Abstract 57. Radiographic reporting in adolescent idiopathic scoliosis: Is there a discrepancy comparing radiologists’ reports and surgeons’ assessments?
Background
Cobb angle measurement is the standard method for quantification of scoliosis in adolescent idiopathic scoliosis (AIS), which guides decision-making for treatment. Primary care physicians rely on radiology reports to follow guidelines provided by specialists to facilitate priority setting of referrals. Accurate clinical and radiographic assessments are key to ensure timely tertiary care referrals for patients who are eligible for bracing, which is a time-sensitive evidence-based intervention. This study investigated the agreement of Cobb angle measurements between radiologists and treating spine surgeons.
Methods
We performed a retrospective audit of 40 randomly selected spinal radiographs of patients with AIS. Cobb angle, Risser stage and “end vertebrae” were recorded from the radiographic reports. Two spine surgeons (blinded to the reports) provided the same measurements on the same set of radiographs, prospectively. We compared measurements by radiologists and spine surgeons. To assess interrater reliability, interclass correlation coefficients (ICC) with 95% confidence intervals (CI) were computed. ICC less than 0.70, 0.70 to 0.79, 0.80 to 89 and 0.90 to 0.99 were considered to indicate poor, fair, good and excellent reliability, respectively.
Results
The agreement of Cobb angle between spine surgeons was excellent (ICC 0.96, 95% CI 0.93 to 0.98). The agreement of Cobb angle between spine surgeons and radiologists was good (ICC 0.83, 95% CI 0.13 to 0.97). The agreement of Cobb angle between spine surgeons and community radiologists and pediatric radiologists was good (ICC 0.88, 95% CI −0.13 to 0.97) and excellent (ICC 0.90, 95% CI 0.55 to 0.97), respectively. However, 15 out of 37 radiographs performed at the community level did not meet the standard requisite criteria. Risser stage was not reported in 21 of the 37 radiology reports. Agreement between spine surgeons and radiologists for Risser stage was poor (ICC 0.42, 95% CI −0.59 to 0.79). There was a 25% agreement between radiologists and spine surgeons for end vertebrae identification. Overall, 1 in every 4 referrals was inappropriate as a result of these discrepancies.
Conclusion
This study demonstrated a substantial disagreement in scoliosis measurements between radiologists and spine surgeons. This may have a negative impact on timely and appropriate referrals to tertiary care for AIS.
Abstract 74. How useful is prereferral spine imaging? A quality improvement project
Background
Primary care physicians rely on radiology reports to confirm a scoliosis diagnosis and inform the need for spine specialist referral. In turn, spine specialists use these reports for triage decisions and planning of care. To be a valid predictor of disease and management, radiographic evaluation should include coronal and lateral views of the spine and pelvis, leading to accurate Cobb angle measurements and Risser staging. The study objectives were to determine (1) the adequacy of index images to inform treatment decisions at initial consultation and (2) the utility of index radiology reports for appropriate triage decisions, by comparing reports with corresponding images.
Methods
We conducted a retrospective chart and radiographic review of patients with idiopathic scoliosis seen for initial consultation, aged 3–18 years, between Jan. 1 and Apr. 30, 2021. Index images were determined to be inadequate if they could not provide accurate Cobb angle measurements and/or skeletal maturity information. Comparisons were made between the index radiology report, associated imaging and new imaging obtained at initial consultation. Major discrepancies were defined by an interreader difference of 15° or greater, treatment change requiring surgery or a brace, or discordant Risser staging. The location of index imaging, hospital- versus community-based private clinic, was evaluated as a risk factor for inadequate or discrepant imaging.
Results
There were 94 patients reviewed, with 79% (n = 74) requiring repeat imaging at initial consultation; for 74 % of these patients (n = 55) this was due to insufficient quality and/or visualization of the sagittal profile, pelvis or rib cage. Risser stage was reported in 14% of the index radiology reports. Comparing these reports with the initial visit evaluation with 60 days or fewer between imaging (n = 49), discrepancies were found in 24.5% (95% confidence interval [CI] 14.6–38.1) of patients, with 18.4 % (95% CI 10.0–31.4) categorized as major discrepancies. In 13.8% (n = 13) of the total cohort, surgical or brace treatment was recommended when not predicted on the basis of the index radiology report. Repeat radiographs (p = 0.001, odds ratio [OR] 8.38) and discrepancies (p = 0.02, OR 7.96) were increased when index imaging was obtained at community-based private clinic. Reevaluation of available index imaging demonstrated that 24.6% (95% CI 15.2–37.1) of Cobb angles were misreported.
Conclusion
Standardization of spine imaging and reporting is imperative to improve measurement accuracy, facilitate triage and decrease unnecessary radiation exposure.
Abstract 82. Early recovery after surgery, predictors of shorter length of stay
Background
To determine if length of stay (LOS) differed by protocol type (enhanced recovery after surgery [ERAS] protocol v. non-ERAS [N-ERAS]). The secondary objectives were to assess differences in pain and opioid consumption by protocol type over time and identify predictors of LOS.
Methods
We conducted a retrospective chart review of 157 consecutive patients with adolescent idiopathic scoliosis treated with posterior spine fusion (PSF). Sample characteristics were compared between groups using independent t tests and χ2 tests. Differences in LOS by protocol type were assessed using regression adjusting for age, sex, body mass index, presurgical Cobb angle, levels fused and year of surgery. Similar models were used to identify predictors of LOS. Linear mixed models were used to assess differences in pain and opioid used over time by protocol type.
Results
Fifty-nine patients from the ERAS group were compared with 81 patients from the N-ERAS group. Seventeen patients were excluded from protocol-specific analysis because of partial exposure to the ERAS protocol but were included in the analysis of predictors of LOS. Patients were comparable in their baseline characteristics with regard to weight, height, number of levels fused and coronal Cobb angel. Median LOS was 3 days (interquartile range [IQR] 3–4 d) for the ERAS group, compared with 5 (IQR 4–5) days for the N-ERAS group (p < 0.001). The ERAS group had a significantly lower adjusted rate of stay (RR 0.75, 95% confidence interval 0.61–0.91). Patients in the ERAS group had significantly lower average pain on postoperative day (POD) 0 (2.73 v. 4.37, p < 0.001), POD1 (3.2 v. 4.4, p < 0.001), POD2 (4.4 v. 4.7, p = 0.028) and POD5 (2.9 v. 4.4, p = 0.038). At each time point, the ERAS group used a significantly lower average morphine equivalent dose (mg/kg) (all p < 0.001).
Conclusion
Adoption of a modified ERAS-based protocol for patients undergoing PSF for adolescent idiopathic scoliosis led to a significant reduction in LOS, pain scores and opioid consumption.
Abstract 68. Gliding screws on early-onset scoliosis: a 5-year experience
Background
Our goal is to present our surgical, radiologic and clinical outcomes related to the utilization of gliding screws for the treatment of early-onset scoliosis in combination with modern Luque trolley, growing rods and magnetically controlled rods over the past 5 years.
Methods
This is an original work based on prospectively collected and retrospectively analyzed data on all pediatric patients treated using gliding screws at a single institution from 2015 to 2020. Data collected were mainly but not limited to age, sex, diagnosis and complications. Preoperative and postoperative data were collected on type of curve, curve magnitude, T1–L1 length, T1–T12 length and T1–S1 length (at each follow-up appointment).
Results
Thirty patients with the diagnosis of early-onset scoliosis with associated pathologies such as a neuromuscular condition, syndrome, or mesenchymal disease such as osteogenesis imperfecta were treated using gliding screws between 2015 and 2020. The average age at the time of surgery was 7 years. Follow-up was between 24 and 50 months. Radiologic outcomes were as follows: the percentage of immediate curve correction postoperatively was approximately 70% and at the latest follow-up it was less than 30% for 15% of patients, 31%–60% for 61% of patients and more than 60% for 23% of patients. Seventy percent of the patients reached 70% of their expected growth. However, 30% of patients reached less than 40% of their expected growth. Curve progression was observed for 70% of the patients and 2 patients required fusion owing to the severity of their curve progression at 24 months after gliding screw insertion. No correlation between curve magnitude and diagnosis, curve magnitude and growth, or expected growth and diagnosis was statistically observed. There were no complications or revision surgeries related to hardware failure.
Conclusion
Gliding screws used in combination with growth modulation techniques such as modern Luque trolley, growing rods and/or magnetically controlled growing rods represent a safe and relatively effective option to achieve and maintain curve correction while preserving growth after at least 24 months follow-up. However, further studies including a larger, more homogeneous sample may be necessary to confirm our results.
Abstract 66. Reliability of radiographic assessment of growth modulation from anterior vertebral body tethering surgery in pediatric scoliosis
Background
Anterior vertebral body tethering (AVBT) in patients who are skeletally immature is thought to lead to asymmetric peri-apical growth. The purpose of this study was to assess the reliability of vertebral height and angular measurements for fusionless techniques.
Methods
Eight independent observers measured radiographs of patients with idiopathic scoliosis treated with AVBT, preoperatively and at 4-year follow-up, obtained from an international pediatric spine registry. Vertebral wedging, disc wedging, convex vertebral body heights and concave vertebral body heights of the 3 apical vertebrae were measured. A period of at least 2 weeks elapsed between the first and second round measurements.
Results
There were 15 patients (87% female, 13% male) with a mean age at surgery of 12 years (standard deviation [SD] 1 yr). Mean preoperative scoliosis was 52° (SD 10°) and 4-year post-operative scoliosis was 31° (SD 7°) (p = 0.000); mean preoperative kyphosis was 26° (SD 10°) and 4-year postoperative kyphosis was 35° (SD 17°) (p = 0.045). Interobserver agreement for preoperative vertebral body heights was good (interclass correlation coefficient [ICC] 0.80, 95% confidence interval [CI] 0.74–0.85). At 4-year follow-up there was moderate agreement (ICC 0.67, 95% CI 0.59–0.74). When we averaged measurements using 4 groups of 2 observers each, there was good agreement both preoperatively (ICC 0.89, 95% CI 0.84–0.92) and at 4 years (ICC 0.78, 95% CI 0.68–0.85). There was poor interobserver agreement for preoperative wedging angle measurements (ICC 0.41, 95% CI 0.32–0.52) and 4-year follow-up (ICC 0.45, 95% CI 0.36–0.56). When averaging measurements this increased to moderate agreement (preoperative ICC 0.62, 95% CI 0.52–0.72; 4-year ICC 0.59, 95% CI 0.48–0.69). Intraobserver agreement ranged from good to excellent per individual (ICC 0.84–0.94) for preoperative vertebral body heights, but this decreased with the 4-year images (ICC 0.52–0.88). Intraobserver agreement was low overall for wedging (preoperative ICC 0.41–0.71; 4-year follow-up ICC 0.41–0.76).
Conclusion
On the basis of 8 individual observers, interobserver agreement ranged from good (preoperative vertebral body heights) to moderate (4-year follow-up vertebral body heights) to poor (all wedging angles). However, when the measurements of at least 2 observers were averaged, the ICC-based agreement classifications increased.
Abstract 16. A dangerous curve: impact of the COVID-19 pandemic on brace treatment in adolescent idiopathic scoliosis
Background
We conducted this study to test the hypothesis that the COVID-19 pandemic associated with social distancing, teleworking and school closures has had a negative impact on patient adherence to brace treatment in adolescent idiopathic scoliosis (AIS).
Methods
We reviewed a database of patients with AIS recruited between March and September 2020. Inclusion criteria were patients with AIS under brace treatment according to SRS criteria. The patients were divided into 2 cohorts: those with self-reported good adherence to treatment and those who voluntarily abandoned treatment during follow-up. Patients with irregular adherence were excluded. Data analysis included age, sex, Risser stage, type of brace, Cobb angles at first visit and last follow-up (mean 11 mo) and percentage of progression. Unpaired Student tests were used for comparison.
Results
A total of 154 patients met the inclusion criteria. Twenty patients were excluded because of irregular adherence. Eighty-nine patients (aged 12.1 [standard deviation (SD) 1.4] yr) reported good adherence to treatment, while 45 patients (12.6 [SD 1.5] yr) abandoned treatment, an abandon rate of 29 %. The cohort of compliant patients started treatment with a mean main thoracic (MT) curve of 26º and finished with 27º. The mean difference between measurements was +0.65º (SD 7.5º); the mean progression rate was 4.6%. However, patients who abandoned treatment started with a mean MT curve of 28º and finished with 33º, with a mean increase of +5º (SD 8º) and a mean progression rate of 11%. The differences between the 2 cohorts were statistically significant (p = 0.002). Five patients from the abandon group were counselled for surgery because of curve progression.
Conclusion
The abandon rate of brace treatment in AIS significantly increased during the COVID-19 pandemic. Patients who voluntarily discontinued treatment had significant increases in curve progression and surgical indication rates.
Abstract 24. Development of a model of interprofessional support interventions to enhance brace adherence in adolescents with idiopathic scoliosis
Background
Brace treatment in adolescent idiopathic scoliosis is recognized as effective, if the brace is worn as prescribed (20 to 23 h/d). Because of its negative biopsychosocial impact on adolescent patients’ quality of life, brace adherence is a common problem (average brace wear of 12 h/d). The purpose of this paper is to develop an interprofessional support intervention model to enhance brace adherence in adolescents with scoliosis.
Methods
Using convenience and snowball recruitment strategies, we enrolled 9 health professionals of different expertise, working with braced patients, to participate in individual interviews (2 orthopedists, 2 physiotherapists, 2 orthotists, 1 clinical nurse, 1 psychologist and 1 social worker). The interview guide was built according to the Information-Motivation-Strategy Model and the Interprofessional Care Competency Framework. Interviews were recorded, transcribed and coded deductively and inductively by 2 authors. Thematic analysis was performed to identify the most relevant concepts for designing the intervention model. A panel of 5 clinical experts was recruited to review and validate the intervention model.
Results
Participants highlighted different brace adherence barriers (physical/functional and psychosocial) experienced by patients and their families. They also brought up organizational, relational and cultural work barriers that might limit their capacity to support patients. Finally, they suggested educational, motivational, functional, psychological and interprofessional teamwork strategies to improve patients’ and parents’ support and potentially increase brace adherence. Using the emerging themes and their relationships, we designed an interprofessional adherence support (IPAS) intervention model that identifies the actors, activities, structure and intended impacts of the intervention. According to the expert panel, the IPAS model is highly relevant to respond to the brace adherence problem and it has potential for implementation in practice.
Conclusion
We designed an interprofessional support intervention model based on professional perspectives in response to the brace adherence problem in adolescents with scoliosis. Plans for implementation of the IPAS model at the scoliosis clinic are under development and considered essential for improving brace treatment outcomes.
Abstract 94. Recognizing the importance of self-image in adult spinal deformity: results from the Prospective Evaluation of Elderly Deformity Surgery (PEEDS)
Background
The importance of self-image has been well studied in the adolescent idiopathic scoliosis population, but it has been considered less important in adult spinal deformity (ASD). More recently, evidence suggests that self-image is an important consideration in ASD and that radiographic parameters such as Cobb angle, sagittal vertical alignment (SVA) and lumbopelvic alignment correlate to self-image scores. The goal of this study is to analyze the importance of self-image in ASD surgery and whether certain radiographic parameters influence outcome.
Methods
We conducted a prospective, multicentre, multicontinental, observational longitudinal cohort study of patients 60 years of age and older undergoing primary spinal fusion surgery of at least 5 levels for coronal, sagittal or combined deformity. Data from the Scoliosis Research Society questionnaire-22 revised (SRS-22r) were collected prospectively at 10 weeks, 1 year and 2 years postoperatively. Three-foot anteroposterior and lateral radiographs were taken preoperatively and at 10 weeks and 2 years postoperatively.
Results
A total of 219 patients met the inclusion criteria, with 214 completing SRS-22r preoperatively and 179 patients completing 2-year follow-up. Self-image had the lowest mean baseline SRS-22r score of 2.32 (standard deviation [SD] 0.8), which improved to 3.47 (SD 0.9) at 10 weeks, 3.66 (SD 0.7) at 1 year and 3.61 (SD 0.8) at 2 years. Preoperatively, 69.6% of patients felt very unhappy with their current back shape. Conversely, at 2 years, 59.3% of patients were either somewhat or very happy with the shape of their back. By Schwab type, 73 (40.1%) were N, 81 (44.5%) L and 28 (15.4%) D. The self-image scores of all groups improved from baseline to 24 months: N, 2.15 (SD 0.68) to 3.53 (SD 0.78); L, 2.32 (SD 0.81) to 3.62 (SD0.84); D, 2.28 (SD 0.74) to 3.89 (SD 0.70). Radiographic changes from baseline to 2 years were as follows: thoracolumbar Cobb 24.9° (95% confidence interval [CI] 1.7° to 102.5°) to 10.7° (95% CI 0.1° to 59.7°), SVA 81.8 (95% CI −76.3 to 327.2) to 37.3 (95% CI −108.1 to 334.1) mm, pelvic tilt 28.7° (95% CI 8.0° to 56.6°) to 21.1° (95% CI −0.2° to 51.5°). Univariate and multivariate linear mixed effect models identified SVA as the strongest radiographic predictor of self-image.
Conclusion
This study clearly demonstrates the importance of self-image in ASD surgery and SVA as the strongest radiographic predictor of outcome.
Abstract 25. Assessing pain as a primary factor in the surgical treatment of adult spinal deformity surgery in patients over 60 years of age
Background
Adult spinal deformity is associated with significant pain and disability, and multiple health-realted quality of life scales have been developed to assess change in pain and function after corrective deformity surgery. The present study utilizes data from an international prospective multicentre cohort study to investigate the effects of deformity surgery on multiple pain scores in geriatric patients.
Methods
Twelve centres recruited 255 patients undergoing a minimum of 5-level fusion. Patients were assessed pre-operatively, as well as postoperatively at 10 weeks, 1 year and 2 years using the numeric rating scale (NRS) for back and leg, and pain subscores were extracted from the Scoliosis Research Society questionnaire-22 revised (SRS-22r), and the Oswestry Disability Index (ODI).
Results
One hundred and seventy-nine patients completed the 2-year study period. Mean baseline NRS scores for back pain were worse than leg pain, with mean scores of 6.1 (standard deviation [SD] 2.7) and 4.3 (SD 3.3), respectively. NRS back pain decreased to 3.5 (SD 2.3) at 10 weeks, 2.8 (SD 2.4) at 1 year and 2.6 (SD 2.6) at 2 years. NRS leg pain decreased to 2.4 (SD 2.6) at 10 weeks and 2.1 (SD 2.5) at 1 year and remained stable at 2.3 (SD 2.7) at 2 years. Mean baseline SRS-22r pain subdomain scores measured 2.7 (SD 0.9), which remained unchanged at 2.9 (SD 0.9) at 10 weeks and improved to 3.7 (SD 0.9) at 1 and 2 years (p < 0.001). All outcome measures demonstrated a significant reduction in the percentage of patients who experienced fairly severe or severe pain (the worst 2 pain scores on the ODI pain intensity scale) from baseline to 2 years, ranging from 39% to 60 % preoperatively to between 5% and 9% at 2 years. Despite this, pain worsened compared with the preoperative level in 15.1% on NRS leg pain and 5.7 % on NRS back pain.
Conclusion
Results of this study demonstrate that all pain measures showed significant improvement at 1 and 2 years postoperatively, which was similar for all investigated age subgroups. Most of the patients concluded the study with some degree of pain, with the majority of patients reporting only mild pain.
Abstract 72. Application of the Ames-International Spine Study Group (AMES) radiographic modifiers to an asymptomatic population. Are the thresholds for “normal” appropriate?
Background
The Ames-International Spine Study Group (AMES) classification is used as a tool for stratifying severity in adult cervical deformity (ACD). The AMES sagittal modifiers include C2–7 sagittal vertical axis (cSVA), T1 slope cervical lordosis mismatch (TS-CL) and chin-brow vertical angle (CBVA). Each modifier stratifies patients as grade 0, 1 or 2. We aimed to classify an asymptomatic normal population by the AMES sagittal modifier grades and assess differences in health-related quality of life and overall sagittal alignment between grades.
Methods
Five hundred healthy asymptomatic volunteers from 20 to 40 years of age without prior spinal pathology or surgery were enrolled. The Oswestry Disability Index (ODI) and Neck Disability Index (NDI) were ascertained before EOS imaging scans.
Results
Four hundred and ninety-seven individuals were scanned, with 20 exclusions for incidental findings of scoliosis or spondylolisthesis. The cSVA was grade 0 (< 4 cm) in 475 individuals. The TS-CL modifier was grade 0 (< 15°) in 472 individuals. The CBVA was grade 0 (1° to 0°) in 65% of individuals and grade 1 (−10° to 0° or 11° to 25°) in 35%. There was no difference between CBVA grades in HRQOL and alignment.
Conclusion
When applied to a normal population the AMES classification sagittal modifier cSVA and TS-CL grade 0 demonstrates normal anatomy. A CBVA grade of 1 is present in 36% of asymptomatic individuals. These individuals have no difference in HRQOL scores or overall alignment. Therefore, surgical correction of “abnormal” grade 1 CBVA in ACD may be inappropriate.
Abstract 109. Exploring the relationship between cannabis and narcotic use on preoperative health considerations in Canadian thoracolumbar patients: a CSORN study
Background
This study aims to examine demographic and health status differences between patients who report the use of (1) cannabis, (2) narcotics, (3) cannabis and narcotics or (4) no cannabis or narcotic use.
Methods
We conducted a retrospective cohort study of thoracolumbar patients enrolled in the Canadian Spine Outcomes and Research Network (CSORN)registry after legalization of cannabis in Canada. Variables included age, sex, modified Oswestry Disability Index (mODI), numerical rating scales (NRS) for leg and back pain, tingling/numbness leg sensation, SF-12 quality of life mental component score, the Patient Health Questionnaire (PHQ-9) and general health state. An analysis of covariance with pathology as the covariate and post-hoc analysis was run.
Results
The majority of the 704 patients enrolled (mean age 59 yr; 46.9% female) were nonusers (41.8%). More patients reported narcotic use than cannabis use (29.7% v. 12.9%), with 13.4% stating concurrent use. MCS scores were significantly lower for patients with concurrent use compared to no use (mean 39.95 v. 47.98, p = 0.001) or cannabis use (mean 45.66, p = 0.043). The narcotic-use cohort had significantly worse MCS scores (mean 41.37, p = 0.001) than the no-use cohort. Patients reporting no use and cannabis use (mean 41.39 v. 42.94) had significantly lower ODI scores than those reporting narcotic use (mean 54.91, p = 0.001) and concurrent use (mean 50.80, p = 0.001). Lower NRS-leg pain was reported in cannabis use (mean 5.72) compared with narcotic use (mean 7.19) and concurrent use (mean 7.03, p = 0.001). Patients with no use (mean 6.31) had significantly lower NRS-leg pain than those with narcotic use (p = 0.011) and significantly lower NRS-back pain (mean 6.17) than those with narcotic use (mean 7.16, p = 0.001) and concurrent use (mean 7.15, p = 0.012). Patients with cannabis use reported significantly lower tingling/numbness leg scores (mean 4.85) than those with no use (mean 6.14, p = 0.022), narcotic use (mean 6.67, p = 0.001) and concurrent use (mean 6.50, p = 0.01). PHQ-9 scores were significantly lower for the no-use (mean 6.99) and cannabis-use (mean 8.10) cohorts than for the narcotic-use (mean 10.65) and concurrent-use (mean 11.93) cohorts. Patients who reported narcotic use reported a significantly lower rating of their overall health state (mean 50.03) than those who reported cannabis use (mean 60.50, p = 0.011) and no use (mean 61.89, p = 0.001).
Conclusion
Patients with preoperative narcotic use or concurrent use of narcotics and cannabis experienced higher levels of disability, pain and depressive symptoms and worse mental health functioning compared with patients with no cannabis or narcotic use and cannabis-only use.
Abstract 36. Metastatic spine disease: Should patients with short life expectancy be denied surgical care? An international retrospective cohort study
Background
Despite our inability to accurately predict survival in many patients with cancer, a life expectancy of at least 3 months is historically necessary to be considered for surgical treatment of spinal metastases. The aim of this study was to compare health-related quality of life (HRQOL) in patients surviving less than 3 months after surgical treatment with that of patients surviving more than 3 months to assess the validity of this inclusion criterion.
Methods
Patients who underwent surgery for spinal metastases between August 2013 and May 2017 were retrospectively identified from an international cohort study. HRQOL was evaluated using generic and disease-specific outcome tools at baseline and at 6 and 12 weeks after surgery. The primary outcome was the HRQOL at 6 weeks after treatment measured by the Spine Oncology Study Group Outcomes Questionnaire (SOSGOQ).
Results
A total of 253 patients were included: 40 patients died within the first 3 months after surgery and 213 patients survived more than 3 months. Patients surviving less than 3 months after surgery presented with lower baseline performance status. Adjusted analyses for baseline performance status did not reveal a significant difference in HRQOL between the 2 groups at 6 weeks after treatment. No significant difference in patient satisfaction at 6 weeks with regard to their treatment could be detected between both groups.
Conclusion
When controlled for baseline performance status, quality of life 6 weeks after surgery for spinal metastasis is independent of survival. To optimize improvement in HRQOL for this patient population, baseline performance status should take priority over expected survival in the surgical decision-making process.
Abstract 91. What do patients expect of palliative treatment for symptomatic spinal metastases? A qualitative study
Background
Patients with spinal metastases often receive palliative surgery or radiation therapy to maintain or improve health-related quality of life. Patients with unrealistic expectations regarding treatment outcomes have been shown to be less satisfied with their post-treatment health status. This study evaluated expectations of patients with spinal metastases scheduled for surgery and/or radiation therapy.
Methods
Individual semistructured interviews were conducted with patients with symptomatic spinal metastases before, and 6 weeks after, surgery and/or radiation therapy. Expectations regarding treatment outcomes were discussed before treatment, and level of fulfillment of these pretreatment expectations was discussed after treatment. Interviews were recorded and transcribed then analyzed according to the thematic analysis method to identify themes.
Results
Before treatment, patients felt they were not informed or were minimally informed about (expected) treatment outcomes, but they felt well informed about treatment procedure and possible complications. Although patients expected pain relief and improvement in daily functioning, they found it difficult to describe any recovery timeline or the impact of these expected improvements on their daily life. Patients generally understood that treatment was not curative, but they lacked insight into the impact of treatment on life expectancy as this was hardly discussed by their surgeon and/or radiation oncologist. Pretreatment expectations regarding pain and daily functioning were only partially met in most patients after treatment.
Conclusion
Patients feel they are not informed or only minimally informed about expected outcomes after surgery and/or radiation therapy for symptomatic spinal metastases. Improvements in patient–physician communication and counselling could help guide patients toward realistic pretreatment expectations.
Abstract 44. Denosumab for giant cell tumours of the spine: molecular predictors of clinical response — a pilot study
Background
Giant cell tumours of bone (GCTs) are locally aggressive and destructive tumours. Denosumab, a monoclonal antibody, has been used either as a neoadjuvant in resectable tumors or as a stand-alone treatment in unresectable tumours. While denosumab has been increasing used, different patterns of response have been observed clinically. The objectives were to establish a cohort of denosumab-treated spinal GCTs and to perform an exploratory study to identify putative molecular biomarkers of treatment response in patients treated with denosumab.
Methods
Pretreatment specimens from patients with spinal GCT treated with denosumab were collected at 2 centres. Clinical data were extracted retrospectively. Total mRNA was extracted by using a formalin-fixed paraffin-embedded extraction kit and we determined the transcript profile of 730 immuneoncology related genes by using the PanCancer Immune Profiling Panel (NanoString). The gene expression was compared between patients with good and poor response to denosumab treatment by using the nSolver Analysis Software (NanoString).
Results
Nine pretreatment specimens were acquired. Of these, 4 had denosumab used as a neoadjuvant treatment, 3 received the treatment as a stand-alone and 2 received denosumab as an adjuvant treatment. Two patients showed poor clinical response with tumour progression and metastasis. The statistical analysis showed clear differences between good and poor responders to denosumab treatment through hierarchical clustering analysis. The specific gene analysis, based on expression of specific cell markers and cytokine pattern, revealed that good responders had a higher innate immune response with a marked macrophage and dendritic cell populations, while poorly responding patients presented with a higher lymphocytic response in the lesions, with marked B-cells, and T-cell populations with enhanced Th1, Th2 and Th17 response.
Conclusion
Despite small numbers, this novel experimental approach revealed differences in the immune response in GCTs that predispose to the clinical response to denosumab. Further validation using a larger independent cohort will be required but these results could potentially identify the patients who would most benefit from denosumab therapy.
Abstract 89. Surgical management and outcomes from “stable” degenerative spondylolisthesis (DS) from the CSORN prospective DS study: What the @#$ % are we doing?
Background
There is growing evidence that fusion is not required in all patients who undergo surgery for degenerative spondylolisthesis (DS). The purpose of this study was to compare management choice and 1-year patient-reported outcomes (PROs) in patients with DS categorized by their treating surgeon as “stable” (defined as little to no back pain, positive radiographic stabilization signs and/or collapsed disc, a lordotic disc, less than 3mm of movement, and no facet effusion).
Methods
This was a subgroup analysis from an ongoing multicentre, prospective study (7 Canadian Spine Outcomes and Research Network [CSORN] sites). Within-group comparisons were performed to compare baseline, 3-month follow-up and 1-year follow-up of decompression (D) alone versus decompression and fusion (F). Standard univariate statistical analyses of baseline as well as follow-up categorical and continuous variables were performed.
Results
There were 111 patients categorized as stable with 1 year follow-up (93%). The mean age was 68.13 years, the mean body mass index was 29.1 and 57.7% were female. There was no significant difference between baseline sociodemographic variables and clinical presentation between D (n = 53) and F (n = 58; 45 interbody) groups. Mean procedure duration and length of stay (LOS) were 83 minutes for the D group versus 178 minutes for the F group (p < 0.001) and 2 (standard deviation [SD]2) days versus 4 (SD 3) days (p < 0.001). There were no intraoperative or perioperative adverse events in the D group, whereas 45% and 40% of the fusion patients suffered at least 1 intraoperative or perioperative adverse event, respectively. Postoperative adverse events were greater in the fusion group at the 3-month follow-up (5.7% for the D group v. 19.0 % for the F group; p = 0.046). There was no difference in adverse events between groups at 1 year. Baseline patient-reported outcome measures (back and leg pain, Oswestry Disability Index, SF-36 Physical Component Summary (PCS) and Mental Component Summary (MCS), EQ-5D) were similar between the groups. There was no difference in patient-reported outcome measures at the 3-month or 1-year point, with the exception of PCS at 3 months (40.2 for the D group v. 35.4 for the F group; p = 0.023).
Conclusion
Fusion may only confer harm and cost compared with decompression alone for patients meeting the defined composite clinical and radiographic criteria of stable DS put forth in this study. We really need to ask ourselves why we continue to fuse this subgroup of patients with DS.
Abstract 33. Economic consequences of waiting for lumbar disc herniation surgery
Background
The difference in the economic impact between early and late surgery for symptomatic lumbar disc herniation remains unknown. The goal of this study was to quantify economic advantages of early surgery compared with late surgery.
