STUDY DESIGN: Prospective observational study. SUMMARY OF BACKGROUND DATA: Tokuhashi score (TS) system was developed in 1990s and widely used for the prediction of survival in patients with spinal metastases. There were no reports about how accurately TS system predicts survival in the current practice. OBJECTIVE: To validate the accuracy of TS system using a recent series of the patients with metastatic spinal tumors, focusing on the primary tumor type. METHODS: Between January 2007 and March 2013, 145 patients who underwent surgical treatment for metastatic spinal tumors were followed up prospectively. The subjects were divided into 3 prognosis group according to TS system. The actual and predicted survivals were compared with calculate the accuracy rate. RESULTS: The mean age was 60.0 years. There were 96 male and 49 female patients. There are 106 patients in poor prognosis group (A), 30 in intermediate prognosis group (B), and 9 in good prognosis group (C). The overall accuracy to predict the survival time was 57.2%. The accuracy rate was especially poor in breast cancer (44.4%) and rectal cancer (42.9%). Forty-three patients (40.6%) in group A and 9 patients (30.0%) in group B survived longer than predicted. The mean survival time was 8.5, 18.1, and 41.6 months for group A, B, and C, respectively. Patients with rectal cancer showed poor survival profile than expected. CONCLUSIONS: The accuracy of TS system was suboptimal (57.2%). The predicted survival time needs to be increased in response to the recent population's data. The score given to the rectal cancer subgroup should be lowered.
STUDY DESIGN: Prospective observational study. SUMMARY OF BACKGROUND DATA: Tokuhashi score (TS) system was developed in 1990s and widely used for the prediction of survival in patients with spinal metastases. There were no reports about how accurately TS system predicts survival in the current practice. OBJECTIVE: To validate the accuracy of TS system using a recent series of the patients with metastatic spinal tumors, focusing on the primary tumor type. METHODS: Between January 2007 and March 2013, 145 patients who underwent surgical treatment for metastatic spinal tumors were followed up prospectively. The subjects were divided into 3 prognosis group according to TS system. The actual and predicted survivals were compared with calculate the accuracy rate. RESULTS: The mean age was 60.0 years. There were 96 male and 49 female patients. There are 106 patients in poor prognosis group (A), 30 in intermediate prognosis group (B), and 9 in good prognosis group (C). The overall accuracy to predict the survival time was 57.2%. The accuracy rate was especially poor in breast cancer (44.4%) and rectal cancer (42.9%). Forty-three patients (40.6%) in group A and 9 patients (30.0%) in group B survived longer than predicted. The mean survival time was 8.5, 18.1, and 41.6 months for group A, B, and C, respectively. Patients with rectal cancer showed poor survival profile than expected. CONCLUSIONS: The accuracy of TS system was suboptimal (57.2%). The predicted survival time needs to be increased in response to the recent population's data. The score given to the rectal cancer subgroup should be lowered.
Authors: Chia-Lin Tseng; Wietse Eppinga; Raphaele Charest-Morin; Hany Soliman; Sten Myrehaug; Pejman Jabehdar Maralani; Mikki Campbell; Young K Lee; Charles Fisher; Michael G Fehlings; Eric L Chang; Simon S Lo; Arjun Sahgal Journal: Global Spine J Date: 2017-04-06