An article entitled, "Spotting Breast Cancer: Doctors are Weak Link" appeared on the front page of the New York Times on June 27, 2002. This lengthy article commented on the perceived lack of skills of mammographers such that the anticipated benefits of screening mammography were compromised. Despite the fact that mammography centers, specifically their personnel, equipment and film quality have been strictly monitored for more than a decade by the federal government, the article pointed out that failure to detect breast cancer was not uncommon and that more regulation may be necessary to combat this deficiency. The author appeared to have had little input from radiologists and made little effort to understand the limitations and frustrations they face daily.

The author and the American public may have a misperception about the art of interpreting mammography. The article noted several times the difficulty in interpreting these exams but never explained why. Typically in medicine, once learning and seeing a disease and its presentation, a physician will feel comfortable recognizing it again in the future. Mammographic interpretation is different. No two mammograms look exactly alike which is different than most imaged body parts. They vary in appearance by the amount and distribution of fat and fibroglandular tissue. Fat on film appears gray/black. Fibroglandular tissue appears white. To complicate matters, cancers also appear white. The difficulty comes when the cancer originates in an area of fibroglandular tissue. The cancer (white) may then be obscured by the adjacent fibroglandular tissue (white). Detecting breast cancer may then be likened to detecting a specific snowball during a snow storm. Mammographers have to therefore rely on other subtle signs of the tumors presence

Another important fact to understand is that breast cancer is a heterogeneous disease and may present in a variety ways mammographically including as calcifications, masses or distortions. These findings may also however, represent benign lesions. Oftentimes biopsy is required to distinguish between the two.

Breast cancer pathologically is known to evolve with time from benign normal cells to those which are atypical and then cancerous. This process occurs on a continuum. There is no absolute point in time when the pathology turns from benign to frankly malignant. Similarly, there is no specific point in time when a mammographic finding can no longer be considered normal or benign. Each radiologist must reach a balance when reading films. The goal is to find cancer. To find cancer the screened patient must be called back for additional evaluation. It is important not to call back too many women for false alarms resulting in unnecessary anguish to the patient and her family. If too few patients are called back, some cancers will be missed. Every mammographer must have their own level of comfort when reading films but it is recommended that fewer than 10% of screening patients be called to return for questionable finding.

Another factor weighing on the radiologist is the presence of prior films. Prior films are extremely important when interpreting films in that stability of a finding suggests benignity. However, when a cancer is found prior films can be reviewed, typically by lawyers and their expert witnesses, to say that it could have been picked up earlier but wasn't, resulting in delay of diagnosis. Interpretation is always easier when looking backwards.

As mammographers, we know cancers can be missed. Reasons include technical or interpretive factors. Technical factors account for a small percentage of problems. These may be caused by poor patient positioning, poor film technique or processing. In most centers these issues are addressed when abiding by the guidelines set forth by the federal government. Of concern in the N Y Times article are interpretive errors made by the radiologist. These errors may be due to misinterpretation of perceived findings and failure to perceive abnormalities. Misinterpretation of an identified abnormality can occur when a finding is seen in only one view and subsequently dismissed as not significant. If an abnormality appears stable when review is made of prior exams it may also be discounted although a significant finding. (Breast cancers typically grow with time.) Occasionally, certain types of breast cancer demonstrate benign characteristics and therefore may not be immediately evaluated until a change is demonstrated.

Failure to perceive can easily occur in a busy practice when many films are being read. Reading requires intense prolonged concentration when looking for the infrequent subtle sign of an early breast cancer. Adjustments can be made in a radiology practice to address this problem. Mammography should be performed only by those truly committed to the process. When the volume of exams permits, dedicated mammographers can commit all their time to breast disease and its diagnosis. Volume of interpretation and close followup is instrumental in becoming expert in interpretation.

Secondly, films can be read by two radiologists. This has been shown to increase the cancer detection rate by up to 15%. Many practices do not have the resources available to commit to double reading. Recently developed is a new technology called computer aided detection. This technology utilizes computers to analyze mammograms in search of early signs of breast cancer which may include calcifications, masses or densities. The computer marks films when abnormalities are resent. The radiologist initially reviews films and then refers to the computer analysis. The mammogram is then reviewed again with knowledge of the computer findings. The computer has been shown to increase cancer detection by up to 23%. The public needs to realize the benefits of screening mammography. Negativism as portrayed in the NY Times article can result in scaring women from obtaining the exam, forcing closure of some centers which may limit access particularly in poor and rural areas. We must remember that the vast majority of exams are interpreted correctly and have been for years. Reduction in patient mortality has been demonstrated many times in numerous studies as a result of accurate interpretation by radiologists. With a negative report, the high likelihood is that the interpretation is correct. The odds are in favor of the value of screening.

To assure a correct diagnosis, women when scheduling a screening mammogram should ask the following questions:

At the Center For Breast Care we are committed to the women of our community and have been for the last twelve years. We have three dedicated mammographers spending nearly 100% of their time reading mammography. We spent more then ten years having every film read by two radiologists. Since May of 2002, computer aided detection has been utilized to review patients films. The center is high volume imaging more than 35,000 cases per year and finding more than 400 new breast cancers. This volume is the second largest in the state of Florida. With volume comes expertise.

