A Prospective Analysis of Prognostic Factors in Patients With Spinal Metastases Use of the Revised Tokuhashi Score

Study Design. Prospective observational cohort study. Objective. To define the utility of the revised Tokuhashi score in relation to predicting survival in patients with spinal metastases regardless of the treatment pathway. Summary of Background Data. The revised Tokuhashi score has been used for the prediction of survival. In this scoring system, however, all the patients were sourced by orthopedic surgeons, and asymptomatic patients were excluded. That might present a significant source of patient selection bias. The treatment plan was also affected by the predicted survival in their system. Methods. All patients within 2 years of diagnosis of spinal metastases, whether symptomatic were recruited. Minimum 1-year follow-up was required. During the study period, a total of 85 patients were analyzed including 44 patients who died within 1 year. The relation between the revised Tokuhashi score and survival were analyzed using the Cox proportional hazard model and Spearman's rank correlation coefficient. Results. The mean age was 60.3 years (range: 35-84) and the median survival was 11.6 months. On multivariate analysis, lower performance status (Karnofsky performance status, 50%-70%) and unresectable organ metastases were significantly associated with poor survival, with hazard ratios of 2.92 and 4.44, respectively. In primary cancer type, lung and kidney cancer were also significantly associated with poor survival, with hazard ratios of 4.25 and 2.60, respectively. The revised Tokuhashi score groups were significantly correlated with the survival groups (rho = 0.530, P < 0.001). In 67 (79%) of 85 patients, actual survival matched the predicted survival. Conclusion. Lower score on performance status, the existence of organ metastases, and primary cancer of the lung and the kidney were significantly associated with poor survival. The revised Tokuhashi score was found to be very useful to predict survival regardless of the treatment pathway. In most patients, actual survival matched their predicted survival.