肿瘤患者CT检查申请与报告模板共识
由意大利医学与介入放射学会(SIRM)与意大利肿瘤内科学会(AIOM)专家工作组通过多轮德尔菲法,制定肿瘤患者CT检查申请模板及分期、疗效评估、随访三种结构化报告模板。共识旨在规范放射科与肿瘤科之间的沟通,使非转诊中心也能实施高质量肿瘤影像管理。两轮后所有条目均达优秀评级,跨学会互评均给满分。
要点
- CT申请须包含肿瘤家族史、相关临床资料、分期、再分期、其他原因、随访六部分。
- 结构化报告按分期、疗效评估、随访分别采用九段、十段、六段模板撰写。
- 在无结构化系统条件下,也可参照模板逐段书写,可减少病灶遗漏并利于疗效对照。
核心结论
- SIRM与AIOM联合制定的CT申请与结构化报告模板能改善放射科与肿瘤科的沟通质量。
- 申请与报告条目经两轮德尔菲后均达到优秀评级,跨学会互评所有条目均获最高分。
研究设计与样本
- 研究类型为多轮德尔菲共识,由AIOM与SIRM专家工作组共同建立。
- CT申请模板分为六个部分;CT结构化报告按分期、疗效评估、随访分别制定三个模型。
关键影像表现/征象
- 分期报告为九部分,疗效评估报告为十部分,随访报告为六部分。
- 报告各部分按病灶分布包括原发肿瘤、淋巴结、肝、肺、脑、其他器官、腹膜、偶发发现与并发症、结论。
诊断标准与关键数值
- 第一轮分期、再分期、随访报告Cronbach α系数分别为0.92、0.95、0.90。
- 第一轮分期部分一致度最高条目为肺转移与偶发发现及并发症。
- 疗效评估一致度最高条目为治疗相关并发症与偶发发现。
- 随访一致度最高条目为病变出现与偶发发现及并发症。
- 第一轮申请模板随访部分均值为2.75±0.62,Cronbach α系数为0.75。
临床应用与阅片要点
- 日常阅片可按模板固定顺序逐项排查,减少对转移灶和非肿瘤性偶发发现的漏报。
- 疗效评估报告应单独列出治疗相关并发症与偶发发现,便于肿瘤科调整治疗方案。
局限性
- 原文未报告专家人数、轮次间隔、样本量及中心数等具体方法学信息。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND: Multidisciplinary management of oncological patients has improved patient outcomes, responding effectively and efficiently to the patient's health needs. A critical element remains adequate communication, also between radiologist and oncologist, which promotes correct patient management. The Italian Society of Medical Oncology (AIOM) and the Italian Society of Medical and Interventional Radiology (SIRM) have created a working group of representative members to improve not only communication between the two categories but, above all, to allow each member of the different categories to benefit from guidelines of good clinical practice both for filling out the examination request form than the radiological report, and therefore, also allowing clinicians and radiologists who do not work in reference centers, to correctly manage patients in the various phases of their oncological path.
MATERIALS AND METHODS: A panel of expert oncologists (AIOM members) and radiologists (SIRM members) was established. Multi-round consensus-building Delphi exercise was performed to create a comprehensive structured report (SR) template and a comprehensive requesting model for computed tomography (CT). The requesting model was divided into six sections: (a) oncological family history; (b) relevant clinical data; (c) staging; (d) re-staging; (e) other reason; and (f) follow-up. Regarding CT SR, 3 models were proposed: (1) for staging; (2) for treatment evaluation, and (3) for follow-up. CT SR in staging phase was divided into nine sections: (a) primary tumor; (b) lymph node metastases; (c) liver metastases; (d) lung metastases; (e) brain metastases; (f) other organs (incl. skeleton) metastases; (g) peritoneum; (h) incidental findings and complications; and (i) conclusion. CT SR in treatment evaluation phase was divided into ten sections: (a) primary tumor; (b) lymph node metastases; (c) liver metastases; (d) lung metastases; (e) brain metastases; (f) other organs (incl. skeleton) metastases; (g) peritoneum; (h) treatment-related complication; (i) incidental findings; and (l) conclusion. CT SR in follow-up phase was divided into six sections: (a) appearance of lesions; (b) node metastases; (c) liver lesions; (d) peritoneum; (e) incidental findings and complications; and (f) conclusion. Cronbach's alpha (Cα) correlation coefficient was used to evaluate internal consistency for each item and the quality analysis according to the average inter-item correlation. Each expert expressed individual comments for each specific template section by using a four-point scale (0 = strongly disagree, 1 = slightly disagree, 2 = modestly agree, 3 = strongly agree).
RESULTS: With regard to the 3 CT SR reports, at the first round, all sections achieved ratings above the "good" level. The staging items showing the highest level of agreement among experts in the first round were lung metastases and incidental findings and complications. For treatment evaluation, the items with the highest agreement were treatment-related complication and incidental findings. For follow-up, the items with the highest agreement were appearance of lesions and incidental findings and complications. At the first round, Cronbach's alpha (Cα) correlation coefficients were 0.92, 0.95, and 0.90 for staging, re-staging, and follow-up, respectively. At the second round, all sections achieved ratings at the "excellent" level. Regarding requesting model, at first round, all sections received an overall score equal to or greater than the level defined as "good" (score = 2). The item that showed the highest level of agreement in the first round was the follow-up section, with a mean value of 2.75 ± 0.62. The correlation coefficient, Cronbach's alpha (Cα), was 0.75. In the second round, for both the CT SR report and the oncologist request template, all elements received "excellent" ratings. During the inter-society agreement, six AIOM members assessed the CT SR model developed by SIRM, while six SIRM members evaluated the CT requesting model proposed by AIOM. In this single Delphi round, all participants assigned the maximum score (3) to every item across all sections.
CONCLUSIONS: The inter-society collaboration between AIOM and SIRM is a milestone in improving communication between radiologists and oncologists. The CT request and reporting documents meet the needs of quality care even outside of referral centers, and although do not represent an obligation, they can facilitate understanding between the different categories of professionals involved in patient management.
引用
- Granata, Serafini, Fusco, et al. Request and reporting models for computed tomography in the multidisciplinary management of cancer patients: consensus between the Italian Society of Medical and Interventional Radiology (SIRM) and the Italian Society of Medical Oncology (AIOM). Radiol Med. 2026. PMID: 42414784. doi:10.1007/s11547-026-02255-4
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。