WEDNESDAY, March 14, 2001
Mammograms: Is a Picture Worth a Thousand Words?
Mammographyís history is quite interesting. In its early stages in the 1970ís, there was a concerted effort to understand its value, with randomized trials done as early as the 1960ís. Early studies compared screening mammography to usual care. As a group these studies did show a reduction in mortality. The answer to the question "Does picking up cancer early save lives?" is a resounding "Yes!" For women who begin screening above the age of 50, it has lowered the death rate by about 30%. One of the difficulties with mammography is that although it is a good test, it is not a perfect test. About 20% of breast lumps will not show up on mammogram, and this number may be higher in younger women. Mammography does better with calcifications, and can pick up very small ductal carcinoma in situ. This is important because we tend to treat all DCIS with excision and radiation even though not all will grow and potentially become invasive cancer. After an abnormal screening, additional mammographic views can be magnified to serve as a diagnostic test. Mammography can also serve as an aid for surgery by using the mammogram to help localize a core needle or wire to help the surgeon find something that can be seen on mammogram but not felt. New developments are exciting and include taking the mammographic image and digitizing it to enhance contrast, which helps to pick up abnormalities. Preliminary studies show that the regular and digital mammograms may be complementary. We here at the Breast Care Center are trying to improve standard techniques and standard viewing of mammograms. It seems that the experience of the mammographer and the way in which they are read (agreement between two radiologists) are two ways to improve the accuracy of the technique in the current context. This will hopefully make it possible to pick up cancers with less false positives and less biopsies.
Dulcy Wolverton, M.D., a UCSF faculty member in the department of radiology is our guest tonight. In screening mammograms, two complementary views of each breast are taken. Each breast is compressed in the mammographic unit labeled "right" or "left" and either "cranial-caudal" (squeezed top to bottom) or "medial lateral oblique" (squeezed side to side). This allows us to determine where an abnormality might be in the breast. It is a two-dimensional view of a three-dimensional object. Breast tissue tends to show up light on the mammogram and fatty tissue shows up dark. The screening mammogram allows us to examine both breasts quickly with good sensitivity (about 80%). If there are any questions, extra pictures can be made to identify fine detail structures more clearly. This is known as magnification and is especially effective in evaluating microcalcifications, which can indicate DCIS. Dense breasts will show up very white on the mammogram and it can be quite difficult to determine if a tumor in present in dense breasts. This is why physical exam and mammography are synergistic.
Is there any correlation between overweight and fatty breasts? Very obese women tend to have a lot of fatty tissue in the breast; this is usually reduced if they lose weight, and will increase the density of the breast tissue. Extremely thin women tend to have dense breasts because they have little body fat. The correlation is not perfect however. Women who are still menstruating or taking estrogen will have a greater percentage of the breast occupied by glandular tissue, which is dense. As they go through menopause, a larger portion of this tissue will be fatty. The fact that younger women tend to have denser breast may account for the higher miss rate in this group, as high as 30%.
What causes calcifications? When you have tissue in the body that doesnít have circulation, calcifications can form. In cancers, a central area of the duct can become necrotic; these dead cells can calcify over time. Certain patterns of these calcifications can correlate with carcinoma. There are also patterns that are benign, like blood vessels.
What does the radiologist look for in a mammogram? The first thing is symmetry. One of the biggest challenges in the training of a radiologist is learning an appreciation for the great range of normal in humans. All mammograms are different, even breast to breast shows differences. Breasts are looked at closely with a magnifying glass to see if there are any asymmetries. The second thing to look for is anything standing up along the edge of the tissue. Most cancers are within one centimeter of the edge. Next, they look for masses that are dense and irregular, without sharp borders. They look for clusters of calcifications that are irregular or may be lined up in a duct. The average size of cancer detected on mammogram is one centimeter (one-half inch). The goal is to not only detect the cancer, but to give the surgeon an accurate idea of the extent of the tumor, so that it can all be removed without any positive margins. If there is some question of a palpable mass, but the tissue is too dense to determine, an ultrasound can be performed. This does not work well as a screening technique, as there are too many false positives. It works well to define a cyst.
Is pain ever a result of cancer? Pain is actually an uncommon presenting sign in breast cancer. Only about 5-10% of cases involves pain. It could be a problem if the cancer has the invaded the chest wall.
Why is breast cancer usually a disease of aging? No one has a clear answer to that, but it is true of many cancers. A partial answer is that cancer is a process that involves mutations or changes to the DNA. As we get older the chances of that occurring multiplies. The incidence of breast cancer increases with age, but plateaus at around 50.
