across 16 centers, 311 pediatric patients enrolled and fit the following inclusion criteria: aged 10-21 years-old, a Cobb angle of at least 100° in either the coronal or sagittal plane or any sized deformity with a planned 3-column osteotomy, and community ambulators with or without assistive devices, with an exception if the patient’s declining motor function was due to myelopathy secondary to spinal deformity related cord compression/stretch across the apex. Thus, the only way for patients to be included in this study without a Cobb angle of at least 100° in either plane was if they were undergoing a 3-column osteotomy. Patients were excluded if they used a wheelchair full-time for mobility or were unable/unwilling to commit to routine follow-up visits. Detailed demographic, preoperative, surgical, and postoperative data was collected at baseline, first postoperative erect, 1-year, and 2-year postoperative.
Data Collection
Demographic variables included age, sex, Risser stage, diagnosis, neurologic function, and surgical history. Radiographic variables assessed include coronal and sagittal Cobb angles, thoracic kyphosis (TK), lumbar lordosis (LL), pelvic incidence (PI), pelvic tilt (PT), coronal and sagittal DARs, sagittal vertical axis (SVA), and coronal vertical axis (CVA). Operative variables included preoperative halo-gravity traction (HGT), staged surgery total levels fused, 3-column osteotomy (3CO), estimated blood loss (EBL), operative duration, intraoperative neuromonitoring changes (IONM), and length of stay. Patient-reported outcomes (PROs) were assessed and compared using the Scoliosis Research Society-22r (SRS22r) questionnaire.
Presence of Myelopathy
A diagnosis of preoperative myelopathy was determined by the treating surgeon based on symptomology, clinical examination, and magnetic resonance imaging (MRI) scan findings. Symptomatology included impaired balance, difficulty walking due to spasticity, or bowel/bladder abnormalities. Clinical examination signs included presence of weakness on the lower extremity motor score (LEMS) which has a 50 point maximum score, 11 sensory deficits, or hyperreflexia (Hoffman’s reflex, clonus, or up-going Babinski test). MRI abnormalities included canal stenosis, spinal cord compression/stretch, spinal cord thinning, Chiari malformations, syringomyelia, and/or tethered spinal cords. If patients met any combination of the symptoms and clinical examination signs along with positive MRI findings, they were classified as myelopathic. Patients not meeting such criteria were allotted into the non-myelopathic group. Of note, non-myelopathic patients could still have motor deficits if there were symptoms or signs of lumbar nerve root compression causing motor weakness.
Outcomes and Statistical Analysis
The primary outcome of interest was postoperative neurologic function determined by LEMS score. Secondary outcomes of interest included operative data, radiographic data, and PROs using the Scoliosis Research Society-22r (SRS22r) questionnaire.
Preoperative, perioperative, and postoperative variables of the myelopathic vs. non-myelopathic groups of patients were compared. Statistical analysis included independent and paired t-tests for continuous variables and chi-square or fisher’s exact test for categorical variables as appropriate. SAS version 9.4 (SAS Institute Inc., Cary, NC) was used for all statistical analysis with a significance level at P < 0.05. Institutional review board approval was obtained at each respective center (protocol #AAAI2200 for lead author’s institution) and informed consent was given by every participant enrolled.
Results
Preoperative Demographics
Of the 311 patients, 29 (9.3%) were myelopathic with an average age of 15.3 ± 2.8 years, and 282 (90.7%) were non-myelopathic with an average age of 14.6 ± 2.7 years. There was no statistical difference in age (P = 0.18), gender (P = 0.09), and Risser Stage (P = 0.06). More patients in the non-myelopathic group had previous surgery (non-myelopathic: 16.1% vs. myelopathic: 3.9%; P = 0.03) along with a diagnosis of scoliosis (non-myelopathic: 44.3% vs. myelopathic: 20.7%; P = 0.017), while the myelopathic patients had a higher frequency of a kyphoscoliosis diagnosis (non-myelopathic: 31.9% vs. myelopathic: 55.2%; P = 0.022) (Table 1).
Table 1.