Methods
This is a retrospective analysis of prospectively collected data from the Canadian Spine Outcomes and Research Network (CSORN) registry. Direct and indirect costs were taken into account. The Ontario Health Insurance Plan (OHIP) claims history database was used for direct costs estimates. Canada Income Statistics and the human capital method were used to calculate indirect costs. Comparisons were made between patients receiving surgery less than 60 days after consent and 60 days or more after consent.
Results
A total of 493 patients were included in this study, with 272 patients (55.2%) receiving their surgery less than 60 days after providing consent for surgery and 221 patients (44.8%) 60 days or more after providing consent for surgery. Demographics and outcomes were similar between the 2 groups. Out of 354 patients working at the time of consent, 174 (49.2%) patients returned to work at 3 months follow-up. The group that waited longer had a significantly longer time from consent to the return to work date (56.7 [standard deviation (SD) 43.8] d for the group that waited < 60 d, 110.7 [SD 72.8] d for the group that waited ≥ 60 d; p = < 0.001). This corresponds to an additional $8,082.18 in indirect costs owing to productivity loss per patient who waited over 60 days for their surgery. At 3 months, patients in the short wait group more commonly visited the emergency department (52.6% v. 25.0%; p = 0.032), used physiotherapy (84.6% v. 72.0%; p = 0.001) and used magnetic resonance imaging (65.2% v. 41.4%; p = 0.043). This corresponds to an additional direct cost of $1,388.06 per patient.
Conclusion
As per our economic analysis, offering early surgery to patients who are working at the time of consent and show health care utilization, a cost saving of $6,694.12 per patient could be observed. This points to a potential avenue for both improved spinal care and cost savings.
Abstract 108. Motor recovery depends on timing of surgery in patients with lumbar disc herniation
Background
Although approximately half of patients undergoing lumbar disk surgery present with motor deficits, the timing of surgery for radicular weakness is largely unclear. The aim of this study was to evaluate the impact of surgical timing on motor recovery in patients with motor deficits from lumbar disc herniation (LDH) and to identify an ideal time window for surgical intervention.
Methods
In a single-centre observational trial, 390 patients with LDH-associated motor deficits were prospectively followed for a minimum of 12 months after nonelective microscopic disc surgery. The duration of motor deficit before surgery was documented. Motor function was graded according to the Medical Research Council (MRC) scale. Statistical analysis of motor recovery applied unbiased recursive partitioning conditional inference tree (URP-CTREE) to determine cut-off times to determine optimal windows of opportunity for surgical intervention. The slope of recovery calculated as change of the MRC grade over time served as the primary outcome.
Results
A preoperative motor deficit of MRC less than or equal to 2/5 and the duration of paresis were identified as the most important predictors of recovery (p < 0·001). Surgery within 3 days was associated with a better recovery both for severe and for moderate/mild deficits (p = 0.017 for MRC ≤ 2/5; p < 0.001 for MRC > 2/5; number needed to treat < 2). A sensitivity analysis in mild motor deficits indicated a cut-off of 8 days.
Conclusion
Timing of surgery is crucial for motor recovery in LDH-associated deficits. Immediate diagnosis, imaging and prompt referral to spine surgeons should be aimed for to allow disc surgery within 3 days in patients with severe and moderate radicular weakness. If functionally disabling, even mild deficits may warrant decompression within a week.
Abstract 106. Outcomes following revision decompression for lumbar spinal stenosis when compared to primary decompression: a matched cohort analysis using the Canadian Spine Outcomes and Research Network registry
Background
Symptomatic lumbar spinal stenosis is a common entity and increasing in prevalence. Limited evidence is available regarding patient-reported outcomes (PROMs) comparing primary versus revision surgery for those undergoing lumbar decompression. Evidence available suggests a lower rate of improvement in the revision group. The aim of this study was to assess PROMs in patients undergoing revision decompression, with or without fusion, when compared with primary surgery.
Methods
Patient data were collected from the Canadian Spine Outcomes and Research Network registry. Patients undergoing lumbar decompression without or without fusion were included. Patients under age 18 years, undergoing discectomy, with greater than 2-level decompressions, with concomitant cervical or thoracic spine surgery were excluded. Demographic data, smoking status, narcotic use, number of comorbidities as well as individual comorbidities were included in our propensity scores. Patients undergoing primary versus revision decompression were matched in a 4:1 ratio according to their scores, while a separate matched cohort was created for those undergoing primary versus revision decompression and fusion. Continuous data were compared using a two-tailed t test, while categorical variables were assessed using a χ2 test.
Results
A total of 555 patients were included, with 444 patients undergoing primary surgery matched to 111 patients undergoing revision surgery, of whom 373 (67%) did not have fusion. Patients undergoing primary decompression with fusion compared with those undergoing revision were more likely to answer yes to “feel better after surgery” (87.8% v. 73.8%, p = 0.023), “undergo surgery again” (90.1% v. 76.2%, p = 0.021) and “improvement in mental health” (47.7% v. 28.6%, p = 0.03) at 6 months. There was no difference in either of these outcomes at 12 or 24 months. There was no difference between the groups’ back or leg pain, Oswestry Disability Index, EQ-5D or SF-12 scores at any time. Patients undergoing primary versus revision decompression alone showed no difference in any PROMs at any time.
Conclusion
In a matched cohort, there appears to be no difference in improvement in PROMS between patients undergoing primary versus revision decompression, with or without fusion, at 2 years. This suggests similar outcomes can be obtained in revision cases.
Abstract 64. Patient engagement, remote monitoring and virtual care — a pilot project in rural and remote patients undergoing elective lumbar surgery
Background
Patients from rural and remote communities face multifactorial barriers in attempting to access perioperative care. This has been further amplified by the current pandemic. Tools such as remote patient-monitoring apps and virtual platforms are increasingly being utilized to bridge these gaps. The primary objective of this study was to assess the engagement of patients from these communities utilizing a remote patient-monitoring app. Secondary objectives included patient satisfaction, reported outcome measures and length of stay.
Methods
We implemented an app-based remote patient virtual platform to facilitate perioperative information and care in patients undergoing elective lumbar surgery in rural and remote centres. Patients are contacted within 4 weeks of the surgery and are followed until 30 days after surgery. The app is monitored by a nurse practitioner, prospectively collects data, and has alerts and health checks built within the system. The platform collects basic demographic data and was designed to increase patient engagement, monitor adverse events in real time and record validated outcome scores up to 1 year postoperatively.
Results
In total, 92.5% of patients activated the app. Sixty-seven patients have been enrolled to date (31 males), with 40 having completed the perioperative course, and 27 patients are currently actively engaged. Average age is 57.3 years. In total, 96.6% of patients would recommend the app, more than 93% preferred the app over paper, and 72% and 93% felt that they were less worried before and after surgery, respectively. In addition, 70.3% reported it helped them avoid at least 1 call to the hospital, and 22.2% reported that it helped avoid at least 1 visit to the emergency department. Visual analogue scale scores for leg pain improved on average by 4.07. Average LOS and emergency department visits decreased by 68.5% and 40.1%, respectively.
Conclusion
The implementation of a nurse practitioner–led remote patient monitoring app was successful in engaging patients with their perioperative care. There was high satisfaction and a decrease in LOS and the rate of return to the hospital. This tool has the potential to be useful in facilitating patient care and initiatives such as enhanced recovery after surgery.
Abstract 84. Development of a frailty index from the Canadian Spine Outcomes and Research Network (CSORN) to predict long-term success of surgery for patients with degenerative pathologies of the spine
Background
The concept of frailty reflects a reduced ability to respond to stressors, and it has been more frequently observed in surgical candidates than in the general population. Frailty indexes (FI) have been developed to aid clinical decision-making but their application is time consuming. We sought to develop a FI based upon routinely collected data for patients undergoing spine surgery and to examine how frailty affected postoperative disability.
Methods
A FI was developed using standard data points collected by the Canadian Spine Outcomes and Research Network (CSORN); a total of 56 variables contributed to the index. These variables included pain scores, EQ-5D, PHQ-9, comorbidities and the SF-12. The FI was used to determine how frailty affected patient disability following surgery. A binary logistic regression was performed for the dependent variable of achieving a minimal clinically important difference (MCID) for the Oswestry Disability Index. Significance was taken at p less than 0.05.
Results
A total of 5829 people were included. Patients demonstrating greater degrees of frailty were more likely to be women (p < 0.001), to be smokers (p < 0.001), to have a higher body mass index (BMI) (p < 0.001) and to be younger (p < 0.001). At 3 months following surgery, greater frailty increased the chance of achieving MCID, with mildly frail (0.2–0.3 FI) people being 1.67 times more likely, moderately frail (0.3–0.4 FI) being 2.07 times more likely and severely frail (> 0.4 FI) being 2.24 times more likely to achieve MCID than the least frail patients (all p < 0.001). At 1 year after surgery, mildly frail patients were 1.50 times (p = 0.002), moderately frail patients were 1.72 times (p < 0.001) and severely frail patients were 1.71 times (p < 0.001) more likely to achieve MCID. At 2 years, only BMI and age were predictive of achieving MCID (p < 0.001 and p < 0.001).
Conclusion
The FI developed through the CSORN database can help predict the chances of a patient achieving MCID for their disability following surgery. The FI is also easy to implement as it is based on commonly collected data for patients. Patients with a higher FI are much more likely to achieve MCID.
Abstract 102. Posterolateral versus posterior interbody fusion for the management of lumbar degenerative spondylolisthesis: analysis from the CSORN prospective LDS propensity score matched study
Background
The benefit of interbody fusion (IF) over posterolateral fusion in the treatment of spondylolisthesis is controversial. Little evidence focuses on the treatment of degenerative spondylolisthesis. The objective of this study was to compare postoperative patient-rated outcomes (PROs) of posterolateral fusion (PLF) versus IF surgery in patients with lumbar degenerative spondylolisthesis.
Methods
This is a retrospective analysis of data from a Canadian Spine Outcomes and Research Network (CSORN) multicentre prospective study on the assessment and management of patients with lumbar degenerative spondylolisthesis (LDS). Inclusion criteria included: LDS at 1 or 2 levels, age 18 years or older, IF or PLF surgery, and minimum 1 year follow-up after surgery. Propensity-score matching was used to match similar cohorts of patients who underwent IF and PLF on the basis of baseline patient characteristics including sex, age, primary symptom type (radiculopathy v. claudication back pain), body mass index, SF-12 mental component score (MCS), spondylolisthesis grade, disc angle (lordotic v. kyphotic/neutral), smoking status, levels fused, back pain and presence of neurologic deficit (motor or sensory). Patient-reported outcomes including numeric rating scale (NRS) back pain, NRS leg pain, Oswestry Disability Index (ODI), and SF-12 MCS and physical component score (PCS) were compared over time using repeated-measures mixed-effects modelling.
Results
Eight centres participated in the study from January 2015 to September 2020. Of the 567 enrolled patients, 278 (56 with PLF, 222 with IF) met the inclusion criteria. After propensity score matching, 48 patients were included in each group. The follow-up rate was 80% at 1 year. Study participants mostly were female and had grade I spondylolisthesis in a single level, neurogenic claudication and symptom duration greater than 2 years. At 3 and 12 months there were no significant differences in the intensity of back pain, leg pain or ODI, MCS or PCS scores between the PLF and IF groups. Operation time, blood loss, adverse events and length of stay were equal between the groups.
Conclusion
Our study found equivalence in patient-reported outcomes between PLF and IF in propensity-matched patients with degenerative spondylolisthesis at 12 months. Longer term follow-up and focused investigation are required to determine if specific indications exist for IF in the management of degenerative spondylolisthesis.
Abstract 31. Impact on patient-reported outcomes of ending the posterior construct proximally at C2 versus C3 in degenerative cervical myelopathy patients
Background
Limited evidence exists regarding proximally ending the posterior construct at C2 versus C3, and its impact on patient-reported outcomes (PROs) remains unknown. The objectives were to evaluate the impact of ending the posterior construct proximally at C2 versus C3 on PROs and modified Japanese Orthopaedic Association (mJOA) scores at 3 and 12 months after surgery and to compare operative time, intraoperative blood loss (IOBL), length of stay (LOS) and adverse events (AEs).
Methods
This is a retrospective study of a prospectively multicentric followed cohort of patients with degenerative cervical myelopathy enrolled between January 2015 and October 2020. Patients who underwent a posterior instrumented fusion greater than 3 levels with the proximal end at C2 or C3 with 12 months follow-up were included. PROs (Neck Disability Index, EQ-5D, SF-12 physical component score [PCS] and mental component score [MCS], numeric rating scale [NRS] arm and neck pain) and mJOA scores were compared using analysis of covariance, adjusted for baseline differences. Operative duration, IOBL, AEs and discharge disposition were compared using χ2 tests for categorical variables and independent-samples t tests for continuous variables.
Results
A total of 208 patients were included, of whom 54 had the upper instrumented level being C2 and 154 being C3. When adjusted for baseline differences, there was no statistically significant difference between the 2 groups for the PROs and mJOA scores at 12 months. At 3 months, NRS neck pain and SF-12 MCS scores were better in the group with the construct ending at C3 (p = 0.03 and 0.015, respectively). Longer operative times and IOBL were associated with the C2 cohort (p < 0.05). The C3 group had significantly more perioperative AEs (p < 0.05). There was no significant difference in LOS and discharge disposition between the groups (p < 0.05).
Conclusion
At 12 months after surgery, there was no statistically significant difference in PROs and mJOA scores whether the construct ended proximally at C2 or C3. The NRS neck pain and SF-12 MCS scores were better in the C3 group at 3 months. While the C3 cohort had higher perioperative AEs, the C2 group had longer operative times and increased IOBL.
Abstract 42. Perioperative factors predict 2-year trajectories of pain and disability following anterior cervical discectomy and fusion
Background
The objectives of this study were to identify postoperative trajectories of disability and neck and arm pain for patients undergoing anterior cervical discectomy and fusion (ACDF) and to determine baseline measures that predict trajectory subgroup membership.
Methods
This was a retrospective cohort study of patients with cervical spondylotic radiculopathy undergoing ACDF. Candidate prognostic factors comprised demographic, health, clinical and surgery-related variables. Study outcomes were trajectories of Neck Disability Index scores and numeric rating scales for neck and arm pain intensity modelled with latent-class growth analysis. Associations between candidate prognostic factors and postoperative trajectory subgroups were explored using robust Poisson models and reported with risk ratios and 95 % confidence intervals.
Results
We modelled data from 352 patients (mean age 50.9 [standard deviation (9.5)] yr, 43.8% female) and identified 3 distinct trajectories for disability (excellent 45.3%, fair 39.2%, poor 15.5 %), arm pain (excellent 24.5%, good 52.0%, poor 23.5%) and neck pain (excellent 13.7%, good 63.1 %, poor 23.2%). Higher physical and mental health-related quality of life were associated with a reduced risk of poor outcome for neck pain, arm pain and disability (per SD: 0.40 [range 0.30–0.53] to 0.80 [range 0.65–0.99]). Conversely, higher risk for depression (per SD: 1.36 [range 1.12–1.65] to 2.26 [range 1.84–2.78]), longer surgical wait time (per 90 d: 1.31 [range 1.05–1.63] to 1.64 [range 1.20–2.24]) and longer procedure time (per 30 min: 1.07 [range 1.03–1.10] to 1.08 [range 1.05–1.12]) were associated with an increased risk of poor outcome for neck pain, arm pain and disability. Poor disability outcome risk was increased with self-reported depression (3.03 [range 1.76–5.21]), greater neck-to-arm pain ratio (2.63 [range 1.28–5.40]), American Society of Anesthesiologists (ASA) score greater than 2 (2.26 [range 1.33–3.83]) and preoperative opiates (2.05 [range 1.18–3.56]), while preoperative physiotherapy (0.51 [range 0.30–0.88]), spinal injections (0.48 [range 0.23–0.98]) and regular exercise (0.44 [range 0.24–0.79]) decreased risk. Receiving compensation (1.56 [range 1.00–2.44]) and smoking (1.75 [range 1.15–2.66]) were associated with a poor outcome for neck pain only. Remaining candidate prognostic factors were not associated with clinical outcome.
Conclusion
A subset of patients experience poor outcomes following ACDF. Perioperative factors were shown to decrease risk of poor outcomes for pain and disability 2 years following ACDF. This knowledge may inform decision-making for policy-makers and clinicians.
Abstract 61. Calculating utilities from the modified Japanese Orthopaedic Association score: a prerequisite for quantifying the value of care for cervical myelopathy
Background
The ability to calculate quality-adjusted life-years (QALYs) for degenerative cervical myelopathy (DCM) would enhance treatment decision-making and facilitate economic analysis. QALYs are calculated using utilities, or health-related quality-of-life (HRQoL) weights. An instrument designed for cervical myelopathy disease would increase the sensitivity and specificity of HRQoL assessments. The objective of this study is to develop a multi-attribute utility function for the modified Japanese Orthopaedic Association (mJOA) score.
Methods
We recruited a sample of 760 adults from a market research panel. Using an online discrete choice experiment (DCE), participants rated 8 choice sets based on mJOA health states. A multi-attribute utility function was estimated using a mixed multinomial-logit regression model (MIXL). The sample was partitioned into a training set used for model fitting and validation set used for model evaluation.
Results
The regression model demonstrated good predictive performance on the validation set with an area under the receiver operating characteristic curve (AUC) of 0.81 (95% confidence interval 0.80–0.82). The regression model was used to develop a utility scoring rubric for the mJOA. Regression results revealed that participants did not regard all mJOA domains as equally important. The rank order of importance was (in decreasing order): lower extremity motor function, upper extremity motor function, sphincter function, upper extremity sensation.
Conclusion
This study provides a simple technique for converting the mJOA score to utilities and quantifying the importance of mJOA domains. The ability to evaluate QALYs for DCM will facilitate economic analysis and patient counselling. Clinicians should use these findings to offer treatments that maximize function in the attributes viewed most important by patients.
Abstract 119. Serum neurofilament light (NF-L) and glial fibrillary acidic protein (GFAP) biomarkers and their association with MRI findings in human acute traumatic spinal cord injury
Background
Injury severity after traumatic spinal cord injury (SCI) in the acute setting is assessed by a standardized neurological examination that is subjective, poorly predictive of outcome and often impossible or severely confounded by pharmacologic sedation or concomitant injuries. Therefore, biomarkers that objectively characterize severity and are able to predict outcome are urgently needed for clinical decision-making, communication with patients and their families as well as study design. Preliminary data show that serologic values of neurofilament light NF-L and glial fibrillary acidic protein (GFAP) are interesting candidates for objective diagnostic and prognostic biomarkers. Previous studies also tried to assess magnetic resonance imaging (MRI) findings as potential surrogate markers for SCI. The following study aims to combine certain MRI findings with serum values of NF-L and GFAP as potential markers to characterize injury severity, predict outcome and eventually monitor treatment response.
Methods
This is a retrospective analysis of prospectively collected data. Patients with acute traumatic SCI were included in an observational trial in which serial serum samples were collected in the first 5 days after injury. The concentrations of NF-L as well as GFAP were analyzed using Simoa technology. American Spinal Injury Association (ASIA) impairment scale (AIS) grade and motor scores were obtained according to the International Standards for Neurological Classification of SCI at presentation and at 6 months after injury. As MRI parameters we assessed the maximal canal compromise (MCC), maximal spinal cord compromise (MSCC), intramedullary lesion length, intramedullary hemorrhage, Brain and Spinal Injury Center (BASIC) score and sagittal grade. We included all patients from our institution who were enrolled in the prospective trial and had preoperative MRI available.
Results
A total of 51 patients met the inclusion criteria and had a preoperative MRI available in our institution. In a multicentre trial, serologic NF-L and GFAP levels predicted AIS improvement with 75%–80% accuracy and were able to discriminate between motor complete and incomplete with 90% accuracy.
Conclusion
Preliminary data indicate that Nf-L and GFAP levels in the serum qualify as potential biomarkers after acute human SCI. This study will evaluate the potential additional prognostic value by adding several MRI parameters.
Abstract 95. The Montreal Acute Classification of Spinal Cord Injury (MAC-SCI): a new tool to detect and characterize spinal cord injury in the trauma patient
Background
Spinal cord injury (SCI) is most often secondary to trauma and frequently presents with associated injuries. A neurologic examination is routinely performed during trauma assessment, such as the Advanced Trauma Life Support (ATLS). However, there is no standard neurologic assessment tool for trauma patients to detect and characterize SCI during the initial evaluation. The International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) is the most comprehensive and popular tool for assessing SCI, but it is not adapted to the acute trauma patients. Therefore, the objective is to develop a new tool that can be used routinely in the initial evaluation of trauma patients to detect and characterize acute SCI.
Methods
Using a modified Delphi technique, we designed the Montreal Acute Classification of Spinal Cord Injury (MAC-SCI). The ability of the MAC-SCI to detect and characterize SCI was validated in a cohort of 33 individuals who had sustained an acute traumatic SCI. The severity grade and neurologic level of injury (NLI) were assessed by 2 independent assessors using the MAC-SCI and compared with those obtained with the ISNCSCI.
Results
The MAC-SCI includes 53 of the 134 original elements of the ISNCSCI. The severity grade of the SCI was identified correctly in 100% of the cases showing perfect agreement between the MAC-SCI and the ISNCSCI. Accurate identification of the NLI within 2 levels of that obtained from the ISNCSCI was observed in 94% of patients. The ability of the MAC-SCI to discriminate between cervical (C0 to C7) versus thoracic (T1 to T9) versus thoracolumbar (T10 to L2) versus lumbosacral (L3 to S5) injuries was 100% with respect to the ISNCSCI.
Conclusion
The MAC-SCI is a simple tool proposed to detect and characterize acute SCI in polytrauma patients, which is specifically adapted to the acute trauma setting. It is accurate for determining the severity grade and localize the NLI. It could be implemented in the initial trauma assessment protocol to guide the acute management of patients with SCI.
Abstract 118. Mechanism of injury is associated with neurologic outcomes after cervical sensorimotor complete acute traumatic spinal cord injury
Background
The aim of this study was to compare the impact of high-energy (HEMI) and low-energy (LEMI) mechanisms of injury in neurologic outcomes of patients with acute traumatic cervical sensorimotor complete spinal cord injury (SCI).
Methods
Adult patients with acute traumatic SCI (aged 18–80 yr) were identified in a combined data set of 4 prospective, multicenter clinical trials and registries. American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade was assessed within 72 hours after injury and followed up between 12 and 52 weeks. Patients were included if they had a cervical and sensorimotor complete (AIS grade A) injury at baseline. Primary outcome was change in AIS grade. Secondary outcomes were changes in light-touch, pinprick, lower extremity motor (LEMS), upper extremity motor (UEMS) and total motor (TMS) scores. Missing information was imputed with multiple imputation. Propensity-score matching between HEMI (e.g., sports, motor vehicle collisions) and LEMI (e.g., falls) groups was performed using a caliper width of 0.2 for baseline variables that were significantly different (p < 0.05).
Results
Of 706 patients eligible for inclusion, 546 experienced HEMIs (77.3%). Patients with HEMI were younger, had lower body mass index, had more associated fractures or dislocations and had lower baseline LEMS. After propensity-score matching of these baseline variables, 142 pairs (n = 284) were matched. Upon matching, there were no significant between-group differences on early surgery or steroid administration. Patients with HEMI experienced significantly worse AIS grade conversion (HEMI 38.7%, LEMI 54.9%; p = 0.009) compared with patients with LEMI. Patients with HEMI had worse improvements in UEMS (HEMI 7.9 [standard deviation (SD) 8.0], LEMI 10.7 [SD 9.6]; p = 0.008) and TMS (HEMI 13.3 [SD 16.2], LEMI 18.8 [SD 21.1]; p = 0.015). Difference in LEMS recovery was not statistically significant (HEMI 5.6 [SD 10.5], LEMI 8.5 [SD 14.0]; p = 0.052). Patients with HEMI also struggled more to recover light-touch (HEMI 20.7 [SD 27.1], LEMI 29.0 [SD 31.1]; p = 0.018) and pinprick (HEMI 16.4 [SD 24.5], LEMI 23.6 [SD 27.9]; p = 0.021) sensory function.
Conclusion
Patients with cervical sensorimotor complete HEMI had significantly poorer motor and sensory recovery than patients with LEMI, which reflects the severity of outcome. Our findings suggest that energy of trauma should be considered in modelling prognosis and in understanding the heterogeneity of outcomes after acute traumatic SCI.
Abstract 13. Patient perspective: diagnosis and prognosis of acute spinal cord injuries
Background
The purpose of this study is to understand the patient perspective after diagnosis of an acute traumatic spinal cord injury (tSCI). Discussing the diagnosis and prognosis of a tSCI with a patient can be a challenging experience for both the patient and the physician. As such, this paper attempts to better understand the patient experience to improve communication when discussing this life--altering injury.
Methods
This study is a qualitative study utilizing grounded theory and semistructured interviews. The interview transcripts were manually coded using manifest and latent content analysis. Major and minor codes were identified and discussed.
Results
In total, 17 interviews were conducted; 15 individuals with tSCI who received acute care at Vancouver General Hospital (VGH) and 11 family members were interviewed. Patient participants were interviewed individually or in a paired interview with a participating family member. Patient participants had varying spinal cord injuries from American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade A to D. Two major themes were identified from the interviews. The first major theme was physician demeanor (general approach and attitude toward patients) and the second major theme was delivery of information (content, timing and source).
Conclusion
This study summarizes the preferences of patients who sustained a tSCI concerning discussions regarding their diagnosis and prognosis in the acute care setting. The goal of this study is to help guide physician interactions at this difficult and vulnerable time for patients with hopes to improve patient care through effective communication.
Abstract 136. Predictive analytics to improve dedicated spine trauma operating room resource allocation
Background
Operating room (OR) scheduling of trauma cases is a complex issue with substantial room for improvement. The dedicated trauma room (DTR) model is increasingly being used to address the challenge of scheduling trauma cases. However, insufficient DTR operating room (OR) time leads to after-hours care, while excess time leads to either unused or last-minute redistribution of OR time. We hypothesize that a machine learning model (MLM) can predict trauma volume and allow for schedule optimization to better match supply and demand for the limited resource of OR time. In our study, we compare the current state of spine DTR scheduling with a proposed future state that may be further optimized using a MLM.
Methods
Demand on OR time was derived from a data set of procedures completed by the orthopedics department at a level 1 trauma center. The following data were extracted using a language processing algorithm: date, number of procedures completed and admissions. A regression model was used to determine the variables that significantly contributed to demand. These variables were integrated into a linear integer optimization to create a proposed future state DTR schedule that minimized the number of after-hours cases. This was then compared with the current DTR schedule.
Results
Seasonality and weekday-versus-weekend significantly affected demand on OR time (p < 0.01). Under the current-state spine DTR model (2 DTR/wk), there is an average of 3.4 (winter), 2.6 (summer), 1.3 (fall) and 0.8 (spring) after-hours cases being completed per week. Under the proposed future-state spine DTR model, (3 DTR/wk in winter and summer), the after-hours cases decrease to 1.4 and 0.6, respectively.
Conclusion
Seasonality and weekday-versus-weekend were determined to be variables that affect demand on OR time. A DTR optimization model has been identified that can meaningfully guide OR scheduling. Next steps of this work would be to predict the demand on the OR with a MLM to optimize the schedule, which has broad implications for hospital resource allocation.
Abstract 138. Machine learning models to predict surgical resident workload at a level 1 trauma centre
Background
Excessive resident duty hours (RDH) are a recognized issue with implications for physician well-being and patient safety. A major component of the RDH concern is on-call duty. However, there is a paucity of research on optimizing resident call scheduling. Call coverage is scheduled manually rather than demand based, which generally leads to overscheduling to prevent a service gap. Machine learning (ML) has been widely applied in other industries to prevent such issues of a supply–demand mismatch. As such, the aim of this study was to use ML models to (1) predict demand on surgical residents at a level 1 trauma centre and (2) identify variables key to demand prediction.
Methods
Daily surgical handover emails over an 8-year period were collected. The following data were used to calculate demand: spine call coverage, date, number of operating rooms (ORs), traumas, admissions and consults completed. Various ML models were trained to predict the workload, and their results were compared with the current scheduling approach. Quality of models was determined by using the area under the receiver operating characteristic curve (AUC) and accuracy. The top 10 most important variables were extracted from the most successful model.
Results
During training, the model with the highest AUC and accuracy was the earth model (AUC 0.78 [standard deviation (SD) 0.03], accuracy 71.7% [SD 3.1%]). During testing, the model with the highest AUC metric and accuracy was the neural network model (AUC 0.81, accuracy 73.7%). All models were better than the current approach (AUC 0.50, accuracy 50.1%). Key variables used by the neural network model were (descending order): spine call duty, year, weekday/weekend, month and day of the week.
Conclusion
To our knowledge, this was the first study attempting to use ML to predict the service demand on surgical residents. Spine call duty was shown to be the most important feature of demand prediction. Future work should look to incorporate predictive models with optimization strategies to match scheduling with demand.
Abstract 139. Machine learning to predict duration of surgery and length of stay for single-level discectomy procedures
Background
Single-level discectomy (SLD) is one of the most common spinal procedures. Two key drivers of their cost of care are duration of surgery (DOS) and postoperative length of stay (LOS). The ability to preoperatively predict DOS and LOS for SLD therefore has substantial implications for hospital finances, scheduling and resource allocation. The goal of this study was to predict DOS and LOS for SLD using machine learning models (MLMs) constructed on preoperative factors using a large North American database.
Methods
The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database was queried for SLD procedures from 2014 to 2019. The data-set splitting (training/validation/testing) was based on year. Various MLMs were used and evaluated according to (1) mean squared error (MSE), (2) buffer accuracy (the number of times the predicted target was within a predesignated buffer), and (3) classification accuracy (the number of times the correct class was predicted by the models). The results of the models were compared with a mean regressor.
Results
A total of 11 525 patients were included in this study. During validation, the neural network model (NNM) had the best MSEs for DOS (0.99) and LOS (0.67). During testing, the NNM had the best MSEs for DOS (0.89) and LOS (0.65). The NNM yielded the best 30-minute buffer accuracy for DOS (70.9%) and less than or equal to 120 minutes, more than 120 minutes classification accuracy (86.8%). The NNM had the best 1-day buffer accuracy for LOS (84.5%) and less than or equal to 2 days, more than 2 days classification accuracy (94.6%). All models were more accurate than the mean regressors for both DOS and LOS predictions.
Conclusion
We successfully demonstrated that MLMs can be used to accurately predict the DOS and LOS of SLD on the basis of preoperative factors. This big-data application has substantial practical implications with respect to surgical scheduling and inpatient bedflow. Incorporating this artificial intelligence technique in real-time hospital operations would be enhanced by including institution-specific operational factors such as surgical team and operating room workflow.
Abstract 9. Outpatient spinal surgery in Manitoba
Background
The aim of this study was to assess the feasibility and safety of an ambulatory spine surgery delivery model at the University of Manitoba with an emphasis on highlighting elements that can be used to reduce length of stay (LOS) and representation to hospital.
Methods
This is a retrospective cohort study utilizing prospectively gathered data from the Canadian Spine Outcomes and Research Network (CSORN) along with independent chart review. The patient population included 200 adult patients who underwent outpatient spinal surgery between 2015 and 2018. Patient demographics, comorbidities, perioperative course, LOS and re-presentations were analyzed.
