Machine learning-based preoperative predictive analytics for lumbar spinal stenosis.

OBJECTIVEPatient-reported outcome measures (PROMs) following decompression surgery for lumbar spinal stenosis (LSS) demonstrate considerable heterogeneity. Individualized prediction tools can provide valuable insights for shared decision-making. The authors aim to evaluate the feasibility of predicting short- and long-term PROMs, reoperations, and perioperative parameters by machine learning (ML) methods.METHODSData were derived from a prospective registry. All patients had undergone single- or multilevel mini-open facet-sparing decompression for LSS. The prediction models were trained using various ML-based algorithms to predict the endpoints of interest. Models were selected by area under the receiver operating characteristic curve (AUC). The endpoints were dichotomized by minimum clinically important difference (MCID) and included 6-week and 12-month numeric rating scales for back pain (NRS-BP) and leg pain (NRS-LP) severity and the Oswestry Disability Index (ODI), as well as prolonged surgery (> 45 minutes), extended length of hospital stay (> 28 hours), and reoperations.RESULTSA total of 635 patients were included. The average age was 62 ± 10 years, and 333 patients (52%) were male. At 6 weeks, MCID was seen in 63%, 76%, and 61% of patients for ODI, NRS-LP, and NRS-BP, respectively. At internal validation, the models predicted MCID in these variables with accuracies of 69%, 76%, and 85%, and with AUCs of 0.75, 0.79, and 0.92. At 12 months, 66%, 63%, and 51% of patients reported MCID; the observed accuracies were 62%, 74%, and 66%, with AUCs of 0.68, 0.72, and 0.79. Reoperations occurred in 60 patients (9.5%), of which 27 (4.3%) occurred at the index level. Overall and index-level reoperations were predicted with 69% and 63% accuracy, respectively, and with AUCs of 0.66 and 0.61. In 15%, a length of surgery greater than 45 minutes was observed and predicted with 78% accuracy and AUC of 0.54. Only 15% of patients were admitted to the hospital for longer than 28 hours. The developed ML-based model enabled prediction of extended hospital stay with an accuracy of 77% and AUC of 0.58.CONCLUSIONSPreoperative prediction of a range of clinically relevant endpoints in decompression surgery for LSS using ML is feasible, and may enable enhanced informed patient consent and personalized shared decision-making. Access to individualized preoperative predictive analytics for outcome and treatment risks may represent a further step in the evolution of surgical care for patients with LSS.

Objective functional assessment using the "Timed Up and Go" test in patients with lumbar spinal stenosis.

OBJECTIVEPatient-reported outcome measures (PROMs) are standard of care for the assessment of functional impairment. Subjective outcome measures are increasingly complemented by objective ones, such as the "Timed Up and Go" (TUG) test. Currently, only a few studies report pre- and postoperative TUG test assessments in patients with lumbar spinal stenosis (LSS).METHODSA prospective two-center database was reviewed to identify patients with LSS who underwent lumbar decompression with or without fusion. The subjective functional status was estimated using PROMs for pain (visual analog scale [VAS]), disability (Roland-Morris Disability Index [RMDI] and Oswestry Disability Index [ODI]), and health-related quality of life (HRQoL; 12-Item Short-Form Physical Component Summary [SF-12 PCS] and the EQ-5D) preoperatively, as well as on postoperative day 3 (D3) and week 6 (W6). Objective functional impairment (OFI) was measured using age- and sex-standardized TUG test results.RESULTSSixty-four patients (n = 32 [50%] male, mean age 66.8 ± 11.7 years) were included. Preoperatively, they reported a mean VAS back pain score of 4.1 ± 2.7, VAS leg pain score of 5.4 ± 2.7, RMDI of 10.4 ± 5.3, ODI of 41.9 ± 16.2, SF-12 PCS score of 32.7 ± 8.3, and an EQ-5D index of 0.517 ± 0.226. The preoperative rates of severe, moderate, and mild OFI were 4.7% (n = 3), 12.5% (n = 8), and 7.8% (n = 5), respectively, and the mean OFI T-score was 116.3 ± 23.7. At W6, 60 (93.8%) of 64 patients had a TUG test result within the normal population range (no OFI); 3 patients (4.7%) had mild and 1 patient (1.6%) severe OFI. The mean W6 OFI T-score was significantly decreased (103.1 ± 13.6; p < 0.001). Correspondingly, the PROMs showed a decrease in subjective VAS back pain (1.6 ± 1.7, p < 0.001) and leg pain (1.0 ± 1.8, p < 0.001) scores, disability (RMDI 5.3 ± 4.7, p < 0.001; ODI 21.3 ± 16.1, p < 0.001), and increase in HRQoL (SF-12 PCS 40.1 ± 8.3, p < 0.001; EQ-5D 0.737 ± 0.192, p < 0.001) at W6. The W6 responder status (clinically meaningful improvement) ranged between 81.3% (VAS leg pain) and 29.7% (EQ-5D index) of patients.CONCLUSIONSThe TUG test is a quick and easily applicable tool that reliably measures OFI in patients with LSS. Objective tests incorporating longer walking time should be considered if OFI is suspected but fails to be proven by the TUG test, taking into account that neurogenic claudication may not clinically manifest during the brief TUG examination. Objective tests do not replace the subjective PROM-based assessment, but add valuable information to a comprehensive patient evaluation.

Cost-effectiveness of adult spinal deformity surgery in a military healthcare system.