Correct interpretation and demands for quality cannot be mandated through government regulation. The desire to achieve quality comes from within the institution and its peopleThe federal government has improved the quality of mammography practices. It must be remembewred that it is an imperfect tool but the best we currently have available to detect early breast cancer. We are fortunate aas a community to have such local expertise and commitment make by both Boca Radiology Group and Boca Raton Regional Hospital.

Breast cancer is the number one cause of cancer in American women. We expect to diagnose over 200,000 new cases this year in the USA. For the last several years we have witnessed for the first time a decrease in mortality from the disease. We believe that this is due to improvements in diagnosis and treatment of the disease in the last decade.

Screening mammography has made possible the diagnosis of breast cancer often years before it is palpable, providing a more favorable prognosis. With the increased use of screeening mammography more non-palpable abnormalities are found. We anticipate 1.2 million breast biopsies to be performed in the year 2000 as a result, but expect that 80% of these will be benign . This means that nearly 1 million women will have to undergo a biopsy to find out that there is no cancer

Traditionally, non-palpable breast lesions were evaluated with surgical biopsy after needle localization. This procedure has been considered the "gold standard" to which all others have been compared since the late 1970's. It is a two-stage procedure. Initially, the patient is sent to mammography where a hook wire is placed adjacent to an abnormality. The patient is then sent to the OR where the surgeon performs excisional biopsy as guided by the localizing wire. Specimen radiograph is often taken to prove the excision of the lesion. The skin is then closed. General anesthesia is often required. A two percent failure rate is quoted in the literature. This is not unexpected as the lesions, which may be calcifications, masses or architectural distortions may not be palpable and the procedure is therefore considered semi-blind.

The diagnosis and treatment of non-palpable breast cancer can require 2 surgical procedures if pre-operative diagnosis is not made. The first is needle localized excision. The second, after diagnosis of cancer is made and options are discussed with the patient, is definitive surgery. The goals of the surgeon in all cases are tumor removal, obtaining free margins and node sampling for those with invasive cancers.

Patients are reluctant to consent to the possibility of extensive surgery before a tissue diagnosis is made. We also know that frozen section of breast lesions has its limitations and should not be performed on non-palpable lesions as diagnosis may change on permanent section. It is for these reasons that percutaneous breast biopsy was introduced as an alternative to surgical biopsy of non-palpable breast lesions. Stereotactic biopsy was first introduced in 1990, followed by ultrasound biopsy in 1993.

Stereotactic breast biopsy is most often and reliably performed on a dedicated prone table. The precise location of a lesion in three dimensions can be determined on the basis of its apparent change in position on two angled or stereotactic images. Varying devices, which we will discuss later can be used for tissue sampling. Typically, microcalcifications, masses not seen with ultrasound and densities are evaluated stereotactically.

Ultrasound biopsy technique is usually freehand which allows great flexibility in the approach to the lesion. The ultrasound transducer is held in the physician's non-dominant hand with the needle in the dominant hand. This provides for unlimited angles of approach to lesions situated in any location in the breast. The needle pierces the skin adjacent to the short end of the transducer after local anesthesia is administered. The needle is advanced real time towards the lesion along the plane of the ultrasound beam. Multiple specimens can be obtained quickly, accurately, comfortably and without the use of ionizing radiation.

Varying biopsy tools have been utilized over the years. Initially and currently with palpable lesion, fine needle aspiration has been performed. This technique utilizes a small, 22-25 gauge needle to sample cells from a lesion. Results have typically been disappointing with this technique. Not infrequently results are non-diagnostic or inconclusive and as tissue is not obtained, tumor type or the presence of invasion can not be determined.

The 14 gauge spring loaded core needle biopsy device has been utilized extensively when diagnosing breast lesions. With this device, each sample requires removing and replacing the needle in the breast. It results in discontinuous sampling of a lesion and oftentimes scanty amounts of tissue/blood is obtained. This technique is known to be very difficult when sampling microcalcifications. However, large core permits better characterization of both benign and malignant breast lesions as a histologic diagnosis is given. There is a much lower incidence of insufficient sampling with this technique.

Since 1995, the Mammotome, a directional, vacuum assisted device has been available for stereotactic and now ultrasound guided breast biopsies. It is an 11 gauge hollow needle which is guided towards the abnormality. Vacuum is used to draw tissue into the sample chamber which is then removed by a high speed rotating cutter. The probe can obtain more than one sample contiguously with a single insertion in the breast. By rotating the aperture toward a different position the next sample is obtained. In my practice, the Mammotome is used for all stereotactic breast biopsies as well as all ultrasound guided procedures for masses less than 1.5 cm. Due to the large size of the specimen, lesion removal is possible and placement of a 2 mm tissue marker can be performed at the end of the procedure for future lesion localization and documentation of the biopsy site.