Results
A total of 217 patients were included in the analysis. The mean LOS was 36.2 hours; 71.98% of patients had a LOS less than 24 hours. A Kruskal–Wallis test by ranks analysis was conducted and identified 4 elements that correlated with prolonged length of stay (over 1 d): age over 55 years (p = 0.027), body mass index (BMI) above 25 (p = 0.045), uncontrolled diabetes (p = 0.015) and preoperative use of opioid medication (p = 0.024). A total of 86.8% of patients with none of these elements (i.e., age < 55 yr, low BMI, normal or controlled diabetes, minimal preoperative opioid use) had LOS less than 1 day, compared with 76.7% of those with 1 out of 3, 62.2% of those with 2 out of 3, and finally 44.4% of those with all three. A total of 15.0% of patients presented to an emergency department within 90 days of their operation, and there was a 6.28% readmission rate.
Conclusion
Ambulatory spine surgery is feasible and safe; however nonadherence to strict patient selection criteria may lead to prolonged postoperative stays.
Abstract 131. Unexpected positive culture in presumed aseptic revision spine surgery: a systematic review and meta-analysis
Background
Unexpected intraoperative positive (UIPC) culture has recently become an increasingly important area in revision spine surgery, and it has been reported as an underestimated underlying cause of revision spine surgery, such as pseudoarthrosis. This study aimed to investigate the prevalence of UIPC in patients following presumed aseptic revision spine surgery.
Methods
Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, multiple databases were searched until October 2021, including reference articles. We searched for studies that reported the rate of unexpected intraoperative culture in assumed aseptic spine revision surgery. The primary outcome was the pooled rate of UIPC, whereas secondary outcomes included the microbiologic profile of UIPC and the relationship of UIPC to the different patient and surgical variables.
Results
Of the 1036 articles identified, 11 studies were eligible for the meta-analysis, with a total of 1055 patients. The pooled rate of UIPC was 28.2% (95% confidence interval [CI] 19.1–39.1, I2 91%, p < 0.01). Subgroup analysis showed that the UIPC rate was higher when both conventional wound culture and sonication were used together compared with sonication alone or conventional wound culture alone. The rates were 31.5%, 28.6% and 18.6%, respectively. The most commonly reported organism was Propionibacterium acnes (50%), followed by coagulase-negative staphylococci (36%). Male patients had a higher UIPC (odds ratio 2.7, 95% CI 1.88–3.92, p < 0.001), as did patients with a longer involved spine segment (mean difference 0.76, 95% CI 0.27–1.25, p < 0.001).
Conclusion
The pooled rate of UIPC in aseptic spine revision surgery was 28.2%. The most common organisms were P. acne and coagulase-negative staphylococci. There is not evidence to recommend specific treatment. Further studies are needed to establish comprehensive treatment guidelines and diagnostic criteria.
Abstract 50. Lumbar anterior vertebral body tethering: biomechanical assessment of the surgical decision factors influencing the immediate and 2 years postoperative correction
Background
Anterior vertebral body tethering (AVBT) is a fusionless surgical technique used to treat pediatric scoliosis. To date, AVBT has been used primarily to treat the thoracic spine. Compared with the limited mobility of the thoracic spine, lumbar AVBT offers an additional value proposition as it has the potential to preserve motion. However, there is still a need to evaluate the effects of lumbar AVBT parameters on growth over time. The objectives of this study were to adapt and validate an AVBT finite element model (FEM) for the lumbar spine instrumentation and to study the effects of cable tension and vertebral level selection on growth modulation.
Methods
Ten patients with adolescent idiopathic scoliosis who had undergone AVBT surgery were retrospectively selected (5 lumbar and 5 double instrumentations). 3D reconstructions were done using low-dose biplanar radiographs. The patient-specific FEM was built and calibrated for each case. The actual AVBT instrumentation and growth modulation over a period of 12–24 months were simulated and validated against postoperative radiographs. Three AVBT surgical parameters were evaluated using an experiment study design: cable tension (150 N, 250 N), and upper/lower instrumented levels (UEV+1, LEV-1). Computed thoracic and lumbar Cobb angles, as well as spine height (T1–L5) were evaluated.
Results
There was no significant difference between simulated and actual Cobb angles at the immediate and 12- to 24- month postoperative evaluations (less than 4°, p > 0.05). Increasing cable tension from 150 N to 250 N significantly increased the simulated correction for thoracic and lumbar Cobb angles after surgery (4°/7°) and after 24 months (6°/9°) (p < 0.05). The inclusion of 1 more upper and lower instrumented level did not improve the postoperative correction.
Conclusion
Increasing the tension was the key surgical parameter providing the most significant effect on correction. Given the difficulty of instrumenting the lumbar segment, especially the most caudal part, these initial results suggest that it is not necessary to extend the instrumentation to significantly increase the correction. The study is ongoing to include more patients and test other parameters.
Abstract 145. Does prolonged symptom duration influence surgical outcomes for cervical radiculopathy?
Background
Multiple studies have shown better outcomes for cervical radiculopathy when anterior cervical discectomy, fusion and plating (ACDFP) is performed within 6 months of symptom onset. In Canada, more than 50% of patients have symptoms longer than 2 years. The objective of this study was to determine if there remains benefit to earlier surgery in a Canadian cohort with longer average wait times.
Methods
This is a retrospective chart review of consecutive patients who underwent single- or two-level ACDFP for cervical radiculopathy due to spinal stenosis or disc herniation between May 1, 2012, and May 1, 2018. Primary outcome measures were as follows: Neck Disability Index (NDI), visual analogue scale pain scores for arm and neck (VASneck, VASarm) and EuroQol-5D (EQ-5D) quality of life pain scores. We analyzed preoperative scores and compared them with early outcomes (6–8 wk) and late outcomes (18–24 mo).
Results
A total of 120 patients were included. The average preoperative symptom duration in the entire cohort was 37 months (range 0–250 mo). We analyzed 2 groups: those with symptoms lasting 24 months or less (group A) and those with symptoms lasting greater than 24 months (group B). There were no other significant baseline differences between the groups. Group A patients reported less VAS neck pain at early follow-up (p = 0.022) and less NDI disability at late follow-up (p = 0.039).
Conclusion
In a Canadian population with long wait times, there are still improved outcomes with earlier surgery. Patients with cervical radiculopathy for less than 24 months saw earlier relief of neck pain and reported less disability up to 2 years after ACDFP.
Abstract 147. A data-driven cluster analysis approach to create homogenous subgroups for traumatic spine injury: toward improving traditional classification
Background
Clinical trials for traumatic spine injury (TSI) are difficult to conduct because of the amount of variation in neurologic impairments and patient characteristics within a population. Identifying homogeneous subgroups may improve the effective implementation of treatment strategies and patient trajectories. We hypothesize that using automated clustering can mimic or improve preexisting (traditional) classification
Methods
Our objective was to identify the features influencing the optimized clusters creating each homogenous group of patients.
Methods
Spectral cluster analysis was deployed to group similar patients on the basis of baseline prognostic factors: location of injury, severity of injury, baseline Functional Independence Measure (FIM) motor score and demographic data (age, body mass index). The FIM motor score at follow-up and the total length of stay were superimposed on identified clusters as outcome variables. Prognostic factors used in spectral cluster analysis and superimposed outcome variables were assessed statistically. The performance of the spectral algorithm was evaluated by the silhouette coefficient, the Davies-Bouldin index and the elbow method to determine the optimal number of clusters. The identified clusters were qualitatively and systematically described to explore patient-focused insights.
Results
A total of 338 patients with TSI from the Rick Hansen Spinal Cord Injury Registry (RHSCIR) were analyzed from baseline to follow-up. Five distinct clusters were identified with a statistically significant difference (p ≤ 0.05) for prognostic factors. Outcome variables were statistically significant among all clusters.
Conclusion
Using automated clustering, we found significant differences between prognostic variables and superimposed outcome variables across all 5 clusters. Qualitative descriptions of distinguishing prognostic variables for each cluster revealed that the spectral cluster analysis provided clinically intuitive groupings of patients. Additionally, we observed a relationship between the location of the injury and the mechanism of injury among all clusters, which adds significant clinical categorizations of patients with TSI in their classification. This data-driven approach identified homogenous subgroups without depending on a priori decisions, which is a step toward improving traditional classification.
Abstract 41. The use of neo-adjuvant denosumab in treatment of giant cell tumours of the spine
Background
Giant cell tumours (GCT) of the spine may be large at presentation and cause severe pain. The current recommended treatment is en bloc excision but it is associated with substantial morbidity and mortality. Denosumab is a monoclonal RANKL inhibitor that may be used neoadjuvantly. The goal of this study was to assess the effect of denosumab on tumour characteristics and symptom relief.
Methods
We performed a retrospective review of 10 patients treated with denosumab as neoadjuvant and stand-alone treatment. Tumour measurements were taken before and after treatment, positron emission tomographic (PET) standardized uptake value (SUV) capitation was measured, and patients were interviewed for subjective pain responses. Clinical response was determined by volumetric reduction in tumour size, PET SUV capitation, the Boriani classification and improvement in pain symptoms.
Results
Following treatment 70% of patients were pain free, with 50% noting improvement within 48 hours. Mean relative volumetric reduction in tumour volume was 40%. All pathology specimens confirmed elimination of giant cells. Improvement in Bilsky grading occurred in 4 of 10 cases and progression was halted in the remainder. Median baseline SUVmax was 14.7, and median SUVmax after treatment was 6.2. Seventy-eight percent of patients demonstrated intra lesional bone formation following treatment.
Conclusion
This study demonstrates that neoadjuvant denosumab facilitates en bloc resection of GCT of the spine, reduces the likelihood of intraoperative morbidity and improves preoperative pain. We recommend routine use when Weinstein–Boriani–Biagini–based criteria are fulfilled for en bloc excision. Assuming that margins are disease free following surgery, we advocate cessation of treatment postoperatively.
Abstract 5. Complications, revision rates and accuracy of robotic-guided S2 alar-iliac screw placement
Background
There are multiple methods described for the placement of S2 alar-iliac (AI) screws: free handed, fluoroscopic guided, navigated and robotic guided. This study will exclusively analyze a large cohort of patients in which the S2AI screws were placed under robotic guidance, specifically looking at complications, revision rates and placement accuracy. Complications and revision rates associated with S2AI screws placed using robotic guidance are hypothesized to be low and the accuracy grading is hypothesized to be high.
Methods
We retrospectively reviewed 51 consecutive patients who underwent S2AI fixation (88 screws), using robotic guidance as part of their planned construct. Of these, 13 screws were placed percutaneously and 38 screws were placed in an open fashion by 3 senior spine surgeons using the Mazor core technology robotic systems (Renaissance: n = 9, Mazor X: n = 34, Mazor XSE: n = 8). Medical charts were extensively reviewed for revisions and complications. All follow-ups were 1 year or greater. Postoperative computed tomographic (CT) scans were obtained in 22 of the 51 patients, allowing for 46 screws to be reviewed by an independent neuroradiologist who graded the screws for accuracy using the Gertzbein and Robbins scale.
Results
No intraoperative or postoperative complications were associated with the S2AI screw placement. There were no revisions found to be related to the S2AI screw placement. All 46 screws evaluated with postoperative CT scans were reported as being at the highest level of accuracy, grade A, with a breach distance of 0 mm (no breach).
Conclusion
The robotic-guided technique for S2AI screw placement is a reliable method forachieving pelvic fixation with low complication and revision rates. In addition, a high degree of accuracy can be achieved without relying on visible and tactile landmarks needed for the freehand technique or the additional radiation associated with fluoroscopic guidance.
Abstract 6. Opioid use after spine surgery: How much are we over-prescribing?
Background
The current opioid epidemic is of particular concern in the field of spine surgery, where postoperative analgesia requirements and consumption are often high. There is a need to standardize prescribing practices and mitigate the quantity of unused pills after surgery. This study sought to better characterize postoperative opioid consumption at 90 days in patients undergoing elective spine surgery.
Methods
Adults (> 18 yr) undergoing elective spine surgery at a multi-surgeon single centre were included. Surgery subgroups included the following: anterior cervical, posterior lumbar decompression and short-segment (< 4 levels) circumferential fusion. During the 90-day postoperative period, prescribed morphine milligram equivalents (MMEs) were calculated from opioid prescriptions, consumed MMEs were calculated from pill counts and then both were compared between subgroups. Consumed MME distributions were analyzed as opioid naïve or tolerant to identify the 50th, 75th and 90th percentiles within each subgroup.
Results
Of 117 patients (48.7% male, mean 52 yr) (n = 48, cervical; n = 28, decompression; n = 41, lumbar fusion), 41.9% were opioid tolerant. The mean difference between 90-day MMEs prescribed and consumed overall was 540.0. The percentage of unused opioids at 90 days was as follows: 22.5% cervical, 33.7% lumbar decompression and 23.7% lumbar fusion (p = 0.002). The 90th percentile of MMEs consumed was as follows: 660 naïve and 6728 tolerant cervical, 300 naïve and 2490 tolerant lumbar decompression, 4995 naïve and 7710 tolerant lumbar fusion.
Conclusion
Greater than 20%–30% of total MMEs prescribed were unused at 90 days across surgical subgroups. This suggests the need to develop standardized prescribing practices for postoperative opiates. While the results suggest the number of MMEs prescribed can be reduced to mitigate the effects of leftover pills, larger studies are needed to standardize opioid prescribing practices across elective spine surgeries.
Abstract 7. Intradiscal injection of autologous bone marrow aspirate concentrate improves low back pain at 1 year
Background
Degenerative disc disease is a common cause of low back pain. Surgical fusion is an invasive, last-line treatment associated with higher costs. There is growing interest in regenerative medicine as a less invasive treatment for low back pain. This study aimed to measure clinical improvement in low back pain with intradiscal injection of autologous bone marrow aspirate concentrate (BMAC).
Methods
Adults diagnosed with discogenic low back pain clinically or by discography underwent injection of autologous intradiscal BMAC at a single, multi-surgeon centre. Smokers and patients without purely discogenic back pain were excluded. Baseline and 1-year postinjection Oswestry Diability Index (ODI), visual analogue scale (VAS) back, VAS leg, and EQ-5D-5L questionnaires were compared. Preprocedural lumbar magnetic resonance images (MRIs) were reviewed for Modic changes and assigned a Pfirrmann grade. Thirteen postprocedural MRIs were available and reviewed using the same parameters.
Results
A total of 32 patients (56.3% male, mean age of 45.9 yr), completed 1 year follow-up. Mean VAS back and leg scores improved from 54.03 to 30.41 (p < 0.001) and from 27.87 to 13.25 (p = 0.005), respectively. Mean ODI scores decreased from 33.54 to 21.10 (p < 0.001), and EQ-5D-5L scores improved from 0.69 to 0.78 (p = 0.001). Using established minimum clinically important difference (MCID) values, 59.4% had clinically significant improvement in VAS back pain, 43.8% in VAS leg pain and 56.3 % in ODI scores. On preprocedural MRIs, 65.6% of patients had Modic type 1 or 2 and 93.8% had a Pfirrmann grade of 3 or higher; 61.5% had no measurable change in Pfirrmann grade on postprocedural MRI.
Conclusion
Intradiscal injection of autologous BMAC was shown to significantly improve pain and quality of life at 1 year. This study meets 88% of the American Academy of Orthopaedic Surgeons (AAOS) reporting standards for mesenchymal stem cell studies. This study suggests that intradiscal injection of autologous BMAC has the potential to be an effective nonsurgical treatment for chronic discogenic low back pain.
Abstract 8. Augmented reality–assisted spine surgery: an early experience demonstrating safety and accuracy with 218 screws
Background
In spine surgery, screw guidance techniques continue to improve safety and accuracy while providing minimally invasive options and improved outcomes. Augmented reality (AR) is a novel technology to assist in screw placement and has shown promising results in early cadaveric and feasibility studies. This study aims to contribute an initial experience to the limited in vivo studies available by evaluating safety and accuracy in the largest cohort of patients to date using a head-mounted device (HMD) AR system.
Methods
Consecutive adult patients undergoing AR-assisted thoracolumbar fusion surgery between 2020 and 2021 with a minimum of 2 week follow-up were included in this multi-surgeon, single-centre prospective cohort study. Preoperative, intraoperative and postoperative data were collected to include demographics, complications, revision surgeries and AR performance. Intraoperative 3D imaging was used to assess screw accuracy using the Gertzbein and Robbins (G-R) grading scale.
Results
Thirty-two patients (40.6% male) were included giving a total of 222 screws executed with the FDA-approved HMD-AR system. Intraoperatively, 4 of the screws (1.8%) were deemed misplaced and were replaced freehand. The remaining 218 (98.2%) screws were placed accurately, there were no intraoperative adverse events or complications, and AR was not abandoned. Of the 208 AR-placed screws with 3D imaging intraoperatively, 97.1% were considered clinically accurate (91.8% grade A, 5.3% grade B). There were no postoperative surgical complications or revision surgeries during the 2 week follow-up.
Conclusion
This early experience study demonstrated that HMD-AR assisted spine surgery is a safe and accurate method of placing pedicle, cortical and pelvic fixation. An accuracy rate of 97.1% among all 3 surgeons novice to AR technology suggests ease of integration into the surgical workflow and minimal learning curve. Larger studies are needed to continue supporting this compelling evolution in spine surgery.
Abstract 22. Comparison of complications, revision rates and fluoroscopy time using the latest technology in robotic-guided surgery with historical fluoroscopic-guided controls
Background
Current literature shows the use of robotic guidance in spine surgery can reduce complications, revision rates, and intraoperative radiation exposure; however, there is a lack of data using the current robotic systems, including the addition of navigation. This study compares postoperative outcomes in robotic-guided lumbar fusions using the latest robotic systems with historical fluoroscopic-guided controls. We sought to assess rates of complications, revision surgery and fluoroscopy time using the latest Mazor X and Mazor X Stealth Edition robotic guidance systems and compare them with historical fluoroscopic-guided controls.
Methods
Primary end points were analyzed at the following time points: intraoperatively, perioperatively and postoperatively at 30 days, 90 days and 1 year. Continuous variables were compared using unpaired t tests and categorical variables were compared using Fisher exact test between the robotic-guided (RG) and fluoroscopic-guided (FG) groups. A p values less than 0.05 was used to determine statistical significance.
Results
In the RG cohort, patients had more complex health needs with higher body mass index and Charlson Comorbidity Index (CCI) scores and underwent more complex procedures with a greater number of instrumented levels and higher number of revision cases, thereby elevating the risk of complication. When compared with the FG control cohort with less complex health needs and a lower risk of complication, the actual complication (9.8% RG v.. 16.4% FG, p = 0.119 at 1 year) and revision rates (7.7% RG v. 5.5% FG, p = 0.481 at 1 year) were not statistically different. However, the fluoroscopy time in the control group was 3 times greater than in the RG cohort (50.9 s RG v. 152.4 s FG, p = 0.000).
Conclusion
These findings further the confidence in the latest Mazor robotic technology using Mazor X and Mazor X Stealth Edition as a tool to place instrumentation safely, despite a patient complex with more complex health needs, while decreasing overall radiation exposure.
Abstract 23. Robotic-guided thoracolumbar fusion experience: a multi-surgeon, single-centre study of 628 patients and 3874 robotic-guided screws from 2012 to 2020
Background
We sought to describe 8 years of experience with robotic guidance at a high-volume, multi-surgeon centre. We hypothesize that robotic guidance will lead to low rates of complication and revision surgery, reduced fluoroscopic radiation exposure and accurate thoracolumbar instrumentation.
Methods
We conducted a retrospective review of complications, revision surgery and readmission rates in patients undergoing thoracolumbar fusion surgery utilizing 3 robotic generations. Secondary analysis was conducted comparing the 3 robotic generations for complications, revision surgery, accuracy and readmission rates along with intraoperative fluoroscopic duration.
Results
A total of 628 patients (3874 robotic-guided screws) aged 12–81 years (56.1% female) were included in the study. At 1 year the cumulative complication incidence was 15.5% with a 10.1% incidence of surgical complications. Surgical complications consisted of 1.2% robot-related complications, 3.7 % wound and 5.4 % non-robot–related complications. At 1 year the revision surgery incidence was 9.4%. There was no statistical difference between complications, readmission or revision surgery after initial admission among the 3 robotic generations. The average intraoperative fluoroscopic duration was 53.75 s (11.41 s per screw and 17.65 s per instrumented level).
Conclusion
Robotic guidance in thoracolumbar instrumented fusions was associated with low complication, revision surgery and readmission rates. Our results suggest robotic guidance can provide accurate guidance with minimal adverse events in thoracolumbar instrumentation.
Abstract 86. A province-wide assessment of the appropriateness of lumbar spine MRI
Background
In Canada, it has been observed that there is an overutilization of imaging for degenerative lumbar conditions. A retrospective chart review was conducted in 2019 to determine the appropriateness of lumbar magnetic resonance imaging (MRI) requisitions at the provincial level. Inappropriate imaging represents an unnecessary and expensive usage of limited health care resources. In this study, we sought to determine the surgical referral rate following inappropriate MRI, and the likelihood these patients would be offered surgery.
Methods
A retrospective chart review analyzed all patients receiving an inappropriate MRI from January to December 2018. Patient charts were examined from the time of the original imaging in 2018 until the summer of 2020 to determine whether patients who received inappropriate imaging were subsequently inappropriately referred to a surgeon. Outcome variables included surgeon specialty, specialty of referral, if surgery was offered and if pain management was referred. χ2 tests of proportion were used for statistical analysis, or Fisher exact tests as appropriate, with significance being observed at p less than 0.05.
Results
A total of 1273 patients from the original review in 2019 were determined to have received inappropriate imaging. Of these, 52.4% were women, and 38.1% of the people who received inappropriate imaging were then referred to a surgeon. Older patients were also more likely to be referred to a surgeon even with inappropriate imaging (p = 0.006). For those who were referred to surgeons, no significant difference was observed between neurosurgeons and orthopedic surgeons offering surgery (p = 0.50). A total of 26.2% of patients referred to a surgeon were offered surgery, which represents 9.8% of those who received inappropriate imaging. Orthopedic surgeons were significantly more likely to refer patients to a pain management clinic (p = 0.003).
Conclusion
A significant portion of requisitions for MRIs are considered inappropriate. Furthermore, referral to a surgeon after inappropriate imaging represents further inefficient use of health care resources. We have observed that a significant proportion of those inappropriately referred for MRI are then also referred on for an opinion of surgical appropriateness.
Abstract 134. Concomitant traumatic spinal cord injury and brain injury diagnoses are more frequent and impactful than expected
Background
Concomitant traumatic spinal cord injury (tSCI) and brain injury (TBI) diagnoses are more frequent and impactful than expected.
Methods
We propose a retrospective study of patients with tSCI admitted to a Canadian level 4 trauma centre between April 2010 and June 2020. Medical charts were analyzed to determine the presence and characteristics of concomitant tSCI and TBI, and its impact on returning home and length of stay.
Results
Of the 494 patients with tSCI, 228 (46.2%) had a concomitant diagnosis of tSCI-TBI. A total of 217 patients with tSCI (95.2%) had concomitant mild TBI, 7 (3.1%) a moderate TBI and 4 (1.7%) a severe TBI. Tetraplegia was significantly more frequent (p < 0.001) for patients with concomitant tSCI-TBI (67% v. 51%). With regard to the severity of tSCI, there was no difference in American Spinal Injury Association (ASIA) Impairment Scale grade. Motor vehicle accidents (MVA) were more frequent (p < 0.001) in patients with concomitant tSCI-TBI. Age, height, weight, body mass index, sex and level of education were not significantly different between the 2 groups. While the rate of transfer to specialized intensive functional rehabilitation was similar between patients with concomitant TBI (80%) and patients without TBI (71%), those with concomitant tSCI-TBI were less likely to return home (11% v. 22%, p = 0.006). In addition, the acute length of stay (31.7 [standard deviation (SD) 22.0] d) and rehabilitation length of stay (78.6 [SD 39.9] d) were significantly longer (p < 0.01) for patients with concomitant tSCI-TBI than for those without TBI (24.1 [SD 17.8] and 66.7 [34.7] d, respectively).
Conclusion
It is important to promptly assess the potential presence of TBI in patients with tSCI, because half of them sustain a concomitant TBI. The great majority of patients with concomitant tSCI-TBI have a mild TBI. Tetraplegia and MVA were more often associated with concomitant tSCI-TBI. The burden of care is increased in patients with concomitant tSCI-TBI, as they are more likely to stay in hospital for a longer period of time in acute and rehabilitation facilities. In addition, these patients are significantly less likely to return home after their injury.
Abstract 45. Spatial and depth mapping of nascent mineralization on Ti6Al4V surfaces demonstrating hierarchical macro-micro-nanoscale surface features
Background
Ti6Al4V (Ti) alloy with macro-micro rough (MM) surface and with nanoroughened macro-micro-nano (MMN) surface has shown improved osteogenic differentiation rates of human mesenchymal stem cells (MSCs) in vitro. However, accurately assaying the mineralization deposition is challenging as Ti is opaque and limited on traditional techniques to view spatial localization of mineralization. We utilized fluorescent microscopy techniques to examine early mineralization events and determine if they correlate with osteogenic differentiation.
Methods
MSCs were cultured on smooth Ti, MM or MMN surfaces in either regular (Reg) growth media (alpha-MEM, 10 % FBS, 1% P/S) or osteogenic (Osteo) media (Reg supplemented with 50 μg/mL ascorbic acid and 3 mM β-glycerophosphate). Samples were stained on day 7 and 14 with phalloidin (MSC cytoskeleton), DAPI (nuclei) and tetracycline (calcium deposits). Calcium deposits were imaged with microscopy, and calcified areas quantified with ImageJ then spatially mapped through the height of the samples on the MM and MMN Ti divided into thirds.
Results
Calcium quantification revealed that mineralization on smooth surfaces was low under all conditions examined, never higher than 0%. The MM and MMN surfaces both demonstrated increases in total calcium with time and Osteo media. However, the only significant increase between time points with similar media occurred in MMN samples with Osteo media. Additionally, the only significant increase between media types within a time point occurred at day 14 in the MMN samples. Furthermore, the MMN surfaces consistently demonstrated increased mineralization as compared with other surfaces regardless of the media. Finally, the dramatic rise in mineralization on the MMN surfaces at day 14 in Osteo media was a result of increased mineralization throughout the depth of the samples.
Conclusion
These results demonstrate MSC mineralization in vitro is influenced by surface roughness. Of the Ti surfaces analyzed, MMN Ti supported the highest increase in mineralization in a cell culture model. Incorporation of hierarchical surface topography on spinal implants could facilitate new bone formation and integration, thereby accelerating arthrodesis.
Abstract 111. Propensity-matched outcomes comparing lumbar interbody fusion and total disk arthroplasty: a Canadian Spine Outcomes and Research Network (CSORN) study
Background
There is a paucity of published and generalizable literature investigating patient-reported outcomes following lumbar disc arthroplasty (LDA) for symptomatic degenerative spondylosis. The objective of this study was to quantify long-term reported outcomes following lumbar disc arthroplasty performed as a stand-alone procedure as compared with a propensity-matched lumbar spine fusion cohort.
Methods
We conducted a national multicentre review of prospectively collected data on patients enrolled by the Canadian Spine Outcomes and Research Network (CSORN) who underwent 1- or 2-level LDA for symptomatic lumbar spondylosis. Patients who underwent arthroplasty patients were propensity-score matched to patients who underwent fusions. Patients had a primary diagnosis of degenerative lumbar spondylosis and a chief complaint of back pain. Patient-reported outcome scores and satisfaction were assessed longitudinally.
Results
Propensity-matched cohorts of 61 LDA and 55 fusion patients were collected with at least 1 year follow-up. Cohorts were matched on age, sex, body mass index, smoking status and number of surgical levels. No differences were found between the groups in terms of patient education, symptom duration, medications and work status. Perioperatively, no differences were found in operative time (LDA 122.5 min, fusion 146 min; p = 0.224). Statistical differences were found in length of hospital stay (LDA 1.44 d, fusion 1.93 d, p < 0.05) and intraoperative blood loss (LDA 292 mL, fusion 160 mL; p < 0.001). A trend was found toward higher perioperative adverse events in the fusion cohort (p = 0.05). Adjusted for baseline pain scores, there was a trend toward greater improvement in leg pain among patients in the fusion group (LDA −1.754, fusion −2.597; p = 0.079). No difference was found in the improvement of back pain scores (LDA −3.722, fusion −4.408; p = 0.15). At 1 year follow up, no differences were found for Oswestry Disability Index scores (LDA −21.905, fusion −26.311; p = 0.22). No statistical differences were noted between groups for EQ-5D, physical component summary and mental component summary scores.
Conclusion
Our national outcomes demonstrate comparable results to international literature and confirm the generalizability of the current published data. For both LDA and fusion, comparable improvements in all outcome scores were observed at 1 year. There was a trend toward greater leg pain improvement and higher perioperative adverse events among patients in the fusion cohort.
Abstract 30. A Canadian-based pilot study of current surgical practice and implant preferences in lumbar fusion surgery
Background
Lumbar fusion surgery is an established procedure for the treatment of several spinal pathologies. Despite numerous techniques and existing devices, common surgical trends in lumbar fusion surgery are scarcely investigated. The purpose of this Canada-based study was to provide a descriptive portrait of current surgeons’ practice and implant preferences in lumbar fusion surgery while comparing findings to those of similar investigations performed in the United Kingdom.
Methods
Canadian Spine Society (CSS) members were sampled using an online questionnaire that was based on previous investigations performed in the United Kingdom. Fifteen questions addressed the various aspects of surgeons’ practice: fusion techniques, implant preferences and bone grafting procedures. Responses were analyzed by means of descriptive statistics.
Results
Of 139 eligible CSS members, 41 spinal surgeons completed the survey (29.5%). The most common fusion approach was via transforaminal lumber interbody fusion (TLIF) with 87.8% performing at least 1 procedure in the previous year. In keeping with this, 24 surgeons (58.5%) had performed 11–50 cases in that timeframe. Eighty-six percent had performed no lumbar artificial disc replacements over their last year of practice. There was clear consistency on the relevance of a patient-specific management (73.2%) on the preferred fusion approach. The preferred method was pedicle screw fixation (78%). The use of standalone cages was not supported by any respondents. With regard to the cage material, titanium cages were the most used (41.5%). Published clinical outcome data were the most important variable dictating implant choice (87.8%). Cage thickness was considered the most important aspect of cage geometry, and hyperlordotic cages were preferred at the lower lumbar levels. Autograft bone graft was most commonly preferred (61.0%). Among the synthetic options, DBX/DBM graft (64.1 %) in injectable paste form (47.5%) was preferred.
Conclusion
Findings from this study are in partial agreement with previous work from the United Kingdom but highlight the variance of practice within Canada and the need for largescale clinical studies aimed to set specific guidelines for certain pathologies or patient categories.
Abstract 32. Local contamination is a major cause of early deep wound infections following open posterior lumbosacral fusions
Background
Postoperative surgical site infection in patients treated with lumbosacral fusion has been believed to be caused by perioperative contamination (perioperative inside-out infections) in patients with comorbidities. With the proximity of these incisions to the perianal region and limited patient mobility in the early postoperative period, local contamination from gastrointestinal and/or urogenital flora (postoperative outside-in infections) should be considered as a major source of complication.
