Association between fusion and clinical outcomes after anterior cervical discectomy at 1-, 2- and 5-year follow-up.

INTRODUCTION: Fusion achievement is considered a crucial factor in recovery following anterior discectomy. Nevertheless, the direct correlation between fusion and clinical outcomes, such as pain and disability, remains ambiguous due to inconsistent fusion measurement methods. Recent advancements in diagnostic fusion criteria now enable a more accurate fusion assessment. This study aimed to assess the association between fusion and clinical outcomes in patients undergoing anterior cervical discectomy. METHODS: This post-hoc analysis was conducted using data from the NEtherlands Cervical Kinematics (NECK) trial (NTR1289). Patients with a single level herniated disc that underwent anterior cervical discectomy between 2010 and 2014 were evaluated at 52, 104, and 260 weeks. Fusion was assessed using dynamic radiographs, applying the de Vries-Vleggeert criterion (≤3.0° Cobb angle and ≤2.0 mm interspinous distance). Clinical outcomes included the Neck Disability Index (NDI) and Visual Analog Scale (VAS) for arm and neck pain. Linear and logistic regressions were performed to evaluate correlations. RESULTS: Fusion was present in 57% (52 weeks), 75% (104 weeks), and 83% (260 weeks) of patients. Linear regression analyses revealed a clear trend suggesting favorable long-term VAS arm and neck pain scores in the patient's demonstrating fusion with statistically significant lower VAS arm pain (mean difference: -18.9, 95% CI -36.9 to -0.9, p = 0.040) in the fusion group at 260 weeks follow-up. At earlier follow-up points, the differences in VAS arm pain did show a trend, but did not reach statistical significance (W52: 6.0, 95% CI -6.6 to 18.6, p = 0.346; W104: -11.5, 95% CI -24.2 to 1.3, p = 0.076). VAS neck pain scores showed a trend, but no statistically significant differences between groups across follow-up (W52: 3.2, 95% CI -8.1 to 14.5, p = 0.572; W104: -1.5, 95% CI -13.8 to 10.8, p = 0.808; W260: -12.7, 95% CI -30.9 to 5.6, p = 0.170). No significant differences were observed in NDI outcomes at any time point (W52: 3.2, 95% CI -4.8 to 11.2, p = 0.431; W104: -3.7, 95% CI -11.6 to 4.2, p = 0.358; W260: 2.5, 95% CI -7.2 to 12.1, p = 0.608). Logistic regression analysis using success rates based on established cut-off values showed a trend towards patients with fusion having markedly higher odds of success long term, with a significant higher odd of success in VAS arm pain at 260 weeks FU (OR 9.88, 95% CI 1.55-62.80, p = 0.015). CONCLUSION: A comparative analysis indicated reduced arm and neck pain in the fusion group at the 260-week follow-up. This finding becomes apparent only in the long-term post-intervention period, suggesting that muscle tension may function as a natural brace during the initial years following surgery. This tension effectively limits excessive flexion-extension movements, thereby mitigating discomfort. These results have potential implications for routine clinical surgical practice. Future studies with larger sample sizes are needed to validate these findings, including short-term follow-up.