Dynamic K-Line Status and Surgical Outcomes in Multilevel Cervical OPLL: A Multicenter Comparative Study

Abstract

Background/Objectives: To evaluate the clinical and radiological outcomes of surgical interventions stratified by dynamic K-line status and to identify predictors of neurological recovery in multilevel cervical ossification of the posterior longitudinal ligament (OPLL). Methods: This study analyzed 535 patients with multilevel cervical OPLL who underwent anterior cervical discectomy and fusion (ACDF), laminoplasty (LP), or laminectomy with fusion (LF), with a minimum 24 months of follow-up. Patients were classified based on dynamic K-line status—neutral (NK-line) and flexion (FK-line)—into three groups: Group 1 (NK-line [+]/FK-line [+]), Group 2 (NK-line [+]/FK-line [−]), and Group 3 (NK-line [−]/FK-line [−]). Radiographic parameters, JOA scores, and VAS were compared, and multivariate regression identified predictors of recovery. A multinomial inverse probability of treatment weighting (IPTW) analysis was conducted to reduce treatment selection bias. Results: Progressive dynamic K-line negativity was associated with greater cervical kyphosis, a higher canal-occupying ratio, reduced FK-line distance, and poorer neurological recovery. After IPTW analysis, ACDF showed higher adjusted recovery across subgroups. In Group 1, younger age and fewer operative levels predicted better recovery. In Groups 2 and 3, LF demonstrated significantly greater neurological recovery than LP. A larger preoperative FK-line distance and a greater postoperative FK-line distance increase were independent predictors of favorable outcomes. Conclusions: Dynamic K-line stratification has good prognostic value in multilevel cervical OPLL. ACDF remains the most effective procedure across dynamic K-line status groups, and LF is preferred over LP in patients with latent or fixed FK-line (−). Incorporating dynamic K-line metrics into surgical planning could improve procedure selection and enhance neurological recovery. © 2026 by the authors.