Methods
We conducted a single-centre, retrospective review of adult patients treated with open posterior lumbosacral fusions between January 2014 and January 2021. We aimed to identify common factors in patients experiencing deep postoperative infections. Oncologic, minimally invasive, primary infection and index procedures carried out at other institutions were excluded.
Results
We identified 489 eligible patients, 20 of whom required débridement deep to the fascia (4.1%). Mean age (62.9 v. 60.8 yr), operative time (420 v. 390 min), estimated blood loss (1772 v. 1790 mL) and median levels fused (8.5 v. 9) were similar between the infected and noninfected groups. There was a higher percentage of patients with deformity (75% v. 29%) and increased body mass index (32.7 v. 28.4) in the infected group. The mean time from primary procedure to debridement was 40.8 days. Four patients showed no growth on culture. Three showed Staphylococcus species (perioperative inside-out infections) requiring débridement at a mean of 100.3 days (95% confidence interval 0–225 d). Thirteen patients showed infection with intestinal or urogenital pathogens (postoperative outside-in infections) requiring débridement at a mean of 20.0 days (95% confidence interval 9–31 d). Postoperative outside-in infections led to débridement 80.3 days earlier than perioperative inside-out infections (p = 0.007).
Conclusion
In this series, 65% of deep infections were due to early local contamination by gastrointestinal and/or urogenital tracts pathogens. These infections were débrided significantly earlier than the Staphylococcus species infections. Owing to the proximity of the incisions to the perianal region, there should be increased focus on postoperative local wound management to ensure these pathogens are away from the wound during the critical stages of wound healing.
Abstract 99. Comparing patient preoperative expectations and postoperative expectation fulfillment between minimally invasive versus open fusion surgery
Background
Previous studies have highlighted the association of preoperative expectations and fulfillment of expectations with patient-reported outcomes and satisfaction following spine surgery. We hypothesize that patients undergoing minimally invasive surgery (MIS) have higher preoperative expectations and are therefore less likely to attain postoperative expectation fulfillment. In this study, we aimed to compare preoperative expectations and postoperative fulfillment for patients undergoing MIS versus open posterior spinal fusions.
Methods
We analyzed Canadian Spine Outcomes and Research Network (CSORN) registry data from 1308 patients (482 MIS and 826 open) who underwent 1- to 2-level fusion surgery between 2015 and 2019 with a primary degenerative spinal diagnosis. Preoperatively, patients reported their expectations as a result of the surgery (0 [no change], 1 [somewhat better], 2 [better] or 3 [much better]) for 7 items: leg pain, back pain, independence, sporting activities, general health, quality of social contacts and mental well-being. Patients also ranked their most important expectation. For each of these domains, patients reported whether the surgery fulfilled their expectations at 1 year follow-up.
Results
Over 80% of patients reported expectations for improvements (at least somewhat better) in all items with the exception of sporting activities (71.3% v. 69.8%) and social encounters (64.3% v. 64.8%, MIS v. open), with no significant differences between groups. The most important expectation identified preoperatively was improvement in back pain (29.5% v. 31.0%, MIS v. open) in both groups. The calculated expectation score (0–100) was 75.9 versus 75.7 (overall score), 80.7 versus 80.4 (pain domains) and 71.1 versus 70.9 (general health domains) for the MIS versus open groups, respectively. At 1-year follow-up, there was a minor yet statistically significant difference in the proportion of patients reporting complete resolution of leg pain favoring the open group (37.6% v. 44.3%, p = 0.026) with no differences identified in any other fulfillment rankings.
Conclusion
While variations in individual patient expectation and fulfillment exist, our data suggest that generally there are no significant differences in patient preoperative expectation or fulfillment between MIS and open fusion procedures.
Abstract 146. Outpatient robotic-assisted lumbar spinal fusion using the Mazor X Stealth Edition
Background
Enabling technologies such as robotic assistance for lumbar fusions are being combined with minimally invasive surgical (MIS) techniques to optimize patient outcomes. During the COVID-19 pandemic, substantial restrictions in performing inpatient spine surgeries were encountered. Thus, an opportunity to consider outpatient lumbar fusions was presented. This report summarizes a single surgeon’s experience of outpatient lumbar fusions using robotic-assisted MIS techniques.
Methods
A retrospective review of prospectively collected cases from a single-surgeon, multi-institution database was performed for all robotic-assisted lumbar fusions. Data including patient demographics, diagnosis, procedure type, operating room time, robot time, blood loss and complications were analyzed.
Results
A total of 98 robotic-assisted lumbar fusions were performed. Of these, 22 (22%) cases were performed on an outpatient basis, at the hospital. Two of the patients were male and 10 were female. Average age was 60 years (range 32–80 yr) and average body mass index was 27.3 (range 20–34). Nineteen of the cases were considered primary surgery, and 3 were revisions. The most common indications for lumbar fusion were spondylolisthesis, spinal stenosis and spondylolysis. Techniques for fusion included: transforaminal lumbar interbody fusion (18), posterior lumbar interbody fusion (3) and oblique lateral interbody fusion (1). All instrumentation was placed using a percutaneous, robotic-assisted technique, using either the Mazor X or the Mazor X Stealth Edition. Twenty cases were 1 level, 1 case was 2 level and 1 case was 3 level. Average surgery time was 146.9 minutes (range 90–224 min), and robot time was 28.3 minutes (range 17–50 min). Average time was similar between the private, community hospital (surgery time 151.7 min, robot time 29.3 min) and the academic medical centre (surgery time 142.1 min, robot time 27.3 min). There were no procedure-related complications or any complications that required admitted to hospital a patient with a planned outpatient stay. Two patients developed urinary retention requiring in/out catheterization before discharge. One male patient returned to the emergency department for urinary retention requiring catheterization and 1 female patient returned for leg edema.
Conclusion
Outpatient lumbar fusions can be safely and successfully performed using MIS and robotic-assisted techniques, especially when hospital bed availability is limited.
Abstract 149. Lessons learned from my first 100 robotic-assisted lumbar fusions using the Mazor X Stealth Edition: surgical synergy with MIS, surgical navigation and robotics
Background
Advances in enabling technologies such as minimally invasive surgery (MIS), surgical navigation and spinal robotics have improved our clinical outcomes for spinal fusion. With the recent addition of robotics, spinal fusions can be performed with greater precision, and with fewer complications related to placement of instrumentation. While early robotic-assisted techniques focused on pedicle screw placement, current techniques now allow for precision computer-assisted planning of the incision, interbody cage placement, simple and complex deformity correction and patient-specific sagittal rod contouring.
Methods
A review of a single-surgeon series of the first 100 cases of robotic-assisted, minimally invasive, image-guided lumbar fusions will be reported.
Results
On the basis of this review, a reflective review of surgeon training, adoption, computer-assisted planning, surgical efficiencies, avoidance of complications and future applications will be presented. Specifically, the issues of skive, navigation variance and anatomic movement will be discussed.
Conclusion
Robotic-assisted lumbar fusions can be performed in conjunction with MIS and surgical navigation techniques to enhance patient outcomes. While robotic-assisted techniques are still evolving, much can be learned from the early experience with the current systems.
Abstract 151. Freehand biomechanical testing for use in lumbar discectomy training
Background
Haptic simulators for training surgeons often are informed by mechanical data that poorly represent the natural movement of a surgeon. Tissue testing often involves controlled robotic motion that does not match the freedom present during surgery. The aim of this study was to build and test a device surgeons could use to perform a discectomy procedure naturally while tracking force and position for eventual application in haptic surgical simulators.
Methods
Two excised cadaveric lumbar spine segments were subjected to freehand mechanical testing. Two spine neurosurgeons performed a lumbar discectomy using a Depuy Synthes Concorde Clear MIS vacuum curette. The curette was attached to an Entact W3D device to track position and an ATI 6-axis load cell to track force. The surgeons performed multiple passes through the tissue while position and force were tracked to record the resistance of the tissue during testing. The resistance was measured by normalizing the force as the surgeon penetrated the tissue. The total resistance magnitude and resistances off-axis with respect to the tool were compared.
Results
Successful testing was performed on all lumbar levels of both spines. Up to 30% of the total linear resistance was off-axis, with an average drop of 7% per 10 N of increasing force. Linear resistance normalized for speed was 0.31 (N.mm)/(mm/s). Statistically significant differences (p < 0.05) were found in linear resistance between surgeons.
Conclusion
A significant percentage of the total resistance present during the discectomy was off-axis, meaning it would not have been captured by controlled robotic testing techniques. The amount of off-axis resistance varied with surgeon speed. These off-axis data are integral to the proper replication of surgery in a simulator and these observations expose limitations to the methods that are often used to capture resistance data for building haptic models. The resistances shown here were captured in a method that is more relevant to the operating room experience, which will allow for the development of more accurate training simulators.
Abstract 48. Spinal pathology and outcome post-THA: Does segment of arthrodesis matter?
Background
The presence of lumbar spine arthrodesis (SA) is associated with abnormal spinopelvic characteristics and inferior outcome after total hip arthroplasty (THA). However, whether patients with upper segment SA are also at increased risk of complications is unknown. This study aims to (1) determine if upper segment SA is associated with inferior THA outcomes; (2) assess spino-pelvic characteristics; and (3) test whether static or dynamic spinopelvic characteristics correlate with outcome after THA.
Methods
In this retrospective, case-matched, cohort study from a tertiary referral centre, 40 patients (59 hips) who had undergone both THA and any level of spinal arthrodesis (49 THA-Lumb and 10 THA-Cerv) were compared with 41 patients (59 hips) who had THA-only without known spinal pathology. Spino-pelvic characteristics (including severity of degenerative disc disease [DDD], spinal balance and stiffness) and outcome, including patient-reported outcome measures (PROMs), at a minimum of 1-year after THA, were assessed.
Results
The THA-Lumb and THA-Cerv groups had a greater number of complications and inferior hip and spinal PROMs than the THA-only group (p < 0.001). Similar spinopelvic characteristics were seen between the THA-Cerv and THA-Lumb groups, which were significantly different from those of the THA-only group. The presence of DDD and unbalanced or stiff spine was associated with increased dislocation and inferior PROMs in the whole cohort.
Conclusion
THA in the presence of SA, regardless of level, is associated with inferior outcomes and an increased risk for dislocation. The presence of a SA is associated with increased risk of adverse spinopelvic characteristics. Such characteristics were strongly associated with increased dislocation risk and inferior PROMs. It is likely that these adverse characteristics are the most important adverse predictor, rather than segment of SA per se. A greater understanding of these characteristics will aid spine and hip surgeons alike, enabling them to better guide the expectations of these higher risk patients before surgery.
Abstract 27. Patient, surgical and institutional factors associated with length of stay in degenerative lumbar spine surgery: national multicentre cohort analysis from the Canadian Spine Outcomes and Research Network (CSORN)
Background
Prolonged length of stay (LOS) is a substantial contributor to the variation in surgical health care costs and resource utilization after elective spine surgery. The primary goal of this study was to identify patient, surgical and institutional variables that influence LOS. The secondary objective was to examine variability in institutional practices among participating centres.
Methods
This is a retrospective study of a prospectively multicentric followed cohort of patients enrolled in the Canadian Spine Outcomes and Research Network (CSORN) between January 2015 and October 2020. A logistic regression model and bootstrapping method was used. A survey was sent to participating centers to assessed institutional-level interventions in place to decrease LOS. Centres with LOS shorter than the median were compared with centres with LOS longer than the median.
Results
A total of 3909 patients were included in the study (1017 discectomies, 1153 laminectomies, 1739 fusions). The median LOS for discectomy, laminectomy and fusion were respectively 0.0 day (interquartile range [IQR] 1.0), 1.0 day (IQR 2.0) and 4.0 days (IQR 2.0). The discectomy group had the smallest variability (standard deviation [SD] 1.9, range 0–41). By site, median LOS varied from 0.0 to 1.5 days. The laminectomy group had the largest variability (SD 4.4, range 0–133 d). Median LOS by site varied from 0.0 to 2.0 days. The fusion group had an average variability (SD 3.6, range 0–64 d). Median LOS by site varied between 2.0 and 5.0 days. On multivariate analysis, individual patient–level independent predictors of longer LOS were higher Oswestry Disability Index score, open surgery and perioperative and intraoperative adverse events in all 3 surgical groups. Fifteen centres completed the survey. Institutional-level factors that reduced LOS were patient’s involvement preoperatively in education and discharge planning in all 3 surgical groups.
Conclusion
In this first study stratifying individual patient–level and institutional-level factors, several independent predictors were identified to enhance the understanding of LOS variability in elective spine procedures. Prospective research could help provide more robust results in the influence of patient education and involvement in discharge planning on LOS preoperatively.
Abstract 28. The impact of the increasing proportion of degenerative spine emergency admissions
Background
The proportion of emergency admissions in degenerative spinal pathologies have increased 22.6% in 2006 to 34.8% in 2019. The primary goal of this study was to compare the outcomes of patients with degenerative spinal conditions who were treated emergently in a decompensated state to those of patients treated electively.
Methods
All consecutive patients who underwent emergent and elective spine surgery for a degenerative spinal condition at our institution between Jan. 1, 2006, and Dec. 31, 2019, were included in the study. Elective and emergency admissions were compared. Patients who had elective surgery were also compared with patients who had been placed on a waitlist but decompensated and were treated emergently.
Results
There were a total of 6217 patients, with 4654 patients (74.9%) who receive an elective procedure and 1563 patients (25.1%) were received an emergent procedure. Patients who had elective surgery were older than those who had emergency surgery (56.0 [standard deviation (SD) 15.4] yr) v. 53.9 (SD 16.9) yr; p < 0.001). Emergency patients had a shorter operative time (2.5 (SD 3.0) h v. 3.0 (SD 3.1) h, p < 0.001). Emergency patients had a longer median length of stay (LOS) of 5.1 days (interquartile range [IQR] 2.7–11.2) versus 3.6 days (IQR 1.3–6.4) (p < 0.001). More elective patients were discharged home (94.2% v. 78.6 %; p < 0.001). On multivariate logistic regressions, patients with an emergency admission were 1.49 times more likely to have any adverse events (95% confidence interval [CI] 1.17–1.89). Emergency patients on a wait list had a longer median LOS of 5.6 days (IQR 11.1) versus 3.6 days (IQR 5.1) in elective patients (p < 0.001). Emergency patients on a wait list were less frequently discharged home (76.5% v. 94.2%; p = 0.001). On multivariate logistic regression, emergency patients on a wait list were 4.6 times more likely to have any adverse events (95 % CI 1.17–18.22).
Conclusion
This study suggests worse outcomes for emergency patients who are on a wait list or not compared with the outcomes for elective patients in degenerative spinal pathologies. In consideration of the human and economic cost of late presentation of spinal surgical disease, strategies that better align primary care, screening programs and surgical intervention may be the next target in decreasing health care costs. This could be explored in future cost-effectiveness analyses.
Abstract 51. Patient’s expectations of surgery for degenerative spondylolisthesis: analysis by site and type of surgery from the Canadian Spine Outcomes and Research Network (CSORN)
Background
Patients with lumbar degenerative spondylolisthesis have higher expectations of pain relief and functional well-being than patients with lumbar stenosis. It is unknown if these expectations vary across institutions and between types of surgery. The goal of this study was to assess variation in expectations between 4 major Canadian sites and between decompression and fusion.
Methods
An analysis was conducted on prospectively enrolled patients with degenerative spondylolisthesis between January 2015 and September 2021. Preoperatively, patients were questioned about their expected changes in leg pain, back pain, balance, paresthesias, independence, sporting activities, general physical capacity, reduced social contacts, improvement in mental well-being, prevention of worsening and their most important change expected.
Results
A total of 456 patients were included in the analysis by site. Patients’ expected change in independence, sporting activities, physical capacity, andreduced social contacts differed by site. Patients’ expected change in improved mental well-being showed a trend toward a difference by site. Patients’ expectations did not differ in terms of prevention of worsening, changes in leg pain, back pain, balance, paresthesias or most important change expected. A total of 469 patients were included in the analysis by type of surgery. A total of 131 patients underwent decompression, and 338 patients underwent fusion. The most important expected change differed between the decompression and fusion groups, with the decompression group expecting more commonly an improvement in leg pain (27.3%) and the fusion group expecting an improvement in back pain (30.6%). A trend was observed for expected prevention of worsening (p = 0.06). No difference was observed in terms of expectations for changes in leg pain, back pain, independence, sporting activities, general physical capacity, reduced social contacts, improvement in mental well-being, balance and paresthesias.
Conclusion
This first study on expectations of surgery for patients with degenerative spondylolisthesis provides very valuable insight in its variation between sites and type of surgery. Patients’ expectations concerning changes in independence, sporting activities, physical capacity and reduced social contacts differed by site. Patients undergoing decompression expect most importantly a change in leg pain and patients undergoing fusion expect most importantly a change in back pain.
Abstract 60. The impact of sex on thoracolumbar surgery outcomes in patients with diabetes — a CSORN study
Background
The purpose of this study was to determine sex differences in baseline health and surgical outcomes of patients with diabetes and lumbar spinal stenosis (LSS) who underwent thoracolumbar decompression with or without fusion.
Methods
This was a retrospective cohort study of patients with diabetes and LSS in the Canadian Spine Outcomes and Research Network (CSORN registry). Primary outcomes of interest were blood loss (BL), length of stay (LOS), adverse events (AEs), wait time, modified Oswestry Disability Index (mODI), numerical rating scales for leg and back pain (NRSL/B), SF-12 Quality of Life Questionnaire – Mental Health Component (MCS) and meaningful change. Time points of interest were baseline, in-hospital, and 3 months and 1 year postoperatively. A 2(sex) × 3(time) mixed analysis of variance with type of surgery as a covariate was used. Continuous variables were compared using independent t tests and categorical variables were compared using χ2 analyses.
Results
Patients (n = 217, 37% female, mean age 68.24 yr ) are reported. Body mass index was found to be different at baseline (males 30.08, females 32.05; p = 0.003). No other baseline variables were significantly different. No difference was found for BL (difference 21.15 mL; p = 0.661), LOS (difference 0.078; p = 0.88), number of patients who had AEs (difference 7.2 %, p = 0.05) nor rate of infection (males 3, females 2; p = 0.017). Sex (p = 0.035) and time (p = 0.001) were significant, while sex × time was not significant (p = 0.90). Females had higher NRS-B scores at baseline (difference 0.78; p = 0.016) and 1 year postoperatively (difference 0.962; p = 0.001), higher mODI scores at baseline (difference 5.81; p = 0.005), at 3 months postoperatively (difference 5.99; p = 0.026) and at 1 year postoperatively (difference 6.8; p = 0.013). Patients’ disability and pain scores improved from baseline to 3 months (mODI: difference 15.861, p = 0.001; NRS-B: difference 3.781, p = 0.001; NRS-L: difference 3.781, p = 0.001; MCS: difference 2.48, p = 0.006) and from baseline to 1 year (mODI: difference 17.078; p = 0.001; NRS-B: difference 3.263, p = 0.001; NRS-L: difference 3.796, p = 0.001; MCS: difference 3.014, p = 0.001). No significant differences were seen for meaningful change for mODI (males 58.1%, females 55.6%; p = 0.41), NRS-B (males 67.7%, females 67.9%; p = 0.55) and NRS-L (males 75.7%, females 72.8%; p = 0.38).
Conclusion
Females with diabetes and LSS reported worse baseline and follow-up disability and back pain. Both cohorts achieved the same degree of improvement in disability and pain. This knowledge may help inform patients of expectations postoperatively.
Abstract 81. The impact of older age on rehabilitation outcomes following functional motor-incomplete traumatic spinal cord injury
Background
Specialized health care centres are facing a major demographic shift in the traumatic spinal cord injury (tSCI) population. As average age at injury increases, the proportion of American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade D injuries also climbs. In this context, gaining a better understanding of the impact of age in patients with AIS grade D injuries is crucial to identify the needs of the future SCI population. This study aims to determine the impact of older age (≥ 65 yr) on long-term functional outcomes and health-related quality of life in patients with AIS grade D injuries.
Methods
A prospective cohort of 105 tSCI patients with an acute AIS grade D injury was considered. The baseline characteristics, functional status (Spinal Cord Independence Measure total and subscores) and health-related quality of life (SF-36v2 physical component summary [PCS] and mental component summary [MCS] scores) at 12 months after injury were compared between individuals aged 65 years and older and those aged less than 65 years in bivariate analyses. Multivariable regression analyses were then performed to determine the relationship between older age and functional/quality of life outcomes while controlling for variables that were significant at the bivariate level.
Results
A total of 34.3% (n = 36) of the cohort was 65 years of age or older. This group displayed higher body mass index, a higher rate of comorbidities, different living arrangements and longer delay to surgery compared with the younger group. Individuals 65 years of age and older also reached maximum self-care, mobility and total scores at follow-up in lower proportions than younger individuals. There was no significant difference in quality of life at follow-up between the groups. At the multivariate level, older age remained significantly associated with poorer functional scores for self-care and mobility at 12 months after injury.
Conclusion
Individuals 65 years of age and older with AIS grade D tSCI have poorer functional outcomes despite longer inpatient rehabilitation length of stay. Age-specific rehabilitation strategies should be implemented to optimize functional outcomes. Specifically, the early involvement of geriatricians in the continuum of care could be envisaged to maximize the functional independence of elderly patients.
Abstract 47. Devise and investigate a novel, intramuscular pressure based, muscle activation strategy in a spine stability model
Background
Spinal stability, a determining factor in examining spinal pathologies, can be assessed via engineering optimization models. As such, the purpose of this research was to leverage a novel validated finite element (FE) model of the spine, inclusive of an accurate representation of intramuscular pressure (IMP), to explore muscle recruitment patterns maintaining spinal stability.
Methods
First a spine model was validated toward its use in exploring activation strategies. The underlying spine FE model comprised most tissues known to contribute to spinal stability, namely paraspinal muscles, thoracolumbar fascia (TLF) and abdominal pressure. The model was then employed to explore the effect of minimizing and maximizing IMP. This was examined by activating both objective functions via the engineering stability test of perturbating the spine to a flexed position. Under physiologic muscle and vertebral constraints, muscle activations were quantified in an effort to retrieve the initial position of the spine.
Results
Validation efforts proved successful. Collected data showed that to maintain high engineering stability, minimum individual muscle forces and IMP of 75 N and 50 mmHg, respectively, was sufficient. On the other hand, excessive increase in individual IMP, with values exceeding an average of 250 mmHg, led to another physiologic stability. This was accompanied with about 93 mmHg of compartmental pressure between muscles, as well as elevated tensional forces in the TLF reaching 56.9 N.
Conclusion
Minimum IMP requirements to achieve stability support the role of agonist and antagonist muscle activations to achieve efficient mobility and stable joints (i.e., maximizing muscular endurance while maintaining a stable spine position). Furthermore, excessive IMP activations showed that muscular fluidic behaviour may play a substantial role in sharing muscular generated loads, radially, with surrounding tissues. Consequently, compartmental pressure between muscles potentially starts to increase, putting the TLF under higher tensional forces and thus supporting TLF’s passive load-storing role. The explored novelties and impactful findings may greatly inform clinical applications such as motion analysis, device design for spine pathologies, and focused rehabilitation.
Abstract 17. 3D radiologic outcomes for patients with moderate idiopathic scoliosis curves treated with internal (anterior vertebral growth modulation) versus external bracing: a 2-year observational study
Background
This study aimed to test the hypothesis that anterior vertebral body growth modulation (AVBGM) achieves 3D deformity correction after 2 years of follow-up while brace treatment limits curve progression for moderate adolescent idiopathic scoliosis (30°–50°).
Methods
A retrospective review of a prospective database of patients with AIS, recruited between 2013 and 2018, was performed. Inclusion criteria were skeletally immature patients (Risser 0–2), with Cobb angles between 30° and 50° and a 2-year follow-up after bracing or AVBGM. 3D radiologic parameters were evaluated. Unpaired t test was used.
Results
Thirty-nine patients (mean age 12.7 [standard deviation (SD) 1.3] yr) with Cobb angles of 30° or greater treated with a brace and 41 patients (mean age 11.8 [SD 1.2] yr) with Cobb angles less than 50° who received AVBGM were reviewed. Statistical analysis of 3D deformity measurements showed that at 2-year follow-up, only the 3D spine length and apical vertebral heights changed significantly with brace treatment. On the other hand, AVBGM treatment achieved statistically significant correction differences in thoracic and lumbar Cobb angles, true kyphosis (segmental derotated kyphosis of T5–T12), 3D spine length and selective left apical vertebra height (p < 0.001). Thirty-five percent of brace patients had a curve progression of more than 5° at final follow-up while it was 0% for patients who received AVBGM.
Conclusion
Even though these 2 cohorts are not fully comparable, bracing seems to control progression for a substantial portion of patients with moderate scoliosis curves, while AVBGM significantly corrected and maintained 3D deformity parameters at 2-year follow-up.
Abstract 18. Is quality of life affected by concomitant isthmic spondylolisthesis when undergoing surgery for adolescent idiopathic scoliosis and nonsurgical management of the spondylolisthesis?
Background
The aim of this study was to test the hypothesis that improvement in quality of life (QoL) scores after surgically treated adolescent idiopathic scoliosis (AIS) is expected when there is a concomitant spondylolisthesis to a similar degree as in those without it.
Methods
A retrospective review of a prospective cohort of 464 individuals undergoing AIS surgery between 2008 and 2018 was performed. All patients undergoing surgery for AIS and with a minimum 2 years of follow-up were included. We excluded patients with prior surgery for spondylolisthesis and patients scheduled to have both conditions addressed at once. QoL scores were measured through the SRS-22 questionnaire. Comparisons were performed using Student t and χ2 tests.
Results
Thirty-six patients (15.2 [standard deviation (SD) 2.6] yr) with concomitant isthmic spondylolisthesis were operated for AIS, and 428 patients (15.5 [SD 2.4] yr) without it. The 2 groups were similar in terms of age, sex, and preoperative and postoperative Cobb angles. Preoperative and postoperative QoL scores were similar between the groups. QoL improved significantly for all domains in both groups, except for pain in patients with spondylolisthesis. There was no need for surgical treatment of the spondylolisthesis and no slip progression during follow-up.
Conclusion
Patients with concomitant nonsurgically managed isthmic spondylolisthesis can expect postoperative QoL scores and improvement after surgical treatment of AIS similar to that of patients without it. There was no progression of the spondylolisthesis or need for spondylolisthesis surgery despite the presence of spinal fusion above the spondylolisthesis level.
Abstract 128. Toward macrostructural and microstructural investigation of the cervical spinal cord through quantitative analysis of T2-weighted and diffusion-weighted imaging
Background
Studies have demonstrated the correlation between fractional anisotropy (FA) and functional outcomes in patients with spinal cord injury (SCI), but there are a few studies correlating microstructural and macroscopic features in the healthy population. The primary objective of this study is to find whether there is any significant relation between macrostructural and microstructural characteristics of cervical spinal cord magnetic resonance (MR) images of the healthy population.
Methods
We conducted a quantitative analysis on the spine generic public data set containing 260 healthy multi-subject MRI scans using the latest spinal cord toolbox. We experimented with T2w images to derive shape characteristics (cross-sectional area of the cord, space available for the cord) and diffusion-weighted imaging (DWI) to extract diffusion metrics (fractional anisotropy, mean diffusivity and radial diffusivity). After the preprocessing of extracted features, we proceeded with the Pearson correlation between each DWI metric and T2w macroscopic measurements; with more interest in finding the correlation between FA and space available for the cord / cross-sectional area (SAC/CSA).
Results
There was moderate correlation between SAC/CSA and FA (r = 0.35), with moderate correlations, highly moderate, moderate, and slightly moderate to low correlations (r = 0.38, r = 0.33, r = 0.48, r = 0.40, r = 0.28) between SAC/CSA and FA in the male and female categories and the subgroups of images acquired with GE, Philips and Siemens MRI machines, respectively. In total, the correlations were computed for 12 various subcategories of images ranging from slightly moderate to low (r = 0.27) to highly moderate (r = 0.53) correlations, with a p value of 0.001 for all experiments.
Conclusion
The relationships between anatomic and microstructural characteristics of the cervical spinal cord in the healthy population are presented. Results demonstrate a potential relationship between FA and SAC/CSA. These findings suggest that microstructural measurements might be used as a screening tool in addition to macroscopic features in populations suspected of SCI. The sensibility and threshold values of DWI metrics would need more clarification in the future. Further studies are required to evaluate the progression of the disease and its correlation with quantitative measurements.
Abstract 26. Minimally invasive versus open thoracolumbar spine surgery for patients who have lumbar spinal stenosis and an ASA score of 3 or above: a CSORN study
Background
The aim of this study was to compare outcomes of minimally invasive (MIS) versus open thoracolumbar surgery for lumbar spinal stenosis in patients with an American Society of Anesthesiologists (ASA) score of 3 or above.
Methods
Data for Canadian Spine Outcomes and Research Network (CSORN) registry patients with an ASA score of 3 or above who underwent thoracolumbar surgery were extracted. Two surgical cohorts were examined: decompression (n = 138) and fusion (n = 120). Outcomes of interest were blood loss (BL), length of stay (LOS), adverse events (AE), modified Oswestry Disability Index (mODI), numerical rating scale (NRS) for back and leg pain, and meaningful change. Meaningful change was defined as achieving minimal clinical important difference or minimum pain/disability from baseline to 24 months. χ2 tests, t tests and a 2 (MIS v. OPEN) × 4 (baseline, 12 wk, 12 mo, 24 mo) mixed-measures analysis of variance was run for each cohort (α = 0.05).
Results
In the decompression cohort, MIS resulted in significantly less BL (difference 187 mL, p < 0.001) and a shorter LOS (difference 1.4 d, p < 0.001). No significant effect of surgical approach on AE (difference 7 patients, p = 0.30) was found. There was no significant effect of MIS (p = 0.15) nor interaction effect of time × MIS (p = 0.42) for mODI, NRS-leg and NRS-back. There was no significant effect of time (p = 0.001). There was no significant difference in the percentage of patients achieving meaningful change (mODI: difference 7%, p = 0.25; NRS-leg: difference 8%, p = 0.17; NRS-back: difference 0%, p = 0.58). In the fusion cohort, MIS demonstrated significantly less BL (difference 240 mL, p < 0.001). There was no significant effect for LOS (difference 0.5 d, p < 0.21) nor in the number of patients who experienced an adverse event (difference 3 patients, p = 0.44). There was a significant effect of MIS (p = 0.023), time (p = 0.001) and time × MIS (p = 0.039). MIS resulted in significantly lower 2-year mODI (difference 1.55, p < 0.001) and NRS-leg scores (difference 1.61, p = 0.003) but not NRS-back scores (difference 0.37, p < 0.44). MIS resulted in significantly more patients achieving a meaningful change in mODI (difference 18%, p = 0.034) and NRS-leg scores (difference 19%, p = 0.025), but not NRS-back scores (difference 8.33%, p = 0.21).
Conclusion
MIS for patients with an ASA score of 3 or above results in improved outcomes when the patient is undergoing a thoracolumbar fusion for spinal stenosis.