OBJECTIVEAdult spinal deformity surgery is an effective way of treating pain and disability, but little research has been done to evaluate the costs associated with changes in health outcome measures. This study determined the change in quality-adjusted life years (QALYs) and the cost per QALY in patients undergoing spinal deformity surgery in the unique environment of a military healthcare system (MHS).METHODSPatients were enrolled between 2011 and 2017. Patients were eligible to participate if they were undergoing a thoracolumbar spinal fusion spanning more than 6 levels to treat an underlying deformity. Patients completed the 36-Item Short Form Health Survey (SF-36) prior to surgery and 6 and 12 months after surgery. The authors used paired t-tests to compare SF-36 Physical Component Summary (PCS) scores between baseline and postsurgery. To estimate the cost per QALY of complex spine surgery in this population, the authors extended the change in health-related quality of life (HRQOL) between baseline and follow-up over 5 years. Data on the cost of surgery were obtained from the MHS and include all facility and physician costs.RESULTSHRQOL and surgical data were available for 49 of 91 eligible patients. Thirty-one patients met additional criteria allowing for cost-effectiveness analysis. Over 12 months, patients demonstrated significant improvement (p < 0.01) in SF-36 PCS scores. A majority of patients met the minimum clinically important difference (MCID; 83.7%) and substantive clinical benefit threshold (SCBT; 83.7%). The average change in QALY was an increase of 0.08. Extended across 5 years, including the 3.5% discounting per year, study participants increased their QALYs by 0.39, resulting in an average cost per QALY of $181,649.20. Nineteen percent of patients met the < $100,000/QALY threshold with half of the patients meeting the < $100,000/QALY mark by 10 years. A sensitivity analysis showed that patients who scored below 60 on their preoperative SF-36 PCS had an average increase in QALYs of 0.10 per year or 0.47 over 5 years.CONCLUSIONSWith a 5-year extended analysis, patients who receive spinal deformity surgery in the MHS increased their QALYs by 0.39, with 19% of patients meeting the $100,000/QALY threshold. The majority of patients met the threshold for MCID and SCBT at 1 year postoperatively. Consideration of preoperative functional status (SF-36 PCS score < 60) may be an important factor in determining which patients benefit the most from spinal deformity surgery.

Defining the minimum clinically important difference for grade I degenerative lumbar spondylolisthesis: insights from the Quality Outcomes Database.

OBJECTIVE Patient-reported outcomes (PROs) play a pivotal role in defining the value of surgical interventions for spinal disease. The concept of minimum clinically important difference (MCID) is considered the new standard for determining the effectiveness of a given treatment and describing patient satisfaction in response to that treatment. The purpose of this study was to determine the MCID associated with surgical treatment for degenerative lumbar spondylolisthesis. METHODS The authors queried the Quality Outcomes Database registry from July 2014 through December 2015 for patients who underwent posterior lumbar surgery for grade I degenerative spondylolisthesis. Recorded PROs included scores on the Oswestry Disability Index (ODI), EQ-5D, and numeric rating scale (NRS) for leg pain (NRS-LP) and back pain (NRS-BP). Anchor-based (using the North American Spine Society satisfaction scale) and distribution-based (half a standard deviation, small Cohen's effect size, standard error of measurement, and minimum detectable change [MDC]) methods were used to calculate the MCID for each PRO. RESULTS A total of 441 patients (80 who underwent laminectomies alone and 361 who underwent fusion procedures) from 11 participating sites were included in the analysis. The changes in functional outcome scores between baseline and the 1-year postoperative evaluation were as follows: 23.5 ± 17.4 points for ODI, 0.24 ± 0.23 for EQ-5D, 4.1 ± 3.5 for NRS-LP, and 3.7 ± 3.2 for NRS-BP. The different calculation methods generated a range of MCID values for each PRO: 3.3-26.5 points for ODI, 0.04-0.3 points for EQ-5D, 0.6-4.5 points for NRS-LP, and 0.5-4.2 points for NRS-BP. The MDC approach appeared to be the most appropriate for calculating MCID because it provided a threshold greater than the measurement error and was closest to the average change difference between the satisfied and not-satisfied patients. On subgroup analysis, the MCID thresholds for laminectomy-alone patients were comparable to those for the patients who underwent arthrodesis as well as for the entire cohort. CONCLUSIONS The MCID for PROs was highly variable depending on the calculation technique. The MDC seems to be a statistically and clinically sound method for defining the appropriate MCID value for patients with grade I degenerative lumbar spondylolisthesis. Based on this method, the MCID values are 14.3 points for ODI, 0.2 points for EQ-5D, 1.7 points for NRS-LP, and 1.6 points for NRS-BP.

Variability in the utility of predictive models in predicting patient-reported outcomes following spine surgery for degenerative conditions: a systematic review.

OBJECTIVEThere is increasing emphasis on patient-reported outcomes (PROs) to quantitatively evaluate quality outcomes from degenerative spine surgery. However, accurate prediction of PROs is challenging due to heterogeneity in outcome measures, patient characteristics, treatment characteristics, and methodological characteristics. The purpose of this study was to evaluate the current landscape of independently validated predictive models for PROs in elective degenerative spinal surgery with respect to study design and model generation, training, accuracy, reliability, variance, and utility.METHODSThe authors analyzed the current predictive models in PROs by performing a search of the PubMed and Ovid databases using PRISMA guidelines and a PICOS (participants, intervention, comparison, outcomes, study design) model. They assessed the common outcomes and variables used across models as well as the study design and internal validation methods.RESULTSA total of 7 articles met the inclusion criteria, including a total of 17 validated predictive models of PROs after adult degenerative spine surgery. National registry databases were used in 4 of the studies. Validation cohorts were used in 2 studies for model verification and 5 studies used other methods, including random sample bootstrapping techniques. Reported c-index values ranged from 0.47 to 0.79. Two studies report the area under the curve (0.71-0.83) and one reports a misclassification rate (9.9%). Several positive predictors, including high baseline pain intensity and disability, demonstrated high likelihood of favorable PROs.CONCLUSIONSA limited but effective cohort of validated predictive models of spine surgical outcomes had proven good predictability for PROs. Instruments with predictive accuracy can enhance shared decision-making, improve rehabilitation, and inform best practices in the setting of heterogeneous patient characteristics and surgical factors.

