【作者】 洪庆山；

【导师】 许乙凯；

【作者基本信息】 南方医科大学 ， 影像医学与核医学（专业学位）， 2014， 博士

【摘要】 研究背景乳腺癌(breast cancer, BC)是严重威胁女性健康的恶性肿瘤之一。乳腺癌从普查发现到进行治疗,大致经历了筛查、诊断、术前评估三个阶段。常用的影像检查手段有钼靶x线、超声、CT和MRI。如何掌握各种影像检查手段的适应症、序贯性运用各种影像检查手段,是个重要的临床课题。①乳腺钼靶X线是乳腺癌筛查和诊断的首选。微钙化是诊断乳腺癌的最具特异性征象,提高微钙化检出率以及研究微钙化影像学特征与病理的关系非常重要。②CT目前仅用于术前了解乳腺癌胸部转移情况。值得注意的是,某些妇女乳腺癌是由于胸部其它疾病行CT扫描发现的。因此,有必要了解乳腺癌CT扫描的影像征象及其鉴别。③随着保乳手术的兴起,如何对早期乳腺癌进行精密分析、确定切除范围,最大程度保留周围正常腺体,成为一项亟需解决的临床难题。而传统的钼靶X线,超声已经无法满足保乳术的特殊评估要求。乳腺MRI能更准确的评估肿瘤的大小、浸润范围、周围子灶情况,具有更好的应用全景。第一部分X线对乳腺微钙化的定量分析及其临床价值目的1.探讨乳腺手术标本X线摄影存在的问题及改良技术。2.探讨乳腺微钙化的X线影像定量分析与病理的关系。方法1.研究对象搜集2009年5月至2010年10月行数字化乳腺X线检查的106例女性乳腺疾病患者为研究对象,年龄25-68岁,平均年龄41.9岁,中位年龄41岁。病变在影像图像中均表现为微钙化病变,并且均经穿刺或手术活检明确病理诊断,其中Mammotome穿刺活检37例,手术活检69例,两者中采用术前钢丝定位穿刺活检切除42例,对手术标本分别进行常规摄影和改良技术摄影。切除的标本须常规X线摄影,确定微钙化病变已切除。2.仪器与设备采用意大利GIOTTO数字化乳腺X线机,配有高清晰数字化快速采集工作站。检查时患者取站立位或坐位,常规拍摄双侧乳腺头尾位(craniocaudal projection, CC位)和内外斜位(]mediolateral oblique projection, MLO位),必要时加照侧位或特殊体位。采用电脑自动曝光条件系统,获取的乳腺X线图像数据须符合乳腺图像质量规范,并以标准的DICOM文件格式存储和传输导入PACS系统。3.乳腺手术标本的获取与摄影技术改良活检系统为配套的Image3D Biospy。乳腺标本的获取：透视下引导钢丝到达病变部位并锚紧,采用手术切除或Mammotome微创旋切系统切除病变。将乳腺标本放在摄影台,摄影自动条件为30-40kv,80-100mAs,1.2-1.4mGy。常规组按常规自动条件进行乳腺X线摄影,分析标本摄影存在的问题。摄影背景台：取延展性能比较好的避孕套,充盈以植物油脂,模拟乳腺的脂肪环境。改良方法：乳腺穿刺活检标本放在制备好的背景台上,使标本和背景台充分接触,采用手动调整合适的摄影参数(20-30kv,25-30mAs,0.3mGy)进行摄影。对于较大的标本,参考文献经验采用直接加压(压力0.5-2kg)后摄影。比较常规和改良两种摄影技术的图像清晰度、对比度,钙化的数目、形态等细节。比较两组技术对钙化灶的显示率。4.微钙化定量分析工具和方法微钙化定量分析工具在计算机辅助检测平台下进行。数字化乳腺摄影机获取DICOM协议的数字图像后,导入MammoCAD2.0(沈阳东软医疗系统有限公司),自动标识微钙化、肿块,结构扭曲和淋巴结。最后结合乳腺影像诊断医师的经验进行分析。详细记录微钙化形态分型、数量、分布方式、密度、分布于何侧乳腺、是否伴有肿块等特点。①形态分型参考Le Gal分型：Ⅰ型：环形(Annular)钙化；Ⅱ型：形态规整的点状(Reuglarly punctiform)钙化;Ⅲ型：泥沙样(Dusty)钙化；Ⅳ型：形态不规整的点状(Irreuglarly punctiform)钙化；Ⅴ型：蠕虫样(Vermicular)钙化。由于型为良性钙化,本研究未纳入。②计算数量时选取钙化最密集分布区域的单位面积内钙化数(个/cm2),按0-10、11-20、21～30和>30为标准分组。③微钙化的分布方式根据2003年美国放射学会的乳腺摄影和报告数据系统(Brest Imaging Reporting and Data System,BI-RADS)分为簇状分布、线样分布、段样分布、区域分布和弥漫分布。④微钙化根据密度分为高密度和低密度(潜钙化)。由两位乳腺影像诊断专业医师(中级职称以上)用共同阅读乳腺X线片,意见不一致时,协商后得出结果。5.统计学处理采用SPSS13.0软件分析。两组计数资料的比较采用卡方检验(根据适用条件不同,选择非校正卡方检验、校正卡方检验及Fisher精确概率法)。等级频数表资料的比较采用两独立样本非参数检验。P≤0.05有显著性差异。结果1.乳腺标本常规摄影存在的问题与技术改良乳腺标本常规摄影常见的问题主要包括：①摄影图像颜色过白或过深,清晰度不够,对比度差,细节模糊,小簇状微钙化或泥沙样钙化易被掩盖；②标本蜷缩,影像细节重叠,导致小结节被掩盖,部分小点钙化可重叠成粗块状钙化,而蜷缩的边缘组织密度升高甚至接近钙化密度。经上述技术改良后,乳腺标本摄影图像清晰度高,对比度好,细节清楚,乳头、皮肤、血管、腺体等组织层次结构清楚,钙化对比度好,没有重叠现象。1例钢丝定位活检的标本摄影改良后,图像清晰度和对比度均较好,显示了术前片的“小簇状钙化”。改良组钙化灶的显示率100%(42/42),显著高于常规组83.3%(35/42),差异有统计学意义(χ2=7.63,P=0.012)。2.微钙化病灶的病理结果本组106例患者中其中有5例为双侧乳腺微钙化病变,因此共获得111例乳腺微钙化病变。