Significant benefit of multimodal imaging: PET/CT compared with PET alone in staging and follow-up of patients with Ewing tumors

Significant benefit of multimodal imaging: PET/CT compared with PET alone in staging and follow-up of patients with Ewing tumors
复制标题

DOI:
10.2967/jnumed.107.045286
复制
发表时间:
2007-12-01
影响因子:
9.3
通讯作者:
Franzius, Christiane
Franzius, Christiane
中科院分区:
医学1区
文献类型:
--
作者:
Gerth, Hans U.;Juergens, Kai U.;Franzius, Christiane

文献摘要

被引文献

相似文献

在尤文肿瘤患者的分期和再分期中,将混合PET/CT与单独使用PET进行比较,以评估联合成像技术的益处。方法:对53例经手术病理证实的尤文瘤患者共163例(年龄4~38岁,中位数16.5岁)进行163F-18FDGPET/CT检查。所有的PET/CT研究包括用于衰减校正的低剂量CT;在91次检查中,还进行了诊断性胸部CT检查。PET和CT数据分别由2名核医学医生和2名放射科医生独立评估。最后,使用软件对两个数据集进行融合,并由所有4位评价者进行分析(共识阅读)。以5分为标准对每个病变进行评分。活检、成像或临床随访作为参考标准。接收器工作特性(ROC)分析用于评估PET和PET/CT的性能特性。为了测量发现和正确定位肿瘤病灶的能力,绘制了PET的定位ROC(L-ROC)曲线。结果:仅用PET就能检出609个病变。PET/CT混合技术导致160个病变(26%)评分改变,其中23个病变评分较高(4%),137个病变评分较低(23%)。PET发现49个病灶(8%),图像融合后需要改变定位。此外,PET/CT发现的病变比单独使用PET多124个(21%),总共有733个病变。根据基于病变的分析,正电子发射计算机断层扫描的灵敏度、特异度和准确度分别为71%、95%和88%;而联合应用正电子发射计算机断层扫描/CT技术的相应数值分别为87%、97%和94%(P<0.0001)。ROC分析的曲线下面积分别为0.82和0.92(P<0.0001),L-ROC分析的曲线下面积为0.66。结论:PET/CT对尤文肿瘤的诊断和定位明显优于单独应用PET,并能改善患者的分期。混合技术在敏感性、特异性和准确性方面优于单独的PET,这主要是因为它可以检测到新的病变。
Hybrid PET/CT was compared with PET alone in the staging and restaging of patients with Ewing tumor to assess the benefit of the combined imaging technique. Methods: A total of 163 F-18-FDG PET/CT studies performed in 53 patients (age: range, 4-38 y; median, 16.5 y) with histopathologically confirmed Ewing tumor were evaluated retrospectively. All PET/CT studies included low-dose CT for attenuation correction; in 91 examinations, additional diagnostic chest CT was performed. PET and CT data were assessed independently by 2 nuclear medicine physicians and 2 radiologists, respectively. Finally, both datasets were fused by use of software and analyzed by all 4 reviewers (consensus reading). Each lesion was scored with a 5-point scale. Biopsy, imaging, or clinical follow-up served as a standard of reference. Receiver operating characteristic (ROC) analyses were performed to evaluate PET and PET/CT performance characteristics. To measure the abilities to detect and correctly localize tumor foci, localization ROC (L-ROC) curves were generated for PET. Results: A total of 609 lesions were detected by PET alone. The hybrid PET/CT technique resulted in a change of score in 160 of these lesions (26%): higher scores in 23 lesions (4%) and lower scores in 137 lesions (23%). In 49 lesions detected by PET (8%), the localization had to be changed after image fusion. Additionally, 124 (21%) more lesions were found by PET/CT than by PET alone, resulting in a total of 733 lesions. As determined by lesion-based analysis, the sensitivity, specificity, and accuracy of PET were 71%, 95%, and 88%, respectively; the corresponding values for the hybrid PET/CT technique were 87%, 97%, and 94% (P < 0.0001). The areas under the curve in the ROC analysis were 0.82 for PET and 0.92 for PET/CT (P < 0.0001), and that in the L-ROC analysis was 0.66 for PET. Conclusion: PET/CT is significantly more accurate than PET alone for the detection and localization of lesions and improves staging for patients with Ewing tumor. The hybrid technique is superior to PET alone in terms of sensitivity, specificity, and accuracy, mainly because of the detection of new lesions.