Mekicar Jernej, Omejc Mirko
Department of Abdominal Surgery, University Medical Centre Ljubljana, Zaloska 7, Ljubljana, Slovenia.
Dig Surg. 2009;26(3):256-61. doi: 10.1159/000227296. Epub 2009 Jul 3.
An optimal management of gastric cancer should be stage-oriented with the main criterion of lymph node status. Nodal status can be assessed preoperatively with modern radiological imaging techniques, sentinel node biopsy, molecular or genetic markers of tumor and computer analysis. The aim of this study was to evaluate the feasibility and accuracy in preoperative prediction of lymph node status with the help of computer analysis.
The computer model was constructed with data from patients with gastric cancer treated at the University Medical Centre Ljubljana between 1993 and 2003. Prediction of the computer model was then tested on 110 patients treated between 2004 and 2005 in the same manner as patients from the database. The goal of the study was to compare prediction for lymph node status in the control group by the computer program with real data collected after surgery.
Accuracy of computerized preoperative predictions of N0/N1 status is 86% (sensitivity 91% and specificity 79%) and of N1/N2 status is 89% (sensitivity 87% and specificity 90%).
Results of accuracy of preoperative predictions of nodal status are very high and allow in each individual case to guide stage-appropriate therapy. Best applicability is obtained with implementation of this new approach with standard diagnostic methods.
胃癌的最佳治疗应根据分期进行，主要依据淋巴结状况。淋巴结状况可通过现代放射影像学技术、前哨淋巴结活检、肿瘤的分子或基因标志物以及计算机分析进行术前评估。本研究的目的是评估借助计算机分析术前预测淋巴结状况的可行性和准确性。
利用1993年至2003年在卢布尔雅那大学医学中心接受治疗的胃癌患者的数据构建计算机模型。然后，以与数据库患者相同的方式，对2004年至2005年接受治疗的110名患者测试计算机模型的预测。该研究的目的是将计算机程序对对照组淋巴结状况的预测与术后收集的实际数据进行比较。
计算机术前对N0/N1状况预测的准确率为86%（敏感性91%，特异性79%），对N1/N2状况预测的准确率为89%（敏感性87%，特异性90%）。
术前淋巴结状况预测的准确率非常高，能够在每个病例中指导合适分期的治疗。将这种新方法与标准诊断方法结合使用可获得最佳适用性。
