Objective To explore the consistency and accuracy of Tomita score, revised Tokuhashi score, Rades score and Oswestry spinal risk index (OSRI) in predicting the prognosis of patients with spinal metastasis.

Methods The medical records of 127 hospitalized patients with spinal metastasis admitted to two comprehensive hospitals in Yibin City from January 2010 to December 2020 were collected, and the deadline for follow-up was June 30, 2022. The Kaplan-Meier method was used to draw the survival curve, the Log-rank Chi-square test and Cox regression analysis were used to analyze the factors affecting prognosis, the Kappa value was used to evaluate the consistency between predicted and actual survival time, and area under the curve (AUC) was used to evaluate the accuracy of four scoring systems in predicting survival time of patients with spinal metastases.

Results The patient′s survival time was 10 (6, 14) months, with a mortality rate of 76.38% during the follow-up period, and the 6-month and 12-month survival rates were 69.3% and 27.6% respectively. The AUC values of Tomita score, revised Tokuhashi score, Rades score and OSRI in predicting patient′s survival time were 0.968, 0.957, 0.930 and 0.798 respectively, with the optimal cut-off values of 4.5, 10.5, 36.5 and 2.5. Cox regression analysis showed that the four factors such as Tomita score, revised Tokuhashi score, Rades score and presence or absence of visceral metastasis were the predictive factors for the prognosis of patients with spinal metastasis; the hazard ratio (HR) values for revised Tokuhashi score ≤ 8 and 9 to 11 were 9.328 and 6.417 respectively, the HR values for Tomita score 2 to 3, 4 to 5 and 6 to 7 were 0.183, 0.197 and 0.398 respectively, the HR values for Rades score ≤ 30 and 31 to 35 were 2.898 and 2.166 respectively, and the HR for resectable and non-resectable visceral metastases were 1.517 and 4.116, respectively. The revised Tokuhashi score predicted a Kappa value >0.40 for 6-month and 12-month survival rates, while the Tomita score, Rades score and OSRI predicted a Kappa value < 0.40. The revised Tokuhashi score had the highest accuracy in predicting patient′s 6-month survival (AUC=0.863), the Tomita score had average accuracy (AUC=0.717), and the Rades score and OSRI had poor accuracy (AUC < 0.70).

Conclusion The prognosis of patients with spinal metastasis is poor, and the prognosis is related to multiple factors. The revised Tokuhashi score has relatively high accuracy and consistency in predicting prognosis of patients.