Risk factors for persistent arm pain 5 years after surgery for cervical spondylotic myelopathy: a Spine CORe™ analysis of QOD data

OBJECTIVE: The purpose of this study was to evaluate patients undergoing surgery for cervical spondylotic myelopathy who presented with severe arm pain to determine which factors are associated with persistent and improved postoperative arm pain. METHODS: Of the patients with CSM included from 14 Spine CORe™ study group sites participating in the Quality Outcomes Database, those who presented with severe arm pain preoperatively (numeric rating scale [NRS] scores 7-10) were selected for analysis. Within this subset, patients who reported persistent severe arm pain (NRS scores 7-10) postoperatively were compared with patients who had moderate and improved/mild arm pain (NRS scores 4-6 and 0-3, respectively) postoperatively. NRS scores for arm pain were recorded at baseline and postoperatively at 3 months, 1 year, 2 years, and 5 years. Demographics, comorbidities, and patient-reported outcome measures (Neck Disability Index [NDI], quality-adjusted life years [QALY], modified Japanese Orthopaedic Association [mJOA] scale, and EuroQol visual analog scale [EQ-VAS]) were evaluated. These factors were compared between patients with persistent and those with improved arm pain at the 5-year follow-up using the Wilcoxon rank-sum test. RESULTS: Of 1085 patients with CSM, 458 reported severe arm pain (NRS scores 7-10) preoperatively. Of these, 60.7% of patients reported mild arm pain (NRS scores 0-3), 20.4% reported moderate arm pain (NRS scores 4-6), and 18.9% reported persistent severe arm pain (NRS scores 7-10) at the 5-year follow-up. Patients with persistent severe pain had significantly higher NDI scores (p = 0.001) and lower mJOA (p = 0.04), QALY (p = 0.02), and EQ-VAS (p < 0.00001) scores at baseline. Patients with mild/improved arm pain were significantly more likely to have postgraduate education (p = 0.04). Patients with early postoperative improvement in arm pain (at 3 months, p < 0.0001) had sustained improvement through the 5-year postoperative time point. However, patients with persistent severe pain at 5 years had consistently higher pain at all postoperative time points (3 months, 1 year, and 2 years), which significantly worsened between the 2- and 5-year time points (mean NRS scores from 4.8 to 8, p < 0.001). CONCLUSIONS: Among patients who underwent surgery for CSM who presented with severe radicular arm pain (NRS scores 7-10), 81.1% reported improvement (NRS scores 0-6), and 18.9% reported persistent severe arm pain 5 years after surgery. Persistent arm pain was associated with increased preoperative disability (mJOA, NDI, QALY, and EQ-VAS), while postgraduate education was associated with improved arm pain. Patients with early postoperative improvement in arm pain demonstrated a durable result through 5 years of follow-up, which could improve further. Conversely, those with severe postoperative arm pain had persistent pain at 2 years, which further deteriorated between 2 and 5 years, thus warranting careful long-term surveillance.