Reliability of a spinal metastasis prognostic score to model 1-year survival

Background Predicting survival after surgery for patients with metastatic spine disease can be challenging, with multiple variables that can influence a patient's overall survival. Predictive models have been developed to assist clinicians in providing a prognosis for patients. Recently, Ghori et al. reported a composite model taking into account a modified Bauer score, preoperative albumin, and ambulatory status of patients with spinal metastasis. Using an independent cohort, we sought to assess the reliability and validity of this composite model to predict 1-year survival in patients diagnosed with metastatic cancer to the spine. Purpose This study aimed to assess the reliability and validity of the Ghori et al. composite model to predict 1-year survival in patients diagnosed with metastatic cancer to the spine, using an independent cohort. Study Design/Setting A retrospective study was carried out. Patient Sample The sample comprised 161 patients with spinal metastasis undergoing surgery. Outcome Measures Patients' modified Bauer score, preoperative albumin, and ambulatory status were assessed. Methods This study used a retrospective analysis of 161 patients with spinal metastasis who underwent surgical management from 2007 to 2013. The ability of this composite model to predict 1-year survival was compared with actual patient survival using multivariable logistic regression to control for confounders, as well as post-regression diagnostics. Results Our analysis revealed significantly lower 1-year mortality among patients with higher composite scores as compared with those with lower scores. Strong associations between scores and survival were appreciated in unadjusted analysis. The final model was able to account for 80% of the variation in the 1-year survival, and there was no evidence of lack of fit. Conclusion This study demonstrates, in an independent cohort of spinal metastases patients, that a composite model taking into account the ambulatory status, serum albumin, and modified Bauer score is able to better predict postoperative survival. These data serve to validate the use of this predictive model in determining the prognosis of patients with spinal metastasis.