Department of Radiology, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada.
Department of Radiology, University of British Columbia, Vancouver, British Columbia, Canada.
J Ultrasound Med. 2021 Dec;40(12):2581-2593. doi: 10.1002/jum.15644. Epub 2021 Feb 12.
To demonstrate the usefulness of contrast-enhanced ultrasound (CEUS) for the evaluation of focal liver masses via a direct comparison to standard ultrasound and computed tomography/magnetic resonance imaging (CT/MRI).
A cohort of 214 patients with previously undiagnosed focal liver masses were included from 5 different centers. Each patient was imaged using CEUS and CT and/or MRI. Anonymized and randomized images were interpreted by 4 separate blind readers from 3 of the participating centers (2 readers for CEUS and 2 readers for CT/MRI). Readers were blinded to patient demographics and past medical history. Readers were asked to decide if the lesion was benign or malignant, provide a final diagnosis for the lesion, and provide a confidence interval. Results were compared to truth standard from pathology or expert consensus.
In determination of malignancy, CEUS had a sensitivity of 95%, specificity of 82%, PPV of 82%, NPV of 95%, statistically better than standard ultrasound (sensitivity 82%, specificity 56%, PPV 60%, NPV 78%) with P < .01 and not statistically different from CT (sensitivity 90%, specificity 73% PPV 81%, NPV 86%) or MRI (sensitivity 85%, specificity 79%, PPV 68%, NPV 91%) with P ≥ .01. In assigning a final diagnosis, CEUS had an accuracy of 78% statistically better than standard ultrasound (46%) with P < .01 and not statistically different from CT (68%) or MRI (71%) with P > .01.
In the evaluation of focal liver lesions, both for determination of malignancy and in accuracy of final diagnosis, CEUS performs better than standard ultrasound and at least equivalent to both CT and MRI.
通过与标准超声和计算机断层扫描/磁共振成像（CT/MRI）的直接比较，展示对比增强超声（CEUS）在评估局灶性肝肿块中的实用性。
从 5 个不同的中心纳入了 214 名患有先前未确诊的局灶性肝肿块的患者。每位患者均接受了 CEUS 和 CT 和/或 MRI 检查。来自 3 个参与中心的 4 位独立盲法读者对匿名和随机图像进行了解读（CEUS 有 2 位读者，CT/MRI 有 2 位读者）。读者对患者的人口统计学和既往病史不知情。读者被要求判断病变是良性还是恶性，为病变提供最终诊断，并提供置信区间。结果与病理或专家共识的真实标准进行比较。
在判断恶性肿瘤方面，CEUS 的灵敏度为 95%，特异性为 82%，PPV 为 82%，NPV 为 95%，明显优于标准超声（灵敏度 82%，特异性 56%，PPV 60%，NPV 78%），差异有统计学意义（P<0.01），与 CT（灵敏度 90%，特异性 73%，PPV 81%，NPV 86%）或 MRI（灵敏度 85%，特异性 79%，PPV 68%，NPV 91%）无统计学差异（P≥0.01）。在确定最终诊断方面，CEUS 的准确率为 78%，明显优于标准超声（46%），差异有统计学意义（P<0.01），与 CT（68%）或 MRI（71%）无统计学差异（P>0.01）。
在评估局灶性肝病变时，CEUS 在判断恶性肿瘤和最终诊断准确性方面均优于标准超声，且与 CT 和 MRI 至少具有同等的效果。