Revisions for screw malposition and clinical outcomes after robot-guided lumbar fusion for spondylolisthesis.

OBJECTIVE The accuracy of robot-guided pedicle screw placement has been proven to be high, but little is known about the impact of such guidance on clinical outcomes such as the rate of revision surgeries for screw malposition. In addition, there are very few data about the impact of robot-guided fusion on patient-reported outcomes (PROs). Thus, the clinical benefit for the patient is unclear. In this study, the authors analyzed revision rates for screw malposition and changes in PROs following minimally invasive robot-guided pedicle screw fixation. METHODS A retrospective cohort study of patients who had undergone minimally invasive posterior lumbar interbody fusion (MI-PLIF) or minimally invasive transforaminal lumbar interbody fusion was performed. Patients were followed up clinically at 6 weeks, 12 months, and 24 months after treatment and by mailed questionnaire in March 2016 as a final follow-up. Visual analog scale (VAS) scores for back and leg pain severity, Oswestry Disability Index (ODI), screw revisions, and socio-demographic factors were analyzed. A literature review was performed, comparing the incidence of intraoperative screw revisions and revision surgery for screw malposition in robot-guided, navigated, and freehand fusion procedures. RESULTS Seventy-two patients fit the study inclusion criteria and had a mean follow up of 32 ± 17 months. No screws had to be revised intraoperatively, and no revision surgery for screw malposition was needed. In the literature review, the authors found a higher rate of intraoperative screw revisions in the navigated pool than in the robot-guided pool (p < 0.001, OR 9.7). Additionally, a higher incidence of revision surgery for screw malposition was observed for freehand procedures than for the robot-guided procedures (p < 0.001, OR 8.1). The VAS score for back pain improved significantly from 66.9 ± 25.0 preoperatively to 30.1 ± 26.8 at the final follow-up, as did the VAS score for leg pain (from 70.6 ± 22.8 to 24.3 ± 28.3) and ODI (from 43.4 ± 18.3 to 16.2 ± 16.7; all p < 0.001). Undergoing PLIF, a high body mass index, smoking status, and a preoperative ability to work were identified as predictors of a reduction in back pain. Length of hospital stay was 2.4 ± 1.1 days and operating time was 161 ± 50 minutes. Ability to work increased from 38.9% to 78.2% of patients (p < 0.001) at the final follow-up, and 89.1% of patients indicated they would choose to undergo the same treatment again. CONCLUSIONS In adults with low-grade spondylolisthesis, the data demonstrated a benefit in using robotic guidance to reduce the rate of revision surgery for screw malposition as compared with other techniques of pedicle screw insertion described in peer-reviewed publications. Larger comparative studies are required to assess differences in PROs following a minimally invasive approach in spinal fusion surgeries compared with other techniques.

Potential of predictive computer models for preoperative patient selection to enhance overall quality-adjusted life years gained at 2-year follow-up: a simulation in 234 patients with adult spinal deformity.

OBJECTIVE Patients with adult spinal deformity (ASD) experience significant quality of life improvements after surgery. Treatment, however, is expensive and complication rates are high. Predictive analytics has the potential to use many variables to make accurate predictions in large data sets. A validated minimum clinically important difference (MCID) model has the potential to assist in patient selection, thereby improving outcomes and, potentially, cost-effectiveness. METHODS The present study was a retrospective analysis of a multiinstitutional database of patients with ASD. Inclusion criteria were as follows: age ≥ 18 years, radiographic evidence of ASD, 2-year follow-up, and preoperative Oswestry Disability Index (ODI) > 15. Forty-six variables were used for model training: demographic data, radiographic parameters, surgical variables, and results on the health-related quality of life questionnaire. Patients were grouped as reaching a 2-year ODI MCID (+MCID) or not (-MCID). An ensemble of 5 different bootstrapped decision trees was constructed using the C5.0 algorithm. Internal validation was performed via 70:30 data split for training/testing. Model accuracy and area under the curve (AUC) were calculated. The mean quality-adjusted life years (QALYs) and QALYs gained at 2 years were calculated and discounted at 3.5% per year. The QALYs were compared between patients in the +MCID and -MCID groups. RESULTS A total of 234 patients met inclusion criteria (+MCID 129, -MCID 105). Sixty-nine patients (29.5%) were included for model testing. Predicted versus actual results were 50 versus 40 for +MCID and 19 versus 29 for -MCID (i.e., 10 patients were misclassified). Model accuracy was 85.5%, with 0.96 AUC. Predicted results showed that patients in the +MCID group had significantly greater 2-year mean QALYs (p = 0.0057) and QALYs gained (p = 0.0002). CONCLUSIONS A successful model with 85.5% accuracy and 0.96 AUC was constructed to predict which patients would reach ODI MCID. The patients in the +MCID group had significantly higher mean 2-year QALYs and QALYs gained. This study provides proof of concept for using predictive modeling techniques to optimize patient selection in complex spine surgery.

Predicting the minimum clinically important difference in patients undergoing surgery for the treatment of degenerative cervical myelopathy.

