癌症電腦斷層或磁振造影診斷報告之AJCC TNM分期之一致性分析研究：以外科病理報告做爲參考依據

Analysis of Uniform Diagnostic Methods on AJCC TNM Staging by CT and MRI for Cancer Patients: Using Surgical Pathological Staging as Gold Standard

何子龍(Tzu-Lung Ho) ； 潘師涵(Shih-Han Pan) ； 詹慧倫(Hui-Lun Zhan) ； 徐守民(Soa-Min Hsu) ； 劉時佐(Shi-Zuo Liu)

《中華放射線醫學雜誌》 35卷4期 (2010/12) Pp. 209-214

摘要

提升癌症電腦斷層(CT)與磁振造影(MRI)的影像診斷品質，有兩個主要的目標：第一、「報告內容的規格化」，即制定報告內容的必要項目，並且要求放射診斷科醫師必須依照這些必要項目來完成CT或MRI的影像報告，其中須包括腫瘤大小、腫瘤位置、區域淋巴腺轉移、遠端轉移、及AJCC的TNM分期。這個部分的整體達成率，我們稱之為「報告內容規格化」的「完整性」。第二、「診斷方法的一致性」，即放射診斷科醫師之間，對於癌症的影像診斷方法應該有一致性，包括對腫瘤大小的測量方法、腫瘤是否侵犯周圍組織？是否有區域淋巴結轉移？以及AJCC的分期等等，應該都有一致性的診斷方法。為了達成這兩項目標，我們追蹤3個月的50個癌症確診病例，抽審其中的25例（抽審比率50%），召開兩次放射診斷科同儕審查會議，一方面檢查「報告內容規格化」的完整性（達成率80%），另一方面則依據外科病理報告的AJCC TNM分期，經由開會討論尋求放射診斷科醫師之間，對於癌症的影像診斷方法的一致性。

並列摘要

There are two goals in the pursuit of improving the image diagnostic quality of Computer-assisted Tomography (CT) and Magnetic Resonance Imaging (MRI) for cancer patients. The first one is ”formulation of image report”, meaning to set up the mandatory parts in a report and to ask all radiologists to contain these mandatory parts in their CT and MRI reports for a cancer patient. The mandatory parts should include tumor size, tumor location, regional lymph node metastasis, distant metastasis, and American Joint committee on Cancer (AJCC) Tumor-Node-Metastasis (TNM) staging. We call it ”completeness” for the overall completion rate of the ”formulation of image report”. The second goal is ”uniform diagnostic methods”, meaning uniform diagnostic methods among radiologists in the CT and MRI image diagnosis for cancer patients, including the way of tumor size measurement, whether there is tumor invasion to nearby structures or organs, whether there is regional lymph node metastasis, and AJCC TNM staging. Radiologists should have the same methods to determine these issues. To reach these two goals, we followed up 50 cancer patients who had definite surgical pathological diagnosis and randomly picked 25 cases for peer review (50% sampling rate). We had two radiological peer review meetings. On the first meeting we examined the completeness of the ”formulation of image report”, and found our completion rate was 80%. On the second meeting with discussion we sought the common opinions and ”uniform diagnostic methods” in the CT and MRI image diagnosis for cancer patients, using surgical pathological AJCC TNM staging as our gold standard.