Demographic and Preoperative Data.
| Variables (mean ± SD or n (%)) | Myelopathy (n = 29) | No myelopathy (n = 282) | P-valuea |
|---|---|---|---|
| Age (yrs) | 15.3 ± 2.8 | 14.6 ± 2.7 | 0.18 |
| Male | 8 (27.6) | 123 (43.6) | 0.09 |
| Risser | 3.7 ± 1.8 | 3.0 ± 1.9 | 0.06 |
| Diagnosis | 0.022 | ||
| Kyphosis | 7 (24.1) | 66 (23.4) | |
| Scoliosis | 6 (20.7) | 125 (44.3) | |
| Kyphoscoliosis | 16 (55.2) | 90 (31.9) | |
| Syndromic comorbidity | 9 (31.0) | 64 (22.7) | 0.32 |
| Prior surgery | 12 (41.4) | 50 (17.7) | 0.03 |
| Preop Halo Gravity Traction | |||
| Used | 16 (55.2) | 110 (39.0) | 0.09 |
| Time (days) | 61.6 ± 48.4 | 64.1 ± 42.4 | 0.87 |
| Preop Motor score | |||
| Upper extremities | 49.4 ± 1.3 | 48.3 ± 8.5 | 0.48 |
| Lower extremities | 40.7 ± 9.9 | 48.2 ± 8.2 | <0.001 |
aIndependent t-test for continuous variables and chi-square test for categorical variables. Boldface indicates significant P values.
Operative and Perioperative Data
There were no significant differences in length of surgery (P = 0.14), levels fused (P = 0.41), EBL (P = 0.60), or staged surgeries (P = 0.34); however, the myelopathic patients had a significantly longer postoperative length of stay (13.0 ± 10.0 vs. 8.6 ± 5.0; P < 0.001) (Table 2). Myelopathic patients underwent a VCR in 20/29 (69.0%) cases, somewhat more commonly than non-myelopathic patients who underwent a VCR in 143/283 (50.5%) cases (P = 0.045). Of the remaining 9 myelopathic patients not treated with a VCR, 6/9 (66.7%) underwent preoperative HGT for an average 105 ± 47.6 days. Surgical treatment consisted of Schwab type 1 12 (3/9) or Schwab type 2 12 (6/9) osteotomies, while 4/9 (44.4%) also underwent a thoracoplasty.
Table 2.
Intraoperative and Postoperative Data.
| Variables (mean ± SD or n (%)) | Myelopathy (n = 29) | No myelopathy (n = 282) | P-valuea |
|---|---|---|---|
| Levels of fusion | 12.7 ± 11.9 | 12.3 ± 12.0 | 0.41 |
| Staged | 6 (20.7) | 64 (22.7) | 0.37 |
| Length of surgery (hrs) | 7.4 ± 2.4 | 6.4 ± 3.6 | 0.14 |
| Estimated blood loss (cc) | 1209 ± 1041 | 1307 ± 940 | 0.60 |
| VCR | 20 (69.0) | 143 (50.5) | 0.045 |
| IONM Changes (n used) | |||
| SSEP (302) | 9 (31.0) | 50 (17.7) | .02 |
| TcMEP (259) | 14 (48.3) | 97 (34.4) | .02 |
| DNEP (55) | 2 (6.9) | 7 (2.5) | .18 |
| Length of stay (days) | 13.0 ± 10.0 | 8.6 ± 5.0 | <0.001 |
aBoldface indicates significant P values.
Radiographic Data
There was no difference between groups in the magnitude of preoperative coronal radiographic major curves (P = 0.32), secondary curves (P = 0.22), or in the coronal-DAR (C-DAR) (P = 0.67) (Table 3). Although, myelopathic patients had a larger maximum kyphosis (non-myelopathic: 90.2° ± 37.7, myelopathic: 112.5° ± 37.6, P = 0.003), as well as a significantly higher sagittal-DAR (S-DAR) in the myelopathic group (19.1 ± 8.5) compared to the non-myelopathic group (13.6 ± 8.5; P = 0.001), emphasizing that greater angular kyphosis was present in the myelopathic patients. There was no difference in preoperative spinopelvic parameters (PI, PT, SS), SVA, CVA, nor trunkshift. Postoperatively, the non-myelopathic group had a significantly lower maximum kyphosis (non-myelopathic: 51.3° ± 24.2, myelopathic: 60.6° ± 26.3, P = 0.035), S-DAR (non-myelopathic: 5.7 ± 3.5, myelopathic: 7.9 ± 4.2, P = 0.0051), and CVA (non-myelopathic: 0.5 ± 2.5 cm, myelopathic: 1.7 ± 4.4 cm, P = 0.027) (Table 3).