Abstract 101. Association between surgeon age and outcomes of spine surgery: a population-based retrospective cohort study
Background
It is unclear if surgeon age has an impact on patient outcomes after spine surgery. More surgeons are working past 65 years of age, and it is critical to evaluate safety given the effects of aging on cognitive and physical abilities. The purpose of this study is to investigate the association between spine surgeon age and outcomes of spine surgery.
Methods
This national study linked data from the Canadian Spine Outcomes and Research Network and a survey distributed to 70 spine surgeons. Four groups based on surgeon age were identified (35–44 yr, 45–54 yr, 55–64 yr, 65 yr and older). Patient-reported outcome measures (PROMs) and secondary outcomes including adverse events, length of stay, surgical time and blood loss were compared between the surgeon age groups.
Results
Fifty-two of seventy (74%) surgeons responded to the survey. Both thoracolumbar (TL) and cervical spine (CR) patients were included (n = 5871). Between each surgeon age group, there was no statistically significant difference in age (p < 0.74), sex (p < 0.89), smoking status (p < 0.64) or body mass index (p > 0.05); however, there was a statistically significant difference in the number of comorbidities per patient for TL patients (3.04 for the 35–44 yr group v. 2.76 for the 55–64 yr group, p < 0.001) and CR patients (0.37 for the 35–44 yr group v. 0.49 for the 45–54 yr group, p < 0.046). For PROMs there were no differences in change of score from baseline to 12 months after surgery. For surgical factors, there were statistically significant differences in operating time (35–44 yr group 171.5min v. 45–54 yr group 151.9 min, p < 0.001; 35–44 v. 55–64 yr group 155.7 min, p < 0.001) and blood loss (35–44 yr group 320.4 mL v. 45–54 yr group 249.1 mL, p < 0.001; 35–44 v. 55–64 yr group 276.3 mL, p < 0.023). The group aged 65 years and older had the lowest percentage of adverse events both intraoperatively and perioperatively (p < 0.013 and p < 0.001, respectively).
Conclusion
With no detectable difference between patient traits categorized into each surgeon age group, it is demonstrated that patients treated by surgeons who are aged 65 years and older have equivalent outcomes. Furthermore, the group of spine surgeons aged 65 years and older had the lowest percentage of intra-peri-operative adverse events. As the pressure on the health care system remains steady, the experienced surgeon is an important asset in delivering safe patient-centred care, and traditional retirement ages may need to be reconsidered.
Abstract 77. Utilizing machine learning methodology to create a short form of the Multi-Morbidity Index in spinal cord injury
Background
Spinal cord injury (SCI) affects multiple body systems and the presence of comorbidities and secondary conditions affects outcomes following injury. A Multi-Morbidity Index (MMI) was developed and validated for traumatic (TSCI) and nontraumatic SCI (NTSCI), to assess the impact on system (health care utilization [HCU]) and patient (health and quality of life [QoL]) outcomes. The objective of this study was to utilize machine learning methodology to create a short form of the MMI.
Methods
Participants with TSCI and NTSCI from the SCI Community Survey (SCICS) data set were included. Variables included demographic characteristics, clinical features (e.g., severity of injury), and MMI-30 (presence/absence of 30 comorbidities and health conditions). Participants were grouped using the 7-item HCU Questionnaire (Group 1: received needed care; Group 2: did not receive needed care). The Short Form-12 (SF-12) physical and mental component scores (PCS/MCS) assessed health and an item from the Life Satisfaction Questionnaire (LiSat-11) measured QoL. Network analyses were utilized. Network nodes represent the 30 comorbidities and health conditions and outcomes, and the edges represent the statistical relationships (positive or negative).
Results
A total of 1549 people living with SCI in Canada completed the SCICS; 412 (26.6%) had NTSCI and 1137 had TSCI. The MMI-30 was shortened to 25 items. The mean of MMI-25 was significantly lower in those who received needed care compared with those who did not receive care (12.1 v. 14.8 in NTSCI, p < 0.001; 11.4 v. 14.1 in TSCI, p < 0.001). A higher MMI-25 score and HCU Group 2 were significant factors for having a lower PCS and MCS (p < 0.001) and in turn were associated with lower QoL (p < 0.001).
Conclusion
A modified version of MMI (MMI-25) was created using machine learning methodology. Multimorbidity and HCU are interrelated and associated with lower physical and mental health, which in turn is associated with lower QoL. Multimorbidity can be managed with appropriate SCI care and the MMI-25 will be piloted as a screening tool to identify patients at risk of having poor outcomes.
Abstract 70. Ten-year reoperation rate and clinical outcome in patients treated surgically for lumbar spinal stenosis
Background
The aim of this study was to determine the reoperation rate in patients treated with lumbar decompression and fusion for spinal stenosis at 10 years after surgery and to identify risk factors for reoperation by comparing patient factors, patient-rated outcomes and x-ray measures between patients who had a reoperation and those who did not.
Methods
The study aims to recruit patients previously referred to the London Spine Centre and surgically treated for degenerative lumbar spondylolisthesis and spinal stenosis who were enrolled in a prospective observational study, between 2006 and 2011. Patient health-related quality of life measures were recorded at the time of initial consultation, preoperatively and postoperatively to 2 years after surgery. We proposed to conduct an ambispective study where patients in the original study are being invited to attend a follow-up (FUP) visit, have x-rays and complete questionnaires to assess their health-related quality of life 10 years after their primary surgery.
Results
Of 212 patients, 81% underwent instrumented fusion, 17% decompression alone and 2% noninstrumented fusion. At 10 years, the overall reoperation rate was 22%, of which 5.2% occurred within 2 years and 16% within 5 years. No difference in reoperation rate was found between the decompression-alone and fusion groups (25% v. 22%, p = 0.83). As preliminary data, among the patients who have already attended the 10-year follow-up meeting (n = 54), the reoperation rate was 31%. Patients who had not undergone reoperation surgery at the 10-year FUP had a better SF-36 MCS at baseline (p = 0.031) and had a better improvement in the SF-36 PCS in relation to the reoperation group (6.61 v. −1.18. p = 0.049). After 10 years, no differences in SF-36, ODI and ZCQ scores were identified. With regard to radiographic parameters at 10 years, patients who underwent revision surgery had clinically worse pelvic incidence–lumbar lordosis (PI-LL) mismatch (21° v. 10°, p = 0.06) and lumbar lordosis (38° v. 48°, p = 0.16) but sagittal vertical axis (SVA) was similar for both groups.
Conclusion
After 10 years, the reoperation rate was higher in patients with a worse SF-36 MCS and the SF-36 PCS did not improve for those who underwent reoperation surgery.
Abstract 105. Assessing the importance of radiographic and clinical parameters when choosing decompression without fusion for LDS: results from the CSORN prospective DS study
Background
The decision to not fuse in degenerative spondylolisthesis surgery is partially based on an assessment of “stability.” Traditionally, stability has been assessed using both radiographic and clinical parameters including little to no back pain, radiographic stabilization signs and/or collapsed disc, a lordotic disc, less than 3 mm movement and no facet effusion. This study examines the utilization of these factors in the decision to not fuse and correlates them to postoperative outcomes and slip progression.
Methods
The Canadian Spine Outcomes and Research Network (CSORN) multicentred prospective study on the assessment and management of lumbar degenerative spondylolisthesis (LDS) database was retrospectively reviewed to identify patients who had decompression surgery. Descriptive statistics summarized stability parameters. Regression analysis will be performed to identify the effect of stability parameters on outcome and slip progression.
Results
Of a total of 567 patients enrolled across 8 centres, 136 underwent decompression alone. Evaluation of preoperative imaging demonstrated the following stability parameters: mean percentage slip was 14.3 mm (4–36 mm); 36% had 3 mm or more of translation between supine and standing x-rays (mean 5.7 mm); 18% had a neutral/kyphotic disc with standing, average disc height was 9.8 mm (range 0–14 mm); facet angle ranged between 32° and 86°; and 37% had facet effusions of which 32% of these joints were distracted. The average preoperative back pain was 6.6/10 and 17% reported minimal to no pain. On the degenerative spondylolisthesis instability classification scheme, 50% of patients who underwent decompression had grade type 1 stable slips.
Conclusion
Despite a large proportion of the stability parameters suggesting less than stable spondylolisthesis, surgeons elected to decompress these patients without fusion. Further analysis will assess this effect on outcomes and determine which parameters correlated with postoperative slip progression.
Abstract 104. Preoperative cannabis use in Canadian thoracolumbar spine surgery patients: a CSORN study
Background
This study aims to (1) determine reported cannabis use among patients waiting for thoracolumbar surgery and to (2) identify demographics and health differences between cannabis users and noncannabis users.
Methods
This observational cohort study is a retrospective national multicentre review using data from the Canadian Spine Outcomes and Research Network registry. Patients were dichotomized as cannabis users and noncannabis users. Variables of interest were age, sex, body mass index, smoking status, education, work status, exercise, modified Oswestry Disability Index (mODI), the numerical rating scales (NRS) for leg and back pain, tingling/numbness scale, SF-12 Quality of Life Questionnaire – Mental Health Component (MCS), and use of prescription cannabis, recreational cannabis and narcotic pain medication. Continuous variables were compared using an independent t test and categorical variables were compared using χ2 analyses.
Results
Cannabis use was reported by 28.4% of preoperative patients (n = 704), 47% of whom used prescription cannabis. Cannabis use was reported most often by patients in Alberta (43.6%), British Colombia (38.1%) and New Brunswick (33.7%). Patients who reported using cannabis were significantly younger (mean 52.9 yr v. mean 61.21 yr). There was a higher percentage of concurrent narcotic use (51.5%) and smoking (21.5%) reported in cannabis users in comparison with non cannabis users (41.1% [p = 0.001] and 9.5% [p = 0.001], respectively). There were significant differences in cannabis use on the basis of pathology (p = 0.01). Patients who reported using cannabis had significantly worse MCS scores (difference 3.93, p = 0.001) and PHQ-8 scores (difference 2.51, p = 0.001). There was a significant difference in work status (p = 0.002), with cannabis users reporting higher rates (20%) of being employed but not working compared with noncannabis users (11.1%). Noncannabis users were more likely to be retired (45.9%) compared with cannabis users (31.3%). There were no significant differences based on cannabis use for sex, education, exercise, NRS-back, NRS-Leg, tingling-leg, mODI or health state.
Conclusion
Thoracolumbar spine surgery patients were utilizing cannabis before surgery both through recreational use and prescription. Patients who were using cannabis preoperatively did not differ in regard to reported pain or disability from non-users, although they did in regard to demographic and mental health variables.
Abstract 142. Post-traumatic ascending myelitis, about 2 cases, etiologic analysis and treatment
Background
Since 1969 when Frankel described subacute posttraumatic ascending myelopathy (SPAM) as a rare condition, fewer than 35 cases have been reported in the literature. Its pathophysiologic mechanisms are not clearly understood, making its true cause and incidence difficult to estimate and confirming that spinal cord injury is a dynamic lesion. In the last year in our service we had 2 patients with SPAM. Here we do a parallel and etiologic analysis considering possible risk factors that could predispose to this condition.
Methods
We present a case report and literature review.
Results
Case 1 was a 24-year-old women with fracture dislocation T10–T11 and complete paraplegia (American Spinal Injury Association [ASIA] A), following surgical reduction and stabilization. She developed progressive sensitive motor symptoms on the sixth postoperative day described as paresthesias in her chest and arms, decreased strength in C8–T1 and magnetic resonance imaging (MRI) findings with T2 high signal intensity level of the spinal cord up to C4, and diffuse cord edema requiring follow-up in the intensive care unit for TAM therapy, with stabilization and progressive improvement of symptoms in the next 3 weeks. Case 2 was a 32-year-old man who had a high-speed racing car accident with a T12 burst fracture with complete spinal cord injury. He underwent surgery for early decompression and fusion. On postrauma day 12 he developed hemodynamic changes with hypotension, chest pain with radiation to the scapula, allodynia, decreased grip and impaired manual dexterity with high signal intensity changes in T2/STIR extending up to C4 with spinal cord expansion and lately ascending to C1 despite clear symptomatic improvement.
Conclusion
SPAM is a subacute complication of spinal cord trauma of high energy mainly in the dorsolumbar region causing severe compression, with hypersignal changes and edema in the MRI and ascending sensorimotor symptoms in young patients (60% aged 20–40 yr) with self-limiting evolution in most cases. Its diagnosis should be always recognized because of the potential of ascending to the brainstem, which could be fatal. There are no treatment guidelines and some of proposed pathophysiologic mechanisms are controversial without clearly defined cause.
Abstract 55. NanoLOCK surfaces enhance osteoblast activities at the cellular level
Background
The nanoLOCK Surface Technology is specifically engineered to create textured features on titanium implants at the micro- and nano-metre levels. These features have been shown to increase osteogenic factor production in human mesenchymal stem cells in vitro. This leads to the hypothesis that nanoLOCK surfaces may enhance bone formation by increasing osteoblast activities at the cellular level. In this study, we tested our hypothesis using an in vitro bone formation model by culturing osteoblast precursor cells directly on titanium materials with and without nanoLOCK.
Methods
Titanium alloy discs were subjected to proprietary surface treatments to create either macro-micro rough surface (MM), macro-micro-nano rough surface (MMN) or micronano rough surface (MN). MC3T3-E1 mouse preosteoblasts were used in the study. Cellular activities were examined over 28 days, including proliferation (WST-1 assay), differentiation (ALP [alkaline phosphatase] and OCN [osteocalcin] assays) and mineralization (Alizarin Red staining). One-way analysis of variance with Tukey pairwise comparison was used to compare measures among test materials with a p value less than 0.05 as statistically significant (n = 6).
Results
Live cells were detected on all test materials. Increased levels of ALP and OCN over time suggested that pre-osteoblast cells differentiated into mature osteoblast cells on each test group. MMN and MN had significantly higher cell numbers (2.5 fold), ALP activities (1.5 fold) and OCN levels (2.9 fold) compared with those of MM (p < 0.05). Mineralization was visible on 1 of the MMN and 2 of the MN samples, but not on any of the MM samples at day 21 (n = 4). Visually, more mineralization was detected on the MMN and MN surfaces compared to those of MM at day 28.
Conclusion
All titanium groups supported cell attachment, proliferation, differentiation and mineralization in the in vitro bone formation assay using the mouse preosteoblast cells. Test materials with the engineered nanoLOCK surface (MMN and MN) increased the osteoblast activities compared with surfaces without nanoLOCK (MM). These results suggest that specifically engineered nanoscale surface textures on titanium implants may facilitate new bone formation and create a stable bone–implant interface.
Abstract 76. Which scoring system is the most accurate for predicting survival in patients undergoing surgery for spinal metastases from lung cancer?
Background
Surgery for spinal metastasis can improve quality of life. However, patients with a short life expectancy will not experience the benefits of surgery. Therefore, spinal surgeons should assess potential survival as accurately as possible before considering surgery, especially in patients with lung cancer in whom survival is poor. Several scores are available to assess survival in patients with spinal metastasis. The Tokuhashi score remains the gold standard while more recent scores consider biologic parameters. The aim of our study was to compare different survival score systems in patients who underwent operations for spinal metastasis from lung cancer.
Methods
In this monocentric study, we retrospectively reviewed a database of patients with lung cancer and spinal metastasis undergoing spinal surgery at our institution from January 2008 to October 2020. We evaluated perioperative data including preoperative biologic parameters (white blood cell count, hemoglobin, albumin). We determined the prognostic value associated with the following scoring systems: revised Tokuhashi, revised Bauer, Skeletal Oncology Research Group (SORG) classic and New England Spinal Metastasis Score (NESMS). Kaplan–Meier survival curve (log-rank test) was used to analyze the 1-year survival and to analyze the survival curves of the prognostic groups of each scoring system. The discrimination was assessed by computing the area under the receiver operating characteristic curve (AUC).
Results
A total of 94 patients who underwent operations for spinal metastasis from lung cancer were included. Mean age was 62 years (32–79 yr); 51% were males. The 1-year survival rate was 18% and the medial survival time was 4 months. The 6- and 12-month AUC were 60% and 76% for revised Tokuhashi, 55% and 58% for revised Bauer, 58% and 63% for SORG classic and 61% and 69% for NESMS, respectively.
Conclusion
Whereas it was not significant, the revised Tokuhashi score seems to be the most accurate score to predict survival in patients undergoing surgery for spinal metastasis from lung cancer. New scores including biologic parameters do not seem to provide further precision in this specific population.
Abstract 11. Pedicle screw insertion using ultrasound-based navigation without intraoperative radiation: feasibility study on porcine cadavers
Background
The aim of this study was to evaluate the feasibility and accuracy of pedicle screw insertion in the lumbar and thoracolumbar spine using a fully automated navigation system based on intraoperative ultrasonography.
Methods
Five porcine cadavers were used to instrument lumbar and thoracolumbar spines using posterior open surgery. Intraoperatively, ultrasound (US) images were acquired and used to establish automatic registration between the anatomy and preoperative computed tomography (CT) images. Navigation was performed on a preoperative CT using tracked instruments. Accuracy of the system (registration error) was measured as the distance of manually collected points to the preoperative CT vertebral surface and compared against fiducial-based registration. A postoperative CT was acquired and screw placements were manually verified. We report breach rates and screw deviations.
Results
A total of 56 screws were inserted (5.5 mm diameter n = 50, 6.5 mm n = 6). Four screws (7.14%) presented a medial breach with an average deviation of 1.35 (standard deviation 0.43) mm and all less than 2 mm. No lateral, anterior, superior or inferior breaches were documented. Among the 4 breaches, 2 were caused by a 6.5-mm diameter screw and 2 by a 5.5-mm screw. The average registration error was 0.96 mm using the US-based registration and 0.97 mm using the fiducial-based registration. For vertebrae instrumented with a 5.5-mm screw, the average axial diameter was 9.29 mm leaving a 1.89-mm margin on each side of the screw. For vertebrae instrumented with a 6.5-mm screw, the average axial diameter was 8.99 mm leaving a 1.24-mm error margin on each side of the screw.
Conclusion
Under experimental conditions, US-based navigation seems to be feasible and practical for pedicle screw insertion in the lumbar and thoracolumbar spine. According to the literature, all breaches recorded (< 2 mm) probably would not be dangerous clinically and revision would not be required. The breach rate (7.1%) was comparable to the figures found in the literature. This method could be used as a low-cost and radiation-free alternative to intraoperative CT and fluoroscopy. More investigation is needed to evaluate consistency, reproducibility and performance in the surgical context.
Abstract 85. What barriers prevent patients being discharged from hospital following elective spine surgery?
Background
Enhanced recovery after surgery (ERAS) protocols are common in many surgical fields; however, they have yet to be formally implemented in spine surgery. Managing efficient discharge from hospital following surgery is a key component in implementing a successful ERAS program. The purpose of this study was to determine which factors are most responsible for continued admission following elective spine surgery.
Methods
Consecutive patients having undergone elective spine surgery for degenerative conditions, who required post-operative admission, were included in this prospective study. The spine service charge nurse was asked twice daily (at 10 am and 3 pm) which factor(s) kept a patient in hospital. If a patient stayed more than 1 night in hospital, they were also approached to determine their impression of the reasons for their ongoing admission. Potential factors included pain management, urinary retention, mobility, nausea, confusion, muscle weakness, and technical or logistical problems. All analyses were conducted using either a χ2 test or a 1-way analysis of variance. Significance was obtained at p less than 0.05.
Results
Eighty-two patients were included in this study and 37 patients consented to provide their input on why they felt they are still in hospital. The average age of the cohort was 60.3 years, and 43 patients were men. Half of the cohort were operated on by neurosurgeons, with the other half by orthopedic surgeons. The average length of stay was 2.74 days, and 90.2% of patients were discharged to their home. The most common factors keeping patients in hospital were mobility (27.5%) and urinary retention (22.5%). Patients most frequently thought pain management (37.1%), or “other” reasons (28.6%) were the most common factors keeping them in hospital. In 62.2% of cases, patients knew the main reason why they were still in hospital. No difference existed in length of stay depending on surgeon specialty (p = 0.58).
Conclusion
Mobility and urinary retention are the most common reasons patients remain in hospital following spine surgery. By targeting these problems, we can potentially help patients toward earlier discharge.
Abstract 15. Propensity-matched comparison of 90-day complications in robotic-assisted versus non-robotic-assisted lumbar fusion
Background
A recent administrative database (PearlDiver) study reported increased 30-day complications with the utilization of robotic-assisted enabling technology. The objective of the current study is to compare 90-day complication rates between robotic-assisted and non-robotic-assisted lumbar spinal fusions in propensity-matched cohorts.
Methods
Patients who underwent 1- to 3-level lumbar fusion surgery for degenerative conditions were identified. Of 146 robotic-assisted cases that met the inclusion criteria, 114 were successfully propensity matched to 114 patients from 214 cases who had 1- to 3-level lumbar fusion without robotic assistance based on age, sex, body mass index, smoking status, American Society of Anesthesiologists (ASA) grade, number of surgical levels, primary versus revision surgery, and surgical approach (posterior only or anterior–posterior). We excluded tumour, trauma, infection and deformity cases. Outcomes included surgical and medical (major/minor) complications at intraoperative, immediately postoperative, and 30- and 90-day postoperative intervals, including re-operations, and readmissions within 90 days.
Results
All-cause intraoperative complication rates were similar between the non-robotic-assisted (5.3%) and robotic-assisted groups (10.5%, p = 0.37). The immediate postoperative medical complication rate was also similar between the non-robotic-assisted (6.1%) and robotic-assisted groups (1.8%, p = 0.09). Thirty-day complication rates, 90-day complication rates, reoperation rates and readmission rates showed no difference between non-robotic-assisted and robotic-assisted groups. There was no difference between return to the operating room for infection between the cohorts (non-roboticassisted 6 [5%] v. robotic-assisted 1 [0.8%], p = 0.12). There was, however, improved length of stay (LOS) in the robotic-assisted group compared with the non-robotic-assisted group (2.5 v. 3.17 d, p = 0.018).
Conclusion
In propensity-matched cohorts, patients undergoing 1- to 3-level robotic-assisted posterior lumbar fusion for degenerative conditions did not have an increased 90-day complication rate, and they had a shorter LOS compared with patients undergoing a non-robotic-assisted procedure. There findings differ from a prior administrative database study as the robotic-assisted group in the current study had a 0% return rate to the operating room for malpositioned screws and a 0.8% return rate to operating room for infection.
Abstract 56. No-tap (2-step) robotic-assisted cortical bone trajectory (RA-CBT) screw insertion is safe and efficient: comparative analysis of 179 patients and 924 RA-CBT screws
Background
Traditionally, the workflow for cortical bone trajectory (CBT) screws has included tapping either line-to-line or under tapping by 1 mm. To our knowledge, no reports describe a non-tapping, 2-step workflow for CBT screw placement. We sought to compare the safety profile and time savings, if any, in this no-tap (2-step) versus tap (3-step) workflow.
Methods
Patients undergoing robotic-assisted (RA) 1- to 3-level posterior spine fusion with CBT screws for degenerative conditions were identified. Patients were separated into either no-tap or tap workflow. Numbers of total screws, screw-related complications, screws malpositioned, aborted or converted, other complications, estimated blood loss (EBL), operating room (OR) time, robotic time and return to OR were analyzed.
Results
We identified 179 RA-CBT patients with 91 patients undergoing 2-step (458 screws) and 88 undergoing 3-step (466 screws) workflow. There was no difference in age, sex, body mass index, revisions status, smoking, American Society of Anesthesiologists (ASA) score, approach (posterior only v. anterior–posterior), number of levels fused or diagnosis between the cohorts. Total robotic time was decreased from 30.3 (tap) to 26.7 (no tap) minutes (p = 0.05), while OR time (181.4 [tap] v. 169.9 [no tap] min) was decreased as well but not significantly (p = 0.24). No patient in either cohort was returned to OR for malpositioned screws. There was no difference in malpositioned screws identified intraoperatively (6 v. 10, p = 0.43), screws converted to freehand (3 v. 3, p = 0.69) or screws abandoned (2 v. 3, p = 1.00). No pedicle fracture or fixation failure was identified in the no-tap cohort with 1 in the tapping cohort (p = 1.00).
Conclusion
The no-tap (2-step) screw insertion workflow for RA-CBT reduced robotic time without increased complications. In both cohorts, no patients were returned to the OR for malpositioned screws and no pedicle fractures or fixation-related complications were identified in the no-tap (2-step) workflow. The no-tap workflow is safe and improves efficiency in RA-CBT screw insertion.
Abstract 124. Developing a Web-based application to promote the adoption of a clinical prediction model for independent walking in people with traumatic spinal cord injury — a protocol
Background
Prediction models using algorithms to estimate future clinical outcomes can inform decision-making and potentially improve patient care. For traumatic spinal cord injury (TSCI), several clinical prediction rules (CPRs) have been published, but few are used in clinical practice. Web-based tools integrating CPRs into more easily accessible and user-friendly “calculators” may help increase their use. We aimed to develop a web-based calculator, called Ambulation, using a validated CPR for independent ambulation 1 year after TSCI and describe the process of developing a digital tool for clinical use.
Methods
A validated CPR was used as the basis for Ambulation. The Web app uses patient data collected within 15 days of injury to produce a risk score (0%–100%) reflecting the probability of walking independently 1 year after TSCI. Ambulation was developed using HTML, CSS and JavaScript. A diverse team of professionals were consulted throughout the design and development process. Feedback was solicited in an iterative fashion and relevant steps were recorded and are presented to improve future prediction model digital implementation efforts.
Results
Ambulation, a Web-based application, was developed for clinicians treating individuals with TSCI. The application underwent 9 revisions. Changes aimed to clarify communication and model output and to improve the design, layout and site navigation for the end user. We consulted a multidisciplinary team of experts (n = 27) including clinicians and researchers (n = 9), people with lived experience of SCI (n = 4), communications, marketing and networking professionals (n = 2), information technology and Web developers (n = 3), a privacy lawyer (n = 1) and various external advisors (n = 8). Important considerations included privacy and licensing, defining the end user, tailoring the user interface for better integration into existing clinical workflows, establishing systems to monitor use and gather ongoing feedback, and preparing a pilot to obtain user feedback.
Conclusion
Ambulation has been successfully developed and will be piloted with a small group of clinicians. Future work will involve monitoring Ambulation website traffic, collating and analyzing users’ feedback and making improvements before releasing the final version.
Abstract 125. Multivariable prediction models for prognostication after traumatic spinal cord injury — a systematic review
Background
Prognosis after traumatic spinal cord injury (TSCI) can help improve treatment strategies, health care resource allocation and counselling for patients. This systematic review aimed to critically examine predictor selection methods and describe common patient-centred outcomes used in multivariable prediction models for TSCI.
Methods
We searched Medline, PubMed, Embase, Scopus and IEEE for the period between database inception and Jan. 1, 2021. English peer-reviewed studies that developed multivariable prediction models to prognosticate patient-centred outcomes in adults with TSCI were included. Reference lists were screened to capture additional relevant literature. Risk of bias was assessed using PROBAST (Prediction Model Risk of Bias Assessment Tool), and articles not meeting a prespecified threshold of quality were excluded. Data extraction was performed using a modified checklist for systematic reviews of prediction modelling studies.
Results
A total of 958 studies were initially retrieved, and 663 abstracts were screened; of these, 21 studies (published between 2009 and 2020) met the inclusion criteria and underwent data extraction. Out of the 4 domains (participants, predictors, outcome and analysis) assessed for methodologic quality using the PROBAST tool, the data analysis domain was most commonly at a high risk of bias (21; 100%). Eighteen (86%) studies retrospectively reviewed data from existing registries, and 3 used data from a prospective cohort design. Predictors commonly considered for model development included age and variables derived from neurologic examination. Overall, 71.4% (15/21) of studies predicted a functional outcome, 9.5% (2/21) a neurologic outcome, 9.5% (2/21) survival and 9.5% (2/21) other. Two (11%) recent studies (from 2016 and 2019) followed established guidelines for transparent reporting of multivariable prediction models.
Conclusion
This systematic review critically examined multivariable prediction models of sufficient methodologic quality that have been validated for prognosticating patient-centred outcomes in patients with TSCI. Outcomes commonly examined were functional recovery and independence after TSCI. Poor methodologic rigour in data analysis, as defined by a validated quality assessment tool for prediction model studies, and inconsistent reporting practices are vital areas for improvement for future researchers.
Abstract 148. Expression of blood serum cytokines in the presentation of acute sciatica
Background
The aim of this study was to investigate systemic inflammatory profiles in the natural history of lumbar disc herniation to understand, and possibly predict, successful spontaneous improvement.
Methods
Patients within 4 weeks of onset of acute sciatica were included. After assessment by an attending spine surgeon, patients provided informed consent and were enrolled if the RAPIDH score was greater than 9. At each follow-up visit (at 0, 4, 16 wk) blood samples were taken, and conservative management was directed, including analgesics and a selective nerve root injection. Luminex multiplex technology was used to quantify a 48-plex panel of serum cytokines at each time point. Pain and disability were quantified at each visit using validated questionnaires. A 2-way analysis of variance was used to investigate the interaction of cytokine expression over time, and between surgical and nonsurgical patients.
Results
After we screened 178 patients, 54 were enrolled. Of the 31 patients who had completed cytokine measurements, 6 required surgical intervention. All nonsurgical patients showed a significant reduction in self-reported pain and disability by the 16-week time point. Surgery was provided to patients for the treatment of pain, not neurologic deficits. A significant interaction was found between surgical and nonsurgical patients; however, a significant interaction with time was not identified. Of the 48 blood serum cytokines investigated, macrophage-derived chemokine (MDC), platelet-derived growth factor (PDGFabbb), RANTES (CCL5) and FLT3-ligand (FLT3L) returned statistical significance (p < 0.05), with Interleukin1-beta (IL1b) showing a trend toward significance (p = 0.07). RANTES and PDGFabbb are elevated in nonsurgical patients, while MDC, FLT3L and IL1b are elevated in surgical patients. Interestingly, cytokines previously associated with sciatica (IL-4, IL-6 and IL-10) were not found to have statistical significance in this study.
Conclusion
The findings presented are preliminary findings of a larger work in progress. Interestingly, the interaction of time and treatment was not significant, suggesting that difference in cytokine levels between nonsurgical and surgical patients may be consistent over time. This finding may aide us in our overall goal of predicting spontaneous improvement in symptomatic disc herniations.
Abstract 150. Do patient-reported outcome scores (PROs) correlate with bundled payment plan performance for elective spine surgeries?
Background
The primary objective of this study was to compare change in patient-reported outcome (PRO) scores and bundled payment plan performance for surgical spine care episodes. Secondary objectives included analysis of episode of care factors that correlate with poor financial performance or change in Disability Index (DI) scores.
Methods
Patients enrolled in the Medicare Bundled Payments for Care Improvement (BPCI) bundled payment plan undergoing elective spine surgery between October 2018 and October 2020 at a single institution were retrospectively analyzed. DI scores (including Oswestry Disability Index and Neck Disability Index) were evaluated preoperatively and at 3 months postoperatively. Statistical analysis was performed using Spearman coefficient for continuous variables and the Fisher exact tests for categorical variables.
