非小细胞肺癌病人术后生存分析的死亡风险分组研究

Risk grouping after survival analysis of non-small-cell lung cancer surgery patients

目的通过非小细胞肺癌手术病人的生存分析获得死亡风险得分,进行死亡风险分组,区分不同死亡风险的病人,为提高病人的术后生存率提供信息。方法收集非小细胞肺癌病人的病历资料及随访数据,利用Cox比例风险模型进行多因素分析,模型中保留P<0.10的变量;根据模型中变量的系数分别赋予相应的风险得分并进行病人分组;利用Kaplan-Meier法估计各组生存曲线。结果获得非小细胞肺癌病历660份并随访5年,5年生存率为55%;Cox模型中保留的变量为年龄,肿瘤直径,病理类型,病理分期,系统性淋巴结清扫;根据风险得分获得低、中、高和极高4个风险组;低风险组5年生存率为66.4%,中等风险组53.4%(OR=1.72),高风险组43.8%(OR=2.53),极高风险组19.2%(OR=8.29),中、高、极高风险组与低风险组相比,差异有统计学意义。结论进行非小细胞肺癌生存分析,并根据死亡风险得分进行分组,有效地区分了不同死亡风险的病人。根据死亡风险分组,为临床上制定不同死亡风险病人的术后治疗方案,预防病人术后死亡提供信息,从而改善病人的术后生存。

Objective We obtained risk score through survival anal）,sis of non-small-cell lung cancer surgery patients, and grouped death risk and discriminated patients with different death risk, so as to provide information to improve postoperative survival of patients. Methods We collected non-small-cell lung cancer medical records and follow-up data, and used Cox proportional hazards model for multivariate analysis, with the final model including factors with P〈0.10. The risk score was designed from coefficients of the significant factorl and patients were divided into different groups according to their total score. Survival curves were estimated using Kaplan-Meier method. Results We collected 660 consecutive medical records and follow-up data in 5 years, and the overall 5-year survival rate was 55%. The final Cox regression model included age, tumor size, patho- logical types, pathological stage and systematic nodal dissection. Patients were grouped into low, medium, high and very high risk groups according to the risk scores. Five-year survival rates of the four groups were 66.4% in the low risk group, 53.4% （OR = 1.72） in the medium group, 43.8% （OR = 2.53 ） in the high risk group and 19.2% （OR = 8.29） in the very high risk group respectively, and the difference was statistically significant different when the low risk group was compared with the other three groups. Conclusion We conducted survival analysis of non-small-cell lung cancer surgery patients and divided patients into different groups according to death risk scores, so as to differentiate patients of different death risks effectively. According to death risk groups, we can provide information for clinic for developing postoperative treatment, preventing postoperative death, and improving survival of non-small-cell lung cancer surgery patients.