Risk of spring-back closure in skipped-level plating for open-door laminoplasty: an insight of its cost-saving potential

Abstract

Introduction: The open-door laminoplasty technique is a commonly adopted posterior approach to decompression for cervical spondylotic myelopathy (CSM). Spring-back phenomenon is a major concern as closure of the lamina opening can cause recurrence of symptoms. Plating has since reduced the risk of spring-back but incurs higher costs. Providing the most cost-effective treatment is necessary at current medicine practice. For laminoplasty, the need for plating at every level of lamina opening is unknown. Hence, the aim of this study is to determine whether skipped-level plating can still avoid the risk of spring-back closure while maintaining adequate neurological recovery. Materials and Methods: Patients with CSM treated by open-door laminoplasty with 2-year postoperative follow-up were recruited. All patients had opening from C3-6 or C3-7 and were divided into skipped-level or all-level plating groups. Japanese Orthopaedic Association (JOA) scores were obtained preoperatively, immediate (within 1 week) postoperatively, and at 2 weeks, 6 weeks, 3, 6 and 12 months postoperatively. Recovery rate was calculated using the formula: recovery rate(%) = (postoperative JOA-preoperative JOA) / (17[full score]-preoperative JOA)x100. Similarly, cervical spine radiographs at each time-point were measured by three independent readers for the canal diameter. Comparisons of measurement between time-points allow determining of any spring-back closure, defined as > 1 mm loss of initial expansion. Statistical analyses included intraclass correlation coefficient (ICC) for inter-rater and intra-rater reliability; chi-square test of independence, Mann-Whitney U test and paired t-test for comparative analysis. Receiver operating characteristic (ROC) analysis was utilized to determine cut-off values of canal expansion with which spring-back occurred. Results: A total of 74 subjects (24% females) were included with mean age of 66.1 ± 11.3 years at surgery. 32 underwent skipped-level plating and 42 underwent all-level plating. Good intra-rater reliability was shown in both pre-operative (ICC: 0.826-0.915, p < 0.05) and post-operative measurements (ICC: 0.878-0.936 and 0.855-0.884 with p < 0.001 for all-level and skipped-level plating respectively) of canal diameter at all vertebral levels. No significant differences existed between the two groups at baseline and follow-up. Spring-back closure was observed in up to 50% of the non-plated levels within 3 months postoperatively. The cut-off for developing spring-back closure was 7 mm canal expansion for C3-6, with an area under the curve of 0.795 to 0.889, a sensitivity and specificity of 80.0 to 84.6% and 75.0 to 83.3% respectively. No differences were observed in JOA scores (p = 0.294-0.850) and recovery rates (p = 0.189-0.864) between the two groups at all postoperative time points. Conclusion: This is a novel study that presents compelling data that highlights the non-inferiority of patients undergoing different combinations of skipped-level plating as compared with all-level plating for laminoplasty. No significant differences between groups were observed for spring-back closure at all levels from C3-7 without plating. This has tremendous impact on saving costs in CSM management as up to two plates per patient undergoing a standard C3-6 laminoplasty may be omitted instead of four plates to every level to achieve similar clinical and radiological outcomes. In addition, a threshold of at least 7 mm canal expansion can be proposed to avoid spring-back closure.