【作者】 王立平；

【导师】 蒋天安；

【作者基本信息】 浙江大学 ， 影像医学与核医学， 2012， 硕士

【摘要】 背景乳腺超声检查因为其安全性、稳定性及可靠性目前以成为临床筛查与诊断乳腺疾病的最常用的方法之一。2003年美国放射学院(ACR)借签了乳腺X线摄影的成功经验,创立并推荐了超声版的乳腺影像报告及数据系统(US-BI-RADS),其采用乳腺肿块的形态、边缘、微小钙化等主要分级依据在超声判断乳腺肿块性质中得到了广泛的证实和承认;超声BI-RADS级不仅很好的解决了乳腺超声描述及诊断规范化的问题,而且其阶梯式的评级标准对乳腺疾病的临床处理提供了重要参考依据。根据美国放射学院的推荐标准,超声对乳腺疾病的不同的BI-RADS分级(USCategory)有着不同的临床评估和管理(Assessment and Management):超声判断乳腺BI-RADS3级(良性可能)建议定期随访,而BI-RADS4级及其上的患者则需穿刺活检或手术切除。所以超声BI-RADS的准确性对临床医疗行为指导及患者疾病风险的控制有着重要的意义。临床和患者都期待超声医生能对乳腺肿物的性质与评分的做出正确的匹配。超声对乳腺肿块可能恶性的回顾性总结和研究比较多,且临床对BI-RADS4级及以上者多能引起足够的重视而得到相应的临床处理；但是超声对乳腺结节良性(BI-RADS3级)判断的可靠性研究却很少见报导,且基于影像学良性的判断,经常出现临床及患者对于乳腺肿物的重视不足、处理保守致病理缺失、失随访率高的特点。2003年美国放射学院推出BI-RADS超声版分级时直接引用X线摄影版本的数据,只含糊的推断是恶性率(PPV)应当小于2%的概率区段(should have a risk of malignancy of less than2%),国外的相关报导比较零星,且统计样本量较少。笔者没有检索到国内有关于BI-RADS3可靠性的研究的相关报导,且国内女性与西方白色人种女性在乳房体积、腺体层厚度及乳腺癌发病率方面存在较为明显差异性,我们亦需要国人自己的数据。所以本研究的目的是在基于大样本的病例的求证,以评价超声诊断乳病灶特别是国内女性乳腺病灶良性的可靠性并且分析影响其乳腺肿块BI-RADS分级的各项可能因素及其程度。方法对参照BI-RADS超声影像学特征描述初步分级为可能良性的乳腺结节(BI-RADS3级),并且有明确病理结果或有二年以上随访记录的患者共792例,进行回顾性研究超声诊断乳腺良性结节的阳性预测值；并且结合已由文献推荐的新的分级标准重新对研究对象超声图像重新判读,探讨可能影响其BI-RADS超声分级的多个因素(患者年龄、病灶大小、病灶数目、体表触诊、医师经验)。结果792例病例中,781例(98.6%)为良性,11例(1.4%)为恶性。重新判读后：35.1%(278/792)的病例被调升至BI-RADS4级,恶性肿瘤的检查出为81.8%(9/11)；低年资组判断的良性结节较高年资组者易被调升至4级(P<0.05)；年龄≥40岁或多发病灶的患者其乳腺结节被调至为4级的可能性大(P<0.01)；乳腺病灶能否触及、病灶大小则对其BI-RADS分级没有明显的影响结论超声乳腺结节BI-RADS3级的阳性预测值(PPV)为1.4%,超声判断乳腺结节良性的可信性较高,医师经验、乳腺结节是否多发以及患者年龄对超声判断良性病灶有影响。更多还原

【Abstract】 BackgroundBecause of its safety、stability and reliability, ultrasound exanimation has become one of the most common methods in clinical screening and diagnosis of breast diseases.In2003, American College of Radiology (ACR) drew lessons from the successful experience of mammography, founded and recommended the ultrasonic version of the Breast Imaging Reporting and Data System (BI-RADS-US), the rules of main categories depend on the characteristics of breast nodes, such as shape、edge and micro-calcification etc.the capability of judging breast diseases by ultrasound has got been widely confirmed and recognized. Ultrasound BI-RADS classification is not only solve the problems of the breasts ultrasonic description and diagnosis standardization, and its ladder-like classification standards provide an important reference for the treatment of breast disease.According to the recommended standards of the American College of Radiology, different BI-RADS-US categories have corresponding different clinical assessment and management.Patients suffer from breast mass of BI-RADS3are suggested regular follow-up, BI-RADS4or higher need to biopsy or surgical resection. Therefore, the accuracy of BI-RADS-US category has a great significance for guidance of clinical behavior and risk control of the disease. Clinical colleague and patients hope for the ultrasound radiologists make the right match of nature and scoring of breast masses.There are many retrospective reviews and studies on ultrasonic characteristics of probable malignant breast lumps, patient with BI-RADS4or higher are able to evoke enough alertness and get corresponding clinical treatment usually; however,there are rare reports on the reliability research of BI-RADS3category (probably benign),and the benign judgment based on imaging without pathological confirmation, often make the clinical colleague and patients pay few attention and generally adopt conservative treatment without clear pathological result,out of trace rate high. in2003edition, the ultrasound version was first introduced by ACR,the PPV directly quoted form the number value of X-ray mammography version, just vague infer that the probability of malignant ratio should be less than2%. foreign reports sporadic, and the statistical sample size small. The author didn’t found domestic report on the reliability of the BI-RADS3, and domestic females and white females in the West there was obvious difference in breast size, gland thickness and breast cancer incidence rate, we also need our own data. The purpose of this study is based on the confirmation of a large sample of cases to evaluate the reliability of ultrasound diagnosis of breast lesions, especially the domestic female benign breast lesions. and analyze the all possible factors and extent of which impact of the BI-RADS classification of breast masses.MethodsThis study contains a total of792cases with initial category3assessments on BI-RADS-Ultrasound categorization consistent with more than2years follow-up or pathologic results. We not only retrospectively study the reliability of probable benign lesions; but also explore several factors that could influence the categorization such as age, breast lumps size, multiplicity, palpability, and US radiologist’s experience referring to ACR-BI-RADS and the new categorizations recommendation.Resultsof792breast nodules,781(98.6%) were benign and11(1.4%) were malignant. If the new categorizations recommendation was applied,35.1%(278/792) of all cased would have been upgraded to BI-RADS4,81.8%(9/11) of malignant tumors would have been found; The breast masses diagnosed by the junior radiologists were more frequently upgraded to BI-RADS4than senior radiologists’ findings(P<0.05), the nodules in women40years or older and multiple lesions were more frequently upgraded to category4(P<0.01). The presence of palpability and mass size do not significantly influence the BI-RADS US categorization of breast nodules.ConclusionsThe positive predictive value(PPV) of the BI-RADS-US is1.4%,ultrasonic diagnosis of breast probably benign nodules (BI-RADS3) is relatively reliable The multiplicity of breast nodules, patient’s age and US radiologist’s experience were found to have significant influences on the categorizations of probably benign breast nodules.更多还原