Laboratory markers as useful prognostic measures for survival in patients with spinal metastases.

BACKGROUND CONTEXT: Laboratory values have been found to be useful predictive measures of survival following surgery. The utility of laboratory values for prognosticating outcomes among patients with spinal metastases has not been studied. PURPOSE: To determine the prognostic capacity of laboratory values at presentation including white blood cell count, serum albumin and platelet-lymphocyte ratio (PLR) in patients with spinal metastases. STUDY DESIGN: Retrospective review of records from two tertiary care centers (2005-2017). PATIENT SAMPLE: Patients, aged 40 to 80, who received operative or nonoperative management for spinal metastases. OUTCOME MEASURES: Survival, complications, or hospital readmissions within 90 days of treatment and a composite measure for treatment failure accounting for changes in ambulatory function and mortality at 6 months following presentation. METHODS: Multivariable Cox proportional hazard regression analysis was used to analyze the relationship between laboratory values and length of survival, adjusting for confounders. Multivariable logistic regression was used in analyses related to 6-month and 1-year mortality, complications, readmissions, and treatment failure. A scoring rubric was developed based on the performance of laboratory values in the multivariable tests. Internal validation was performed using a bootstrap simulation that consisted of sampling with replacement and 1,000 replications. RESULTS: We included 1,216 patients. Thirty-seven percent of patients received a surgical intervention and 63% were treated nonoperatively. Median survival for the cohort as a whole was 255 days (interquartile range 93-642 days). The PLR (hazard ratio [HR] 1.53; 95% confidence interval [CI] 1.29, 1.80; p<.001) and albumin (HR 0.54; 95% CI 0.45, 0.64; p<.001) were significantly associated with survival, whereas WBC count (HR 1.08; 95% CI 0.86, 1.36; p=.50) was not associated with this outcome. Similar findings were encountered for 6-month and 1-year mortality as well as the composite measure for treatment failure. The PLR and albumin performed well in our scoring rubric and findings were preserved in the bootstrapping validation. CONCLUSIONS: Individuals with low serum albumin and elevated PLR should be advised regarding the impact of these laboratory markers on outcomes including survival, irrespective of treatments received. An effort should also be made to optimize nutrition and PLR, if practicable, before treatment to minimize the potential for development of adverse events.

Duke Scholars

Author Peter Passias Orthopaedic Surgery