OBJECTIVE The minimum clinically important difference (MCID) is defined as the minimum change in a measurement that a patient would identify as beneficial. Before undergoing surgery, patients are likely to inquire about the ultimate goals of the operation and of their chances of experiencing meaningful improvements. The objective of this study was to define significant predictors of achieving an MCID on the modified Japanese Orthopaedic Association (mJOA) scale at 2 years following surgery for the treatment of degenerative cervical myelopathy (DCM). METHODS Seven hundred fifty-seven patients were prospectively enrolled in either the AOSpine North America or International study at 26 global sites. Fourteen patients had a perfect preoperative mJOA score of 18 and were excluded from this analysis (n = 743). Data were collected for each participating subject, including demographic information, symptomatology, medical history, causative pathology, and functional impairment. Univariate log-binominal regression analyses were conducted to evaluate the association between preoperative clinical factors and achieving an MCID on the mJOA scale. Modified Poisson regression using robust error variances was used to create the final multivariate model and compute the relative risk for each predictor. RESULTS The sample consisted of 463 men (62.31%) and 280 women (37.69%), with an average age of 56.48 ± 11.85 years. At 2 years following surgery, patients exhibited a mean change in functional status of 2.71 ± 2.89 points on the mJOA scale. Of the 687 patients with available follow-up data, 481 (70.01%) exhibited meaningful gains on the mJOA scale, whereas 206 (29.98%) failed to achieve an MCID. Based on univariate analysis, significant predictors of achieving the MCID on the mJOA scale were younger age; female sex; shorter duration of symptoms; nonsmoking status; a lower comorbidity score and absence of cardiovascular disease; and absence of upgoing plantar responses, lower-limb spasticity, and broad-based unstable gait. The final model included age (relative risk [RR] 0.924, p < 0.0001), smoking status (RR 0.837, p = 0.0043), broad-based unstable gait (RR 0.869, p = 0.0036), and duration of symptoms (RR 0.943, p = 0.0003). CONCLUSIONS In this large multinational prospective cohort, 70% of patients treated surgically for DCM exhibited a meaningful functional gain on the mJOA scale. The key predictors of achieving an MCID on the mJOA scale were younger age, shorter duration of symptoms, nonsmoking status, and lack of significant gait impairment.

Psychometric analysis and critical appraisal of the original, revised, and modified versions of the Japanese Orthopaedic Association score in the assessment of patients with cervical spondylotic myelopathy.

OBJECTIVE Cervical spondylotic myelopathy (CSM) is the most common cause of nontraumatic spinal cord impairment and disability in the world. Given that the Japanese Orthopaedic Association (JOA) score is the most frequently used outcome measure in clinical research and practice for treating patients with CSM, this review was undertaken to comprehensively and critically evaluate the psychometric properties of the JOA score. METHODS The authors identified studies (published in the period of January 1975 to November 2015) on the psychometric properties of the original, revised, and modified versions of the JOA score in Medline, PsycINFO, Excerpta Medica dataBASE (EMBASE), American College of Physicians Journal Club, and Cochrane Database of Systematic Reviews. Additional publications were captured in a secondary search of the bibliographies in both original research articles and literature reviews identified in the original search. The JOA scores were evaluated for item generation and reduction, internal consistency, reliability, validity, and responsiveness. This review included all those versions of the JOA score whose psychometric properties had been reported in at least 2 published studies. RESULTS The primary search strategy identified 59 studies, of which 9 fulfilled the inclusion and exclusion criteria. An additional 18 publications were captured in the secondary search and included in the analysis. The key findings from the 27 studies analyzed indicated the following: 1) the original JOA score (1975) was the source for the revised JOA score (1994) and 3 modified versions (1991, 1993, and 1999 JOA scores) reported or used in at least 2 published studies; 2) the revised and modified versions of the JOA score are markedly different from each other; 3) only the revised JOA score (1994) was validated with the original JOA score; and 4) the 1975 JOA score is the most appropriate instrument for assessing patients in Asian populations (especially from Japan) because of its psychometric attributes, and the 1991 JOA score is the most appropriate version for use in Western populations. CONCLUSIONS The authors' results indicate that the original (1975), revised (1994), and modified (1991, 1993, and 1999) versions of the JOA score are substantially different from each other in terms of their content and have been incompletely examined for their psychometric properties and cultural sensitivity. Whereas the 1975 JOA score is the most appropriate version for assessing individuals from Asian populations (particularly those eating with chopsticks), the 1991 JOA score is most suitable for evaluating patients in Western populations. Nonetheless, further investigation of the psychometric properties of the 1975 and 1991 JOA scores is recommended because of a paucity of studies reporting on the responsiveness of these 2 scoring instruments.

Prediction model for outcome after low-back surgery: individualized likelihood of complication, hospital readmission, return to work, and 12-month improvement in functional disability.

OBJECT Lumbar spine surgery has been demonstrated to be efficacious for many degenerative spine conditions. However, there is wide variability in outcome after spine surgery at the individual patient level. All stakeholders in spine care will benefit from identification of the unique patient or disease subgroups that are least likely to benefit from surgery, are prone to costly complications, and have increased health care utilization. There remains a large demand for individual patient-level predictive analytics to guide decision support to optimize outcomes at the patient and population levels. METHODS One thousand eight hundred three consecutive patients undergoing spine surgery for various degenerative lumbar diagnoses were prospectively enrolled and followed for 1 year. A comprehensive patient interview and health assessment was performed at baseline and at 3 and 12 months after surgery. All predictive covariates were selected a priori. Eighty percent of the sample was randomly selected for model development, and 20% for model validation. Linear regression was performed with Bayesian model averaging to model 12-month ODI (Oswestry Disability Index). Logistic regression with Bayesian model averaging was used to model likelihood of complications, 30-day readmission, need for inpatient rehabilitation, and return to work. Goodness-of-fit was assessed via R(2) for 12-month ODI and via the c-statistic, area under the receiver operating characteristic curve (AUC), for the categorical endpoints. Discrimination (predictive performance) was assessed, using R(2) for the ODI model and the c-statistic for the categorical endpoint models. Calibration was assessed using a plot of predicted versus observed values for the ODI model and the Hosmer-Lemeshow test for the categorical endpoint models. RESULTS On average, all patient-reported outcomes (PROs) were improved after surgery (ODI baseline vs 12 month: 50.4 vs 29.5%, p < 0.001). Complications occurred in 121 patients (6.6%), 108 (5.9%) were readmitted within 30 days of surgery, 188 (10.3%) required discharge to inpatient rehabilitation, 1630 (88.9%) returned to work, and 449 (24.5%) experienced an unplanned outcome (no improvement in ODI, a complication, or readmission). There were 45 unique baseline variable inputs, derived from 39 clinical variables and 38 questionnaire items (ODI, SF-12, MSPQ, VAS-BP, VAS-LP, VAS-NP), included in each model. For prediction of 12-month ODI, R(2) was 0.51 for development and 0.47 for the validation study. For prediction of a complication, readmission, inpatient rehabilitation, and return to work, AUC values ranged 0.72-0.84 for development and 0.79-0.84 for validation study. CONCLUSIONS A novel prediction model utilizing both clinical data and patient interview inputs explained the majority of variation in outcome observed after lumbar spine surgery and reliably predicted 12-month improvement in physical disability, return to work, major complications, readmission, and need for inpatient rehabilitation for individual patients. Application of these models may allow clinicians to offer spine surgery specifically to those who are most likely to benefit and least likely to incur complications and excess costs.