经病理结果证实,50例为恶性病变(45.0%),61例为良性病变(55.0%),其中双侧微钙化的病变均为良性病变。3.微钙化影像学特征与病理学的关系微钙化的形态、分布、单位面积内的数量以及是否伴有肿块对微钙化病灶良恶性的鉴别诊断均有统计学差异。随着形态和单位面积数量的变化,病灶的恶性程度是依次递增的,Le Gal Ⅱ型、Ⅲ型、Ⅳ型、V型的恶性比例分别为29.6%、30.1%、87.0%、100.0%;单位面积数量(个/cm2)0～10、11～20、21～30和>30分组的恶性比例分别为14.6%、48.0%、58.5%、75.8%。本组111例病灶中,37例伴有肿块,占全部病灶的比例为33.3%。在37例钙化伴有肿块病灶中,恶性病灶的比例为75.7%(28/37),明显高于良性病灶24.3(9/37),P<0.001。但微钙化的密度、乳腺的部位(左、右侧)则对良恶性鉴别无明显影响,P>0.10。3、微钙化的Le Gal分型与病理的关系本组微钙化的Le Gal分型如下：Ⅱ型54例,Ⅲ型29例,Ⅳ型23例,Ⅴ型5例。随着微钙化Le Gal分型的推进,病灶的恶性比例依次递增,Ⅱ型、Ⅲ型、Ⅳ型、V型恶性比例分别为29.6%、30.1%、87.0%、100.0%。Ⅱ型和Ⅲ型对良恶性的病理分类并无显著性差异(χ2=0.730,P=0.393)。将两者合为一体,其诊断恶性的准确率只有30.1%(16+9)/(54+29)。Ⅳ型和V型对良恶性的病理分类并无显著性差异(χ2=0.003,P=0.955)。将两者合为一体,其诊断恶性的准确率达到89.3%(20+5)/(23+5)。“微钙化单位面积的数量”分别对Ⅱ型、Ⅲ型微钙化的良恶性鉴别诊断有统计学意义(P=0.027；P=0.050)。“是否合并肿块”分别对Ⅱ型、Ⅲ型微钙化的良恶性鉴别诊断有统计学意义(χ2=6.862,P=0.016；χ2=7.741,P=0.010)。结论1.数字化乳腺摄影、CAD以及标本摄影技术改良有助于提高微钙化的检出率。2.微钙化的影像定量分析,包括形态、分布、数量、密度及是否伴肿块,对乳腺癌的诊断有重要参考价值。3.微钙化的Le Gal形态分型与病灶的良恶性具有良好的相关性,结合其它影像征象能提高诊断的准确性。第二部分乳腺癌淋巴结转移的MSCT表现与分子分型等相关因素分析目的1.探讨MSCT对乳腺癌及早期淋巴结转移的应用价值。2.探讨乳腺癌淋巴结转移与分子分型等因素的关系。方法1.一般资料搜集手术病理证实的乳腺浸润性导管癌30例为研究组,临床可触及明显肿块,平均年龄(49.7±11.8)岁。对照组来源于同期手术切除的30例乳腺纤维腺瘤患者,平均年龄(48.6±10.2)岁。两组年龄无显著性差异。另外搜集20例无淋巴结转移的乳腺癌病例。2.仪器和设备乳腺摄影设备为GITTO乳腺数字化摄影机。MSCT为SIEMENZ Somatom128排螺旋CT,高压注射器medrad injection system,扫描参数120kv,100mAs,0.5s/slice,0.5cm/slice, FOV400mm(L)。造影剂为30%碘海醇,1.5ml/kg,流速3-4ml/s,注射造影剂25s(动脉期),55s(静脉期)后进行扫描,记录肿块的强化规律。3. MSCT影像评价指标肿块：形状分为规则(圆形、椭圆形)和不规则(分叶等)；边缘分为光滑和毛刺。淋巴结直径的测量取经过淋巴结门的短径。CT强化的分度：增强前后差值<30HU为无明显强化,30-50HU轻度强化,>50HU为明显强化。强化方式分为环形强化和均匀强化,静脉期强化CT值大于动脉期30%以上为持续性强化。4.分子分型检测乳腺癌标本切除后,进行ER, PR及Her-2免疫组化染色。乳腺癌各分子分型有Luminal A型、Luminal B型、Her-2过表达型、Basal-like型(又称三阴型)。Luminal A型：ER和／或PR+, Her-2-。Luminal B型：ER和／或PR+, Her-2+。Her-2过表达型：ER和／或PR-, Her-2+。Basal-like型(三阴型)：ER和／或PR-, Her-2-。5.乳腺癌淋巴结转移的相关因素搜集本组30例乳腺癌中出现淋巴结转移的病例。另外搜集20例无淋巴结转移的乳腺癌病例作比较,对比两者在年龄、病理类型、分子免疫分型、VEGF评分的差异。6.统计分析统计处理软件为SPSS13.0。计数资料比较采用卡方检验。比较乳腺癌肿块和纤维腺瘤的MSCT征象差异。比较乳腺摄影与MSCT发现腋下LNM的差异。四格表资料采用卡方检验,等级资料采用两独立样本的非参数检验。P≤0.05有显著性差异。结果1.IDC的MSCT表现：肿块多为形状不规则,边缘毛刺,动脉期呈明显强化,静脉期呈持续性强化,可见环形强化。两组在肿块形状、边缘、强化方式等存在显著性差异(P<0.05)。MSCT发现淋巴结转移22例,6例无明显增大,仅增强后明显强化。2.淋巴结转移(LNM)的MSCT表现：本组共发现LNM22例。部位分布如下：乳腺内2例,腋下区15例,内乳区2例,纵隔2例,锁骨上窝1例。MSCT表现为淋巴结直径增大或轻度增大,淋巴结门脂肪间隙消失,变圆钝,增强扫描呈明显强化。MSCT与乳腺摄影对LNM探测能力存在很大差异,以腋下区LNM为例,乳腺摄影仅发现10例,MSCT发现15例,两者有显著性差异(P<0.05)。3.LNM的分布与MSCT征象的关系：22例LNM中,有6例直径小于10mm,包括乳腺内1例,腋下区3例,内乳区2例,仅表现为密度升高、淋巴结门圆钝及明显强化。4.乳腺癌淋巴结转移的相关因素：乳腺癌年龄(≧50岁)组淋巴结转移比例76.5%(13/17),高于<50岁年龄组36.0%(9/25),差异有统计学意义,P=0.010。