自动乳腺超声与手持超声作为乳腺MRI二次检查的诊断效能
本研究为单中心前瞻性研究,纳入54例患者的66个MRI检出乳腺病灶,比较ABUS与HHUS作为二次超声技术的检出效能。HHUS检出率84.9%(56/66),高于ABUS的69.7%(46/66),差异有统计学意义(p=0.010)。两者敏感度均为90%,但ABUS阳性预测值(39.1%)高于HHUS(32.1%)。
要点
- HHUS对MRI检出病灶的二次超声检出率高于ABUS,应优先使用HHUS进行二次评估。
- ABUS阳性预测值更高,仅ABUS检出的病灶中恶性比例较高,需重视ABUS额外检出灶。
- 两种方法均未检出的7个病灶无一为恶性,双阴性结果支持随访观察。
核心结论
- HHUS在二次超声评估中对MRI检出病灶的检出率优于ABUS,但两者敏感度均为90%,癌症检出率相似。
- ABUS的阳性预测值(39.1%)高于HHUS(32.1%),主要因仅ABUS检出的病灶中恶性比例较高。
- HHUS与ABUS具有互补性,在同时具备两种设备的机构联合使用可提高病灶总检出率。
研究设计与样本
- 单中心前瞻性研究,纳入54例患者共66个乳腺MRI检出病灶,均转诊行MRI引导活检。
- 所有患者均接受HHUS和ABUS二次超声评估,由不同放射科医生操作并按BI-RADS词典评估。
- 参考标准为组织病理或影像随访;超声检出病灶行超声引导活检或术前定位,余者行MRI引导活检或定位,拒绝者行MRI随访至少2年。
关键影像表现/征象
- HHUS检出66个病灶中的56个(84.9%),ABUS检出46个(69.7%)。
- 仅HHUS检出的13个病灶中2个(15.4%)为恶性;仅ABUS检出的3个病灶中2个(66.7%)为恶性。
- 两种方法均未检出的7个病灶(10.6%)中无一为恶性。
诊断标准与关键数值
- HHUS对MRI检出病灶的检出率为84.9%(56/66),ABUS为69.7%(46/66),差异有统计学意义(p=0.010)。
- HHUS与ABUS的敏感度均为90%。
- ABUS的阳性预测值为39.1%,HHUS为32.1%。
- 病灶大小、深度、类型(肿块/非肿块)、位置、实质密度、动力学特征及形态学表现对两种方法检出率的影响均无统计学差异(p>0.05)。
临床应用与阅片要点
- 对乳腺MRI检出的可疑病灶行二次超声时,应首选HHUS以获得更高检出率。
- HHUS未检出而ABUS检出的病灶恶性比例较高,ABUS可作为HHUS的有益补充。
- 两种方法均未检出的病灶恶性可能性低,可结合MRI特征考虑短期随访而非立即活检。
局限性
- 单中心研究,样本量较小,共54例患者66个病灶。
- ABUS与HHUS由不同放射科医生操作,存在操作者依赖性。
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
RATIONALE AND OBJECTIVES: To compare the efficacy of automatic breast ultrasound (ABUS) with handheld ultrasound (HHUS) as second-look US techniques in the identification of lesions that are detected on breast MRI, and to determine factors that affect lesion detection on second-look US examinations.
MATERIAL AND METHODS: This single-center prospective study included 54 patients with 66 MRI-detected breast lesions referred for MRI-guided biopsy. All patients underwent second-look evaluation with both HHUS and ABUS. Histopathology or imaging follow-up served as the reference standard.US-guided biopsy or preoperative surgical localization was performed for lesions detected by US; while the rest of the patients underwent MRI-guided biopsy or localization procedures. Patients who refused to undergo MRI-guided procedures were followed-up with MRI for at least 2 year. ABUS and HHUS examinations were performed by different radiologists and evaluated based on BI-RADS lexicon.
RESULTS: Out of the 66 lesions detected on MRI examinations, 30 (45.5%) were benign, 16 (24.2%) were high-risk lesions, and 20 (30.3%) were malignant. HHUS demonstrated 56/66 (84.9%), while ABUS demonstrated 46/66 (69.7%) of them; and the difference was statistically significant (p=0.010). Two out of 13 (15.4%) lesions detected only by HHUS, and 2/3 (66.7%) lesions detected only by ABUS were malignant. None of the 7 lesions (10.6%), that could not be detected by either method, were malignant. There was no statistically significant difference between the number of lesions detected on HHUS and ABUS in terms of lesion size, depth, lesion type (mass/non-mass), lesion localization, parenchymal density, kinetic features or morphological findings (p>0.05). Sensitivity was similar for HHUS and ABUS (90% for both). However, the positive predictive value (PPV) was higher for ABUS (39.1%) than for HHUS (32.1%) (notably due to a higher proportion of malignant lesions among ABUS-only detected findings).
CONCLUSION: HHUS was superior to ABUS in detecting lesions in second-look US evaluation. However, cancer detection rates were similar, resulting in a higher positive predictive value for ABUS. Our results show that the two methods are complementary to each other and have the potential to increase lesion detection rate when used together in clinics where both methods are available.
引用
- Parlakkilic, Esen, Tokat, et al. Diagnostic Efficiency of Automatic Breast Ultrasound and Handheld Breast Ultrasound as Second Look Ultrasound Techniques for Suspicious Lesions Detected on Breast MRI. Acad Radiol. 2026. PMID: 41580317. doi:10.1016/j.acra.2025.12.056
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。