东南大学学报（医学版）
東南大學學報（醫學版）
동남대학학보（의학판）
JOURNAL OF SOUTHEAST UNIVERSITY(MEDICAL SCIENCE EDITION)
2015年
1期
92-95
,共4页
甲状腺癌%复发%危险因素%Cox回归模型
甲狀腺癌%複髮%危險因素%Cox迴歸模型
갑상선암%복발%위험인소%Cox회귀모형
thyroid cancer%recurrence%risk factors%Cox regression model
目的：：探讨甲状腺癌术后复发的相关危险因素,构建甲状腺癌复发的预测模型。方法：回顾性分析2009年6月至2012年12月手术治疗且经病理确诊的甲状腺癌患者的临床资料,并对患者的术后复发情况进行追踪随访；采用Kaplan-Meier法计算患者的术后复发率,绘制患者的无复发生存曲线；运用多因素Cox回归模型分析患者术后复发的独立危险因素,并建立复发预测模型。结果：共纳入符合条件的甲状腺癌患者376例,男86例,女290例；年龄16~81岁,平均(43.5±8.1)岁。随访1~58个月,中位随访时间24个月；随访期间97例(25.8%)患者复发,无复发生存时间1~58个月,1年、2年、3年累积无复发生存率分别为96.8%、92.5%、87.9%。多因素Cox回归模型分析发现,肿瘤最大直径≥4 cm、手术方式为单侧腺叶加峡部切除、病理类型为未分化癌、未作淋巴结清扫是甲状腺癌患者术后复发的独立危险因素(P<0.05)。预测模型为：h(t)=h0exp(3.798 x1+2.721 x2+5.972 x3+2.679 x4)。结论：肿瘤体积大、单侧腺叶加峡部切除术、病理组织学恶性程度高及未作淋巴结清扫是甲状腺癌患者术后复发的主要危险因素。
目的：：探討甲狀腺癌術後複髮的相關危險因素,構建甲狀腺癌複髮的預測模型。方法：迴顧性分析2009年6月至2012年12月手術治療且經病理確診的甲狀腺癌患者的臨床資料,併對患者的術後複髮情況進行追蹤隨訪；採用Kaplan-Meier法計算患者的術後複髮率,繪製患者的無複髮生存麯線；運用多因素Cox迴歸模型分析患者術後複髮的獨立危險因素,併建立複髮預測模型。結果：共納入符閤條件的甲狀腺癌患者376例,男86例,女290例；年齡16~81歲,平均(43.5±8.1)歲。隨訪1~58箇月,中位隨訪時間24箇月；隨訪期間97例(25.8%)患者複髮,無複髮生存時間1~58箇月,1年、2年、3年纍積無複髮生存率分彆為96.8%、92.5%、87.9%。多因素Cox迴歸模型分析髮現,腫瘤最大直徑≥4 cm、手術方式為單側腺葉加峽部切除、病理類型為未分化癌、未作淋巴結清掃是甲狀腺癌患者術後複髮的獨立危險因素(P<0.05)。預測模型為：h(t)=h0exp(3.798 x1+2.721 x2+5.972 x3+2.679 x4)。結論：腫瘤體積大、單側腺葉加峽部切除術、病理組織學噁性程度高及未作淋巴結清掃是甲狀腺癌患者術後複髮的主要危險因素。
목적：：탐토갑상선암술후복발적상관위험인소,구건갑상선암복발적예측모형。방법：회고성분석2009년6월지2012년12월수술치료차경병리학진적갑상선암환자적림상자료,병대환자적술후복발정황진행추종수방；채용Kaplan-Meier법계산환자적술후복발솔,회제환자적무복발생존곡선；운용다인소Cox회귀모형분석환자술후복발적독립위험인소,병건립복발예측모형。결과：공납입부합조건적갑상선암환자376례,남86례,녀290례；년령16~81세,평균(43.5±8.1)세。수방1~58개월,중위수방시간24개월；수방기간97례(25.8%)환자복발,무복발생존시간1~58개월,1년、2년、3년루적무복발생존솔분별위96.8%、92.5%、87.9%。다인소Cox회귀모형분석발현,종류최대직경≥4 cm、수술방식위단측선협가협부절제、병리류형위미분화암、미작림파결청소시갑상선암환자술후복발적독립위험인소(P<0.05)。예측모형위：h(t)=h0exp(3.798 x1+2.721 x2+5.972 x3+2.679 x4)。결론：종류체적대、단측선협가협부절제술、병리조직학악성정도고급미작림파결청소시갑상선암환자술후복발적주요위험인소。
Objective:To explore risk factors of thyroid cancer recurrence and establish prediction model for recurrent thyroid cancer. Methods: We retrospectively reviewed the patients diagnosed pathologically with thyroid cancer during the period from June, 2009 to December, 2012 and followed up the patients. Kaplan-Meier method was used to estimate recurrence rates and to plot relapse-free survival curves of patients at different levels of predictive factors. Multivariate Cox proportional hazards model was used to analysis inde-pendent risk factors of patients’ relapse and establish prediction model for recurrent thyroid cancer. Results:376 cases of thyroid cancer patients were included, male 86 cases, female 290 cases, age 16-81 years old, mean age (43. 5 ± 8. 1)years old, followed up 1 to 58 months, with a median follow-up time of 24 months. 97 cases of (25. 8%) patients relapse, relapse-free survival time from 1 to 58 months, one year, two years, three years cumulative recurrence-free survival rates were 96 . 8%, 92 . 5%, 87 . 9%. Multivariate Cox proportional hazards regression analysis found that maximum tumor diameter> 4 cm, lobectomy plus isthmus resection, undifferentiated carcinoma, did not undergo lymph node dissection were independent risk factors for recurrence with thyroid cancer ( P <0 . 05 ) . The risk of recurrence in patients with thyroid cancer function model expression was:h(t) =h0exp(3. 798 x1 +2. 721 x2 +5. 972 x3 +2. 679 x4). Conclusion:Tumor size, lobectomy plus isthmectomy, high degree of malignancy histopathology, and without lymph node dissection are the main risk factors for thyroid cancer recurrence.