It is found that there is less lesion histologic underestimation with 11 g. Mammotome as compared to 14 g. core. 20-56% of patients found to have atypical ductal carcinoma on 14 g. biopsy are found to have cancer at surgical excision. 0-38% of Mammotome patients diagnosed with atypical ductal hyperplasia are found to cancer at surgical excision. It is for this reason that when the diagnosis of ADH is made at percutaneous breast biopsy, the patient must be sent for excisional biopsy. Other lesions such as Radial scar, atypical lobular hyperplasia, lobular carcinoma in situ and some papillary lesions are somewhat more controversial in their management.

Certainly, it is the biopsy physician's responsibility to determine if the pathologic diagnosis is concordant with the imaging findings when recommending follow up for the patient.

Currently, as percutaneous biopsy is a sampling technique, we recommend a six month follow up for all patients found to have benign, concordant pathology. The accuracy of percutaneous breast biopsy has been found to be similar to that of needle localized excision with a 2% miss rate. Certainly, performing large numbers of cases improves accuracy.

It has been found that a single surgical procedure was performed significantly more often in patients whose non-palpable breast cancer was diagnosed at core biopsy rather than at excision. Core biopsy provides the information necessary to plan surgical therapy and can decrease the number of surgical procedures required in patients with non-palpable breast cancer. A surgeon's approach is different when performing a diagnostic biopsy than a therapeutic operation after core diagnosis of cancer.

At our center, surgical consultation is obtained prior to all percutaneous biopsies. This is of benefit to the patient who will obtain a thorough physical exam to ensure that no suspicious palpable abnormalities are present and to establish a relationship with the patient should further follow up be necessary.

Percutaneous breast biopsy has become the standard of care for nearly all non-palpable breast abnormalities in our community. The majority of lesions are BIRADS Category 4 of which 20-40% are found to be cancer. Due to the high rate of finding a benign lesion in this population, surgical excision should be avoided. The BIRADS Category 5 patients are found to have a cancer 75-90% of the time. A preoperative diagnosis improves the care of these patients. Occasionally, we are asked to biopsy w patient with BIRADS Category 3 abnormality. 0.5-2% of these patients are found to have a cancer. These patients are only done if follow-up is considered unlikely, prior to breast augmentation or pregnancy or in the high risk patient or one with extreme anxiety.

We have found percutaneous breast biopsy to change the way we approach patients with non-palpable breast lesions. It offers a less invasive, less costly alternative to surgical excision with high accuracy, lack of cosmetic and mammographic deformity and few complications. Patients also benefit from fewer surgeries if cancer is found . We look forward to continued evolution of the diagnostic and therapeutic techniques for breast cancer with the hopes for improved outcomes for our patients.

In late January 2000, the Food and Drug Administration approved for immediate use the General Electric Full Field Digital Mammography system in facilities that are MQSA certified. Up to this time, all mammography was performed using specialized screen film combinations, collecting images using an analog technique and radiographic film. Digital imaging appears to be the greatest breakthrough in breast imaging in the last 25 years.

Digital mammography improves the radiologist's ability to read mammograms, as there is improved contrast between normal and abnormal tissues making diagnosis easier. Images are obtained using digital technique and are displayed on a computer monitor for review. The physician then has the opportunity for image manipulation using enhancement methods not available with standard imaging technique. Contrast can be changed. Magnification can be utilized and overall there appears to be better visualization of the breast especially at the skin line, chest wall and in areas of dense breast tissue. Using computer technique there appears to be better differentiation of abnormal from normal tissue.

Like digital cameras, this technique does not use film. Images are created on a computer screen and are available for review within 10 seconds after the patient exposure. These images are available for physician review within minutes. It has been found that the exam can be performed in half the time of a standard mammogram (yes, compression of the breast is still required!) and in certain patients at a reduced x-ray exposure.

The equipment appears as a standard mammography unit except for the presence of a computer and monitor where images are displayed. It has been found that there is improved physician confidence in interpreting digital screening mammograms resulting in fewer patient callbacks. Additionally, with the advancement of Tele-medicine, these computerized images can be sent anywhere in the world electronically for review. The images are stored electronically making it more efficient for hospitals to archive and retrieve records.

These new digitized images are also suitable for review by a computer with the hopes of detecting missed lesions. The computer in the case acts as a "second reader". One study found that the computer flagged more than 75% of previously missed lesions. Therefore, since even the best dedicated mammographers miss up to 15% of cancers which are detected by a second human double reader, there is tremendous potential for computer assistance in reading mammography.

Several studies have been conducted using digital mammography both in the screening and diagnostic setting. Results indicate the two, film screen and digital technology, are equivalent in their ability to detect breast cancer. The equipment costs about four times that of standard mammography, in the range of 4 to 500,000 dollars.

Thirty-two and one half million mammograms are performed annually. Over 200,000 new breast cancers will be diagnosed in the USA this year. We know that the earlier we find these cancers, the better chances are for patient survival. We hope through the use of this new and exciting technology that more women will be screened and smaller, curable breast cancers can be found.

RECENT LITERATURE