乳腺密度（一）

乳房的组成

就X射线成像而言，乳房可分为两种基本组织：1.脂肪；2.纤维腺体组织。纤维腺体组织，对于X射线成像，具有相近的X射线吸收衰减，但在组织学上可进一步细分为： 

a.为乳房提供结构和支持的纤维结缔组织（库珀韧带）

b.管状的分支乳管

c.小叶，即导管末端分泌乳汁的腺体

乳房是有一层皮下脂肪的，本质上，每个女性的乳腺脂肪分布都是独一无二的，纤维腺体组织也是如此。腺体组织是乳腺管末端的小叶。乳管从乳头上的开口向后延伸，形成分支，最终到达被小叶盖住的终末导管。一个“终末导管”和它的小叶被称为“终末导管小叶单元”（TDL）。据信大多数乳腺癌的起源与TDL有关。

X射线衰减

X射线成像中，脂肪是“透射线的”，X射线穿过脂肪受到阻碍很小。一般来说，乳腺X射线摄影中，脂肪在图像中基本上是黑色的。所有其他“纤维腺体”结构都会更加减弱X射线，因为它们含有水分，并在乳房X光片上投射白色或灰色的“影子”。由于乳腺癌主要是细胞性的（水“密度”），很少含有脂肪，因此其X射线衰减与纤维腺体组织产生的衰减相似。这是一个问题，因为在乳腺X射线摄影图像上被纤维腺组织包围的癌症，如果边缘没有脂肪，在图像上很难看到，除非它周围的结构发生扭曲，或者产生的图像看起来有可疑的钙化沉积。

乳腺密度与乳腺癌检测

因为乳房中的脂肪（实际上是大多数女性乳房的主要成分）在乳腺X射线摄影图像中就像“玻璃窗”，乳腺癌几乎完全是细胞性的（通常脂肪含量最低，如果还有的话），当它们被脂肪包围时，很容易在X射线图像上看到。即使它们与脂肪组织有部分界面，也很容易被看到。当乳腺组织不均匀（脂肪和纤维腺组织混合在一起）时，癌症仍然也很容易被发现，但风险在于癌症隐藏在这些纤维腺乳腺组织的“群岛”中间或后面的时候。当乳房中有大量致密的乳腺组织时，风险就是癌症会被隐藏起来。数字乳腺体层合成（DBT）大大降低了癌症隐藏在致密乳腺组织后面被漏诊的风险。但如果癌症病灶周围如果几乎没有脂肪，甚至在DBT上也可能会被遮挡。

一些致密组织中的癌症在超声上更为明显，超声可以将癌症与可能发生癌症的纤维腺体组织区分开来。无论组织类型如何，静脉注射造影剂的MRI是检测乳腺癌的最佳方法。

乳房的“密度”

20世纪70年代，John Wolfe，开发乳腺干板照相术的先驱，将所看到的脂肪和纤维腺体（致密）组织的类型分为4类。他的分类，从“完全脂肪型”（几乎没有可见的纤维腺体组织）称之为“N1”，到他的第四个分型，“DY”，他（错误地）称之为“发育不良”，其中乳房包含大量“非常密集”的纤维腺体组织。最后一个类型的命名用词不当，和“发育不良”无关。这些被证明只是正常的、水密度高的组织，主要是库珀韧带的纤维支撑结构，与导管和腺体组织混合在一起。这些组织在一些女性的乳房中非常厚，从而产生“致密”的乳房，而在另一些女性乳房中，它们甚至不存在。介于两者之间的是“P1”分型，它主要是脂肪，但有零星的纤维腺体组织。 当纤维腺组织的体积使乳房“不均匀”致密时，Wolfe将其归类为“P2”型乳房。这就是Wolfe对于腺体类型的主观评价：N1、P1、P2、DY。

Wolfe是最先提出乳房的“密度”与女性患乳腺癌的风险有关 ¹。在他的分析中，他（错误地）声称DY型发展为乳腺癌的可能性是正常人群的40倍（见下文）。

BIRADS和腺体类型

美国放射学会乳腺成像报告和数据系统继续承认腺体分类。为避免与最终BIRADS诊断评估类别混淆，最新修订的BIRADS（ACR BI-RADS®Atlas第五版）将四种乳腺腺体类型标记为“a”、“b”、“c”和“d”：

a. 乳腺内几乎全部是脂肪

b. 乳腺内散在纤维腺体密度区域 

c. 乳腺组织密度不均匀，可能会遮挡小肿块

d. 乳腺组织极其致密，使X线敏感性降低

根据BIRADS分类，具有每种腺体类型的女性的百分比约为：

a. 10%

b. 40%

c. 40%

d. 10%

而与腺体类型相关的真正重要问题是：在乳腺X射线摄影检查中，癌症可能被致密组织遮挡。

下期我们将就测定乳房密度、乳房密度和乳腺癌的风险等内容继续分享，期待大家的持续关注！

英文对照｜ENGLISH COMPARISON

BREAST DENSITY-UPPER

BREAST COMPOSITION

With regard to X-ray imaging, the breast can be divided into two basic tissues:

1. fat 

2. fibroglandular tissue  

The fibroglandular tissues have the same X-ray attenuation, but can be further subdivided, histologically, into 

a. the fibrous connective tissues that provide structure and support for the breast (Cooper’s ligaments)

b. the tubular branching milk ducts

c. the lobules that are the milk producing glands at the end of the ducts.  

