影响儿童肝母细胞瘤预后的多因素分析

Multivariate prognostic analysis of hepatoblastoma in children.

目的探讨影响肝母细胞瘤患儿预后的因素。方法收集2002年1月至2011年6月青岛大学医学院附属医院小儿血液科42例肝母细胞瘤患儿临床资料,其中未手术组8例,手术组34例,对其实验室检查、病理分型、治疗前病变范围(PRETEXT)分期及预后情况进行回顾性分析及随访。Kaplan-Meier法计算5年生存率并绘制生存曲线,Log-rank检验进行各因素亚组间生存率比较;Cox多因素回归对差异有统计学意义的因素进行多因素分析。结果 42例中存活29例,5年生存率为64.4%,其中手术组34例中死亡5例,5年生存率81.4%。血C反应蛋白(CRP)、甲胎蛋白(AFP)、乳酸脱氢酶(LDH)、病理分型及PRETEXT分期的不同亚组患儿生存率间的比较,差异均有统计学意义(P<0.05),其中CRP升高、LDH升高、AFP>10000μg/L、胚胎型、PRETEXTIV期及转移组的生存率明显降低(P<0.05);手术组Cox多因素分析示病理分型、分期及AFP水平对患儿生存率的影响差异均有统计学意义(P<0.05)。结论病理分型、PRETEXT分期及血清AFP水平为影响肝母细胞瘤患儿预后的独立因素。熟悉肝母细胞瘤临床和实验室检查特点并尽早明确诊断及分期,积极手术及规范化疗有利于改善患儿预后。

To explore the prognostic factors of hepatoblastoma （HB） patients in children. Methods For- ty-two patients diagnosed with hepatoblastoma between Jan.2002 and Jun.2011 in our department were divided into two groups ： 8 in the unoperated group and 34 in the operated group.The results of their laboratory examinations, pathology, staging and prognosis were reviewed and analyzed by SPSS 17.0.Kaplan-Meier method was used to evaluate survival rates and to draw the survival curve ; Log-rank test was used for univariate analysis and Cox regression was used for multivariate analysis. Results The 5-year survival rate in the total 42 cases and in the operated group was 64.4% and 81.4%, respectively. The univariate analysis showed that factors including elevated levels of CRP and LDH, AFP 〉 10 000μg/L, embryonal histology, PRETEXT 1V and presence of metastases could significantly reduce the 5-year survival rate.Cox multivariate analysis showed that the 5-year survival rate of patients in different subgroups of histology, PRETEXT staging and the AFP level were significantly different （P 〈 0.05）. Conclusion The histology, PRETEXT staging and the AFP level are independent factors influencing the prognosis of hepatoblastoma （HB）patients.Early diagnosis and active treatment based on the grasping of the clinical and laboratory characters of the disease can improve the prognosis of the patients.