Table 3.
Preoperative and Postoperative Radiographic Data.
| Preoperative | Postoperative | |||||
|---|---|---|---|---|---|---|
| Variables (mean ± SD or n (%)) | Myelopathy | No myelopathy | P-value* | Myelopathy | No myelopathy | P-valuea |
| Regional Parameters | ||||||
| Major Curve Cobb Angle (°) | 79.3 ± 39.3 | 87.5 ± 42.3 | 0.32 | 42.7 ± 17.8 | 43.4 ± 24.2 | 0.90 |
| Secondary Curve Cobb Angle (°) | 51.3 ± 30.6 | 57.9 ± 26.3 | 0.22 | 32.5 ± 19.3 | 33.1 ± 20.2 | 0.89 |
| Maximum Kyphosis (°) | 112.5 ± 37.6 | 90.2 ± 37.7 | 0.0031 | 60.6 ± 26.3 | 51.3 ± 24.2 | 0.035 |
| Coronal DAR | 11.5 ± 7.1 | 12.1 ± 6.3 | 0.67 | 5.5 ± 2.9 | 5.8 ± 3.6 | 0.76 |
| Sagittal DAR | 19.1 ± 8.5 | 13.6 ± 8.5 | 0.001 | 7.9 ± 4.2 | 5.7 ± 3.5 | 0.0051 |
| Spinopelvic Parameters | ||||||
| Pelvic Incidence (°) | 35.7 ± 20.3 | 40.1 ± 13.3 | 0.17 | 40.5 ± 16.5 | 43.4 ± 13.5 | 0.41 |
| Pelvic Tilt (°) | 14.4 ± 11.7 | 10.3 ± 9.3 | 0.058 | 9.8 ± 8.9 | 10.4 ± 9.0 | 0.80 |
| Sacral Slope (°) | 36.1 ± 23.3 | 35.2 ± 15.9 | 0.85 | 39.1 ± 17.3 | 36.4 ± 15.3 | 0.43 |
| Global Alignment Parameter | ||||||
| Sagittal Vertical Axis | 2.7 ± 4.6 | 2.0 ± 4.4 | 0.48 | 0.85 ± 3.9 | 0.24 ± 4.8 | 0.30 |
| Coronal Vertical Axis | 0.9 ± 3.8 | 0.9 ± 3.5 | 0.98 | 1.7 ± 4.4 | 0.49 ± 2.5 | 0.027 |
| Trunkshift | 1.8 ± 1.8 | 2.9 ± 2.5 | 0.06 | 1.7 ± 2.3 | 1.4 ± 1.5 | 0.57 |
aIndependent t-test. Boldface indicates significant P values.
Neurologic Function—Pre and Post LEMS
Mean LEMS in myelopathic patients increased significantly at every postoperative time point compared to preoperative baseline LEMS: Baseline 40.7 ± 9.9; Immediate postop: 46.0 ± 7.1, P = 0.02; 1-year: 48.2 ± 3.7, P < 0.001; 2-year: 48.2 ± 7.7, P < 0.001; while the non-myelopathic group were largely normal at baseline (48.2 ± 8.2) and didn’t experience any significant postoperative changes. The non-myelopathic group had significantly higher LEMS immediately postoperative (non-myelopathic: 49.1 ± 3.7, myelopathic: 46.0 ± 7.1, P = 0.0007), but by 1 year postoperative the myelopathic patients recovered (non-myelopathic: 49.3 ± 3.6, myelopathic: 48.2 ± 3.7, P = 0.10) such that there was no difference between LEMS scores. This was maintained at 2-years postoperative (non-myelopathic: 49.2 ± 3.3, myelopathic: 48.2 ± 5.7, P = 0.09) (Table 4) (Figure 1).
Table 4.