There is always a layer of subcutaneous fat and then, essentially, individualized volumes of fat that are unique in distribution to each woman.  

The fibroglandular tissues are exactly that. The glandular tissues are the lobules at the ends of the milk ducts. The milk ducts going back from openings on the nipple, branch, to ultimately reach the terminal ducts that are capped off by the lobules. A “terminal duct” and its lobule are known as the “terminal duct lobular unit”(TDL). It is believed that most cancers originate in association with the TDL.

X-RAY ATTENUATION

Fat is “radiolucent”. X-rays pass through fat with little impediment. By convention, on mammography, fat is, essentially, black on the mammogram. All the other “fibroglandular” structures attenuate X-rays because they contain water and cast white or gray “shadows” on the mammogram. 

Since breast cancers are primarily cellular (water “density”) and rarely contain fat, their X-ray attenuation is similar to that produced by the fibroglandular tissues. This is a problem since a cancer that is surrounded by fibroglandular tissues on a mammogram, with no fat along its margin, will be, essentially, invisible on the mammogram unless it distorts the surrounding architecture, or produces calcium deposits that look suspicious on the mammogram.  

BREAST DENSITY AND DETECTING BREAST CANCER

Because fat in the breast (actually its major component in most women) is like “window glass” on a mammogram, breast cancers, which are almost completely cellular (usually with a minimum of fat, if any) are easily visible on X-rays when they are surrounded by fat. Even if they have a partial interface with fatty tissue, they are easily seen. When the breast tissues are heterogenous (fat and fibroglandular tissues mixed together) cancers are still fairly easily seen, but the risk is that they will be hidden in, or behind, these “islands” of fibroglandular breast tissue. When a breast has large volumes of dense breast tissue, the risk is that a cancer will be hidden. Digital Breast Tomosynthesis (DBT) greatly reduces the risk that a cancer will be hidden behind dense breast tissue, but if there is little if any fat abutting the cancer, it can even be hidden on DBT.

Some cancers in dense tissues are much more evident on Ultrasound which can differentiate cancers from the fibroglandular tissues in which they may develop. Magnetic Resonance Imaging (MRI), with intravenous contrast, is the best way to detect breast cancer regardless of tissue pattern.

BREAST “DENSITY”

In the 1970’s John Wolfe, a pioneer who helped develop Xeroradiography for mammography, divided the patterns of fat and fibroglandular (dense) tissues that he was seeing into 4 categories. His categories ranged from “all fat”, with little if any visible fibroglandular tissue, which he called “N1”, to his 4th pattern, “DY”, which he (incorrectly) termed “dysplastic” in which the breast contained major volumes of “extremely dense” fibroglandular tissues. This last category is a misnomer. There was no “dysplasia” involved. These prove to be nothing more than normal, water density tissues that are, predominantly, the fibrous supporting structures of Cooper’s ligaments intermixed with ducts and the glandular tissues. In some women these are extremely thick and produce “dense” breasts while in others they are fine to even nonexistent. In between was the “P1” pattern which was largely fat, but had scattered fibroglandular tissues. When the volume of fibroglandular tissues made the breast “heterogeneously” dense it was classified by Wolfe as a “P2” breast. These were his subjective evaluations–N1, P1, P2, DY.   

Wolfe was the first to suggest that the “density” of breast related to its risk of developing breast cancer ¹. In his analysis he (incorrectly) claimed that the DY pattern was almost 40 times as likely to develop breast cancer (see below).

BIRADS AND TISSUE PATTERNS

The American College of Radiology Breast Imaging Reporting and Data System has continued to recognize the various tissue patterns. To avoid confusion with Final Assessment categories, the most recent revision of BIRADS (ACR BI-RADS® Atlas 5th Edition) labels the four general breast tissue patterns as a, b, c, and d.

a. “The breast is almost entirely fat.”

b. “There are scattered areas of fibroglandular density

c. “The breasts are heterogeneously dense which may obscure small masses.”

d. “The breasts are extremely dense which lowers the sensitivity of mammography”

According to the BIRADS Atlas the percentages of women with each pattern are:

a. 10%

b. 40%

c. 40%

d. 10%

The most important problem revolving around tissue patterns is that cancers can be hidden on mammography in areas of dense tissues.

REFERENCES

1.Wolfe JN. Risk for breast cancer development determined by mammographic parenchymal pattern. Cancer. 1976 May;37(5):2486-92. doi: 10.1002/1097-0142(197605)37:5<2486::aid-cncr2820370542>3.0.co;2-8. PMID: 1260729.