The Michigan Spine Surgery Improvement Collaborative: a statewide Collaborative Quality Initiative.

OBJECT Given the scrutiny of spine surgery by policy makers, spine surgeons are motivated to demonstrate and improve outcomes, by determining which patients will and will not benefit from surgery, and to reduce costs, often by reducing complications. Insurers are similarly motivated. In 2013, Blue Cross Blue Shield of Michigan (BCBSM) and Blue Care Network (BCN) established the Michigan Spine Surgery Improvement Collaborative (MSSIC) as a Collaborative Quality Initiative (CQI). MSSIC is one of the newest of 21 other CQIs that have significantly improved-and continue to improve-the quality of patient care throughout the state of Michigan. METHODS MSSIC focuses on lumbar and cervical spine surgery, specifically indications such as stenosis, disk herniation, and degenerative disease. Surgery for tumors, traumatic fractures, deformity, scoliosis, and acute spinal cord injury are currently not within the scope of MSSIC. Starting in 2014, MSSIC consisted of 7 hospitals and in 2015 included another 15 hospitals, for a total of 22 hospitals statewide. A standardized data set is obtained by data abstractors, who are funded by BCBSM/BCN. Variables of interest include indications for surgery, baseline patient-reported outcome measures, and medical history. These are obtained within 30 days of surgery. Outcome instruments used include the EQ-5D general health state score (0 being worst and 100 being the best health one can imagine) and EQ-5D-3 L. For patients undergoing lumbar surgery, a 0 to 10 numeric rating scale for leg and back pain and the Oswestry Disability Index for back pain are collected. For patients undergoing cervical surgery, a 0 to 10 numeric rating scale for arm and neck pain, Neck Disability Index, and the modified Japanese Orthopaedic Association score are collected. Surgical details, postoperative hospital course, and patient-reported outcome measures are collected at 90-day, 1-year, and 2-year intervals. RESULTS As of July 1, 2015, a total of 6397 cases have been entered into the registry. This number reflects 4824 eligible cases with confirmed surgery dates. Of these 4824 eligible cases, 3338 cases went beyond the 120-day window and were considered eligible for the extraction of surgical details, 90-day outcomes, and adverse events. Among these 3338 patients, there are a total of 2469 lumbar cases, 862 cervical cases, and 7 combined procedures that were entered into the registry. CONCLUSIONS In addition to functioning as a registry, MSSIC is also meant to be a platform for quality improvement with the potential for future initiatives and best practices to be implemented statewide in order to improve quality and lower costs. With its current rate of recruitment and expansion, MSSIC will provide a robust platform as a regional prospective registry. Its unique funding model, which is supported by BCBSM/BCN, will help ensure its longevity and viability, as has been observed in other CQIs that have been active for several years.

Clinical predictors of achieving the minimal clinically important difference after surgery for cervical spondylotic myelopathy: an external validation study from the Canadian Spine Outcomes and Research Network.

OBJECTIVE: Recently identified prognostic variables among patients undergoing surgery for cervical spondylotic myelopathy (CSM) are limited to two large international data sets. To optimally inform shared clinical decision-making, the authors evaluated which preoperative clinical factors are significantly associated with improvement on the modified Japanese Orthopaedic Association (mJOA) scale by at least the minimum clinically important difference (MCID) 12 months after surgery, among patients from the Canadian Spine Outcomes and Research Network (CSORN). METHODS: The authors performed an observational cohort study with data that were prospectively collected from CSM patients at 7 centers between 2015 and 2017. Candidate variables were tested using univariable and multiple binomial logistic regression, and multiple sensitivity analyses were performed to test assumptions about the nature of the statistical models. Validated mJOA MCIDs were implemented that varied according to baseline CSM severity. RESULTS: Among 205 patients with CSM, there were 64 (31%) classified as mild, 86 (42%) as moderate, and 55 (27%) as severe. Overall, 52% of patients achieved MCID and the mean change in mJOA score at 12 months after surgery was 1.7 ± 2.6 points (p < 0.01), but the subgroup of patients with mild CSM did not significantly improve (mean change 0.1 ± 1.9 points, p = 0.8). Univariate analyses failed to identify significant associations between achieving MCID and sex, BMI, living status, education, smoking, disability claims, or number of comorbidities. After adjustment for potential confounders, the odds of achieving MCID were significantly reduced with older age (OR 0.7 per decade, 95% CI 0.5-0.9, p < 0.01) and higher baseline mJOA score (OR 0.8 per point, 95% CI 0.7-0.9, p < 0.01). The effects of symptom duration (OR 1.0 per additional month, 95% CI 0.9-1.0, p = 0.2) and smoking (OR 0.4, 95% CI 0.2-1.0, p = 0.06) were not statistically significant. CONCLUSIONS: Surgery is effective at halting the progression of functional decline with CSM, and approximately half of all patients achieve the MCID. Data from the CSORN confirmed that older age is independently associated with poorer outcomes, but novel findings include that patients with milder CSM did not experience meaningful improvement, and that symptom duration and smoking were not important. These findings support a nuanced approach to shared decision-making that acknowledges some prognostic uncertainty when weighing the various risks, benefits, and alternatives to surgical treatment.

