Abstract
INTRODUCTION:Surgery for cervical spondylotic myelopathy (CSM) aims to halt disease progression. However, some patients will also ultimately note relief from their myelopathic symptoms. This study aims to identify the factors that predict whether a patient will improve from moderate or severe myelopathy to mild or no myelopathy through an analysis of mJOA scores.METHODS:Data was obtained from the Quality Outcomes Database CSM dataset, a prospective registry cohort of 1141 patients from 14 sites. Patients were excluded if they had missing baseline or 24-month mJOA scores, or baseline mJOA greater than 14. Remaining patients were partitioned into a training (n = 487) or test set (n = 119). Logistic regression and random forest models, with and without principal component analysis (PCA), were trained to predict whether patients achieved mild/no myelopathy at 24 months.RESULTS:Overall, 606 patients met inclusion criteria with a mean age of 61.7 ± 11.3, 47.2% female, mean BMI 30.3 ± 6.3, and mean baseline mJOA 11.2 ± 2.4. All models performed similarly with AUROCs near 0.62. Significant positive predictors of improvement to mild/no myelopathy were low leg numbness (OR 1.55, 95% CI: 1.20-1.99, p<0.001), participation in outdoor activity (OR 2.34, 95% CI: 1.13-4.82, p=0.022), and higher SES index (OR 1.28, 95% CI: 1.01-1.62, p=0.043). A significant negative predictor was symptoms lasting longer than 12 months (OR 0.39, 95% CI: 0.18-0.86, p=0.021).CONCLUSIONS:In a large prospective registry cohort of patients operated for CSM, factors associated with myelopathic symptom relief were higher socioeconomic status, fewer leg numbness symptoms, higher baseline activity levels, and a shorter symptom duration (<12 months). The finding that longer symptom duration was associated with a decreased propensity for myelopathic symptom relief suggests that earlier surgery may be beneficial for patients with moderate to severe myelopathy.