Lower Extremity Motor Score Outcomes.
| Variables | Preoperative | Immediate postoperative | 1-Year postoperative | 2-Year postoperative | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| (Mean ± SD, or n (%)) |
Myelopathy (n = 29) |
No myelopathy (n = 232) |
P-valuea | Myelopathy (n = 24) |
No myelopathy (n = 232) |
P-valuea | Myelopathy (n = 19) |
No myelopathy (n = 149) |
P-valuea | Myelopathy (n = 21) | No myelopathy (n = 81) | P-valuea |
| LEMS | 40.7 ± 9.9 | 48.2 ± 8.2 | <0.001 | 46.0 ± 7.1 | 49.1 ± 3.7 | <0.001 | 48.2 ± 3.7 | 49.3 ± 3.6 | 0.10 | 48.2 ± 5.7 | 49.2 ± 3.3 | 0.093 |
| Decline from baseline | 2 (6.9) | 19 (6.8) | - | 0 (0) | 7 (2.5) | - | 2 (6.9) | 6 (2.1) | - | |||
| Improved from baseline | 16 (55.2) | 13 (4.6) | - | 13 (44.8) | 12 (4.3) | - | 16 (55.2) | 2 (0.7) | - | |||
| P-valueb | - | 0.02 | 0.14 | - | <0.001 | 0.05 | - | <0.001 | 0.06 | - | ||
aIndependent t-test.
bpaired t-tests compared to baseline LEMS.
Figure 1.
Comparison of postoperative LEMS between myelopathic and non-myelopathic patients.
Patient Reported Outcomes
Myelopathic patients had significantly worse preoperative SRS22r scores in every domain, except for satisfaction (P = 0.18), compared to the non-myelopathic group (Table 5) (Figure 2). Both groups had significant increases in every SRS domain 2-years postoperative compared to preoperative, while the non-myelopathic patients had significantly higher outcomes in mental health (P = 0.0028), function (P = 0.020), and total score (P = 0.011) compared to the myelopathic group (Table 6). However, the myelopathic patients improved to the point where their SRS22r scores were not different from the non-myelopathic patients in the domains of pain (P = 0.12), self-image (P = 0.08), and satisfaction (P = 0.83). Sub-analysis of outcome scores between the myelopathic patients treated with a VCR compared to myelopathic patients treated without a VCR did not exhibit differences in outcomes except for total SRS score (P = 0.028), self-image domain (P = 0.016), and satisfaction (P = 0.035) (Table 7). A case example is shown in Figures 3–6.
Table 5.
Preoperative and 2-Year Postoperative SRS-22r Patient-Reported Outcomes.
| Preoperative | 2-Year Postoperative | |||||
|---|---|---|---|---|---|---|
| Variables (mean ± SD or n (%)) | Myelopathy | No myelopathy | P-valuea | Myelopathy | No myelopathy | P-valuea |
| SRS-22r Domains | ||||||
| Total | 3.1 ± 0.6 | 3.5 ± 0.6 | 0.0006 | 4.0 ± 0.6 | 4.3 ± 0.5 | 0.011 |
| Pain | 3.5 ± 0.8 | 3.9 ± 0.9 | 0.022 | 4.0 ± 0.9 | 4.3 ± 0.8 | 0.12 |
| Self-image | 2.3 ± 0.7 | 2.7 ± 0.8 | 0.020 | 3.9 ± 0.7 | 4.1 ± 0.7 | 0.08 |
| Function | 3.1 ± 0.8 | 3.7 ± 0.8 | 0.0002 | 4.0 ± 0.7 | 4.3 ± 0.6 | 0.020 |
| Mental health | 3.4 ± 0.8 | 3.8 ± 0.7 | 0.0067 | 3.8 ± 0.7 | 4.3 ± 0.6 | 0.0028 |
| Satisfaction | 2.9 ± 1.3 | 3.2 ± 1.3 | 0.18 | 4.5 ± 0.7 | 4.5 ± 0.7 | 0.83 |
aIndependent t-test. Boldface indicates significant P values.
Figure 2.
Preoperative to 2-year postoperative SRS outcome scores in myelopathic patients.
Table 6.
Comparisons of Myelopathic Patients’ Outcome Scores From Preoperative to 2-Year Postoperative.
| Variables (mean ± SD or n (%)) | Preoperative | 2-Year postoperative | P-valuea |
|---|---|---|---|
| SRS-22r Domains | |||
| Total | 3.06 ± 0.58 | 3.96 ± 0.62 | <0.0001 |
| Pain | 3.47 ± 0.79 | 3.97 ± 0.93 | 0.035 |
| Self-image | 2.32 ± 0.70 | 3.86 ± 0.71 | <0.0001 |
| Function | 3.07 ± 0.83 | 3.97 ± 0.73 | 0.0002 |
| Mental health | 3.44 ± 0.77 | 3.84 ± 0.68 | 0.017 |
| Satisfaction | 2.86 ± 1.25 | 4.45 ± 0.67 | <0.0001 |
aPaired t-tests.