VEGF评分≧2分组淋巴结转移比例75.0%(15/20),高于<2分组31.8%(7/22),差异有统计学意义,P=0.005。将免疫分型的Luminal A与LuminalB合并,Her-2过表达型与Basal-like型合并,进行比较,结果发现Luminal A+B型淋巴结转移比例40.7%(11/27),低于后者73.3%(11/15),差异有统计学意义,P=0.043。结论1.乳腺癌肿块在MSCT双期增强扫描上具有良好的影像特征；2.LNM的判断必须综合淋巴结直径、结门形态、密度及CT强化值等因素,MSCT双期增强扫描可作为是一种全面、灵敏的检查方法。3.乳腺癌年龄较大、VEGF评分较高和分子分型Her-2过表达型+Basal-like型,应该更关注其淋巴结转移的可能。第三部分乳腺磁共振成像对乳腺癌诊断和保乳术前评估的应用目的1.评价术前MRI检查对乳腺癌保乳术中切缘阳性率的影响。2.评价术前MR]对乳腺癌病人的保乳手术治疗方案更改。方法1.一般资料搜集2012年3月-2013年5月期间所有符合保乳术指症拟行保乳的乳腺癌病人共66例,其中2例单行新辅助化疗,最后行保乳术者64例作为研究组。2010年6月～2011年8月期间所有拟行保乳术治疗的乳腺癌病人共76例,其中3例患者拒绝保乳而行全切术,最后行保乳术者共73例作为历史对照组。两组病人术前均行钼靶及超声检查,研究组病人术前加做乳腺MRI检查。所有病灶均经活检或手术病理证实。准入标准：年龄<65岁；未经任何治疗；同时有超声和钼靶摄影检查资料。排除标准：复发病灶；晚期乳腺癌姑息疗法行保乳术治疗；已行放化疗。2.影像学检查技术研究组病人术前均行MRI检查,采用Aurora公司1.5T乳腺专用磁共振扫描机及整合于检查床中的射频线圈。病人俯卧于检查床上,双乳自然悬垂于线圈内。扫描前先进行预扫描,对比增强前采用GRE T1WI及FSE压脂T2WI横断位扫描2个期相,分别获取图片64张及40张。后采用3D Aurora SPIRAL(双侧螺旋采样)及RODEO(脂肪、液体和腺体组织抑制)技术横断位扫描,层厚、层距1.125mm,平扫、动态增强横断面SPIRAL RODEO动态增强扫描5回合,注射对比剂后于90s行第2回合增强扫描,第2-5期增强扫描时间间隔均为180s。动态增强对比剂采用钆喷酸葡甲胺(Gd-DTPA, Gadopentetate dimeglumine),注射剂量为0.2mmol/kg体重,以2mL/s的速度快速推注,检查扫描总时间约为28min；钼靶采用意大利GIOTTO高频钼靶乳腺X线机,取站立位,常规头尾位(CC)、内外斜位(MLO)分别两次曝光；超声使用Esaote My Lab60型,My Lab90型及GE Logiq9型超声诊断仪,高频线阵探头。患者取仰卧位,全面检查双侧乳腺及腋窝。3.图像后处理与资料分析MRI图像的后处理在Aurora的后处理工作站完成,可自动获取减影、伪彩图像、制作时间-信号强度曲线(time-intensity curve, TIC)以及选取任意期相行多平面重组,在同一界面同时观察矢状位、冠状位及轴位图像。对病灶影像学的评估采用美国放射学会提出的乳腺影像报告和数据系统(breast imaging and reporting data system, BI-RADS)标准并结合临床情况,将BI-RADS4b级及其以上病变作为恶性病变；结合所有影像检查手段,将乳腺纤维腺体分型分为两类,即：致密型乳腺(ACR标准3-4型)和非致密型乳腺(ACR标准1-2型)。钼靶上显示的钙化由一位专门从事乳腺疾病诊断工作的放射科医师及两位副主任以上放射科医师盲法阅片,对少数判断不一致的病例共同讨论进行最终的判定。4.手术计划与切缘阳性所有病人最初的手术计划是根据病史、临床触诊、钼靶及超声征象来制定的。研究组病人进行术前乳腺MRI检查后,由外科医生、超声科医生、病理科医生、放射科医生组成的讨论组,结合研究组病人之前的临床及影像资料以及随后的MRI检查结果共同讨论一个最终的治疗方案。切缘阳性：是指根据术前讨论的切除范围,将病灶切除送检,并将切缘的组织残端平均分6点进行取材切片,距切缘5mm以内有癌细胞者判定为切缘阳性。5.统计学方法采用SPSS13.0统计软件进行数据分析。计量资料以x±s表示。对原始数据及函数转换后的数据进行正态性检验(Normality test)。①计量资料显著性检验：符合正态分布特点的计量资料,采用两独立样本t检验；不符合正态分布的,采用非参数检验。②计数资料：四格表资料采用卡方检验,等级资料采用两独立样本的非参数检验。P≤0.05有显著性差异。结果1.两组一般情况研究组64例,共77个病灶,年龄21-59岁,平均年龄40.19±0.91岁。历史对照组73例,共81个病灶,年龄20～62岁,平均年龄39.14±0.92岁。两组患者的年龄、致密型腺体、触诊阳性、钼靶显示恶性钙化、各病理类型之间都没有显著性差异。但两组之间病灶大小比较,研究组的肿块大于历史对照组,差异有统计学意义(P=0.02)。2、两组不同病理类型的肿块大小比较比较两组不同病理类型的肿块大小,结果显示,研究组中的浸润性导管癌要比历史对照组的IDC直径更大,差异有统计学意义(P=0.019)。导管原位癌及其他病理类型的大小两组间并没有显示出明显的统计学差异。2.切缘阳性率比较研究组的切缘阳性率(9.6%)比历史对照组(24.7%)低,差异有统计学意义(P=0.032)。3.手术计划的变更本研究病例中,出现了因钼靶及超声检查发现乳腺癌多发病灶,以及MRI检查发现额外病灶,经讨论后认为病灶的浸润周围扩大,最终扩大了保乳术的切除范围。其中,研究组12例(18.8%)病人因发现多发病变、癌周浸润或对侧病变而改变手术方式。其中2例病人在钼靶、超声发现多发病变(2个病灶)的同时在MRI上又发现了额外病变(1个病灶),最终行单侧乳房全切术。术后病理证实这种治疗方案是正确的。6例病人因在MRI上发现病灶周围导管样强化或斑点状强化而疑为病灶周围存在浸润,测得的病变范围也较钼靶、超声更大,术后病理证实有4例MRI的判断是正确的,其中1例为粘液癌伴癌周导管内癌。另有2例病人经术前MRI检查后应行全乳切除术,但病人保乳意愿强烈,遂行保乳术,其中1例术中病理显示切缘阳性。1例病人因在MRI上发现对侧乳腺癌而同时行对侧切除术。结论1.术前MRI能更准确显示肿块大小、范围,从而有助于降低乳腺癌保乳术中的切缘阳性率。2.MRI能够在保乳术前提供乳腺癌浸润范围、子灶、双侧乳腺癌等更多信息,从而使手术方案发生更改。更多还原