Results
Sixty-two patients were included in the analysis. Forty-seven patients had episodes of care with total costs below target bundle price and 15 had costs above target bundle price. Mean overall financial performance (actual cost – bundle repayment) was $2920 below target price. Mean improvement in DI scores was 18.5 (standard deviation 14.6). There was a statistically significant positive correlation between financial performance (target price – actual cost) and change in DI (rs = 0.328, p < 0.05). Patients with poor bundle performance had a significantly higher rate of emergency department visits (33.3% v. 6.4 %, p = 0.016) and readmissions (20% v. 0 %, p = 0.012) within the bundle period. Poor financial performers also had lower mean improvement in DI scores (12.9 v. 20.4) that trended toward significance (p = 0.06).
Conclusion
As we look to improve care value, it is important to assess how this affects patient outcomes. To our knowledge, this is the first analysis to determine the correlation between PRO scores and financial performance within a defined episode of care. In this patient population, there was a positive correlation between financial performance and PRO scores. Subgroup analysis by financial performance suggests that patients who “beat” bundle pricing might have improved outcomes compared with those who “bust” the bundle.
Abstract 46. Effects of delayed neurosurgery on anxiety, depression and economic burden
Background
Patients waiting for surgery in Canada often have their operation delayed for a variety of reasons. Surgical postponement generally has negative effects on patients, but there are still aspects of this experience that are not well understood. This study examines how delayed neurosurgical operations affects patients’ emotional experience and economic factors.
Methods
Patients with a delayed neurosurgery in Edmonton, Alberta, from March 2019 to May 2021 completed surveys evaluating psychological and economic impacts, based on modified versions of the Hospital Anxiety and Depression Scale (HADS) and the Ambulatory and Home Care Record. Demographic, lifestyle and clinical variables were also collected. Ordinal regression was used to find what factors were associated with anxiety (HADS-A) and depression (HADS-D) scores. Economic impact was assessed with descriptive statistics and Mann–Whitney U tests.
Results
The survey was completed by 83 of 199 patients identified. On average, cancellations happened 6 days before the operation and delayed surgery 51.8 days. Cases or borderline cases of anxiety presented in 56.6% of patients, while 49.4% were classified as cases or borderline cases of depression. Anxiety was associated with age, receiving conflicting information, accompaniment to appointments, impact on work capacity and retained domestic capacity. Depression was associated with decreased desire for surgery, worsening symptoms and domestic capacity. On average, patients attended 10.1 additional appointments costing Can$502 and required 55.2 days off work. Additionally, 16.9% of patients paid for domestic help and 28.9% required caregiving from family or friends. These economic burdens disproportionately affected some. Further, approximately half of patients and caregivers made arrangements for the cancelled surgery that could not be changed. Those most economically affected were female, employed or accompanied to appointments, had worsening symptoms, prior cancellation or changed work status or received conflicting information.
Conclusion
Patients experienced moderate levels of anxiety and depression from delayed neurosurgical operations. Cancellations imposed economic burdens that disproportionately affected some. Our findings suggest ways to minimize the impact of delayed surgeries by anticipating which patients will be the most affected.
Abstract 69. Care close to home — a retrospective analysis of patients undergoing elective lumbar surgery in a rural satellite hospital
Background
Patients from rural communities face multiple barriers in accessing operative care. Given the recent reduced access to the operating room in tertiary centres, alternative strategies are necessary to provide patients with the care they need. In March 2021 our region expanded elective operative interventions into a satellite centre to be performed by the tertiary surgical team. Patients from rural communities and those willing to travel were carefully selected. The primary objective was to report the incidence of adverse events in patients undergoing elective lumbar surgery at this centre. Secondary objectives were to assess length of stay, readmissions and wait times.
Methods
Patients who underwent elective lumbar spine surgery in satellite centres were identified through surgeon billings. All patients were monitored perioperatively by a nurse practitioner–led virtual platform. This collected demographic data such as age, sex, body mass index (BMI) and smoking status. It also prospectively collected patient-reported outcome measures, and adverse events up to 30 days. Adverse events were additionally retrospectively collected from patient records.
Results
Twenty-one patients have received surgery to date; the average age was 54 years (standard deviation [SD] 11.84 yr). Average BMI and American Society of Anesthesiologists (ASA) were 27.4, and 2.14 respectively. Thirteen patients underwent at least a single-level fusion. Average operative time was 74.9 minutes (SD 27.6 min). All simple decompressions were performed as outpatients, and patients who underwent fusions were discharged on postoperative day 1. Six patients (28%) experienced at least 1 adverse event. All were considered minor, with the exception of 1 delayed presentation dural tear requiring revision surgery, and 1 postoperative reduction of muscle strength in dorsiflexion that has since improved. No transfusions, cardiopulmonary events, emergency transfers or deaths occurred. Average wait time to surgery was only 86.1 days.
Conclusion
Elective lumbar spine surgery performed in rural satellite centres on selected patients by an experienced surgical team has a low rate of adverse events. It allows rural patients to have surgery close to home, and it has an average time to surgery of less than 3 months. This has the potential to improve access to surgical care and reduce cost of delivery in a safe manner.
Abstract 110. Surgical adverse events for primary tumours of the spine and their impact on prognosis and outcomes: a PTRON study
Background
Best available evidence supports highly invasive en bloc resection for primary tumours of the spine to decrease local recurrence. These operations are however challenging and associated with a high rate of adverse events (AEs). The impact of adverse events on patient-reported outcomes is unknown and is critical to the shared decision-making process and to quality improvement initiatives. Our objectives were to assess the rate of surgical AEs from a large multicentre registry and their impact on patient-reported outcomes and prognosis.
Methods
The Primary Tumor Research and Outcome Network (PTRON) is a multicentre international prospective registry. We selected adult patients with documented surgical treatment and available follow-up data at 3 months. Our primary outcome was the rate of AEs (total, intraoperative and postoperative). The secondary outcomes were quality of life (SOSGOQ, EQ-5D), local recurrence, readmission, reoperation and mortality. We performed a descriptive synthesis of our results as well as a multivariate logistic regression model to assess the impact of adverse events on outcomes.
Results
From the 895 patients enrolled in PTRON, 327 met the inclusion criteria. The most frequent histology was chordoma (33.3%) followed by malignant peripheral nerve sheath tumors (15%) and chondrosarcoma (11%). Lumbar and thoracic spine were the most frequent locations with 34.6% and 32.7%, respectively, followed by sacrum (24.2%) and cervical (17.1%). Sixty-six percent of the patients did not have prior surgical treatment. Sixty-four patients (19.6%) experienced at least 1 intraoperative AE and 122 patients (37.3%) had at least 1 postoperative AE. Overall, 150 patients (45.9%) experienced AEs, either intraoperative or postoperative. Statistical analyses assessing the impact of AEs on quality of life and prognosis are still pending.
Conclusion
The rate of surgical AEs is considerable in this patient population, but lower than what was previously reported in single-centre studies. Results concerning the impact of AEs on outcomes are pending and will be of great interest when discussing these operations with patients.
Abstract 80. Spinal cord stimulation research in the restoration of function for individuals living with spinal cord injuries: a scoping review
Background
Spinal cord stimulation (SCS) is emerging as a promising treatment to restore neurologic function following spinal cord injury (SCI). Given the heterogeneity of SCI and the intricacies of SCS programming, there is a need to utilize rigorous study methods and transparent reporting to establish the efficacy and safety in SCI. The objective of this study was to conduct a scoping review describing SCS research targeting motor, sensory and autonomic function in SCI to inform future research studies.
Methods
A search of 8 databases with permutations of the search terms “spinal cord stimulation,” “epidural spinal cord stimulation” (ESCS), “transcutaneous spinal cord stimulation” (TSCS), “magnetic spinal cord stimulation” (MSCS) and “neuromodulation” was conducted. Original human research published through Aug. 31, 2020 was included. Data extracted included authors, publication year, participant characteristics, purpose, study design, methods, primary outcomes, results and adverse events.
Results
The search identified 5700 articles, and 76 articles were included. They included 41 studies on ESCS, 23 on TSCS and 12 on MSCS, with 68% of the articles conducted in the United States. Common study designs included case studies and series (n = 35) and quasi-experimental designs (n = 21), and only 1 single randomized clinical trial. The most common primary outcome was motor recovery for ESCS and TSCS studies (28 of 64), and bowel and bladder outcomes were the main focus for MSCS studies (10 of 12). Sample size in 65% of the studies was 10 or fewer participants. Common adverse events reported for ESCS were infection and pain at the implant site and for TSCS it was unusual sensations and increased spasticity.
Conclusion
Results from this review identified the need for more robust study designs, larger sample sizes, studies comparing the different types of SCS, and the alignment of outcomes with priorities of the SCI community. Improved reporting in publications (e.g., study design, stimulation parameters, adverse events) is also needed. Advancement in these areas will strengthen the evidence on the efficacy and safety of SCS and facilitate commercialization so individuals with SCI can benefit from this technology.
Abstract 132. Workup and management of asymptomatic extracranial traumatic vertebral artery injury: a Canadian Neurosurgery Resident Research Collaborative study
Background
Extracranial traumatic vertebral artery injury (eTVAI) occurs in approximately 10% of nonpenetrating head and neck traumas. Most cases are initially asymptomatic with an increased risk for stroke. Given that consensus is lacking regarding screening, treatment and follow-up of asymptomatic patients with eTVAI, our objective was to investigate national practice patterns reflecting these domains.
Methods
An electronic, case-based survey was distributed by the Canadian Neurosurgery Resident Research Collaborative to members of the Canadian Neurological Sciences Federation and the Canadian Spine Society. Case 1 involved an asymptomatic patient, a low-energy injury mechanism and radiographic imaging demonstrating a nondisplaced cervical lateral mass fracture extending into the foramen transversarium. Case 2 involved an asymptomatic patient, a high-energy injury mechanism and radiographic imaging demonstrating a unilateral oblique C2 body and contralateral pars fracture. In both cases, screening questions were followed by angiographic findings. Case 1 angiography: stratified on the basis of luminal diameter reduction and the presence or absence of intimal flap. Case 2 angiography: traumatic pseudoaneurysm dissection. Survey responses were analyzed using descriptive statistics.
Results
One hundred and eight of 182 participants (59%), representing 20 academic sites, completed the survey. The majority (71%) of the respondents do not use a criterion-based screening tool to diagnose eTVAI. Regarding case 1, 78% of respondents would screen for eTVAI; of those, 97% would screen with computed tomographic angiography (CTA) and 88% would screen immediately. Ninety-seven percent of respondents would initiate medical therapy; the majority would treat with aspirin (89%) for 3–6 months (46%). Most respondents would follow up clinically (89%) or radiographically (75%), every 1–3 months, respectively. Regarding case 2, 73% would screen for eTVAI, of whom 96% would screen with CTA and 88% would screen immediately. Ninety-four percent would initiate treatment. Fifty-four percent would treat with aspirin for 3–6 months, and 36% of respondents would utilize endovascular therapy. Most would follow up clinically (97%) or radiographically (89%), every 1–3 months, respectively.
Conclusion
This study highlights Canadian practice patterns for eTVAI. These findings may inform the development of national practice guidelines, which are currently lacking.
Abstract 12. A surgical treatment algorithm for restoring pelvic balance and health-related quality of life in high-grade lumbosacral spondylolisthesis
Background
The purpose of this study is to propose an evidence-based surgical algorithm for achieving normal pelvic balance while optimizing health-related quality of life (HRQOL) in young patients with high-grade spondylolisthesis.
Methods
This retrospective study assessed a multicentre cohort of 61 patients (aged 14.4 [standard deviation 2.7] yr) with high-grade L5–S1 spondylolisthesis followed for a minimum of 2 years after surgery. Classification and regression tree analysis was used to identify objective criteria associated with achieving a pelvic balance and HRQOL after surgery.
Results
The most important predictor for achieving a balanced pelvis after surgery was a postoperative L5 incidence less than or equal to 63.5°. With a postoperative L5 incidence less than or equal 63.5°, preserving a residual slip percentage greater than 9%, and performing an L5–S1 posterior lumbar interbody fusion or transforaminal lumbar interbody fusion (PLIF/TLIF) increased the likelihood of achieving a balanced pelvis after surgery. When the L5 incidence was greater than 63.5°, a balanced pelvis was most likely observed with proximal extent of fusion limited to L5, residual slip percentage less than or equal to 40%, and residual lumbosacral angle greater than 98° between the L5 upper endplate and the posterior border of S1. Predictors of postoperative HRQOL scores were the preoperative HRQOL score, L5 incidence and slip percentage.
Conclusion
A surgical algorithm is proposed to achieve normal pelvic balance, while optimizing HRQOL. The first step is to assess the L5 incidence, which integrates into a single parameter the lumbosacral kyphosis and abnormal pelvic morphology typically observed in high-grade spondylolisthesis. If L5 incidence is less than 65°, the next step will depend on preoperative pelvic balance. With a preoperative balanced pelvis, it is important not to reduce completely the slip percentage by leaving a residual slip percentage of 10% or greater. When the preoperative pelvis is unbalanced, performing a TLIF/PLIF at L5–S1 is strongly recommended because it will facilitate correcting the angular deformity at L5–S1. If L5 incidence is 65° or greater, a TLIF/PLIF at L5–S1 should also be carried out to correct the angular deformity at L5–S1, and fusion should ideally end at L5 proximally, in addition to performing gradual reduction of the slip percentage. If fusion up to L4 is required, achieving a lumbosacral angle of 100° or greater is key.
Abstract 38. Effectiveness of 6 surgical approaches for minimally invasive lumbar interbody fusion: 1-year follow-up results from a global multicentre study
Background
Minimally invasive lumbar interbody fusion (MI-LIF) procedures are an effective treatment for patients with degenerative lumbar disease (DLD). However, consensus does not exist among surgeons for selecting 1 approach over another. The objectives were to collect patient-reported, surgical and fusion outcome data at 1 year after surgery for patients receiving either anterior lumbar interbody fusion (ALIF), direct lateral interbody fusion (DLIF), oblique lumbar interbody fusion (OLIF), posterior lumbar interbody fusion (PLIF), transforaminal lumbar interbody fusion (TLIF) or midline lumbar interbody fusion (MIDLF) for DLD.
Methods
A total of 340 patients with DLD were consecutively enrolled in a prospective, global, multicentre cohort study (MASTERS-D2; NCT02617563). Patients were treated according to the surgeon’s choice with 1 of 6 MI-LIF procedures. Outcome data for disability (Oswestry Disability Index), back and leg pain (Visual Analogue Scale), quality of life (EQ-5D) were collected at baseline, 4 weeks, 3 months and 12 months. Demographic, surgical and safety data were also recorded. Fusion status was assessed by using computed tomography or x-rays at 1 year (plus or minus 6 mo) after surgery. Paired sample t tests were used to test for improvement from baseline.
Results
One year after surgery, patients attained clinically significant improvements on all patient-reported outcome measures regardless of approach used. Patients who were selected to receive an ALIF comprised the highest proportion of smokers, were the youngest and had the longest operating time, but low fluoroscopy exposure. Anterolateral (ALIF, DLIF, OLIF) compared with posterior (MIDLF, PLIF, TLIF) approaches had the least amount of blood loss, despite similar or longer surgical times. Within 1 year of follow-up, 7 device-related and 7 surgery-related serious adverse events (SAEs) had been recorded. Assessment of fusion was hindered by the effects of COVID-19. In total, 196 out of 340 (57.6%) patients were assessed. The aggregate fusion rate for anterolateral approaches was 88.1% and for posterior approaches 85.1% at 12 months of follow-up.
Conclusion
All 6 approaches for MI-LIF surgery demonstrate favourable patient-reported and surgical outcomes for patients with DLD. Continuing data collection up to 5 years after surgery will yield information on long-term effectiveness, safety, health economics and revision surgery and on the long-term impact of surgeons’ choice of approach.
Abstract 39. Clinical outcomes and fusion success in patients with degenerative lumbar disease without spondylolisthesis: comparing anterolateral to posterior MIS approaches from a global multicentre study
Background
The aim of this study was to examine if differences in patient-reported outcomes and fusion success exist at 1 year between patients with degenerative lumbar disease (DLD) without spondylolisthesis following minimally invasive lumbar interbody fusion surgery (MI-LIF).
Methods
Patients with DLD without spondylolisthesis requiring 1- or 2-level anterolateral (anterior lumbar interbody fusion, direct lateral interbody fusion, oblique lumbar interbody fusion) or posterior (midline lumbar interbody fusion, posterior lumbar interbody fusion, transforaminal lumbar interbody fusion) MI-LIF surgery were consecutively enrolled (n = 109) in a multicentre prospective clinical study (NCT02617563). Each procedure was completed by an MI-experienced surgeon who chose the approach per patient. Disability (Oswestry Disability Index [ODI]), back and leg pain (Visual Analogue Scale) and quality of life (EQ-5D) data were collected at baseline, 4 weeks, 3 months and 12 months after surgery. Fusion status was assessed at 1 year (plus or minus 6 mo). Paired t tests and analysis of covariance were used to test for improvements from baseline and between-group differences.
Results
Fifty of 109 patients received anterolateral MI-LIF (58% 1-level, 42% 2-level) and 59 patients received posterior MI-LIF (86% 1-level, 14% 2-level). Stenosis was less common in patients who underwent anterolateral (60% first level, 52% second level) compared with posterior (78% at first level, 88% at second level) MI-LIF procedures. At 1 year after surgery, patients, on average, reported clinically important improvements on all outcome measures compared with baseline regardless of approach used. There were no statistically significant differences in the level of improvement between patients for ODI, leg pain and EQ-5D index score. Patients treated with a posterior approach had a greater reduction in back pain (p = 0.017). Overall, 87% of patients assessed for fusion achieved fusion by 1 year. One MI-LIF procedure-related and 1 device-related serious adverse event in the anterolateral group were reported.
Conclusion
Despite the anterolateral group having a higher percentage of patients requiring 2-level MI-LIF compared with the posterior group, the fusion rates and patient-reported outcome measures were similar, except for the reduction in back pain, which favoured the posterior group. It is not clear if surgeons are matching the approach to the patient, or if it does not matter which approach is used.
Abstract 40. Anterolateral versus posterior approaches to minimally invasive interbody fusion for patients with spondylolisthesis: results at 1-year follow-up from a global multicentre study
Background
The aim of this study was to compare fusion rates and patient-reported outcomes for anterolateral and posterior approaches at 1 year after surgery.
Methods
Fusion was assessed by computed tomography or radiography. Disability, quality of life, and back and leg pain were assessed using the Oswestry Disability Index (ODI), EQ-5D and visual analogue scales (VAS), respectively. Patients with degenerative spondylolisthesis (DS) (n = 231) requiring 1- or 2-level minimally invasive lumbar interbody fusion surgery (MI-LIF) were consecutively enrolled in a prospective, international multicentre clinical study (NCT02617563). Patients received an anterolateral or posterior approach on the basis of their surgeon’s decision. Fusion status at 1 year (plus or minus 6 mo) was assessed by the surgeon or radiologist and patient-reported outcomes (PROs) were assessed at baseline, 4 weeks, 3 months, and 1 year after surgery. Paired sample t tests and analysis of covariance were used to test for improvements from baseline and group differences.
Results
At 1 year after surgery, the overall fusion rate was 86%, and patients demonstrated clinically important improvements from baseline on ODI, EQ-5D and VAS scores. For reductions in back and leg pain and fusion rate there was no differences between groups. Patients who received posterior compared with anterolateral MI-LIF demonstrated greater reductions in disability (p = 0.028) and improvement in quality of life (p < 0.001). The improvement on all PROs remained stable from 3 months to 1 year of follow-up. Up to 1 year after surgery, 6 MI-LIF procedure-related (3 anterolateral, 3 posterior) and 6 device-related serious adverse events (4 anterolateral, 2 posterior) were recorded.
Conclusion
Patients with DS treated with either an anterolateral or posterior approach for MI-LIF demonstrated substantial improvement in quality of life, alleviation of disability and back and leg pain, good rates of fusion and a good safety profile 1 year after surgery. Further research should evaluate surgeons’ rationale for technique selection and the indications for anterolateral versus posterior techniques (e.g., indirect v. direct decompression).
Abstract 73. Benefit of minimally invasive lumbar interbody fusion versus traditional interbody fusion versus posterolateral spinal fusion in lumbar degenerative spondylolisthesis: a propensity-matched analysis from the CSORN prospective LDS study
Background
In Canada, fusion for lumbar degenerative spondylolisthesis (LDS) is trending toward interbody fusion despite the evidence to justify the larger magnitude procedure. One indication for an interbody device is minimally invasive technique. The objective of this study was to compare patient-rated outcomes (PROs) in patients treated with minimally invasive interbody fusion (MIS IF), posterolateral fusion (PLF) or open IF (open IF).
Methods
The Canadian Spine Outcomes and Research Network CSORN multicentre prospective study on LDS was retrospectively reviewed to identify patients aged 55 years and older, who had single-level MIS IF, open IF or PLF surgery between 2015 and 2020, and were 1 year after surgery. Using a propensity score, patients who had PLF or open IF were matched to patients who had MIS IF. Matching was based on 8 demographic, clinical and radiographic covariates. Back and leg pain, Oswestry Disability Index and SF-12 mental and physical scores were measured preoperatively and at 6–18 weeks and 1 year postoperatively. Between-group differences were compared using repeated-measured mixed-effects modelling.
Results
Out of a total of 567 patients enrolled across 8 centres, 278 met the inclusion criteria including 56 PLF, 61 MIS IF and 161 open IF patients. Propensity score matching yielded 40 matched pairs of patients who were age 65 years and mostly female, with primarily neurogenic claudication and grade 1 spondylolisthesis. Blood loss was lowest in the MIS IF group versus open IF group (mean difference 145 mL, p = 0.023). Length of stay, operating time and adverse events were similar between the groups. Patients who had MIS IF reported worse back pain at 6–18 weeks (3.7 [standard deviation (SD) 0.3) versus open IF (2.5 [SD 0.3], p = 0.002) versus PLF (2.7 [SD 0.3], p = 0.004), which resolved at 1 year. All other PROs were similar between the 3 groups to 1 year.
Conclusion
When considering MIS IF, open IF or PLF, patients and surgeons can expect the same functional recovery out to 1 year after surgery. MIS IF provides less blood loss. Future studies on the cost-effectiveness of these techniques is warranted.
Abstract 67. The effect of fusionless pediatric scoliosis surgery on 3D radiographic spinopelvic alignment
Background
Anterior vertebral body tethering (AVBT) and posterior dynamic deformity correction (PDDC) are motion-sparing procedures for the treatment of idiopathic scoliosis. Currently, comparative data between AVBT and PDDC remain limited. We aim to compare the effects of AVBT and PDDC on perioperative 3D spinopelvic alignment.
Methods
Through a single-centre study, patients treated with either AVBT or PDDC were identified. Major scoliosis, minor scoliosis, thoracic kyphosis (TK), lumbar lordosis (LL), pelvic incidence (PI), sacral slope (SS), pelvic tilt (PT), thoracic apical vertebral rotation (AVR thoracic) and thoracic apical vertebral translation (AVT thoracic) were measured pre- and post-operatively using 3D biplanar x-ray imaging and reconstruction. Independent t tests compared correction rates and absolute changes in spinopelvic parameters between groups.
Results
Twenty-two patients (9 PDDC, 13 AVBT) with a mean age of 13.7 (10.3–17.8) years at index surgery were identified. The correction rate of major scoliosis was 50% with PDDC and 40% with AVBT (p = 0.14). The correction rate of minor scoliosis was 48% using PDDC and 29% using AVBT (p = 0.07). The total change of TK was +10° for PDDC and −1° in AVBT (p = 0.007) postoperatively. LL was +5° with PDDC and −8° with AVBT (p = 0.008). SS was +1° with PDDC and −6° with AVBT (p = 0.05). PI and PT were both similarly stable between groups (p = 0.3, p = 0.2). Postoperative AVR thoracic improved by 3.9° for PDDC and 4.9° for AVBT (p = 0.75). AVT thoracic had a postoperative improvement of 3.8 cm for PDDC and 2.2 cm for AVBT (p = 0.04).
Conclusion
Fusionless surgeries improved both major and minor scoliosis, as well as thoracic apical rotation. AVBT was less kyphogenic and decreased LL compared with PDDC, whereas PDDC had better correction of coronal vertebral translation.
Abstract 62. Calculating utilities from the Neck Disability Index: a prerequisite for quantifying the value of cervical spine care
Background
The ability to calculate quality-adjusted life-years (QALYs) for patients with degenerative cervical disease would enhance treatment decision making and facilitate economic analysis. QALYs are calculated using utilities, or health-related quality-of-life (HRQoL) weights. An instrument designed for cervical degeneration would increase the sensitivity and specificity of HRQoL assessments. The objective of this study is to develop a multi-attribute utility function for the Neck Disability Index (NDI).
Methods
We recruited a sample of 2875 adults from a market research panel. Using an online discrete choice experiment (DCE), participants rated 12 choice sets on the basis of NDI health states. A multi-attribute utility function was estimated using a mixed multinomial-logit regression model (MIXL). The sample was partitioned into a training set used for model fitting and validation set used for model evaluation.
Results
The regression model demonstrated good predictive performance on the validation set with an area under the receiver operating characteristic curve (AUC) of 0.77 (95% confidence interval 0.76–0.78). The regression model was used to develop a utility scoring rubric for the NDI. Regression results revealed that participants did not regard all NDI domains as equally important. The rank order of importance was (in decreasing order): pain intensity, work, personal care, headache, concentration, sleeping, driving, recreation, lifting and lastly reading.
Conclusion
This study provides a simple technique for converting the NDI to utilities and quantifying the importance of NDI domains. The ability to evaluate QALYs for cervical degeneration will facilitate economic analysis and patient counselling. Clinicians should use these findings to offer treatments that maximize function in the attributes viewed most important by patients.
Abstract 63. The psychometric properties of the mJOA for quality-of-life assessments in cervical myelopathy
Background
Literature examining the economic and quality-of-life value of treatments for spinal disorders is often inconclusive. This may be a consequence of psychometric limitations of generic health measures, such as the EQ-5D, in spinal cord injury and musculoskeletal disorders. We developed a utility (quality-of-life weight) score for the Modified Japanese Orthopaedic Association scale (mJOA) that can be used by policy-makers, funders, surgeons and patients to support decision-making. We compared the psychometric properties of utilities derived from the mJOA and EQ-5D to determine if a disease-specific instrument is more valid and responsive for the myelopathy population.
Methods
EQ-5D and mJOA data were collected from patients with baseline and 1-year data in the Canadian Spine Outcomes and Research Network (CSORN) Myelopathy Study and used to calculate utilities. Bland–Altman analysis was used to quantify convergent validity between EQ-5D and mJOA utilities. In accordance with best practice guidelines, responsiveness, which is the ability of an instrument to detect change, was quantified using 2 measures. The ability to detect minimum clinically important difference (MCID) was quantified using Guyatt’s responsiveness index (GRI). The ability to detect real change, defined by patients’ self-reported clinical improvement, was quantified using receiver operating characteristic (ROC) curves.
Results
A total of 485 patients were eligible for analysis: 427 (88%) surgical and 85 (12%) nonoperative. The 95% limits of agreement for utility change derived from the mJOA relative to the EQ-5D were (−0.30–0.33) indicating no bias with the mJOA technique. Utilities calculated from the mJOA tended to be more responsive than those from the EQ-5D. The GRI was 3.83 (2.84–5.68) and 1.73 (0.75–5.60) and the AUC was 0.70 (0.65–0.75) and 0.69 (0.64–0.74) for the mJOA and EQ-5D, respectively.
Conclusion
Utilities calculated from the mJOA are valid and tend to be more responsive than those calculated from the gold-standard EQ-5D. Therefore, utilities derived from the mJOA should be incorporated in economic analysis and decision-making for cervical myelopathy to increase the sensitivity of assessments and demonstrate the value of surgical care.
Abstract 59. Low radius of curvature growth friendly implants increases the risk of developing clinically significant proximal junctional kyphosis
Background
Clinically significant proximal junctional kyphosis (PJK) occurs in 20% of children treated with posterior distraction-based growth friendly surgery. It has been theorized biomechanically that low radius of curvature (ROC) implants (i.e., more curved rods) may increase postoperative thoracic kyphosis and may pose a higher risk of developing PJK. We tested the hypothesis that patients with early-onset scoliosis (EOS) treated with low-ROC (more curved rods) distraction-based treatment will have a greater risk of developing PJK compared with those treated with high-ROC (straighter) implants.
Methods
This is a retrospective review of prospectively collected data obtained from a multicentre EOS database on children treated with rib-based distraction with minimum 2-year follow-up. Variables of interest included implant ROC at index (220 or 500 mm), patient age, preoperative scoliosis, preoperative kyphosis and scoliosis cause. In the literature, PJK has been defined as clinically significant if revision surgery with superior extension of the upper instrumented vertebrae was performed.
Results
In 148 patients with scoliosis, there was a higher risk of clinically significant PJK with low ROC (more curved) rods (odds ratio 2.6, 95% confidence interval 1.09–5.99; χ2 (1, n = 148) = 4.8, p = 0.03). Patients had a mean preoperative age of 5.3 years (4.6 y 220 mm v. 6.2 y 500 mm, p = 0.002). A logistic regression model was created with age as a confounding variable, but it was determined to be not significant (p = 0.6). Scoliosis causes included 52 neuromuscular, 52 congenital, 27 idiopathic and 17 syndromic with no significant differences in PJK risk between causes (p = 0.07). Overall, patients had preoperative scoliosis of 69° (67° 220 mm v. 72° 500 mm, p = 0.2) and kyphosis of 48° (45° 220 mm v. 51° 500 mm, p = 0.1). The change in thoracic kyphosis pre-operatively to final follow-up (mean 4.0 [standard deviation (SD) 0.2] yr) was higher in patients treated with 220-mm implants compared with 500-mm implants (220 mm: 7.5° [SD 2.6°] v. 500 mm: −4.0° [SD 3.0°], p = 0.004).
Conclusion
Use of low ROC (more curved) posterior distraction implants is associated with a significantly greater increase in thoracic kyphosis, which probably led to a higher risk of developing clinically significant PJK in patients with EOS.
Abstract 144. Very long–term outcome of single-level minimally invasive lumbar microdiscectomy with a tubular retractor
Background
Minimally invasive lumbar discectomy using tubular retractor systems is a common procedure, with short-term outcomes similar to open techniques. Only 1 report of outcomes at 5 years has been published, showing similar outcomes to the open technique. To our knowledge, no longer term outcome studies have been reported.
Methods
We performed a retrospective study of 103 minimally invasive microdiscectomy procedures from 2011 to 2014. We selected patients with a minimum of 8 years of follow-up. We excluded subjects with cauda equina syndrome, revision surgery, fusions and multilevel decompressions. Long-term telephone follow-up was performed in 2021. Demographic information including body mass index, smoking status and need for revision surgery was collected. We used 1-way analysis to compare visual analogue (VAS), Oswestry Disability Index (ODI), SF-36 and EQ-5D scores at preoperative, 6-week postoperative, 2-year follow-up and long-term follow-up.