Table 7.
Preoperative and 2-Year Postoperative LEMS and SRS-22r Patient-Reported Outcomes of Myelopathic Patients Treated With A VCR and Those Treated Without a VCR.
| Variables (mean ± SD or n (%)) | Preoperative | 2-Year postoperative | ||||
|---|---|---|---|---|---|---|
| VCR (n = 20) | No VCR (n = 9) | P-valuea | VCR (n = 14) | No VCR (n = 8) | P-valuea | |
| LEMS | 42 ± 10 | 38.8 ± 9.3 | 0.44 | 49 ± 4.0 | 47.3 ± 4.0 | .40 |
| SRS-22r Domains | ||||||
| Total | 2.8 ± 0.5 | 3.2 ± 0.6 | 0.12 | 3.6 ± 0.6 | 4.2 ± 0.5 | 0.028 |
| Pain | 3.5 ± 0.6 | 3.5 ± 0.9 | 0.92 | 3.6 ± 0.9 | 4.2 ± 0.9 | 0.13 |
| Self-image | 1.8 ± 0.5 | 2.6 ± 0.6 | 0.002 | 3.4 ± 0.7 | 4.1 ± 0.6 | 0.016 |
| Function | 3.0 ± 0.6 | 3.1 ± 0.9 | 0.90 | 3.6 ± 0.6 | 4.2 ± 0.7 | 0.067 |
| Mental health | 3.0 ± 0.5 | 3.6 ± 0.8 | 0.056 | 3.6 ± 0.7 | 4.0 ± 0.6 | 0.18 |
| Satisfaction | 2.6 ± 1.1 | 3.0 ± 1.3 | 0.43 | 4.1 ± 0.8 | 4.7 ± 0.5 | 0.035 |
aIndependent t-test. Boldface indicates significant P values.
Figure 3.
Case example of a myelopathic patient. 14 Year-old female with 20 previous spinal surgeries presented with a lower extremity motor score of 35/50 and several beats of clonus bilaterally. Deformity progression impacted her ambulatory status to the point where she relied on assistive devices and steadily increasing time spent in a wheel-chair. Immediately postoperative her LEMS improved to 50/50 that was maintained at both 1-year and 2-year postoperative. At 1-year postoperative clonus was present bilaterally, though by 2-years postoperative her clonus was resolved and she had a completely normal neurologic exam. A, Preoperative AP radiograph. Major Cobb of 78°, secondary Cobb of 49°, and a Coronal DAR of 8.7. B, 1-Year postoperative AP radiograph. Major Cobb of 42° and secondary Cobb of 24°. C, 2-Year postoperative AP radiograph. Major Cobb of 37° and secondary Cobb of 25°.
Figure 4.
A, Preoperative lateral radiograph. Maximum Kyphosis angle of 116°, T12-S1 Lordosis of −90°, and Sagittal DAR of 12.9. B, 1-Year postoperative lateral radiograph preoperative lateral radiograph. Maximum Kyphosis angle of 60° and T12-S1 Lordosis of −83°. C, 2-Year postoperative lateral radiograph. Maximum Kyphosis angle of 73° and T12-S1 lordosis of −73°.
Figure 5.
Preoperative sagittal MRI highlighting the stretched spinal cord over the apex of the kyphosis (yellow circle).
Figure 6.
Preoperative 3D CT scan highlighting the pseudarthrosis at T7-8 leading to the segmental angular deformity contributing to her myelopathy (yellow circle).
Discussion
In this study, we assessed the perioperative and 2-years postoperative outcomes, neurologic function, and PROs in myelopathic versus non-myelopathic ambulatory patients undergoing major pediatric spinal deformity reconstructive surgery. By 1-year postoperative, the neurologic status reflected by the LEMS of the myelopathic group was not different compared to the non-myelopathic group (non-myelopathic: 49.3 ± 3.6, myelopathic: 48.2 ± 3.7, P = 0.10), a finding that was maintained at 2-years postoperative (non-myelopathic: 49.2 ± 3.3, myelopathic: 48.2 ± 5.7, P = 0.09). In addition to the improved neurologic outcomes, both groups had significant increases in every SRS domain compared to preoperative.
