Heijnen Luc A, Maas Monique, Lahaye Max J, Lalji Ulrich, Lambregts Doenja M J, Martens Milou H, Riedl Robert G, Beets Geerard L, Beets-Tan Regina G H
Department of Radiology, Maastricht University Medical Centre, PO Box 5800, 6202 AZ, Maastricht, The Netherlands.
Eur Radiol. 2014 Aug;24(8):1845-52. doi: 10.1007/s00330-014-3231-9. Epub 2014 Jun 5.
Our primary objective was to evaluate diagnostic performance of gadofosveset T1-weighted magnetic resonance imaging (T1W MRI) for discriminating between ypT0-2 and ypT3-4 tumours after chemoradiation therapy (CRT) for rectal cancer compared with T2W MRI for a general and expert reader. Second objectives included assessing the value of multiplanar reformatting (MPR) and interobserver agreement.
A general and expert reader evaluated 49 patients for likelihood of ypT0-2 tumour after CRT on T2W, gadofosveset T1W MRI, and gadofosveset T1W MRI + T2W MRI. The general reader scored with and without MPR. Confidence level scores were used to construct receiver-operating characteristic (ROC) curves. Area under the curve (AUC) values and diagnostic parameters were calculated and compared.
Gadofosveset T1W MRI + T2W MRI showed slightly superior sensitivity than T2W MRI for the general but not the expert reader. Specificity was higher for the expert on gadofosveset T1W MRI only compared with T2W MRI only (100% vs. 82%). MPR did not increase diagnostic performance. Interobserver agreement was highest for the combination of gadofosveset-enhanced T1W imaging plus T2W MRI.
The sole use or addition of gadofosveset-enhanced T1W MRI to T2W MRI did not increase significantly diagnostic performance for assessing ypT0-2 tumours. Adding gadofosveset-enhanced T1W MRI slightly increased sensitivity for the general reader and specificity for the expert reader, but this increase was not significant for more accurate clinical decision making. MPR did not improve diagnostic performance.
• ycT restaging with MRI in rectal cancer is challenging. • Gadofosveset-enhanced T1W MRI has shown promise for nodal restaging. • Gadofosveset-enhanced T1W MRI did not significantly increase diagnostic performance for assessing ypT0-2-tumours. • Addition of the gadofosveset sequence to T2W MRI slightly increased sensitivity for the general reader. • MPR did not improve diagnostic performance of ycT staging.
我们的主要目的是评估钆弗塞特T1加权磁共振成像(T1W MRI)在直肠癌放化疗(CRT)后鉴别ypT0 - 2和ypT3 - 4肿瘤方面的诊断性能,并与T2W MRI进行比较,评估对象包括普通读者和专业读者。次要目的包括评估多平面重组(MPR)的价值以及观察者间的一致性。
一名普通读者和一名专业读者对49例患者在CRT后基于T2W、钆弗塞特T1W MRI以及钆弗塞特T1W MRI + T2W MRI评估ypT0 - 2肿瘤的可能性。普通读者在有和没有MPR的情况下进行评分。使用置信水平评分构建受试者操作特征(ROC)曲线。计算并比较曲线下面积(AUC)值和诊断参数。
对于普通读者而非专业读者,钆弗塞特T1W MRI + T2W MRI显示出比T2W MRI略高的敏感性。仅在钆弗塞特T1W MRI上,专业读者的特异性高于仅使用T2W MRI时(100%对82%)。MPR并未提高诊断性能。钆弗塞特增强T1W成像加T2W MRI组合的观察者间一致性最高。
单独使用或在T2W MRI基础上加用钆弗塞特增强T1W MRI在评估ypT0 - 2肿瘤时并未显著提高诊断性能。加用钆弗塞特增强T1W MRI对普通读者的敏感性略有提高,对专业读者的特异性略有提高,但这种提高对于更准确的临床决策而言并不显著。MPR并未改善诊断性能。
• 直肠癌MRI的ycT再分期具有挑战性。• 钆弗塞特增强T1W MRI在淋巴结再分期方面显示出前景。• 钆弗塞特增强T1W MRI在评估ypT0 - 2肿瘤时并未显著提高诊断性能。• 在T2W MRI上加用钆弗塞特序列对普通读者的敏感性略有提高。• MPR并未改善ycT分期的诊断性能。