Cost-effectiveness of postoperative rehabilitation after surgery for lumbar disc herniation: an analysis based on a randomized controlled trial.

OBJECTIVE: The aim of this study was to examine whether routine referral to municipal postoperative rehabilitation is cost-effective in comparison to no referral after surgery for lumbar disc herniation (LDH). METHODS: One hundred forty-six patients scheduled for primary discectomy due to LDH were included. This secondary analysis, based on data from a previous randomized controlled trial, compared costs and quality-adjusted life years (QALYs) between two groups of patients recovering from LDH surgery: one group of patients received a referral for municipal physical rehabilitation (REHAB) and the other group was sent home without a referral to any postoperative rehabilitation (HOME). Primary outcomes were QALYs calculated from the EQ-5D utility score, societal costs, and incremental cost-effectiveness ratios (ICERs). The main cost-effectiveness analysis used intention-to-treat data, whereas sensitivity analyses included as-treated data. Questionnaires were collected after 1, 3, 6, 12, and 24 months postoperatively. RESULTS: The main cost-effectiveness analysis showed a small, insignificant incremental QALY of 0.021 and an incremental cost of €211.8 for the REHAB group compared to the HOME group, resulting in an ICER of €10,085. In the as-treated sensitivity analysis, the REHAB group had poorer outcomes and higher costs compared to the HOME group. CONCLUSIONS: Routine referral to municipal physical rehabilitation in patients recovering from LDH surgery was not cost-effective compared to no referral.Clinical trial registration no.: NCT03505918 (clinicaltrials.gov).

Microendoscopic decompression for lumbosacral foraminal stenosis: a novel surgical strategy based on anatomical considerations using 3D image fusion with MRI/CT.

OBJECTIVE: Persistent lumbar foraminal stenosis (LFS) is one of the most common reasons for poor postoperative outcomes and is a major contributor to "failed back surgery syndrome." The authors describe a new surgical strategy for LFS based on anatomical considerations using 3D image fusion with MRI/CT analysis. METHODS: A retrospective review was conducted on 78 consecutive patients surgically treated for LFS at the lumbosacral junction (2013-2017). The location and extent of stenosis, including the narrowest site and associated pathology (bone or soft tissue), were measured using 3D image fusion with MRI/CT. Stenosis was defined as medial intervertebral foraminal (MF; inner edge to pedicle center), lateral intervertebral foraminal (LF; pedicle center to outer edge), or extraforaminal (EF; outside the pedicle). Lumbar (low-back pain, leg pain) and patient satisfaction visual analog scale (VAS) scores and Japanese Orthopaedic Association (JOA) scores were evaluated. Surgical outcome was evaluated 2 years postoperatively. RESULTS: Most instances of stenosis existed outside the pedicle's center (94%), including LF (58%), EF (36%), and MF (6%). In all MF cases, stenosis resulted from soft-tissue structures. The narrowest stenosis sites were localized around the pedicle's outer border. The areas for sufficient nerve decompression were extended in MF+LF (10%), MF+LF+EF (14%), LF+EF (39%), LF (11%), and EF (26%). No iatrogenic pars interarticularis damage occurred. The JOA score was 14.9 ± 2.6 points preoperatively and 22.4 ± 3.5 points at 2 years postoperatively. The JOA recovery rate was 56.0% ± 18.6%. The VAS score (low-back and leg pain) was significantly improved 2 years postoperatively (p < 0.01). According to patients' self-assessment of the minimally invasive surgery, 62 (79.5%) chose "surgery met my expectations" at follow-up. Nine patients (11.5%) selected "I did not improve as much as I had hoped but I would undergo the same surgery for the same outcome." CONCLUSIONS: Most LFS existed outside the pedicle's center and was rarely noted in the pars region. The main regions of stenosis were localized to the pedicle's outer edge. Considering this anatomical distribution of LFS, the authors recommend that lateral fenestration should be the first priority for foraminal decompression. Other surgical options including foraminotomy, total facetectomy, and hemilaminectomy likely require more bone resections than LFS treatment. The microendoscopic surgery results were very good, indicating that this minimally invasive surgery was suitable for treating this disease.

Effect modifiers for patient-reported outcomes in operatively and nonoperatively treated patients with adult symptomatic lumbar scoliosis: a combined analysis of randomized and observational cohorts.

OBJECTIVE: Adult symptomatic lumbar scoliosis (ASLS) is a common and disabling condition. The ASLS-1 was a multicenter, dual-arm study (with randomized and observational cohorts) examining operative and nonoperative care on health-related quality of life in ASLS. An aim of ASLS-1 was to determine patient and radiographic factors that modify the effect of operative treatment for ASLS. METHODS: Patients 40-80 years old with ASLS were enrolled in randomized and observational cohorts at 9 North American centers. Primary outcomes were the differences in mean change from baseline to 2-year follow-up for the SRS-22 subscore (SRS-SS) and the Oswestry Disability Index (ODI). Analyses were performed using an as-treated approach with combined cohorts. Factors examined were prespecified or determined using regression tree analysis. For each potential effect modifier, subgroups were created using clinically relevant cutoffs or via regression trees. Estimates of within-group and between-group change were compared using generalized linear mixed models. An effect modifier was defined as a treatment effect difference greater than the minimal detectable measurement difference for both SRS-SS (0.4) and ODI (7). RESULTS: Two hundred eighty-six patients were enrolled and 256 (90%) completed 2-year follow-up; 171 received operative treatment and 115 received nonoperative treatment. Surgery was superior to nonoperative care for all effect subgroups considered, with the exception of those with nearly normal pelvic incidence-lumbar lordosis (PI-LL) match (≤ 11°). Male patients and patients with more (> 11°) PI-LL mismatch at baseline had greater operative treatment effects on both the SRS-SS and ODI compared to nonoperative treatment. No other radiographic subgroups were associated with treatment effects. High BMI, lower socioeconomic status, and poor mental health were not related to worse outcomes. CONCLUSIONS: Numerous factors previously related to poor outcomes with surgery, such as low mental health, lower socioeconomic status, and high BMI, were not related to outcomes in ASLS in this exploratory analysis. Those patients with higher PI-LL mismatch did improve more with surgery than those with normal alignment. On average, none of the factors considered were associated with a worse outcome with operative treatment versus nonoperative treatment. These findings may guide future prospective analyses of factors related to outcomes in ASLS care.