When is magnification warranted with mammography? Magnification mammography is an extra step from the routine views. It is only done when there is something that requires a closer look, or if there is anything new compared with the older views. Sometimes spot compression views are taken at the site of a lumpectomy, fat necrosis calcifications, or benign calcified lesions that can develop at the site of radiation.
What is the difference between a screening mammogram and a diagnostic mammogram? A screening mammogram is performed on a woman with no symptoms of breast disease. Diagnostic mammogram is extra pictures performed for problem solving: a palpable mass, nipple discharge, recent diagnosis of cancer, implants, etc.
Do MRIs add anything? MRI is a relatively new technique for imaging the breast. It is not like a regular MRI machine, but rather uses magnetic coils made especially for the breast. The technique is still considered to be investigational. It is most effective to help the surgeon with planning after diagnosis, or if someone is getting chemotherapy before surgery, it helps to monitor what is going on. It is the most sensitive technique we have imaging-wise to find breast cancer, and it does find very small cancers. The program needs some tweaking before it can be used for screening because the sensitivity is such that it can actually show too much. It can be non-specific in up to 50% of cases, meaning that every abnormality has a 50/50 change of being either malignant or benign.
Will inflammatory breast cancer show up on mammogram? Yes, usually the first sign is skin thickening. There is also size asymmetry. The breast would also be much harder to compress.
What is a digital mammogram? This is not film, but an electronic receptor, which captures the information and displays it on a computer screen. The radiologist can then change the contrast or magnify certain areas to see how certain parts of the breast look. The digital images can also travel on phone lines to remote locations almost instantaneously. Women who have digital mammography may have a slightly lower callback rate, but it too is experimental at this time, not yet FDA approved for all machines. There is no evidence to date that digital is actually superior to conventional mammography. What has been shown is that some types of cancers are picked up on digital and not conventional and visa versa. This may change over time, as technicians become more familiar with the technology.
Why are 20% of tumors missed on mammography? This is partially due to dense breasts. Some types of cancers like lobular (found in 10% of the cases) are harder to see on mammogram, but they are also harder to feel. They can be particularly insidious, percolating through the breast tissue rather than forming a mass.
What about the recent Canadian study that compared mammography to really good physical exams? One of the difficulties that radiologists in the U.S. have with the Canadian study is that physical examinations are done differently in the two countries. In Canada, specially trained nurse practitioners do physical exams, which are lengthy procedures. However, heir mammography standards have been slow to catch up with the U.S. There are also issues of quality and amount of training; and both techniques are highly operator dependent. It is not reasonable to expect that a specialist in breast cancer will examine every woman in this country. Another problem about the specificity and sensitivity of mammogram versus physical exam is that some cancers are better found by one technique or the other. The procedures should be complementary.
Do breast self-exams fill this gap? Breast self-exam is controversial. It is advocated because the patient finds 75-80% of lumps. But BSE has never shown that breast masses are picked up any earlier, or saves lives. A large study of 300,000 women in China showed no difference in early detection and mortality in women who were trained in BSE versus those who were not.
What are the current statistics on incidence of breast cancer? Breast cancer incidence has actually been declining over the last 3-4 years. For many years new screening techniques have added more and more cases. There were more cases identified, not necessarily more cases absolutely. Now that we have maximized the number of people who are being screened (60-70% of eligible people are now being screened), the curve on detection has started to go down. Another explanation is that incidence may truly be going down due to changes in lifestyle: fewer women using birth control pills, women using lower doses of estrogen, improving diet in childhood, etc.
What about PET scans? PET scans pick up sites of excess metabolism in the body by means of a glucose analog that is tagged to a particle and which is then detected by a scanner. This is usually done by the nuclear medicine department rather than the department of radiology. PET scans are particularly useful for metastatic involvement outside the breast. The resolution is not small enough or high enough to detect small lesions within the breast.
Are certain types of tests better for bone, soft tissue, etc.? This is really an art. For prostate and breast cancer, bone scans are good for surveying the entire bone. MRI is best for the pelvis, spine, hip and the brain. CT scans are best for the chest and abdomen, also the liver and skull. PET is best for lung, also for looking at specific spots in the breast. Brain and liver are hard to image (with PET) because they are hypermetabolic, except after radiation.
Topic for next time: Radiation therapy: treatment and complications, Wednesday, April 11, 2001.
Return to Forum Page