Introduction: Predicting outcomes after surgery for cervical spondylotic myelopathy (CSM) remains a challenge. This study introduces the use of multivariate linear regression (MLR) and support vector regression (SVR) models to predict postoperative outcomes in a cohort of patients who underwent surgery for CSM.
Methods: Patients were recruited who had a diagnosis of CSM and required decompressive surgery. Fine motor function was tested preoperatively and postoperatively with a handgrip-based tracking device that has been previously validated, yielding mean absolute accuracy (MAA) results for two tracking tasks (sine and step). All patients completed Oswestry Disability Index (ODI) and modified Japanese Orthopaedic Association (mJOA) questionnaires preoperatively and postoperatively. Age, duration of symptoms, narrowest spinal cord diameter, and preoperative MAA scores were utilized in MLR and SVR models to predict postoperative ODI. Predictions were compared to the actual ODI scores with the coefficient of determination (R^2) and mean absolute difference (MAD).
Results: 17 patients met the inclusion criteria and completed follow-up at least two months after surgery (average = 7.9 months). There were 11 males and 6 females, and the average age was 62.5 years. With the MLR model, a combination of the preoperative ODI score and preoperative MAA (sine task) yielded the best prediction of postoperative ODI (R^2 = 0.20, MAD = 0.119, p = 0.032). With the SVR model, a combination of preoperative ODI score and preoperative MAA (step task) yielded the best prediction of postoperative ODI (R^2 = 0.32, MAD = 0.092, p < .001).
Conclusions: Using a combination of preoperative ODI score and performance on a task assessing fine motor function, postoperative ODI scores calculated by MLR and SVR models correlated well with the actual postoperative ODI score in patients who underwent surgery for CSM.
Patient Care: Previous studies have shown that 10% of patients undergoing surgery for cervical spondylotic myelopathy do not experience functional improvement. Our study introduces novel models for predicting surgical outcome that can be used to optimize patient selection for surgery.
Learning Objectives: By the conclusion of this session, participants should be able to:
1. Describe the use of multivariate linear regression (MLR) and support vector regression (SVR) models in predicting surgical outcomes for cervical spondylotic myelopathy (CSM).
2. Identify the combination of preoperative factors that were most predictive of surgical outcome in MLR and SVR models.
3. Discuss the importance of predicting surgical outcomes and how the results can be used to optimize patient selection for the surgical management of CSM.