Influence of comorbid knee osteoarthritis on surgical outcome and sagittal spinopelvic/lower-extremity alignment in elderly patients with degenerative lumbar spondylolisthesis undergoing transforaminal lumbar interbody fusion.

OBJECTIVE: This retrospective study aimed to clarify the influence of comorbid severe knee osteoarthritis (KOA) on surgical outcome in terms of sagittal spinopelvic/lower-extremity alignment in elderly patients with degenerative lumbar spondylolisthesis (DLS). METHODS: In total, 110 patients aged at least 65 years (27 men, 83 women; mean age 74.0 years) who underwent short-segment lumbar fusion were included in the present study. Using the Kellgren-Lawrence (KL) grading system, patients were categorized into those with no to mild KOA (the mild-OA group: KL grades 0-2), moderate KOA (moderate-OA group: KL grade 3), or severe KOA (severe-OA group: KL grade 4). Surgical results were assessed using the Japanese Orthopaedic Association (JOA) scoring system, and spinopelvic/lower-extremity parameters were compared among the 3 groups. Adjacent-segment disease (ASD) was assessed over a mean follow-up period of 4.7 years (range 2-8.1 years). RESULTS: The study cohort was split into the mild-OA group (42 patients), the moderate-OA group (28 patients), and the severe-OA group (40 patients). The severe-OA group contained significantly more women (p = 0.037) and patients with double-level listhesis (p = 0.012) compared with the other groups. No significant differences were found in mean postoperative JOA scores or recovery rate among the 3 groups. The mean postoperative JOA subscore for restriction of activities of daily living was only significantly lower in the severe-OA group compared with the other groups (p = 0.010). The severe-OA group exhibited significantly greater pelvic incidence, pelvic tilt, and knee flexion angle (KFA), along with a smaller degree of lumbar lordosis than the mild-OA group both pre- and postoperatively (all p < 0.05). Overall, the rate of radiographic ASD was observed to be higher in the severe-OA group than in the mild-OA group (p = 0.015). Patients with ASD in the severe-OA group exhibited significantly greater pelvic tilt, pre- and postoperatively, along with less lumbar lordosis, than the patients without ASD postoperatively (all p < 0.05). CONCLUSIONS: A lack of lumbar lordosis caused by double-level listhesis and knee flexion contracture compensated for by far greater pelvic retroversion is experienced by elderly patients with DLS and severe KOA. Therefore, corrective lumbar surgery and knee arthroplasty may be considered to improve sagittal alignment, which may contribute to the prevention of ASD, resulting in favorable long-term surgical outcomes.

Establishing the minimum clinically important difference in Neck Disability Index and modified Japanese Orthopaedic Association scores for adult cervical deformity.

OBJECTIVE: It is being increasingly recognized that adult cervical deformity (ACD) is correlated with significant pain, myelopathy, and disability, and that patients who undergo deformity correction gain significant benefit. However, there are no defined thresholds of minimum clinically important difference (MCID) in Neck Disability Index (NDI) and modified Japanese Orthopaedic Association (mJOA) scores. METHODS: Patients of interest were consecutive patients with ACD who underwent cervical deformity correction. ACD was defined as C2-7 sagittal Cobb angle ≥ 10° (kyphosis), C2-7 coronal Cobb angle ≥ 10° (cervical scoliosis), C2-7 sagittal vertical axis ≥ 4 cm, and/or chin-brow vertical angle ≥ 25°. Data were obtained from a consecutive cohort of patients from a multiinstitutional prospective database maintained across 13 sites. Distribution-based MCID, anchor-based MCID, and minimally detectable measurement difference (MDMD) were calculated. RESULTS: A total of 73 patients met inclusion criteria and had sufficient 1-year follow-up. In the cohort, 42 patients (57.5%) were female. The mean age at the time of surgery was 62.23 years, and average body mass index was 29.28. The mean preoperative NDI was 46.49 and mJOA was 13.17. There was significant improvement in NDI at 1 year (46.49 vs 37.04; p = 0.0001). There was no significant difference in preoperative and 1-year mJOA (13.17 vs 13.7; p = 0.12). Using multiple techniques to yield MCID thresholds specific to the ACD population, the authors obtained values of 5.42 to 7.48 for the NDI, and 1.00 to 1.39 for the mJOA. The MDMD was 6.4 for the NDI, and 1.8 for the mJOA. Therefore, based on their results, the authors recommend using an MCID threshold of 1.8 for the mJOA, and 7.0 for the NDI in patients with ACD. CONCLUSIONS: The ACD-specific MCID thresholds for NDI and mJOA are similar to the reported MCID following surgery for degenerative cervical disease. Additional studies are needed to verify these findings. Nonetheless, the findings here will be useful for future studies evaluating the success of surgery for patients with ACD undergoing deformity correction.

Objective functional impairment in lumbar degenerative disease: concurrent validity of the baseline severity stratification for the five-repetition sit-to-stand test.