It is well known that decreased preoperative neurologic function is significantly associated with postoperative neurologic deficits and intraoperative neuromonitoring data changes, making these patients at high-risk for further worsening of neurologic function.2,3,9 While the majority of the literature focuses on outcomes in patients with normal preoperative neurologic function or a non-ambulatory, neuromuscular cohort, the “in-between” high-functioning, ambulatory patients with impaired preoperative neurologic status are frequently omitted.2,3,6,9 Severe pediatric spinal deformity patients often have a substantial component of their deformity in the sagittal plane, which we found to be true in our cohort as well, with the myelopathic patients having a significantly higher rate of kyphoscoliosis diagnoses, as well as larger maximum kyphosis angles and S-DARs compared to the non-myelopathic group preoperatively.5,6,13 These results strengthen and align with preexisting theories that sagittal plane angular kyphosis can often cause neurologic impairment due to the increased compression and/or stretch placed on the spinal cord as it passed the sharp bony bend. 14
While we know that surgical treatment for myelopathic patients is associated with higher risks of intra- and postoperative complications, surgical intervention is necessary to first and foremost, decompress the spinal cord, and second, correct the deformity as safely as possible.2,8,9,14 Techniques to treat severe spinal deformity and subsequent spinal cord stretch/compression can entail preoperative HGT, use of complex osteotomies (i.e. VCR), spinal instrumentation and fusion, and/or multiple staged surgical procedures.1,4,5,7,10,15 In the present analysis, the majority of myelopathic patients were treated with a VCR (20/29, 69%). Removing the apical vertebra(e) allows for circumferential decompression of the spinal cord, relief of any compression on the spinal cord, and realignment of the spine. In addition to all dorsal bony elements, a VCR prevents any compression from adjacent pedicles, which can often cause cord impingement at the apex. The remaining 9 myelopathic patients not treated with a VCR involved a combination of strategies such as preoperative HGT (6/9, 66.7%) to lessen curve magnitude and potentially obviate the need for a VCR, use of less complex osteotomies intraoperatively (Schwab type 1: 3/9, 33.3% and type 2: 6/9, 66.7%), and/or surgical reduction techniques with subsequent spinal instrumentation and fusion (9/9, 100%).
Understanding the 2-year postoperative outcomes of patients with preoperative myelopathy can significantly aid in patient counseling and managing expectations. While the myelopathic patients continued to have significantly lower LEMS compared to non-myelopathic patients at the immediate postoperative appointment, over half of these patients improved their LEMS at this visit compared to baseline. Additionally, by 1 year postoperative, the myelopathic patients had recovered lower extremity strength to the point where there was no difference in LEMS as compared to the non-myelopathic patients, and the improvement was maintained at 2-years postoperative. A similar study of 13 adult and pediatric patients undergoing a VCR for correction of severe rigid kyphoscoliosis found that patients who were ambulatory with support/aids or non-ambulatory preoperative improved their functional status postoperatively, though the authors did not mention presence or absence of myelopathy nor LEMS. 16 The SCOLI-Risk study prospectively assessed LEMS status in an international cohort of 272 complex adult spinal deformity patients, 68/272 (25%) having a preoperative abnormal LEMS (<50); at discharge 49/66 (74%) either maintained or improved LEMS and 40/46 (87%) by 2-years postoperative; however, these were in adult patients only and there was no mention of myelopathy specifically, only motor deficits. 11 Regardless, these results suggest that patients with severe spinal deformity and evidence of spinal cord dysfunction can reverse the neurologic deficits and improve significantly in both LEMS and functional status after undergoing complex spinal deformity surgery.11,16
While postoperative PROs showed a significantly higher SRS22r function score in the non-myelopathic group, there was no difference between group outcomes in the pain, mental health, and satisfaction SRS22r domains. Though they may have different overall postoperative function, the myelopathic group significantly improved in the function domain by 2-years as compared to their preoperative status. Analysis of SRS22r scores for 50 pediatric patients that underwent a revision surgery stated an average postoperative function score of 3.9 ± 0.8, a slightly smaller magnitude compared to our myelopathy cohort with 4.0 ± 0.7, albeit none of the indications for the revision surgeries were due to neurologic deficit nor was preoperative neurologic status mentioned. 17 Self-image and pain scores improved significantly from baseline in both groups and were not different between the 2 cohorts at 2-years postoperative. A cohort of pediatric patients that underwent a posterior VCR had similar significant improvements in self-image, from 3.3 ± 0.8 preoperatively to 4.2 ± 0.6 at 2-years postoperative; though this group did not experience any significant changes in pain scores, had an average major curve Cobb angle of 71.9 ± 53.6, and only 4/23 (17%) had a preoperative motor deficit making direct comparisons challenging. 7 Considering that preoperative neurologic deficits can impact preoperative planning, anesthetic and blood pressure management, and the amount of sagittal plane correction obtained, it may not be appropriate to compare PROs to a neurologically intact cohort and shift the focus to the significant improvements made in every domain from preoperative to postoperative.
