Bonatti Matteo, Valletta Riccardo, Corato Valentina, Proner Bernardo, Hoxha Laurjan, Odoferdi Luca, Steinkasserer Martin, Ninkova Roberta Valerieva, Avesani Giacomo, Vingiani Vincenzo, Manganaro Lucia
Department of Radiology, Hospital of Bolzano (SABES-ASDAA), Teaching Hospital of Paracelsus Medical University (PMU), Bolzano, Italy.
Department of Gynecology and Obstetrics, Hospital of Bolzano (SABES-ASDAA), Teaching Hospital of Paracelsus Medical University (PMU), Bolzano, Italy.
Eur Radiol. 2025 Aug 20. doi: 10.1007/s00330-025-11923-4.
To assess the reproducibility and diagnostic accuracy of the Node Reporting and Data System 1.0 (Node-RADS) for detecting pelvic nodal metastases by endometrial cancer (EC) using CT and MRI, among readers with different levels of expertise.
This IRB-approved, single-center retrospective study included 128 patients with EC who underwent preoperative MRI at our Institution (Jan 2020-Dec 2023). Six readers with different levels of expertise in pelvic MRI (2 dedicated pelvic radiologists, 2 residents in their fourth year of training, and 2 residents in their second year of training) independently evaluated preoperative CTs and MRIs and assigned Node-RADS scores. Inter-observer agreement and inter-method agreement were calculated. Node-RADS was compared with post-surgical pathology data.
At surgery, pelvic nodal metastases were detected in 12.5% of the patients. Interobserver agreement in nodal status assessment using Node-RADS varied from κ = 0.783 to κ = 0.426 using MRI, and from κ = 0.936 to κ = 0.295 using CT, with worse results among less experienced readers. MRI and CT were concordant in the N definition in 94-98% of the cases. Using MRI, the most experienced readers showed 63% sensitivity and 100% specificity in the detection of nodal metastases, compared to 44% sensitivity and 96% specificity for poorly experienced readers. Using CT, the most experienced readers showed 50% sensitivity and 100% specificity; the less experienced readers showed 43% sensitivity and 94% specificity.
Node-RADS is a reproducible and accurate tool for locoregional nodal staging of EC, but only for readers with specific experience in pelvic imaging. MRI outperforms CT in nodal assessment.
Question Preoperative assessment of nodal metastases by EC is difficult, but it may help in tailoring the best surgical approach for each patient. Findings Node-RADS is a reliable tool for assessing the presence of pelvic nodal metastases by EC, both on CT and MRI, among experienced readers. Clinical relevance The use of Node-RADS among experienced readers enables detection of nodal metastases with good sensitivity and excellent specificity; MRI should be preferred over CT due to its higher sensitivity.
评估节点报告与数据系统1.0(Node-RADS)在不同专业水平的读者中,利用CT和MRI检测子宫内膜癌(EC)盆腔淋巴结转移的可重复性和诊断准确性。
本研究经机构审查委员会批准,为单中心回顾性研究,纳入了128例在我院(2020年1月至2023年12月)接受术前MRI检查的EC患者。6名在盆腔MRI方面具有不同专业水平的读者(2名专门的盆腔放射科医生、2名四年级住院医师和2名二年级住院医师)独立评估术前CT和MRI,并给出Node-RADS评分。计算观察者间一致性和方法间一致性。将Node-RADS与术后病理数据进行比较。
手术中,12.5%的患者检测到盆腔淋巴结转移。使用Node-RADS评估淋巴结状态时,观察者间一致性在使用MRI时κ值从0.783到0.426不等,在使用CT时κ值从0.936到0.295不等,经验较少的读者结果较差。MRI和CT在94%-98%的病例中N定义一致。使用MRI时,经验最丰富的读者检测淋巴结转移的灵敏度为63%,特异性为100%,而经验不足的读者灵敏度为44%,特异性为96%。使用CT时,经验最丰富的读者灵敏度为50%,特异性为100%;经验较少的读者灵敏度为43%,特异性为94%。
Node-RADS是一种用于EC局部区域淋巴结分期的可重复且准确的工具,但仅适用于在盆腔成像方面有特定经验的读者。在淋巴结评估中,MRI优于CT。
问题EC术前评估淋巴结转移困难,但有助于为每位患者制定最佳手术方案。发现Node-RADS是经验丰富的读者在CT和MRI上评估EC盆腔淋巴结转移存在情况的可靠工具。临床意义在经验丰富的读者中使用Node-RADS能够以良好的灵敏度和出色的特异性检测淋巴结转移;由于MRI灵敏度更高,应优先于CT使用。