Study Design: Retrospective/prospective study. Objective: Models based on preoperative factors can predict patients’ outcome at 1-year follow-up. This study measures the performance of several machine learning (ML) models and compares the results with conventional methods. Methods: Inclusion criteria were patients who had lumbar disc herniation (LDH) surgery, identified in the Danish national registry for spine surgery. Initial training of models included 16 independent variables, including demographics and presurgical patient-reported measures. Patients were grouped by reaching minimal clinically important difference or not for EuroQol, Oswestry Disability Index, Visual Analog Scale (VAS) Leg, and VAS Back and by their ability to return to work at 1 year follow-up. Data were randomly split into training, validation, and test sets by 50%/35%/15%. Deep learning, decision trees, random forest, boosted trees, and support vector machines model were trained, and for comparison, multivariate adaptive regression splines (MARS) and logistic regression models were used. Model fit was evaluated by inspecting area under the curve curves and performance during validation. Results: Seven models were arrived at. Classification errors were within ±1% to 4% SD across validation folds. ML did not yield superior performance compared with conventional models. MARS and deep learning performed consistently well. Discrepancy was greatest among VAS Leg models. Conclusions: Five predictive ML and 2 conventional models were developed, predicting improvement for LDH patients at the 1-year follow-up. We demonstrate that it is possible to build an ensemble of models with little effort as a starting point for further model optimization and selection.

Study Design. Propensity-matched retrospective study of patients prospectively enrolled in Danespine.Objective. The aim of this study was to report 5-year patient reported outcome in lumbar spinal stenosis (LSS) patients who underwent wide laminectomy (WL), segmental bilateral laminotomies (SBL), or unilateral hemilaminectomy (UHL) with bilateral decompression. Summary of Background Data. The optimal procedure for LSS remains controversial. Studies have shown no difference in short term outcomes among micro-laminectomy, hemi-laminotomies, broad laminectomy, and laminectomy with instrumented fusion. Methods. Patients with spinal stenosis who were enrolled in DaneSpine at two spine centers from January 2010 until May 2014 and underwent WL0, SBL, or UHL with bilateral decompression were identified. Patients completed standard questionnaires preoperatively and 1, 2, and 5 years after surgery. Patients in the three cohorts were propensity-matched using age, sex, body mass index (BMI), smoking status, number of surgical levels, American Society of Anesthesiologists (ASA) score, and patient-reported outcome measures (PROMs). Results. Propensity matching produced 62 cases in each group. There were no differences in PROM among the three cohorts at five years follow up. Twelve patients were re-operated at the index level. The most frequent indication of reoperation was repeat decompression after SBL. Regression analysis revealed no statistical significant associations between the incidence of reoperation and age, sex, number of operated levels, ASA score, BMI, center, smoking status, or having a dural tear at index operation. Conclusion. This study revealed no significant difference PROMs, reoperation rates or time to reoperation at five years follow up between SBLs, UHL, or WL in patients operated for central LSS.

No abstract