新英格兰脊柱转移瘤评分系统在脊柱转移瘤患者生存期预测中的价值

The clinical value of the New England spinal metastases score system in predicting the survival of patients with spinal metastases

目的探讨新英格兰脊柱转移瘤评分系统(New England spinal metastasis score,NESMS)对脊柱转移瘤患者生存期预测的临床价值。方法回顾性收集2008年1月至2018年12月多中心收治的脊柱转移瘤住院患者179例,男108例(60.3%)、女71例(39.7%);年龄(59.79±10.88)岁(范围27~84岁)。分析患者人口学特征、原发肿瘤类型、脊柱转移部位和节段、椎体病理骨折情况、神经功能Frankel分级、身体功能状态Karnofsky评分(Karnofsky performance scale,KPS)、疼痛视觉模拟评分(visual analogue score,VAS)、脊柱肿瘤不稳定评分(the spinal instability neoplastic score,SINS)、改良Bauer评分及NESMS评分、Tomita评分及改良Tokuhashi评分,评价NESMS评分、Tomita评分及改良Tokuhashi评分对脊柱转移瘤患者预后生存期的预测价值,并通过Cox比例风险回归模型分析影响脊柱转移瘤患者生存期的独立因素。结果61~75岁年龄段脊柱转移瘤构成比最高(45.3%,81/179);原发肿瘤类型中肺癌最多见(46.9%,84/179);40.8%(73/179)的病例有脊柱多发转移,单发转移患者中胸椎受累比例最高(26.3%,47/179);28.5%(51/179)合并内脏转移,52.0%(93/179)合并脊柱外骨转移,31.3%(56/179)存在受累椎体病理性骨折。114例接受手术治疗(63.69%),确诊后3、6、12个月死亡率分别为22.4%(40/179)、51.4%(92/179)、77.1%(138/179)。NESMS评分0~3分者中位生存期分别为3、4、8、10个月,生存期分别为(3.60±2.10)、(6.77±3.39)、(9.69±5.71)和(10.53±6.25)个月;1年内患者死亡率分别为100%(13/13)、87.5%(42/48)、71.6%(63/88)和66.7%(20/30)。NESMS评分、Tomita评分及改良Tokuhashi评分预测所有患者生存期的C指数分别为0.63、0.58和0.55;对脊柱转移瘤患者3个月(AUC=1.00,0.63,0.42)、6个月(AUC=0.71,0.63,0.45)生存期预测准确度,NESMS评分优于Tomita评分及改良Tokuhashi评分;预测12个月生存期时Tomita评分准确度优于NESMS评分及改良Tokuhashi评分(AUC=0.66,0.61,0.38)。多因素Cox比例风险回归模型分析显示原发肿瘤生长速度、神经功能Frankel评分、白蛋白水平及接受手术治疗是影响脊柱转移瘤患者生存时间的独立因素(P<0.05)。结论NESMS评分预测脊柱转移瘤患者预后生存的一致性和准确度优于Tomita评分和改良Tokuhashi评分,尤其在预测3和6个月生存期方面更具优势;原发肿瘤生长速度、神经功能Frankel评分、白蛋白水平及接受手术治疗是影响脊柱转移瘤患者生存时间的独立因素。

Objective To evaluate the clinical value of the New England spinal metastasis score(NESMS)in predicting the prognosis of patients with spinal metastases by retrospectively analyzing the medical records of multicenter spinal metastases in China.Methods The data of 179 patients with spinal metastases from January 2008 to December 2018 were retrospectively collected.There were 108 males(60.3%)and 71 females(39.7%)with an average age of 59.79±10.88 years old(range 27-84 years).The patient demographic characteristics,primary tumor type,spinal metastases and segments,vertebral pathological fractures,neurological Frankel classification,physical function status,Karnofsky performance scale(KPS),visual analogue score(VAS),the spinal instability neoplastic score(SINS),modified Bauer score,NESMS score,Tomita score and modified Tokuhashi score were collected.The clinical value of NESMS score,Tomita score and modified Tokuhashi score in predicting the survival of patients with spinal metastases were compared.The independent factors affecting survival in these patients were analyzed by Cox proportional hazards regression model.Results Among the 179 patients,the peak incidence of spinal metastases was in the age group of 61-75 years(45.3%,81/179)of all patients.Lung cancer was the most common primary tumor(46.9%,84/179).40.8%(73/179)of patients had multi-segment metastasisand thoracic spine was the most common site with single-site metastasis(26.3%,47/179).28.5%(51/179)of the patients had visceral metastases and 52.0%(93/179)of the patients had extraspinal bone metastases.31.3%(56/179)of the patients had pathological fractures of the involved vertebral bodies.114 patients received surgical treatment(63.4%).The mortality rates in 3-months,6-months and 1-year were 22.4%(40/179),51.4%(92/179)and 77.1%(138/179),respectively.The median survival time of patients with NESMS score of 0-3 was 3,4,8,and 10 months respectively with the mean survival time was 3.60±2.10,6.77±3.39,9.69±5.71 and 10.53±6.25 months.The 1-year mortality rates were 100%(13/13),87.5%(42/48),71.6%(63/88)and 66.7%(20/30)respectively.The consistency of NESMS score,Tomita score and modified Tokuhashi score in predicting survival of all patients was 0.63,0.58 and 0.55,respectively.For patients with spinal metastases,the NESMS score was better than the Tomita score and modified Tokuhashi score in predicting survival at 3-months(AUC=1.00,0.63,0.42)and 6-months(AUC=0.71,0.63,0.45).But the accuracy of Tomita score was best in predicting survival at 1-year(AUC=0.66,0.61,0.38).Multivariate Cox proportional hazards regression model analysis showed that growth rate of primary tumor,neurological function Frankel score,albumin level and surgical treatment were independent factors affecting the survival time of patients with spinal metastases(P<0.05).Conclusion The consistency and accuracy of NESMS score in predicting survival of patients with spinal metastases are better than Tomita score and modified Tokuhashi score,especially in predicting 3-and 6-month survival.The growth rate of primary tumor,Frankel classification,albumin level and surgical treatment were independent factors affecting the survival time of patients with spinal metastases.