【Abstract】 BACKGROUNDBreast cancer (breast cancer, BC) is one of the serious malignant tumors threating to women’ health. From detection to treatment, the patients experience3stages, screening, diagnosis, preoperative evaluation. Mammography, ultrasound, CT and MRI are applied commonly as imaging examination. It is very important to master a variety of indications for imaging examination, and to undergo sequentially applying various imaging methods.①Mammography is the first choice for breast cancer screening and diagnosis. Microcalcifications (MC) is the most specific signs for diagnosis of breast cancer. It is important to improve MC detection rate and explore the relations between MC imaging features and pathology.②CT is currently only used for detect breast cancer metastasis in the chest. It is worth noting that breast cancer is found by chest CT scan due to other diseases. Therefore, it is necessary to differential understand the CT imaging signs of breast cancer.③With the rise of breast-conserving surgery (BCS) for early breast cancer, it is urgent for doctors to determine the extent of resection, maximize retention normal glands surrounding the tumor. The traditional X-ray mammography, ultrasound has been unable to meet the special assessment requirements. Mammary MRI can more accurately assess the size, infiltration range, lesions surrounding of tumor. PART1Quantity analysis on breast micro calcifications in X-rays and the clinical evaluationObjectives1. To evaluate the innovation of X-ray photography technique for breast wire-oriented biopsy specimen.2. To evaluate the relations between mammography MC features and pathology in breast lesions.Methods1. SubjectsFrom May2010to October2012,106women with breast disease were collected, aged25-68years old, average age41.9years, median age41years old. FFDM images were characterized as MC lesions. Pathology diagnosis results were obtained by biopsy or tumor resection. Specimens were undergone mammography routinely, which determines MC lesions had been totally removed.2. EquipmentsFFDM underwent all the patients before