Results
The mean long-term follow-up was 9.9 (standard deviation [SD] 1.3) years. The mean preoperative back pain VAS score was 6.23 (SD 2.63), 6-week post-operative follow-up VAS was 3.21 (SD 2.49) and long-term follow-up VAS was 2.56 (SD 2.45). The mean preoperative leg pain VAS score was 7.66 (SD 1.99), 6-week follow-up VAS was 3.56 (SD 2.79) and final follow-up VAS was 2.20 (SD 2.57). The mean preoperative ODI score was 60.41 (SD 13.97), falling to 32.54 (SD 20.57) at 6-week follow-up and further to 24.50 (SD 20.97) at long-term follow-up. The mean baseline EQ-5D quality of life score was 46.4 (SD 18.1), 68.9 (SD 20.2) at 6-week follow up and 69.3 (SD 20.3) at final review.
Conclusion
We report good outcomes for minimally invasive microdiscectomy that are as durable as published results using the open technique. To our knowledge, ours is the longest reported follow-up of minimally invasive microdiscectomy.
Abstract 112. Metal implant hypersensitivity in patients undergoing spinal surgery: a literature review and case report
Background
The aim of this study was to determine the incidence of metal hypersensitivity from implants in patients undergoing spinal surgery and to describe an unusual case of cobalt hypersensitivity following spinal deformity surgery using chrome cobalt rods.
Methods
A PubMed search was conducted to review the literature describing metal hypersensitivity in spine surgery secondary to implant usage. Articles retrieved were classified on the basis of the level of evidence and type of surgery. We describe the case of a 62-year-old female patient who underwent spinal arthrodesis for neurogenic claudication and degenerative scoliosis. We present the surgical technique and implants used as well as the delayed reaction leading to progressive and intermittent neurologic deficits with cauda equina–like symptoms. We display the chronological radiographic, laboratory and histopathologic findings that led to the diagnosis of cobalt hypersensitivity. We will outline the investigations and workup required to confirm a metal hypersensitivity reaction preoperatively and postoperatively.
Results
After we conducted a literature search, 9 papers were included for analysis. The number of papers discussing metal hypersensitivity for the following surgeries was determined: disc arthroplasty 4, anterior fusion surgery 2, posterior fusion surgery 2 and other spine surgeries 2. Two of these papers were literature reviews, 2 were case series and 6 papers were case reports. There were no prospective or high-quality studies investigating this issue. Most of the research published is regarding disc arthroplasties as it involves bearing surfaces. To our knowledge, this case report is the first to describe a rod component as a reason for a delayed hypersensitivity reaction.
Conclusion
There is a paucity of literature about metal hypersensitivity in spinal surgery. This highlights the possible benefit of prospective registries that could document such complications more thoroughly. Exposing the spine surgery community to this issue could lead to more proactive preoperative screening of patients at higher risk of a metal hypersensitivity reaction when indicated.
Abstract 43. Diagnostic value of the lumbar spinal stenosis (SSHQ) survey in virtual care provided at a tertiary spine program
Background
With increased restrictions following the COVID-19 pandemic, use of virtual care has shown an appreciable rise in clinical practice. Use of self-administrated surveys in triaging patients with low back pain assists with prioritizing care. The objective of this study was to assess the diagnostic value of a self-administrated, self-reported history questionnaire (SSHQ) in diagnosing patients with lumbar spinal stenosis (LSS), using combined clinical and imaging results as the gold standard. The SSHQ has 4 questions that all need to be true for a positive diagnosis of LSS.
Methods
Patients with low back pain with and without leg symptoms who had a telephone interview with an advanced practice physiotherapist as a part of a new virtual care initiative were included. A score of 4 points on Q1–Q4 was indicative of the presence of LSS. The relationship between the SSHQ and the Oswestry Disability Index (ODI), the STarT Back questionnaire and the 5-repetition sit-to-stand Test was explored.
Results
Data for 57 patients, of whom 26 were female (46%), with a mean age of 57 (standard deviation 17) years were analyzed. The majority of the patients had a diagnosis of radiculopathy (19 [33%] claudication type and 19 [33%] disc related). Of the remaining patients, 9 (16 %) had a mechanical low back pain, 5 (9%) had neurogenic claudication with bilateral symptoms, 3 (5 ) had degenerative disc disease and 2 (4 %) had peripheral joint involvement with referred pain. The SSHQ score did not correlate with the ODI, STarT Back or performance measure scores. The SSHQ did not differentiate the claudication-type radiculopathy or neurogenic claudication from the disc-related radiculopathy or other diagnoses. Patients whose symptoms were reduced by bending forward as the typical sign of LSS had a significantly lower STarT Back score (p = 0.015), indicating less risk of physical or psychological disability.
Conclusion
The SSHQ did not clearly identify patients with and without LSS, potentially because 3 out of 4 SSHQ questions refer to symptoms that overlap with other diagnostic categories. The STarT Back risk categories appear to be more informative in the diagnosis of LSS.
Abstract 54. Is the Calgary Postoperative Pain After Spine Surgery (CAPPS) score correlated with long-term outcomes after lumbar fusion?
Background
Some evidence suggests that poorly controlled pain in the early period after spine surgery increases the risk of developing chronic pain. The Calgary Postoperative Pain After Spine Surgery (CAPPS) score is a validated prediction tool for postoperative pain. The objective of this study is to determine if CAPPS scores correlate with long-term patient self-reported pain, functional outcomes or quality of life scores after lumbar spine fusion.
Methods
A retrospective review of prospectively collected data for adults (aged > 18 yr) who underwent elective lumbar fusion surgery between 2011 and 2019 was performed for 263 patients. Eight variables were used to calculate the CAPPS score (sex, age, preoperative opioid use, number of motion segments involved within the surgery, numeric pain rating scale, patient depression scores and various outcome measures). A modified 8-tier risk-based score further simplified into the 3-tier CAPPS score was assigned. Outcomes measures include Oswestry Disability Index (ODI), visual analogue scale pain scores for back and leg (VASback, VASleg) and EuroQol-5D (EQ-5D) quality of life scores at 6–8 weeks and 18–24 months postoperative.
Results
There was little correlation on early follow-up (6–8 weeks postoperative) with the CAPPS score. However, at final follow-up (18–24 months postoperative), both the full 8-tier and the collapsed 3-tier CAPPS scores were correlated with VASback and VASleg pain scores, ODI and EQ-5D state of heath, as well as most of the subscores for the ODI and EQ5D.
Conclusion
The CAPPS score was associated with long-term pain, function and quality of life scores after lumbar fusion surgery. Future studies should be done to determine whether aggressive, early management of postoperative pain improves long-term outcomes for lumbar spine fusion.
Abstract 4. Development of a single-entry referral pathway for patients with spinal conditions in Manitoba: a cross-sectional review of impact and potential way forward for Canadian spine programs
Background
Single-entry models (SEMs) are a promising solution for improving wait times. SEMs utilize pooled waiting lists, a centralized intake process and triage using a prioritization scheme. A SEM was introduced for Manitoba spine surgery referrals in 2017. Within this study, we aim to evaluate its effectiveness and areas for potential improvement.
Methods
We conducted a retrospective review of databases capturing all elective outpatient spine surgery consultations referred to a tertiary care center in Manitoba from Sept. 1, 2017, to Dec. 31, 2019. All adult patients (> 18 yr) referred for outpatient evaluation of spinal pathology to a spine surgeon were included. We calculated wait times along the referral pathway from time of referral to time of consultation or surgery.
Results
A total of 8200 patients were referred to spine subspecialists over the time frame of interest. On average, 71 referrals were received per week. The median wait time to consult with a spine surgeon was 131 (interquartile range 79–191) days. As of Jan. 3, 2021, 1926 patients from the original referrals are pending appointments with a spine surgeon. There were a total of 1487 spine procedures performed, of which 575 procedures were off of elective wait lists. There were statistically significant differences in wait times to surgery on the basis of geographic location in Manitoba (p = 0.033).
Conclusion
A retrospective analysis of wait times for outpatient referrals of patients with spinal disorders was performed. A SEM for triage has had limited success in Manitoba owing to a lack of resources available for triage, clinic management and surgical procedures.
Abstract 113. Automatization of bone age calculation
Background
The goal of this project is to automatize bone age calculation of patients with scoliosis on the basis of their respective full-body radiographs. The objective of this clinical preliminary study is to create a large cohort of patients to train an automatization program.
Methods
We performed bone age calculations according to various methods in 267 children with scoliosis and gave this information as a comparative basis to the artificial intelligence (AI) experts. The chosen bone age calculation techniques were the following: Sander (hands’ epiphyses closure observation), Risser (iliac crest process ossification and fusion) and finally the ABC scale method combining Risser and closure of the tri-radiated cartilage. The latter is a promising new route for bone age calculation. There are other methods to calculate bone age; however, we did not choose them as they are less precise or reliable clinically. A minimum of 100 patients were needed for precision. With this sample size, accuracy for the program’s AI was deemed sufficient.
Results
We gathered the bone age calculation for the complete cohort. With manually calculated bone ages combined with full-body radiographs, we will be able to develop an automatization program based on AI and neuronal networks. Therefore, it will allow us to calculate Sanders, Risser and tri-radiated cartilage’s closure bone age automatically o the basis of only 1 radiograph per patient. Validation is currently underway.
Conclusion
A census of 267 patients who were seen in clinic at our institution between Mar. 2 and June 10, 2021, with an average age of 13.54 years was performed. Other parameters have been integrated, such as demographic data (age, height) and measures (Cobb angle and type of convexity). The project’s relevance is in the significant gain of time in bone age calculation for the entire medical team and less subjectivity with more precision.
Abstract 123. An effectiveness and quality-of-life analysis of conservative care versus surgery for moderate and severe cervical myelopathy
Background
There is a paucity of comparative evidence supporting surgical treatment of degenerative cervical myelopathy (DCM). Only 1 randomized controlled trial has been conducted, which found no benefit to surgery. Systematic reviews of observational comparative studies have also not found surgery to be superior to nonoperative care. In this paper we present a decision analysis on the role of surgery.
Methods
A probabilistic patient-level simulation with 100 000 replications was developed using parametric survival models for neurologic progression in cervical spondylotic myelopathy (CSM) and the survivorship of cervical procedures fitted using rigorous meta-regression studies on cohorts of 674 and 98 075 patients, respectively. Utilities (quality-adjusted life-year [QALY] weights) for the Modified Japanese Orthopaedic Association (mJOA) score were obtained from a general population utility valuation study. Overall survival was obtained from life tables published by Statistics Canada. In analysis 1, we calculated the time to neurologic progression (TTNP) for patients with DCM treated conservatively, and time to second surgery (TTSS) for patients undergoing cervical surgery. In analysis 2, we compared the QALYs lived by patients with CSM treated conservatively and surgically.
Results
When comparing conservative care versus anterior cervical discectomy and fusion (ACDF), artificial disc replacement and laminoplasty (ADR), surgery benefitted only 50 % of patients aged 60 years or older (i.e., TTNP < TTSS). With an infinite expected survival, only 50.4% of patients benefitted from surgery. When comparing conservative care versus laminectomy and fusion, at no time horizon did 50% of patients benefit from surgery. For baseline moderate and severe CSM patients, neither ACDF, ADR, and laminoplasty nor laminectomy and fusion imparted a significant QALY benefit over conservative care.
Conclusion
For patients older than 60 years of age, the risk of second surgery is greater than neurologic progression from DCM. For patients younger than 60 years of age treated with ACDF, ADR and laminoplasty, the risk of second surgery is less than the risk of neurologic progression. Neither moderate nor severe CSM patients derive a QALY benefit from surgery. The decision to undergo cervical surgery to prevent neurologic decline must involve shared decision- making with the patient.
Abstract 133. Long-term survivorship of cervical spine procedures: a survivorship meta-analysis and meta-regression
Background
The aim of this study was to synthesize available data on long-term survivorship for commonly performed cervical spine procedures using cutting-edge meta-analytic techniques.
Methods
A systematic review of the Ovid Medline, Embase and CENTRAL databases was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Title, abstract and full-text screening was done in duplicate with kappa scores used to measure agreement. Only articles with cohorts of greater than 20 patients followed for a minimum of 36 months and with available survival data were included. Procedures included were anterior cervical discectomy and fusion (ACDF), cervical disc arthroplasty (CDA), laminoplasty and posterior laminectomy and fusion. Reconstructed individual patient data were pooled across studies using a Bayesian random-effects meta-regression.
Results
Our search identified a total of 1456 citations after exclusion of duplicates. After a full-text review of 156 titles, we included 20 studies including 8 randomized controlled trials. The total sample included 43 487 patients undergoing ACDF, 6800 patients undergoing laminoplasty, 2075 patients undergoing CDA and 32 871 patients undergoing posterior laminectomy and fusion. Our meta-regression of synthesized follow-up data with summary curves to 150 months for each of the aforementioned procedures. Posterior laminectomy and fusion was found to have significantly higher rates of secondary reoperation. At 60 months, implant survival was 94% (95% confidence interval 80%–95%) for laminoplasty, 87% (65%–96%) for laminectomy and fusion, 93% (83 %–97%) for CDA and 92% (80%–96 %) for ACDF. At 150 months, implant survival was 90% (68%–93%) for laminoplasty, 82% (55%–92%) for laminectomy and fusion, 90% (70%–95%) for CDA and 89% (68%–93%) for ACDF.
Conclusion
This is the first study to use rigorous statistical techniques to generate statistical models for time to failure of cervical spine procedures analyzing a total of over 85 000 patients. On the basis of long-term survival data from high-quality publications, posterior laminectomy and fusion appears to have a significantly higher risk of secondary surgery when compared with ventral approaches or posterior laminoplasty. The granular survival curves produced by this study can be used to inform key decision-makers, counsel patients and aid in future trial design.
Abstract 137. Natural history of degenerative cervical myelopathy: a meta-analysis and neurologic deterioration survival curve synthesis
Background
The aim of this study was to synthesize accurate time-based estimates of meaningful neurologic decline in patients with degenerative cervical myelopathy (DCM) managed conservatively.
Methods
Following the Cochrane Prognosis Methods Guidelines, we conducted a broad systematic search of Medline, Embase and CENTRAL. We included English-language studies reporting time-to-event analysis of neurologic deterioration in adult patients with DCM treated without surgery. Title, abstract and full-text screening was done in duplicate. On the basis of the best available evidence, we utilized a change in Modified Japanese Orthopaedic Association (mJOA) score of 2 to signify meaningful neurologic decline.
Results
A total of 9570 studies were identified for title and abstract screening, ultimately yielding 6 studies for inclusion. Our final analysis included 2 level-1 prospective randomized control studies, 1 level-2 prospective cohort study, 2 level-3 retrospective cohort studies and 1 level-4 retrospective case series. Overall, 674 patients underwent conservative treatment for cervical myelopathy (range 35–247 per study) with an average mild–moderate initial modified Japanese Orthopaedic Association (mJOA) score of 14.23 (range 8.5–18) among the 5 studies where this was reported. The average reported age among the 6 included studies was 56.88 (range 28–80), with 50% males. Mean final follow-up was 24.5 months (range 5–208 mo). Our survival curves reveal a slow decline in neurologic function in a population of patients with mild–moderate mJOA scores. They show 86% survival at 1 year, 70% survival at 3 years and a slow decline to 48% survivorship with no meaningful neurologic deterioration at 20 years.
Conclusion
To our knowledge, this systematic review and meta-analysis is the first synthesis of data on the natural history of DCM. Given the paucity of high-quality evidence supporting surgical intervention in mild–moderate DCM, with the relatively slow progression shown by our pooled survival estimates, longitudinal follow-up of conservatively managed moderate DCM patients is warranted to better understand the natural history of this pathology. Ultimately, a shared patient-centred approach should incorporate a careful review of possible prognostic factors for deterioration including nonremissible factors, and informed decision-making can be guided by the use of our proposed survivorship curves.
Abstract 14. Does intraoperative vancomycin powder affect postoperative infections in adolescent idiopathic scoliosis?
Background
The routine use of intraoperative vancomycin powder to prevent postoperative wound infections has not been borne out in the literature in the pediatric spine population. The goal of this study is to determine the impact of vancomycin powder on postoperative wound infection rates and determine its potential impact on microbiology.
Methods
A retrospective analysis of 1269 adolescent idiopathic scoliosis patients in the Harms Study Group was performed. Patients who underwent a posterior fusion from 2004 to 2018 were analyzed. A comparative analysis of postoperative infection rates was done between patients who received vancomycin powder and those who did not. Statistical significance was determined using χ2 tests. Additionally, the microbiology of infected patients was examined.
Results
A total of 765 patients in the vancomycin group (VG) were compared with 504 patients in the nonvancomycin group (NVG). NVG had a significantly higher rate of deep wound infection (p < 0.001) and reoperation rate compared with VG (p < 0.001). The groups were compared for age, sex, race, weight, surgical time, blood loss, number of levels instrumented and preoperative curve magnitude. There were significant differences between the groups for race (p < 0.001), surgical time (p = 0.003) and blood loss (p = 0.002). In terms of microbiology, VG grew Propionibacterium acnes (n = 2) and serratia (n = 1), whereas NVG grew P. acnes (n = 1) and gram-positive bacilli (n = 1). The remaining cultures were negative.
Conclusion
Intraoperative vancomycin powder use in adolescent idiopathic scoliosis appears to contribute significantly to deep wound infection prevention and reduction in associated reoperations. On the basis of the limited culture data obtained, vancomycin does not seem to alter the microbiology of deep wound infections.
Abstract 37. The clinical impact of nano-surface technology on postoperative opioid consumption in patients undergoing anterior lumbar interbody fusion
Background
The aim of this study was to determine if published cellular advantages linked with nano-surface technology on interbody fusion implants are associated with clinical benefits of less opioid consumption postoperatively after anterior lumbar interbody fusion (ALIF).
Methods
This is a retrospective cohort study of consecutively treated patients undergoing ALIF. Patients received control or investigational interbody fusion implants without or with nano-surface technology, respectively. Patients were assessed at 3 postoperative follow-up visits for opioid use and total daily morphine milligram equivalents (MME) consumption with a minimum follow-up of 3 months. A logistic multivariate regression was performed to identify factors predictive of opioid use at final follow-up. Two-way repeated measures analysis of variance was used to compare the groups at each time point.
Results
A total of 120 patients were included in the analysis, with 61 patients receiving the control interbody fusion device without nano-surface technology and 59 receiving the investigational fusion device with nano-surface technology. Regression analysis demonstrated that the surface of the interbody fusion implants and smoking status were statistically significant predictors of opioid use. Patients with the control interbody implant had 4.50 times higher odds to use opioids at the 3-month follow-up compared with patients with the investigational interbody implant. Daily MME was statistically significantly higher at baseline and at each postoperative follow-up in patients with the control interbody implant compared with patients with the investigational interbody implant (13.4 mg/L v. 5.42 mg/L, p = 0.005, baseline; 46.83 mg/L v. 9.63 mg/L, p < 0.001, postoperative 1; 23.68 mg/L v. 1.46 mg/L, p < 0.001, postoperative 2; 10.92 mg/L v 1.00 mg/L, < 0.001, postoperative 3). Lower preoperative MME and absence of preoperative opioid usage were not significant predictors of reduced opioid consumption at final follow-up.
Conclusion
These results suggest that patients implanted with interbody fusion implants with nano-surface technology during ALIF may have a contributory effect on lowering the amount and duration of opioid medications in the early postoperative time period, compared with standard interbody fusion implants without nano-surface technology.
Abstract 130. Design and implementation of a comprehensive perioperative complex spine communication tool
Background
Existing care pathways for complex spine surgery include mitigation of modifiable patient factors, multidisciplinary team determination of appropriateness, and 2 attending surgeries. Opportunities for, and benefit of, effective communication and team dynamics are areas that have received less attention, despite demonstrated effectiveness in other surgical specialties. We report our early experience in developing, initiating and utilizing a comprehensive perioperative communication tool for complex spine surgeries.
Methods
A comprehensive spine-specific communication tool was developed combining existing local tools with published and validated care pathways using a Delphi process that included operating and recovery room technicians and nursing staff, anesthesiology, respiratory therapists, radiographers, neuromonitoring and cell-salvage technicians, residents, fellows and spine surgeons. Multiple plan-do-study-act (PDSA) cycles resulted in the current tool, consisting of a standardized surgical time-out in attestation format, a whiteboard to facilitate intraoperative communication and personnel handovers and a postoperative recovery room handover tool. Pre-, peri- and post-implementation surveys were distributed to all operating room personnel after each PDSA cycle. Once established, intraoperative and post-operative adverse events (using SAVES), length of stay and patient-reported outcome measures will be compared with a matched historical cohort to examine the effectiveness of the tool.
Results
Before development of the tool, poor communication, role uncertainty and lack of continuity were highlighted concerns in multiple staff surveys. Our initial pilot study consisted of 14 complex deformity surgeries, of which 3 were 2-day cases. Each case served as a separate PDSA cycle, allowing adaptation and refinement of the communication tool. Improvement in teamwork was reported by 94%, and a decrease in unnecessary interruptions and duplications was reported at a rate of 89% and 78%, respectively. Improvement in communication at handover and in overall operating room team dynamics, as well as role identification, were cited as primary benefits of the communication tool in qualitative feedback.
Conclusion
With engagement of a multidisciplinary team, a comprehensive complex spine care communication tool can be successfully implemented at a tertiary care facility, improving perioperative communication, handover efficiency, role assumption and team dynamics.
Abstract 87. Stratifying low back pain patients in an inter-professional education and self-management model of care: results of a latent class analysis
Background
Patient outcomes for low back pain (LBP) are highly variable. We aimed to identify clinically relevant “classes” among primary care patients with LBP and assess the performance of class-specific prediction for pain and disability outcomes.
Methods
Patients completed detailed intake questionnaires; clinicians used the CORE Back Tool to classify LBP patterns. Numeric pain rating scale (NPRS) and Oswestry Disability Index (ODI) scores were collected at intake and 6-month follow-up. A latent class model (LCM) was developed considering intake and outcome measures. Clinical reasoning and fit statistics were used to determine the optimal number of classes. Most likely class membership was determined by applying the above LCM. Outcome prediction models were then generated for clinically important improvement (CII; NPRS 2 pt, ODI 10 pt) and minimal pain or disability (NPRS 0–3, ODI 0–20). Performance was assessed using the c-statistic.
Results
A total of 1330 participants (58% female, mean age 53 yr) were randomly allocated to discovery (n = 667) and prediction (n = 663) samples. Four classes emerged. They varied widely in 6-month outcomes: class 1 had substantial improvement, class 2 had moderate improvement, class 3 had mild symptoms at intake and follow- up, and class 4 had persistent symptoms. The class-based prediction model had relatively good performance. Overall c-statistics were 0.72 and 0.67 for 6-month CII in NPRS and ODI, and 0.65 and 0.78 for minimal pain and disability at 6 months.
Conclusion
Patients with LBP can be grouped into 4 phenotypes that have distinct trajectories of pain and function. Class-dependent outcome prediction performed well. Identification of these 4 classes by frontline clinicians may facilitate improved patient–clinician communication about LBP prognosis, without needing more advanced point-of-care prediction tools.
Abstract 88. Cohort accuracy versus confidence at the patient level: clinical challenges for AI-based prediction of low back pain outcomes
Background
Our objective was to investigate how the performance of an artificial intelligence (AI)-based ensemble model for patient-reported low back pain (LBP) outcome varies across clinical subgroups and how this can be used to measure prediction confidence at a patient level.
Methods
Patients referred from primary care to an interprofessional LBP program were assessed (training: n = 1330, validation: n = 406). Relevant demographic, health and psychosocial factors were collected. The primary outcome was clinically important improvement (CII) in disability: 10-point improvement in Oswestry Disability Index from intake to 6-month follow-up. A set of classifiers (logistic regression, random forest; naïve Bayes, k-nearest neighbour, case-based reasoning, support vector machine) was trained and combined as an ensemble, then validated in the independent validation samples. Patients in the validation cohort were categorized on the basis of the number of classifiers with correct CII prediction: high-confidence correct, correct, weak, undecided, wrong, and high-confidence wrong by unweighted voting.
Results
Among the ensemble models considered, the best had high predictive ability (validation c-statistic of 0.834). However, performance was not uniform across patients: 38% of validation cases were correctly predicted by most classifiers (high-confidence correct or correct), 38% had conflicting or uncertain results (weak or undecided) and 23% were incorrectly predicted (high-confidence wrong or wrong). Comparing these results with those from a previous latent class model (classes: sciatica-type, younger discogenic-type; older osteoarthritis-type and persistent symptoms-type) revealed that while the ensemble classifiers tended to correctly predict outcomes in sciatica-type patients, they tended to underpredict success in discogenic- and osteoarthritis-type groups, with the majority of improved patients in the undecided, wrong and high-confidence wrong categories. In the persistent symptoms-type group, ensemble classifiers did not show a strong trend in classification accuracy.
Conclusion
Our results indicate that a model can perform well at the cohort level, while performing poorly for subsets of the target population. Before implementing AI-based prediction tools into clinical practice, subgroup performance and individual prediction confidence must be assessed and understood in combination with model explanation.
Abstract 96. Preoperative disc angle is an important predictor of segmental lordosis after degenerative spondylolisthesis fusion
Background
The aim of this study was to determine the effect of interbody cages inserted via the posterior approach on segmental lordosis in the setting of preoperative lordotic versus kyphotic/neutral disc spaces in patients with lumbar degenerative spondylolisthesis (LDS).
Methods
Five consecutive years of retrospective data from a Canadian Spine Outcomes and Research Network (CSORN) prospective study on the assessment and management of patients with LDS were collected from 2 contributing centres of consecutively enrolled patients. Patients were analyzed preoperatively and at 12-month follow-up with standing lumbar radiographs. At the spondylolisthesis level, segmental lumbar lordosis (SLL) was measured from the upper end plate of the proximal vertebra to the lower end plate of the distal vertebra. Patients were stratified into 4 groups on the basis of the index level disc angle and the type of procedure performed: preoperative lordotic posterolateral fusion (PLF) (group 1), preoperative neutral/kyphotic PLF (group 2), preoperative lordotic interbody fusion (IF) (group 3) and preoperative neutral/kyphotic IF (group 4).
Results
A total of 100 of 111 (90 %) patients completed 1-year follow-up. Twenty-three patients underwent PLF with 18 (18%) in group 1 and only 5 (5%) in group 2. Eighty-eight patients underwent IF, with 40 (40%) in group 3 and 48 in group 4 (48%). Among patients with preoperatively lordotic disc angles, group 3 patients had a greater magnitude of worsening in SLL than group 1 patients, with significant differences persisting at 1 year (mean difference 2.3°, 95% confidence interval 0.3 to 4.3, p = 0.029). Patients in group 4 were more likely to achieve improvement in SLL at 1 year than those in group 3 (67% v. 44%, p = 0.046), with similar mean improvement magnitude between groups 3 and 4 (−1.1, 95% CI −3.7 to 1.6, p = 0.42).
Conclusion
In the setting of an index-level preoperative lordotic disc angle, the magnitude of segmental lordosis worsening was more pronounced when an interbody cage was used versus PLF. Patients who have kyphotic or neutral disc space preoperatively are more likely to gain lordosis when an interbody cage is used.
Abstract 97. Preoperative depression, functional and radiographic outcomes after surgery for degenerative lumbar spondylolisthesis
Background
The objective of this study was to assess for the relationship of preoperative depression, disability and associated functional outcome after surgery for degenerative lumbar spondylolisthesis (DLS).
Methods
Retrospective analysis of the prospectively collected cohort of consecutively enrolled patients in the Canadian Spine Outcomes and Research Network (CSORN) prospective DLS database was performed. Patients enrolled between 2015 and 2020 and who were 1 year after surgery were included. Patients were divided into groups according to the presence of preoperative depression on the basis of the Patient Health Questionnaire- 9 (absent, PHQ-9 < 10 v. presence, PHQ-9 score 11–25) and preoperative severity of disability (low severity, Oswestry Disability Index [ODI] < 40 v. high severity, 41–100). Preoperative patient, clinical and radiographic factors associated with depression and disability were determined using logistic regression.
Results
One-hundred and twenty-four (34%) patients had depression and 229 (63%) patients had a high ODI score before surgery. Patients with depression and patients with high disability were more likely to be female, younger, overweight, smoking, physically inactive and living alone, to have more comorbidities and to not be working. Logistic regression revealed female sex, increasing body mass index, working status and smoking status were not associated with a worse preoperative PHQ-9. However, female sex (p = 0.007), inactivity (p = 0.02) and unemployment (p = 0.017) were associated with worse preoperative ODI. Patients with depression had a higher preoperative ODI score (39 [standard deviation (SD) 14] points v. 55 [SD 13] points). At 1 year, patients with depression continued to have a clinically meaningful higher ODI score (21 [SD 17] v. 31 [SD 19], p < 0.001) meeting a minimum clinically important difference of 10. A similar proportion of patients had PI-LL greater than 9° mismatch in the depressed (54.3% vs 57%, p = 0.73) and high-disability (44% v. 48%, p = 0.50) groups preoperatively and at 1 year (depressed 44% v. 48 %, p = 0.496; high disability: 47% v. 44%, p = 0.61).
Conclusion
Patients with low or high preoperative depressive symptom presence functionally improve after LDS surgery. However, the presence of greater preoperative depressive symptoms is associated with worse functional outcomes among these groups and represents an important avenue of preoperative optimization to consider and counsel patients undergoing LDS surgery.
Abstract 116. A CSORN study of functional outcomes after surgery for lumbar degenerative spondylolisthesis
Background
The objective of this study was to identify the radiographic parameters that correlate with functional outcomes after degenerative lumbar spondylolisthesis (DLS) surgery.
Methods
We performed a retrospective analysis of the cohort of consecutively enrolled patients in the Canadian Spine Outcomes and Research Network (CSORN) prospective study on the assessment and management of DLS. Patients underwent decompression in isolation or with posterolateral or interbody fusion. Patient-reported outcomes (PROs) were severity of leg and back pain on a 10-point numeric rating scale and the Oswestry Disability Index (ODI). Global and regional alignment was assessed via sagittal vertical axis (SVA), pelvic incidence (PI) and lumbar lordosis (LL). PROs and upright radiographs were obtained preoperatively and at 1 year postoperatively. Correlation between patient outcomes and radiographic parameters was assessed using multivariate linear regression adjusted for age, body mass index, sex and presence of preoperative depression using the Patient Health Questionnaire-9 (PHQ- 9) score.
Results
A total of 363 patients were included. Two-hundred and forty-one patients were available for analysis at 1 year. Among participants, mean age was 66 years with 63% female and a primary surgical indication of neurogenic claudication in 78% of patients. Multiple linear regression with adjustment for covariates showed that as PI-LL increased (worsened), ODI (R2 0.160, p = 0.004), back pain (R2 0.154, p < 0.01) and leg pain (R2 0.004, p = 0.014) scores also increased (worsened) at 1 year. Reduction of LL was associated with a higher ODI (R2 0.150, p = 0.027), back pain (R2 0.116, p = 0.007) and leg pain (R2 0.031, p = 0.038) score at 1 year. Increasing change in SVA score from baseline to 1 year (worsening) was associated with a higher (worse) ODI (R2 0.195, p = 0.013), back pain (R2 0.109, p = 0.037) and leg pain score (R2 0.035, p = 0.024).