OBJECTIVE: The five-repetition sit-to-stand (5R-STS) test provides a new dimension of clinical assessment by capturing objective functional impairment (OFI). Through the utilization of data from two prospective studies, the authors sought to evaluate the concurrent validity of the proposed 5R-STS baseline severity stratification (BSS) for OFI with the following levels based on time to completion in seconds: none, ≤ 10.4; mild, 10.5-15.2; moderate, 15.3-22.0; and severe, > 22.0 seconds. METHODS: Patients with degenerative diseases of the spine performed the 5R-STS test and completed visual analog scales (VASs) for back and leg pain, the Oswestry Disability Index (ODI), the Roland-Morris Disability Questionnaire (RMDQ), and EQ-5D questionnaires. The degree of OFI severity was assessed based on the previously proposed BSS, and its association with patient-reported scales was evaluated using ANOVA as well as crude and adjusted linear regression models. RESULTS: Our sample included 240 patients, of whom 101 exhibited no OFI, whereas 80, 34, and 25 were judged to have mild, moderate, and severe OFI, respectively. A higher baseline severity was strongly associated with loss of working ability (p < 0.001), as well as results of all patient-reported scales (p ≤ 0.001), with the exception of the VAS for leg pain (p = 0.556). Crude and adjusted regression analyses corroborated these findings, although only patients with moderate and severe OFI as judged by using the 5R-STS BSS demonstrated clinically relevant differences compared with patients without OFI. CONCLUSIONS: The degree of OFI-based on the 5R-STS BSS-is strongly associated with measures of back pain, subjective functional impairment, and health-related quality of life. However, leg pain severity is not reflected within the dimension of OFI measured by the 5R-STS. The proposed BSS appears to be a concurrently valid and clinically relevant measure of OFI in patients with degenerative spinal pathologies.

Effects of preoperative obesity and psychiatric comorbidities on minimum clinically important differences for lumbar fusion in grade 1 degenerative spondylolisthesis: analysis from the prospective Quality Outcomes Database registry.

OBJECTIVE: Medical comorbidities, particularly preoperatively diagnosed anxiety, depression, and obesity, may influence how patients perceive and measure clinical benefit after a surgical intervention. The current study was performed to define and compare the minimum clinically important difference (MCID) thresholds in patients with and without preoperative diagnoses of anxiety or depression and obesity who underwent spinal fusion for grade 1 degenerative spondylolisthesis. METHODS: The Quality Outcomes Database (QOD) was queried for patients who underwent lumbar fusion for grade 1 degenerative spondylolisthesis during the period from January 2014 to August 2017. Collected patient-reported outcomes (PROs) included the Oswestry Disability Index (ODI), health status (EQ-5D), and numeric rating scale (NRS) scores for back pain (NRS-BP) and leg pain (NRS-LP). Both anchor-based and distribution-based methods for MCID calculation were employed. RESULTS: Of 462 patients included in the prospective registry who underwent a decompression and fusion procedure, 356 patients (77.1%) had complete baseline and 12-month PRO data and were included in the study. The MCID values for ODI scores did not significantly differ in patients with and those without a preoperative diagnosis of obesity (20.58 and 20.69, respectively). In addition, the MCID values for ODI scores did not differ in patients with and without a preoperative diagnosis of anxiety or depression (24.72 and 22.56, respectively). Similarly, the threshold MCID values for NRS-BP, NRS-LP, and EQ-5D scores were not statistically different between all groups. Based on both anchor-based and distribution-based methods for determination of MCID thresholds, there were no statistically significant differences between all cohorts. CONCLUSIONS: MCID thresholds were similar for ODI, EQ-5D, NRS-BP, and NRS-LP in patients with and without preoperative diagnoses of anxiety or depression and obesity undergoing spinal fusion for grade 1 degenerative spondylolisthesis. Preoperative clinical and shared decision-making may be improved by understanding that preoperative medical comorbidities may not affect the way patients experience and assess important clinical changes postoperatively.

Predictors of favorable quality of life outcome following kyphoplasty and vertebroplasty.

OBJECTIVE: In 2009, 2 randomized controlled trials demonstrated no improvement in pain following vertebral augmentation compared with sham surgery. However, a recent randomized trial demonstrated significant pain relief in patients following vertebroplasty compared to controls treated with conservative medical management. This study is a retrospective review of prospectively collected patient-reported quality of life (QOL) outcomes. The authors hypothesized that vertebral augmentation procedures offer a QOL benefit, but that this benefit would be diminished in patients with a history of depression and/or in patients undergoing vertebral augmentation at more than 1 level. METHODS: Multivariable linear regression was used to identify predictors of postoperative pain assessed using the Pain Disability Questionnaire (PDQ), Patient Health Questionnaire 9 (PHQ-9), and EQ-5D scores. Eleven candidate predictors were selected a priori: age, sex, smoking history, coronary artery disease, depression, diabetes, procedure location (thoracic, lumbar), BMI, prior spine surgery, procedure indication (metastases, osteoporosis/osteopenia, other), and number of levels (1, 2, 3, or more). RESULTS: A total of 143 patients were included in the study. For each 10-year increase in age, postoperative PDQ scores decreased (improved) by 9.7 points (p < 0.001). Patients with osteoporosis/osteopenia had significantly higher (worse) postoperative PDQ scores (+17.97, p = 0.028) than patients with metastatic lesions. Male sex was associated with higher (worse) postoperative PHQ-9 scores (+2.48, p = 0.010). Compared to single-level augmentation, operations at 2 levels were associated with significantly higher PHQ-9 scores (+2.58, p = 0.017). Current smokers had significantly lower PHQ-9 scores (-1.98, p = 0.023) than never smokers. No predictors were associated with significantly different EQ-5D score. CONCLUSIONS: Variables associated with worse postoperative PDQ scores included younger age and osteoporosis/osteopenia. Variables associated with decreased (better) postoperative PHQ-9 scores included female sex, single operative vertebral level, and positive smoking status (i.e., current smoker). These clinically relevant predictors may permit identification of patients who may benefit from vertebral augmentation.