Risk factors related to adjacent segment degeneration: retrospective observational cohort study and survivorship analysis of adjacent unfused segments

Object: Adjacent segment disease (ASD) has been described as a frequent complication after a lumbar spinal fusion procedure, though its incidence and the factors related to its appearance are not well established. The radiographic signs that identify ASD in unfused segments may be a consequence of biomechanical changes induced by the fusion procedure. This study sought to analyse the incidence of radiographic changes (radiographic ASD) in all adjacent unfused segments, the clinical changes that require a second procedure (clinical ASD), and the risk factors of their appearance evaluated at different follow-up times. Methods: We conducted a retrospective cohort study of patients fused for degenerative spine disease and instability to analyse ASD risk factors using actuarial estimation, comparison of the Kaplan-Meier survival curves of each variable, and Cox proportional-hazards regression analysis. Results: Among the 263 patients included in the study, radiographic changes were observed in 57.4% and related clinical changes in 20.2%. The univariate analysis showed a higher risk of ASD in patients with smaller post- vs. pre-operative lumbar lordosis (p = .018), diagnosis of lumbar canal stenosis (p = .019), fusion of three or more vs. fewer levels (p = .009) and those fixed with top-loading screws vs. side-connecting screws (p = .001). Cox proportional-hazards regression analysis showed that the use of top-loading pedicle screws and three or more levels of fusion led to a 3- and 2-fold higher risk of degeneration in adjacent unfused segments respectively. Conclusions: Risk of a second surgical procedure due to clinical changes is 3-fold higher in patients with three or more levels of fusion, and 2.5-fold higher in patients intervened with top-loading pedicle screws.