Conclusion
Preoperative emphasis on regional and global spinal alignment parameters must be considered to optimize surgical procedure indication and functional outcome in lumbar degenerative spondylolisthesis treatment.
Abstract 121. A CSORN study of the effect on radiographic alignment outcomes with different surgery type for degenerative lumbar spondylolisthesis
Background
The aim of this study was to assess the effect of decompression versus posterolateral fusion versus interbody fusion on spinal alignment among patients undergoing surgery for degenerative lumbar spondylolisthesis (DLS).
Methods
Retrospective analysis of the Canadian Spine Outcomes and Research Network (CSORN) prospective study on the assessment and management of DLS was performed. Patients who had decompression alone (D), posterolateral fusion (PLF) or interbody fusion (IB) between 2015 and 2020 and were 1 year after surgery were included. Sagittal vertical axis (SVA), lumbar lordosis (LL) and pelvic incidence (PI) were measured preoperatively and 1 year postoperatively. Patients were divided into 2 groups on the basis of whether their 1-year postoperative spinal alignment improved or stayed the same or deteriorated by subtracting the 1-year measure from baseline. The proportion of patients and the magnitude of the change was compared among surgery types.
Results
Two-hundred and forty-eight patients had LL (D = 69, IB = 147, PLF = 32), 192 patients had SVA (D = 52, IB = 114, PLF = 26) and 243 patients had PI-LL (D = 71, IB = 137, PLF = 32) measures at 1 year. The majority of patients saw an improvement in alignment and the proportion of patients who improved was similar among surgery groups (SVA: 50%, 35%, 41%, p = 0.15; PI-LL: 64%, 66%, 56%, p = 0.62; LL: 62%, 71%, 63%, p = 0.39 for D, IB and PLF, respectively). Lumbar lordosis deteriorated an average of 6.1° (standard deviation [SD] 5.7°) and was similar among surgery types. Patients undergoing decompression only saw a smaller improvement in LL compared with IB (p = 0.039; D = 6.5°[SD 6.0°], IB = 9.7° [SD 7.6°], PLF = 9.1° [SD 7.0°]). Sagittal vertical axis deteriorated an average of 14 (SD 18) mm and was similar among surgery types. The PLF cohort saw the greatest improvement in SVA (48 [SD 51] mm v. D 24 [SD 39] mm, p = 0.002, v. IB 18.1 [SD 23], p = 0.003). The PI-LL deteriorated an average of 7.1 (SD 6.0) mm and improved an average of 11.1 (SD 9.2) mm and did not differ among surgery types.
Conclusion
Overall spinal alignment either remains the same or improves with the vast majority of patients undergoing surgery for DLS. In this large patient cohort, more invasive surgical intervention for DLS was not associated with a statistically significant between-group alignment improvement.
Abstract 79. Spinal cord stimulation to restore neurological function: a costing analysis
Background
Spinal cord stimulation (SCS) is used to manage pain and recently was reported to restore neurologic function following spinal cord injury (SCI). Clinical studies are underway and the objective of this study was to conduct a cost analysis to identify the types and magnitude of costs incurred in SCS therapies for individuals with SCI to facilitate reimbursement.
Methods
A literature search was conducted using Ovid Medline, Embase, the Cochrane Library, Centre for Reviews and Dissemination Databases, EBSCO CINAHL, INAHTA HTA Database and EconLit as well as a review of the grey literature. Studies involving individuals treated with SCS published in English between 2010 and September 2020 with cost information were included. Cost estimates were supplemented by key informant interviews (i.e., researchers, study participants, clinicians, industry) and clinical trial protocols to reflect therapy for SCI.
Results
The search identified 1556 full-text articles, and 71 studies were reviewed. Twelve studies (6 peer reviewed, 6 grey literature) met the inclusion criteria. Eight interviews were conducted. Estimated costs in 2019 Canadian dollars for pretreatment diagnostics was $3182. Epidural treatment costs were estimated to be $42 000–$87 000 for autonomic function (6–12 h for programming/training), $176 000–$221 000 for moderate motor function (1040 hours in first yr) and $312 000–$356 000 for enhanced motor function (2080 hours in 2 yr). For transcutaneous SCS, estimated costs are the same for pretreatment diagnostics ($3182). Treatment costs ranged from $5200 (based on 40 h of treatment) and $15 600 (based on 120 h of treatment). The associated rehabilitative therapy varies on the basis of treatment protocols and participant response. Estimates did not include the cost of adverse events requiring treatment.
Conclusion
Existing evidence and expert opinion were used to estimate the costs for epidural and transcutaneous SCS in SCI. In comparing the 2 approaches (epidural and transcutaneous) it is important to consider that they may have different treatment objectives, participants respond differently and the 2 approaches may be complementary. These costs provide a reference to inform future research that will support requests for reimbursing SCS expenses in Canada.
Abstract 107. Biomechanical properties of a novel morselized bone graft cage
Background
Posterior lumbar interbody fusion (PLIF) is performed using various interbody spacers. Wang and colleagues (2014) created an interbody cage by compressing morselized corticocancellous bone chips. They concluded that the newly morselized bone interbody fusion (MBIF) cage can provide appropriate stiffness at the physiologic loads. The purposes of this study were to replicate Wang and colleagues’ study by creating the MBIF cage and in case of failure, to redesign the MBIF cage and assess its biomechanical properties in comparison with the PEEK cage.
Methods
Lamina and spinous processes of fresh frozen spine segments were morselized and placed in a bullet-shaped mold and compressed with 8 kN force. When we redesigned the MBIF cage, the mold was lined with a thin layer of stainless-steel mesh acting as a scaffold. The redesigned MBIF (n = 6) and PEEK (n = 6) cages were place between 2 blocks of solid polyurethane foam, simulating healthy bone, and underwent axial compression while we recorded compressive force and displacement curve. The experiment was repeated with polyurethane foam simulating osteoporotic bone.
Results
The MBIF cage collapsed under axial compression. In the healthy bone group, peak force at 3-mm displacement was significantly lower in the redesigned MBIF cage compared with the PEEK cage. At 5-mm displacement, peak force did not differ significantly between the 2 cages. At lower levels of displacement, the redesigned MBIF construct failed by loss of height of the cage, while the PEEK cage construct failed by destruction of polyurethane foam contact surface. In osteoporotic bone, peak forces at 3 and 5 mm were significantly higher in the redesigned MBIF cage than in the PEEK cage constructs.
Conclusion
The results of Wang and colleagues (2014) were not reproducible in our study. The redesigned MBIF cage showed comparable biomechanical properties to those of the PEEK cage in healthy bone construct and outperformed the PEEK cage in osteoporotic bone construct. The redesigned MBIF can be a viable option instead of a synthetic cage in patients with poor bone quality.
Abstract 93. Optimizing spine surgery instrument trays to immediately increase efficiency and reduce costs in the operating room
Background
Surgical tray configurations are rarely audited, which results in an often-overlooked source of perioperative inefficiency and unnecessary costs. The challenge in determining the ideal contents of a tray lies in the balance between clinical needs, costs, ease of batching, prevention of adverse clinical events and minimization of wear and tear. In our study, we developed a novel hybrid approach (HA) based on surveys and cost analyses to reduce the number of instruments on the laminectomy tray (LT) and the basic neurosurgery tray (BNT). We compared the reduction in the number of instruments as well as cost savings from our approach to 2 approaches previously described in the literature: clinician review (CR) and mathematical optimization (MO).
Methods
Three approaches were tested: MO, CR and HA. A MO was performed via the mathematical analysis of 25 observations of the use of a LT and BNT tray. For the CR and HA approaches, we interviewed 5 neurosurgeons and 3 orthopedic surgeons who performed a total of 5437 spine cases, requiring the use of the LT and BNT, from Apr. 1, 2017, to Feb. 15, 2021, at Sunnybrook Health Sciences Centre. In the CR approach, surgeons suggested which surgical instruments should be removed. The HA approach was performed via a structured survey of the surgeon’s estimated instrument usage, followed by a cost-based inflation point analysis.
Results
The MO, CR and HA resulted in a total instrument reduction of 38%, 41% and 35%, respectively. Similarly, the total cost savings per annum were $44 417.60, $50 211.20 and $46 348.80, respectively.
Conclusion
While hospitals continue to examine perioperative services for potential inefficiencies, surgical inventory will be increasingly scrutinized. Despite MO being the most accurate methodology to do so, our results suggest that savings were similar across all 3 methods. CR and HA are significantly less laborious and thus are viable alternatives.
Abstract 103. Machine learning models can predict subsequent publication of North American Spine Society Annual General Meeting abstracts
Background
Academic meetings serve as an opportunity to present and discuss novel ideas, with manuscript publication in a peer-reviewed journal being the eventual goal of presented research. Previous studies have identified factors predictive of publication without generating predictive models. However, machine learning (ML) presents a novel tool capable of generating these models. As such, the objective of this study was to use ML models to predict subsequent publication of abstracts presented at a major spine surgery meeting.
Methods
All abstracts from the North American Spine Society (NASS) annual general meetings (AGM) from 2013–2015 were reviewed. The following information was extracted: number of authors, institution, location, conference category, subject category, study type, data collection methodology, human subject research and US Food and Drug Administration (FDA) approval. Abstracts were searched on the PubMed, Google Scholar, and Scopus databases for publication. ML models were trained to predict whether the abstract would be published or not. The quality of models was determined by using the area under the receiver operating characteristic curve (AUC). The top 10 most important factors were extracted from the most successful model.
Results
A total of 1119 abstracts were presented, with 553 (49%) abstracts published. During training, the model with the highest AUC and accuracy metrics was the partial least squares (AUC of 0.77 [standard deviation (SD) 0.05], accuracy of 75.5% [SD 4.7%]). During testing, the model with the highest AUC and accuracy was the random forest (AUC of 0.69, accuracy of 67%). The top 10 features for the random forest model were (descending order): number of authors, year, conference category, subject category, human subjects research, continent and data collection methodology.
Conclusion
This was the first study attempting to use ML to predict the publication of complete articles after abstract presentation at an academic conference. Our study used ML to identify key predictive features, combining them to create a potent predictive model. This technique can be used to improve the quality of scientific meetings while demonstrating the potential for broader applications of ML in academia and health care.
Abstract 117. The use of primary sacroiliac joint fusion for lower back pain due to sacroiliac joint pathology: a systematic review and meta-analysis
Background
Sacroiliac joint (SIJ) pathologies are well known causes of chronic low back pain; however, conservative versus operative management (CM) has been a point of contention. As such, the aim of this study was to quantify the effect of sacroiliac joint fusion (SIJF) on patient-reported outcomes in patients with chronic low back pain caused by SIJ pathology.
Methods
Databases were searched before Aug. 18, 2020, for the treatment of SIJ pathology. The primary outcome measure was the Visual Analogue Scale (VAS) for lower back pain. The secondary outcome measure was the Oswestry Disability Index (ODI). For single-cohort studies, treatment outcomes were analyzed using the pooled mean VAS and ODI scores. For studies comparing SIJF with CM, treatment outcomes were analyzed using the standardized mean difference (SMD) for VAS and ODI scores.
Results
A total of 558 patients and 6 studies were included: 2 randomized controlled trials and 4 prospective cohort studies. Five out of the 6 studies were industry funded. Mean pooled VAS scores at baseline were 76.3 (95% confidence interval [CI] 68.4 to 84.1) and at 6-month follow-up 31.0 (95% CI 27.9 to 34.0). Mean pooled ODI score at baseline was 56.7 (95% CI 53.1 to 60.2) and at 6-month follow-up was 32.4 (95% CI 30.4 to 34.4). According to VAS scores, the SMD between SIJF and CM at 6 months follow-up was −1.5 (95% CI −1.8 to −1.1). The SMD between SIJF and CM ODI scores at 6 months was −1.1 (95% CI −1.6 to −0.5).
Conclusion
SIJF showed potential as a surgical treatment option for SIJ pathology. However, owing to the paucity of independent trials with long-term follow-up, definitive conclusions about SIJF efficacy are limited. This work supports further exploration of SIJF to controls through multicentre comparative studies with long-term follow-up.
Abstract 141. How to make the most of your operative time by optimizing surgical scheduling
Background
Hospital cost containment efforts are increasingly restricting operative time. As such, attempting to maximize the efficiency of limited operative time is critical. Unfortunately, scheduling is predominantly performed manually, which requires reasoning about unknowns such as case-specific length of surgery and variability while attempting to maximize throughput. This often leads to operating room (OR) overtime and suboptimal throughput rate. In contrast, scheduling using mathematical optimization methods can produce maximum system efficiency, and it is extensively used in the business world. As such, the purpose of our study was to compare the efficiency of (1) manual and (2) optimized OR scheduling at a major Canadian academic hospital.
Methods
OR data were collected from 2016 2021. The scheduling, surgical duration, overtime and number of OR days were extracted, with the data configured to represent the historic manual scheduling process. This was then inputted into an integer linear programming model with the goal of determining the minimum number of OR days to complete the same number of cases while not exceeding the historic overtime values.
Results
All surgeons saw a median 12% (range 10.2%–13.1%) reduction in the number of OR days needed to complete their annual case-load compared with their historical scheduling practices. Meanwhile, the OR overtime varied by a maximum of 5%. The daily OR configurations differed from historic configurations in 67% of cases. In addition, the number of configurations per surgeon was reduced from an average of 5 to 4.
Conclusion
Our results demonstrate a significant increase in OR throughput rate (12%) with no change in operative time required. This has considerable implications in terms of cost reduction, surgical wait lists and surgeon satisfaction. The results of our model are also easily interpretable as a guideline for surgeons to book elective cases to maximize utilization of their OR time. Further optimization may be achieved by taking into account factors that could predict case duration such as surgeon experience, patient characteristics and institutional attributes via machine learning.
Abstract 126. Altering physician referral practices remains a challenge: a spine assessment clinic quality improvement study
Background
Inappropriate referrals contribute to delayed access to specialists and surgical wait lists. Referral guidelines and physician education have been shown to decrease such clinical inefficiencies. The goal of this study was to reduce the volume of inappropriate referrals to the Royal Alexandra Hospital Spine Assessment Clinic (SAC) via implementation of a quality improvement (QI) initiative. Nurse practitioners (NPs) triage incoming referrals via phone and decide surgical appropriateness on the basis of this screening. We hypothesized that we could improve surgical referral accuracy by 25% following guideline distribution.
Methods
A 3-phase QI methodology was implemented. In phase 1, baseline data were collected from pre-intervention referrals by noting the physician’s reason for referral and if certain information (i.e. inciting events, clinical diagnosis and physical examination findings) deemed relevant for an appropriate referral were included. In phase 2, a referral guideline, outlining when and how to refer, was developed and distributed via Alberta Health Services and primary care networks to family physicians in Northern Alberta on 2 occasions (Mar. 1 and May 18). In phase 3, Postintervention referrals and NP patient assessments were collected as in phase 1 and described using quantitative statistics and the Fisher exact test (p = 0.05).
Results
A total of 406 NP patient assessments (162 pre-intervention and 244 post-intervention) were collected. The proportion of referred patients who were deemed appropriate surgical consultations was 22.2% pre-intervention, with this number increasing to 26.8% in the first 3 months post-intervention (p = 0.18). During the following 3-month period this value further increased to 36.3%, significantly improved relative to pre-intervention (p = 0.039).
Conclusion
We found a positive trend in the proportion of patients deemed surgical candidates post-intervention. This study demonstrates the challenge associated with changing physician behaviour. Translating guidelines into practice is a recognized issue; it requires time and multiple exposures. Active forms of medical education and multifaceted interventions have been demonstrated to be the most effective means of implementing guidelines into practice, an approach that could further address SAC referral inadequacies.
Abstract 152. Outcomes of workers’ compensation patients undergoing neuromodulation for persistent neuropathic pain conditions
Background
Workers’ compensation patients have long been identified as having worse outcomes with interventional and surgical therapies. Patients who develop persistent neuropathic pain conditions may benefit from spinal cord stimulation or dorsal root ganglion stimulation, which has been shown to be more beneficial than conventional medical management. The efficacy of these modalities in workers’ compensation patients has been questioned, and it remains to be seen how this population responds to neuromodulation in various functional domains. The St. Paul’s Hospital Neuromodulation Program is one of the largest neuromodulation programs in Canada, and it sees a large volume of workers’ compensation patients with persistent neuropathic pain conditions.
Methods
We performed a prospective cohort study administering the PROMIS-29 questionnaire to workers’ compensation patients, before and after spinal cord stimulation and dorsal root ganglion implant. The PROMIS-29 questionnaire measures 7 health-related quality of life domains (physical functioning, anxiety, depression, fatigue, sleep disturbance, social functioning, pain intensity and interference). Questionnaires were administered 1 week pre-trial and 8 weeks post-implant.
Results
Five patients were followed (3 for SCS, 2 for DRG stimulation), and PROMIS-29 scores were obtained 1 week pre-trial and 8 weeks post-implant. There was a statistically significant improvement in all PROMIS-29 domains, with the greatest improvements noted in sleep disturbance, social functioning and pain interference.
Conclusion
In this small prospective cohort study, workers’ compensation patients demonstrated a significant improvement in all PROMIS-29 domains, suggesting that appropriately selected patients show improvements in multiple domains when offered neuromodulation therapies. Following these patients longitudinally will help determine if these therapies improve return to work.
Abstract 90. Expectations of treatment outcomes in patients with spinal metastases: What do we tell our patients? A qualitative study
Background
Realistic pretreatment expectations are important and have been associated with post-treatment health-related quality of life (HRQOL). Physicians are the primary resource for information for patients. Patient expectations regarding treatment outcomes are therefore greatly influenced by physicians. This study aimed to explore the communication practices of physicians regarding treatment outcomes for patients with spinal metastases, and physician experiences with patients’ pretreatment expectations.
Methods
An international qualitative study using semistructured interviews including physicians routinely involved in the treatment of patients with metastatic spine disease (spine surgeons, radiation and medical oncologists, and rehabilitation specialists) was conducted. Physicians were interviewed about the content and extent of information they provide to patients with spinal metastases regarding treatment options, risks and treatment outcomes. Interviews were transcribed verbatim and analyzed using a thematic coding network.
Results
After 22 interviews, data saturation occurred. The majority of the physicians indicated that they currently do not establish patients’ pretreatment expectations, despite acknowledging the importance of these expectations. Spine surgeons often believe that patient expectations are disproportionate. Physicians expressed that they currently manage expectations by detailing the most common risks and providing broad but nonspecific information on treatment outcomes. While physicians believe that patients know treatment is palliative, their perception is that the implications of a palliative treatment intent remain elusive to most patients.
Conclusion
This study highlights the current gap in patient–physician communication regarding the expectations of treatment outcomes of patients with spinal metastases. These results warrant further research to improve communication practices and determine the effect of patient expectations on patient-reported outcomes in this population.
Abstract 52. Fluoroscopically guided radiofrequency ablation of the superior cluneal nerve: preliminary outcomes data for a minimally invasive approach for treating superior cluneal neuralgia
Background
Superior cluneal neuralgia (SCN) has been termed pseudo-sciatica and can be an overlooked differential diagnosis when patients are otherwise presenting with low back and radicular pain. Radiofrequency ablation (RFA) is commonly used for denervation of the medial branches of the dorsal rami for facetogenic or sacroiliac arthralgia. RFA outcomes for SCN have not previously been described. Hence, the objective herein was to provide outcomes data from our centre for SCN RFA.
Methods
A research ethics board–approved retrospective chart review was used to collect data for all SCN RFA cases from Jan. 1, 2018, to Feb. 8, 2021. Fluoroscopically guided SCN ablations were performed for patients after diagnosis by physical examination with a positive Maigne sign reproductive of their back and gluteal/leg pain. Sensory stimulation was utilized to confirm RF cannula-probe placement adjacent to the SCN (paresthesia-pain overlay), and motor testing was used to confirm no distal motor response before monopolar RF ablation with a Halyard RF Generator (100 mm curved 20 G 10 mm active tip RF cannulae). Charts were reviewed for duration to follow-up, duration and degree of analgesia, improvements in patients’ activity and gait, and complications, as well as changes in analgesic medication utilization.
Results
A total of 51 patients underwent SCN RFA, 5 of whom were lost to follow up. Of the remaining 46 patients (29 women, 17 men, mean age 59.4 yr), 78.3% of patients had ongoing analgesia at a mean of 92.1 days follow-up (range 13–308 d), with a mean of 92.3% analgesia (standard deviation 15.0%). A total of 21.7% of patients reported return of pain at follow-up (mean of 111.2 d [range 42–201 d] of 95% [SD 6.7 %] analgesia). A total of 41.3% (n = 19) of patients reported or were observed with improved activity and gait, 17.4% reported improved mood (n = 8) and 8.7% reported decreased medication use (n = 4). Five patients had minor complications including bruising, site soreness, or myofascial pain or spasm, and 1 required trigger point injections.
Conclusion
RFA is a suitable intervention for the treatment of SCN. One hundred percent of cases had analgesia after their procedure (mean > 90 d with > 90% analgesia). Further prospective investigation is recommended.
Abstract 21. Single-stage posterior approach for en bloc resection and spinal reconstruction of T4 Pancoast tumour invading spine
Background
The aim of this study was to evaluate the outcomes of patients with T4 Pancoast tumors invading the spine who underwent en bloc resection and spinal stabilization through a single-stage posterior approach.
Methods
All patients with T4N0M0 Pancoast tumours invading the spine who underwent a single-stage posterior approach were retrospectively recruited. Data collected consisted of demographics, tumour histology, preoperative and postoperative therapy, complications, spinal reconstruction technique, extent of tumour resection, survival time and recurrence of disease.
Results
Eighteen patients were included. The mean population age was 61 (standard deviation [SD] 17) years old. The most common pathologic type was adenocarcinoma (61.1%). Complete resection (R0) was obtained in 15 (83.3%) patients, and positive surgical margins (R1) were found in the remaining 3 (16.7%) patients. The 90-day mortality rate was 0%. Twelve (66.7%) patients had postoperative major complications, all of whom required reoperation. The mean survival time was 67 (SD 24) months and the median survival time was not reached (10 patients [55.6%] still alive at the end of the study). The actual 2-year and 5-year survival rates were 59% (95% confidence interval [CI] 35.7%–82.3%) and 52.5% (95% CI 28.4%–76.6%), respectively.
Conclusion
En bloc resection and spinal stabilization through a single-stage posterior approach might be effective for T4 Pancoast tumours invading the spine.
Abstract 34. Predictors of sacral ulcers in patients with complete spinal cord injury
Background
Despite current best practices, sacral ulcers (SU) remain a devastating hospital-acquired complication for patients with acute traumatic spinal cord injuries (SCIs). The objective of this study is to examine associations between norepinephrine dose and duration and development of SUs in patients with complete cervical and thoracic spinal cord injury patients (American Spinal Injury Association [ASIA] classification A) as well as to identify other demographic risk factors for SU development in this patient group.
Methods
This retrospective study included adults with acute complete SCIs who were admitted to a level-1 trauma centre between 2014 and 2018. Data on patient and injury characteristics were recorded, including age, sex, location of SCI (cervical v. thoracic), length of stay (LOS) and mortality; presence or absence of SU during their acute hospital stay; as well as treatment factors such as spinal surgery, mean arterial pressure (MAP) targets and vasopressor treatment. Potential associations with SU were examined using multivariable logistic regression analysis.
Results
Eighty-two patients met the inclusion criteria and had complete data. Of these patients, 30 (37%) had developed SUs. Patient and injury characteristics, including age (mean 50.6 yr [standard deviation (SD) 21.3]), location of SCI (48 cervical, 59%) and Injury Severity Score (mean 33.1 [SD 11.8]), did not differ between the SU and non-SU groups. Results of the multivariable logistic regression analysis revealed that male sex (odds ratio [OR] 34.1, 95% confidence interval [CI] 2.3–506.5, p = 0.010) and increased LOS (log-transformed; OR 20.5, 95% CI 2.8–149.9, p = 0.003) were associated with increased risk of SU. Having an order for a MAP greater than 80 mmHg (OR 0.05, 95% CI 0.01–0.30, p = 0.001) was associated with a reduced risk of SU. There were no significant associations between SU and the log10 mean rate or log10 duration of norepinephrine treatment.
Conclusion
Norepinephrine treatment parameters were not associated with development of SU. This study provides evidence of the association between hospital LOS and risk of unavoidable pressure ulcers in SCI patients. Unavoidable pressure ulcers are defined as those that develop in spite of consistent documentation of evidence-based preventive interventions. Therefore, efforts to reduce LOS are recommended. This requires a systematic approach that involves assessing an organization’s readiness for change.
Abstract 135. Targeting geographic wait time disparities in Canada: a rapid review of domestic and international strategies to reduce orthopedic wait times in the Maritimes
Background
There is considerable geographic inequity regarding the average time a patient is made to wait, from referral to consultation to treatment, across provinces, particularly for elective spine procedures. This review aims to analyze current wait time mitigation strategies within Canada and high-performing European jurisdictions to provide evidencebased recommendations for improving the timely access of orthopedic care in the Maritimes.
Methods
Primary literature published between January 2010 to May 2021 was considered. Four medical databases were utilized: CINAHL, Cochrane Library, Medline and PubMed. Screening was conducted in 2 stages. Wait time reduction strategies were analyzed using the supply-demand-performance management framework.
Results
After we removed duplicates, 678 studies entered title and abstract review, from which 88 full-text studies were assessed for eligibility, and 27 articles were included. About half of the initiatives referenced focused on increasing the availability of services to address waitlists for orthopedic surgery through funding initiatives and optimizing patient pathways in the surgical journey. Other initiatives focused on reducing the demand for elective orthopedic surgery through improving surgical triage and patient education and prevention methods. In the Maritimes, wait time issues can be categorized as concerns pertaining to (1) a lack of human resources, (2) system inefficiencies, (3) a changing health care landscape and (4) inequitable hospital funding structure. Thus, the following strategies were recommended: (1) specialized triage clinics for appropriate referral screening, (2) prevention initiatives targeting chronic disease factors and (3) local performance wait-time targets (particularly for spine procedures).
Conclusion
Overall, wait-time mitigation strategies should take a contextualized approach to encourage uptake and sustained change. Evidence on the relative efficacy of these initiatives at reducing wait times or their effects on quality of life, safety outcomes and economic cost is limited. Unlike hip and knee procedures, there is currently no established national wait-time benchmark for elective spine procedures from which provincial targets for improvements can be reviewed or incented. Developing such a benchmark can encourage transparency and accountability, especially if federal and provincial governments want to collaborate to combat the substantial surgical backlog.
Abstract 143. The influence of coronal plane parameters on patient-reported outcome measures in patients undergoing decompression for lumbar spinal stenosis
Background
We aim to determine the effect of degenerative scoliosis on clinical outcomes in patients undergoing decompression surgery for lumbar spinal stenosis. We hypothesize that patients with associated coronal plane deformities of the spine tend to have poorer outcomes than those without.
Methods
This was a retrospective chart analysis of prospectively gathered data in patients who underwent lumbar decompression. Upright AP lumbar or 3-foot standing radiographs were analyzed. Coronal Cobb angles were measured in thoracic and lumbar spine. Patient-reported outcomes (PROMs) include Oswestry Disability Index (ODI), SF-12 mental and physical component scores (MCS, PCS) and visual analogue scales for back and leg pain (VAS-B, VAS-L). Baseline measurements were compared with outcomes at 6 weeks and 1 year.
Results
Scoliosis in our study was defined as any Cobb angle more than 10°. A total of 113 patients were included in preliminary analyses. A total of 60.2% (68/113) of patients had scoliosis. Baseline demographics were comparable in patients with and without scoliosis. Mean baseline measurements in the nonscoliosis group were ODI 42.1, SF12-MCS 48.3, PCS 34.1, VAS-B 6.7 and VAS-L 6.9. Mean baseline measurements in the scoliosis group were ODI 36.6, SF12-MCS 51.9, PCS 33.3, VAS-B 5.1 and VAS-L 5.9. At 6 weeks follow-up in the nonscoliosis group, mean measurement were ODI 26.2, SF12-MCS 49.6, SF12-PCS 36.8, VAS-B 3.5 and VAS-L 3.2. Measurements in the scoliosis group were ODI 26.9, SF12-MCS 50.8, PCS 37.2, VAS-B 2.9 and VAS-L 2.9. At 1 year follow-up in the nonscoliosis group, mean measurements were: ODI 24.6, SF12-MCS 51.3, SF12-PCS 40.5, VAS-B 3.3 and VAS-L 2.8. Measurements in the scoliosis group were ODI 23.9, SF12-MCS 51.6, PCS 42.4, VAS-B 3.3 and VAS-L 2.7.
Conclusion
Preliminary analysis did not demonstrate that the presence of lumbar scoliosis influenced PROMs at all time points. Additional patients will be added to the analysis to permit subgroup analyses, which include subdividing Cobb measurements into 3 categories (10–20, 21–30, >31), effect of C7 plumb line (C7PL) to central sacral vertical line (CSVL) distance, and presence and degree of lateral listhesis.
Footnotes
The Canadian Spine Society is a collaborative organization of spine surgeons advancing excellence in research, education and patient care.
Accreditation: This event is an Accredited Group Learning Activity (Section 1) as defined by the Maintenance of Certification Program of the Royal College of Physicians and Surgeons of Canada, approved by The Canadian Orthopaedic Association.
Course Background: The 2022 Annual Virtual Scientific Conference of the Canadian Spine Society will be held in conjunction with the Canadian Paediatric Spine Society. Every year the Canadian Spine Society in conjunction with the Canadian Paediatric Spine Society holds its Annual Scientific Conference. This year, because of the COVID-19 pandemic, our meeting will be totally virtual. The format will be a distributed meeting, hosted on the Canadian Spine Society website, with two-and-a-half-hour sessions on Apr. 10, 13, 20 and 27, 2022. Each event will include live paper presentations followed by an online question period. Presentations include cause, clinical presentations and current surgical and nonoperative treatment. The first of three symposia will focus on managing the psychological and medicolegal consequences of surgical complications. Another symposium will investigate the present state of neuromodulation for pain control. The final symposium will look at the value of spine surgery with particular reference to oncology and deformity as well as its overall impact on health care. The conference will host debates on the optimal surgery for spinal deformity in the elderly, whether a black disc on magnetic resonance imaging is justification for operative intervention and the place for early discectomy versus routine care. The COVID pandemic has heightened awareness of the gap between virtual and in-person interactions with both patients and colleagues. Posters and podium presentations will address the pros and cons and offer suggestions for future engagement. Timely access to care remains a concern and the conference will advance constructive solutions. The program offers ample opportunity for professional contact, sharing ideas and problems. Although our Annual Scientific Conference must be a totally virtual experience, it remains the most important spine meeting in Canada.
Disclosure of competing interests: Available for all speakers who presented abstracts at the 22nd Annual Virtual Scientific Conference of the Canadian Spine Society. To view, please contact Jennifer Edwards, Director of Operations, Canadian Spine Society, at css{at}spinecanada.ca.
This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) licence, which permits use, distribution and reproduction in any medium, provided that the original publication is properly cited, the use is noncommercial (i.e., research or educational use), and no modifications or adaptations are made. See: https://creativecommons.org/licenses/by-nc-nd/4.0/