This study is not without limitation. Firstly, as this was an international multicenter study the motor and sensory exams may differ across examiners possibly introducing subjective bias, though studies have shown that the ASIA exams are reliable across experienced examiners. 18 Secondly, as this was an observational long-term follow-up study some patients have been lost to follow-up and did not complete their 1-year (12.9%) or 2-year (36.3%) postoperative visits. Third, differences between surgeons could not be accounted for in surgical plan and operative technique. Fourth, we did not collect the duration of myelopathic symptoms and thus we cannot assess the effect that preoperative symptomology had on postoperative recovery; though the a-priori objective of this analysis was to assess recovery of postoperative neurologic deficits in patients with preoperative myelopathy following complex surgical treatment, not to assess risk factors for recovery of postoperative neurologic deficits. Fifth, this cohort is comprised of patients with multiple different pathologies and the myelopathic group had a small sample size. Future studies are warranted to assess the weight of risk factors for recovery of neurologic deficits in a larger sample of preoperative myelopathic patients.
Conclusion
In severe spinal deformity pediatric patients presenting with preoperative myelopathy undergoing spinal reconstructive surgery, myelopathic patients can expect significant improvement in neurologic function postoperatively. Neurologic function at 1-year postoperative measured via LEMS, demonstrated there was no difference between the 2 groups. Thus, myelopathic patients undergoing severe deformity correction can attain similar neurologic function compared to non-myelopathic patients, which was maintained at 2-years postoperative. While non-myelopathic patients had significantly higher postoperative outcomes in SRS mental health, function, and total score, both groups had significantly superior outcomes in every SRS domain as compared to preoperative scores.
Acknowledgment
Fox Pediatric Spinal Deformity Study Group.
Authors’ Note: Columbia University Institutional Review Board Approved Protocol # AAAI2200 for Informed Consent.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received financial support for the research, authorship, and/or publication of this article: This study was supported by Fox study group.
ORCID iDs: Meghan Cerpa, MPH 
https://orcid.org/0000-0002-5931-7067
Michael P. Kelly, MD, MS
https://orcid.org/0000-0001-6221-7406
Munish Gupta, MD
https://orcid.org/0000-0002-4711-4377
References
- 1.Lenke LG, Sides BA, Koester LA, Hensley M, Blanke KM. Vertebral column resection for the treatment of severe spinal deformity. In: Clin Orthop Relat Res. 2010;468(3):687–699. doi:10.1007/s11999-009-1037-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cho SK, Lenke LG, Bolon SM, et al. Progressive myelopathy patients who lack spinal cord monitoring data have the highest rate of spinal cord deficits following posterior vertebral column resection surgery. Spine Deform. 2015;3(4):352–359. doi:10.1016/j.jspd.2014.11.009 [DOI] [PubMed] [Google Scholar]
- 3.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(1):149–156. doi:10.1007/s00586-013-2793-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lenke LG, O’Leary PT, Bridwell KH, Sides BA, Koester LA, Blanke KM. Posterior vertebral column resection for severe pediatric deformity: minimum two-year follow-up of thirty-five consecutive patients. Spine (Phila Pa 1976). 2009;34(20):2213–2221. doi:10.1097/BRS.0b013e3181b53cba [DOI] [PubMed] [Google Scholar]
- 5.Bogunovic L, Lenke LG, Bridwell KH, Luhmann SJ. Preoperative halo-gravity traction for severe pediatric spinal deformity: complications, radiographic correction and changes in pulmonary function. Spine Deform. 2013;1(1):33–39. doi:10.1016/j.jspd.2012.09.003 [DOI] [PubMed] [Google Scholar]