operation, equipped with a high-definition digital rapid acquisition station and X-ray induced biopsy system (GIOTTO Corp., Italy). Patients were examined standing or sitting. Mammography regularly shot bilateral breast, included craniocaudal (CC) and mediolateral oblique (MLO) projection, besides side or special projection if necessary. Computer system implemented automatic X-ray exposure conditions, and breast X-ray images must be acquired conforming to mammography image quality standard, stored with standard DICOM file format and imported into PACS.3. Problems in DR routine photography for wire-oriented biopsy specimens were analyzed. Photography technique was reformed to improve image quality. Detection rate of calcification was compared between the two photography techniques (Routine Group and Reforming Group).4. MC image featuresMC image features in FFDM were recorded, included breast density, MC shape, distribution, quantity (number of per unit area), MC Density, location and whether accompanying by mass, etc.(?)MC shape was divided into4types, referring Le Gal classification in1984. And type I was typical benign calcification, followed with no intervention in clinic. Consequently, type I was not explored in this study.㎝C distribution, was classified into cluster, linear, segmental, regional, and diffuse, according to Brest Imaging Reporting and Data System (BI-RADS), formulated by American Radiology Association in2003(?)MC quantity, was calculated by counting the MC numbers in a unit area (usually1cm2) in the most concentrated area, and was divided into4groups,0～10,11～20,21～30, and>30.(?)MC Density was divided into high density and low density (hidden calcification).(?) MC location is in left or in right breast,.(?)Whether accompanied by mass:yes or not.4. Statistical processingSPSS13.0software was applied for statistical analysis. Single factor analysis was utilized by x2test (Fisher’s exact test), and multiple factors analysis by Logistic regression analysis. Comparison difference is statistically significant when P<0.05.Results1. Shortages of breast specimens DR images and innovationThe shortages of breast specimens DR images included signs overlap, poor contrast, calcification or glandular with blurred details. The best images were obtained by using the background sets, appropriate pressure, manually adjustment of exposure parameters. Routine Group had a high calcification detection