摘要
目的验证基于西方人样本资料的Partin tables预测国人前列腺癌术后病理特征的准确性。方法2000年6月至2005年12月,连续167例临床局限性前列腺癌患者接受根治性耻骨后前列腺切除术,术后病理报告是否器官局限性癌及有无包膜侵犯、精囊侵犯或淋巴结转移。根据术前PSA、临床分期及Gleason评分,应用Partin tables预测上述病理特征。统计学处理采用受试者工作特征(ROC)曲线分析的方法检验Partin tables预测术后病理特征的准确性,ROC曲线下面积〉0.7时判断相应病理特征的预测准确性具有理想的临床使用价值。结果本组器官局限性癌、包膜侵犯、精囊侵犯及淋巴结转移的发生率分别为54%、33%、14%及10%。应用Partin tables预测本组患者器官局限性癌、包膜侵犯、精囊侵犯及淋巴结转移的ROC曲线下面积分别为0.713(95%CI,0.636-0.790)、0.665(95%CI,0.579-0.750)、0.810(95%CI,0.717-0.903)、0.768(95%CI,0.641-0.895)。结论Partin tables对国人前列腺癌术后器官局限性癌、精囊侵犯及淋巴结转移3个病理特征预测较准确,但对包膜侵犯的预测准确性较低。
Objective To validate the accuracy of Partin tables in prostate cancer Chinese population. Methods From June 2000 to December 2005, 167 consecutive patients who were diagnosed clinically localized prostate cancer and without preoperative treatment were treated with radical retro pubic prostatectomy. Prediction made with Partin tables using pretreatment PSA, Gleason score and clinical stage were compared with actual pathologic features. Receiver operating characteristic (ROC) analysis was performed to assess predictable value of the tables. Results The pathological outcomes analysis revealed an organ confined disease rate of 54 % and the established capsular penetration, semi nal vesicle involvement and lymph node involvement rate were 33%, 14%, and 10%, respectively. The area under the ROC (AUC) for organ-confined disease prediction was 0.713. The AUC for established capsular penetration, seminal vesicle involvement and lymph node involvement predictions were 0.665, 0.810, and 0.768, respectively. Conclusions The Partin tables are able to accurately predict the pathologic features in our study population, especially for the predictions of organ confined disease, seminal vesicle involvement and lymph node involvement.
出处
《中华泌尿外科杂志》
CAS
CSCD
北大核心
2007年第9期614-618,共5页
Chinese Journal of Urology