- 6.Boachie-Adjei O, Yagi M, Sacramento-Dominguez C, et al. FOCOS Spine Research Group. Surgical risk stratification based on preoperative risk factors in severe pediatric spinal deformity surgery. Spine Deform. 2014;2(5):340–349. doi:10.1016/j.jspd.2014.05.004 [DOI] [PubMed] [Google Scholar]
- 7.Riley MS, Lenke LG, Chapman TM, Jr, Sides BA, Blanke KM, Kelly MP. Clinical and radiographic outcomes after posterior vertebral column resection for severe spinal deformity with five-year follow-up. J Bone Joint Surg Am. 2018;100(5):396–405. doi:10.2106/JBJS.17.00597 [DOI] [PubMed] [Google Scholar]
- 8.Sielatycki JA, Cerpa M, Baum G, et al. A novel MRI-based classification of spinal cord shape and CSF presence at the curve apex to assess risk of intraoperative neuromonitoring data loss with thoracic spinal deformity correction. Spine Deform. 2020;8(4):655–661. doi:10.1007/s43390-020-00101-9 [DOI] [PubMed] [Google Scholar]
- 9.Lewis NDH, Keshen SGN, Lenke LG, et al. The deformity angular ratio: does it correlate with high-risk cases for potential spinal cord monitoring alerts in pediatric 3-column thoracic spinal deformity corrective surgery? Spine (Phila Pa 1976). 2015;40(15):E879–E885. doi:10.1097/BRS.0000000000000984 [DOI] [PubMed] [Google Scholar]
- 10.Iyer S, Boachie-Adjei O, Duah HO, et al. FOCOS Spine Research Group. Halo gravity traction can mitigate preoperative risk factors and early surgical complications in complex spine deformity. Spine (Phila Pa 1976). 2019;44(9):629–636. doi:10.1097/BRS.0000000000002906 [DOI] [PubMed] [Google Scholar]
- 11.Lenke LG, Shaffrey CI, Carreon LY, et al. AO Spine International and SRS Scoli-RISK-1 Study Group. Lower extremity motor function following complex adult spinal deformity surgery. J Bone Joint Surg Am. 2018;100(8):656–665. doi:10.2106/jbjs.17.00575 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schwab F, Blondel B, Chay E, et al. The comprehensive anatomical spinal osteotomy classification. Neurosurgery. 2014;74(1):112–120. doi:10.1227/NEU.0000000000000182o [DOI] [PubMed] [Google Scholar]
- 13.Ramirez N, Olivella G, Rodriguez O, et al. Incidence of complications in the management of non-ambulatory neuromuscular early-onset scoliosis with a rib-based growing system: high- versus low-tone patients. Eur J Orthop Surg Traumatol. 2020;30(4):621–627. doi:10.1007/s00590-019-02614-0 [DOI] [PubMed] [Google Scholar]
- 14.Cheh G, Lenke LG, Padberg AM, et al. Loss of spinal cord monitoring signals in children during thoracic kyphosis correction with spinal osteotomy: why does it occur and what should you do? Spine (Phila Pa 1976). 2008;33(10):1093–1099. doi:10.1097/BRS.0b013e31816f5f73 [DOI] [PubMed] [Google Scholar]
- 15.Suk SIl, Chung ER, Kim JH, Kim SS, Lee JS, Choi WK. Posterior vertebral column resection for severe rigid scoliosis. Spine (Phila Pa 1976). 2005;30(14):1682–1687. doi:10.1097/01.brs.0000170590.21071.c1 [DOI] [PubMed] [Google Scholar]
- 16.Modi HN, Suh SW, Hong JY, Yang JH. Posterior multilevel vertebral osteotomy for severe and rigid idiopathic and nonidiopathic kyphoscoliosis: a further experience with minimum two-year follow-up. Spine (Phila Pa 1976). 2011;36(14):1146–1153. doi:10.1097/BRS.0b013e3181f39d9b [DOI] [PubMed] [Google Scholar]
- 17.Yagi M, King AB, Kim HJ, Cunningham ME, Boachie-Adjei O. Outcome of revision surgery in pediatric spine deformity patients. Spine Deform. 2013;1(1):59–67. doi:10.1016/j.jspd.2012.10.002 [DOI] [PubMed] [Google Scholar]
- 18.Savic G, Bergstrombergstrom E, Frankel HL, Jamous MA, Jones PW. Inter-rater reliability of motor and sensory examinations performed according to American Spinal Injury Association standards. Spinal Cord. 2007;45(6):444–451. doi:10.1038/sj.sc.3102044 [DOI] [PubMed] [Google Scholar]