rate than Reforming Group (100%vs.83.3%, P<0.05)2. Pathology of106patients with MC111MC lesions were found in106cases with breast disease, including5cases with MC in both side breasts. There were50malignant lesions (45.0%) and61benign lesions (55.0%), confirmed by pathology.3. Single factor analysisMC features factors, including shape, distribution, and whether accompanied by mass, number of per unit area, had significant influence on proportions of malignance. Furthermore, as the elevations classification of MC shape and number per unit area, proportions of malignance is increasing, in turn, respectively29.6%,30.1%,87.0%,30.1%and87.0%,48.0%,58.5%,75.8%. When MC is accompanied with mass, malignant lesion rate was75.7%, significantly higher than without mass29.7%(P<0.001). But MC density and mammary gland density type had no significant effect for calcification identification.Conclusions1.Full digital mammography, computer-aided detection and X-ray innovation will improve the detection rate of calcification.2. MC image features, including shape, distribution, density and whether accompanied by mass, is valuable for breast cancer differential diagnosis.3. Le Gal classification of micro calcification is highly relative to malignancy. Diagnosis accuracy will be improved if combining with other imaging signs. PART2Application of multi-sliced CT on breast cancer and related factors analysis such as molecular typing on lymph node metastasisObjectives1. To investigate the application of multi-sliced CT on breast cancer and early lymph node metastasis (LNM).2. To investigate the relation between molecular typing and lymph node metastasis.Methods1.30cases with infiltrative ductal carcinoma (IDC) confirmed by pathology were collected and set as Study Group, with an average age(49.7±11.8)yr, and30fibromas as Control Group, with an average age(48.6±10.2)yr. There was no significant difference in age. In addition,20breast cancer cases without lymph node metastasis were collected.2. Equipments.128-slice CT plain scan and2phrase enhance was undergone. Image signs of mass of two groups were compared. Image signs of early lymph node metastasis were analyzed.ResultsIn MSCT images, IDC masses had irregular shape, burr rim.2-phrase MSCI demonstrated obvious enhancement in artery phrase, persistence enhancement in vein phrase and ring shape enhancement. There were significant different between2groups in mass shape, rim