1460. A Partial Least Squares Analysis of Functional Status, Disability, and Quality of Life after Surgical Decompression for Degenerative Cervical Myelopathy
Authors: Jetan Badhiwala, MD; Omar Khan, BSc; Adam Wegner, MD, PhD; Frank Jiang, MD; Jamie Wilson, MD; Benjamin Morgan; George Ibrahim, MD, PhD; Jefferson Wilson, MD, PhD; Michael Fehlings, MD, PhD (Toronto, Canada)
Introduction: Previous studies aimed at identifying predictors of clinical outcomes following surgical decompression for degenerative cervical myelopathy (DCM) are limited by multicollinearity among predictors, whereby the high degree of correlation between covariates precludes detection of potentially significant findings. We apply a data-driven method, partial least squares (PLS), to model multidimensional variance and dissociate distinct patient phenotypes that are associated with functional, disability, and QOL outcomes. Methods: This was a post-hoc analysis of patients with DCM enrolled in the prospective, multi-center AOSpine CSM-NA/CSM-I studies. Baseline clinical covariates evaluated as predictors included demographic (e.g., age, sex, BMI, education, comorbidities), clinical presentation (e.g., symptoms, signs, duration of myelopathy, causative pathology), and surgical treatment (e.g., approach, number of levels, operative duration) characteristics. Outcomes evaluated included change in functional status (∆mJOA), disability (∆NDI), and QOL (∆SF-36) at 2 years. PLS was used to derive latent variables (LVs) that relate specific clinical covariates with specific outcomes. Bootstrapping was used to estimate statistical significance. Results: A total of 478 patients met eligibility criteria. PLS identified 3 significant LVs. LV1 demonstrated that the patients with intrinsic hand muscle atrophy who were not treated with laminectomy alone had greater improvement in physical health-related QOL (SF-36 Physical Component Summary, Physical Functioning, Role Physical) at 2 years. For LV2, the patient phenotype associated with lesser improvement in mJOA was dominated by respiratory, rheumatologic, and psychiatric comorbidities. Finally, LV3 revealed that a phenotype of a patient with more severe myelopathy (lower mJOA) presenting with a history of gait impairment and exam findings of a broad-based, unstable gait was associated with poorer mental health-related QOL outcomes (SF-36 Mental Component Summary, Mental Health, Vitality, Social Functioning). Conclusion: By applying a data-driven, multivariate approach, we identified robust associations between DCM clinical presentations and functional, disability, and QOL outcomes.