卵巢癌患者预后因素的Cox回归分析

背景与目的：卵巢癌发病率高，生存率低，远期疗效依然不理想。本研究目的是探讨影响卵巢癌患者生存预后的相关因素。方法：收集1999～2004年在陕西省肿瘤医院收治的103例卵巢癌患者资料进行回顾性研究。随访100个月．以寿命表法计算生存率，变量分析以Kaplan-Meier法进行分析，用log-rank检验生存率差异，采用Cox多因素模型进行多因素分析，并对预后生存指标进行评价。结果：卵巢癌患者1年、3年、5年生存率分别为87．7％、50．8％、31．1％。单因素分析显示临床分期、分化程度、手术残余瘤的大小和化疗疗程是影响卵巢癌患者预后的因素：Cox比例风险模型分析显示，临床分期、分化程度、手术残余瘤的大小和化疗疗程是影响预后的指标。结论：卵巢癌在早期尽可能地切除病灶．术后残留灶〈2cm，术后足量疗程的化疗可改善预后。

Prognostic analysis of ovarian cancer using the Cox regression model

Background and Objective： cause of death among gynecologica Ovarian cancer is the most common malignancy. This study was to investigate the prognostic factors for ovarian cancer using the Cox regression model. Methods： Data of 103 patients, initially operated at Shaan＇xi Tumor Hospital from 1999 to 2004, were retrospectively analyzed. All patients were operated by experienced surgeons, and underwent adjuvant chemotherapy afterwards. Follow-up lasted for 100 months. Kaplan-Meier survival curves, long-rank test and the Cox regression model were used for analysis. Results： The 1-, 3- and 5-year survival rates of patients with ovarian cancer were 87.7%, 50.8% and 31.1% , respectively. Univariate analysis and the Cox regression model showed that clinical stage, differentiation, tumor residues and chemotherapy cycle were the major prognostic factors of ovarian cancer. Conclusions： Sufficient removal of the ovarian cancer at the early stage, residual tumor of less than 2 cm in diameter and sufficient adjuvant chemotherapy after surgery could improve the survival rates of ovarian cancer.