and enhancement mode (P<0.05). MSCT demonstrated22LNM, but6of them had no swelling but obvious enhancement. Lymph node (LN) metastasis showed round performance, lost of normal lymph node gate structure and enhanced significantly, including15in axillary LN,2in internal mammary chain,2in supraclavicular LN,8in mediastinal LN.Conclusions1. MSCT2-phrase enhance scan plays an important role in assessing breast cancer and early lymph node metastasis.2. There is relation between molecular typing and lymph node metastasis. Part3Application of preoperative MRI on breast cancer diagnosis and evaluation before breast-conserving surgeryObjectivesTo evaluate the influence of preoperative MRI on tumor-positive rate of resection margins and surgery treatment programs in breast cancer patients eligible for breast-conserving surgery.MethodsFrom Mar2012to May2013,64cases with breast cancer eligible for breast-conserving surgery (BCS) were collected as STUDY GROPU (SG). From Jun2010to Aug2011,73cases with breast cancer eligible for BCS were collected as history CONTROL GROUP (CG). The two groups were undergone with mammography and ultrasound, and were eligible for BCS after assessment. In addition, SG was undergone with breast MRI preoperatively.ResultsSG included64patients, with a total of77lesions, mean age40.19±0.91yr; historical SG included73patients, with a total of81lesions, mean age39.14±0.92yr. Tumor size differences between the two groups (P=0.020), between the two groups infiltrating ductal carcinoma (Infiltrating ductal carcinoma, IDC), there are differences in the size (P=0.022). While age, the gland type, whether positive palpation, mammography is showing malignant calcification, pathologic type between the two groups did not show a significant difference. Compared with the historical CG, SG had lower positive margin rate (9.6%and24.7%; P=0.04). In addition, preoperative MRI demonstrated multiple lesions in12cases (18.8%), peritumoral infiltration and contralateral lesions. Therefore, the original surgical resection plan was changed.Conclusions1. Preoperative MRI, demonstrating the size, infiltration range of breast cancer, therefore effectively reduces positive margin rate for the patients who underwent breast-conserving surgery.2. Preoperative MRI, demonstrating more multiple lesions, peritumoral infiltration and contralateral lesions, therefore changes the original surgical resection plan.更多还原