| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Surgery, Ohio State University, N908 Doan Hall, 410 West 10th Avenue, Columbus, OH 43210. e-mail: burak.1@osu.edu; American Society of Breast Surgeons, Dickstein Cancer Center, White Plains, NY 10601
Editor:
The article by Dr Dennis and colleagues in the April issue of Radiology (1) nicely illustrates the importance of quality breast imaging in symptomatic women. We congratulate the authors on their work; however, we have substantial concerns about how this information will be interpreted and applied in clinical practice.
The authors clearly point out that the 100% negative predictive value of mammography and ultrasonography (US) in imaging palpable breast lumps may be a reflection of their practice setting and the exceptional expertise of the radiologists involved in the study. This is of paramount importance and should be taken into account should other, less experienced mammographers try to reproduce these results.
More important, the use of the phrase palpable lumps is misleading. The majority (58%) of the physical finding descriptors (304 of 522) were classified as thickening, while only 22% (113 of 522) were classified as discrete. As the imaging results revealed, all of these lumps were variants of normal breast tissue, which brings into question whether the term lump is an appropriate descriptor of the physical finding.
The lump was most frequently reported by the patient (47%), while only 5% of patients were referred by a surgeon. It is unclear from this article what percentage of patients underwent an appropriate clinical breast examination (a prerequisite to any radiologic study) prior to imaging, because the descriptors of the breast abnormality were obtained from the radiology request form or by asking the patient. It would seem that a great number of these lumps might not have required this additional radiologic (US) work-up had a thorough physical breast examination been performed. Because of this lack of information, it is difficult to draw conclusions regarding the negative predictive value obtained with use of an appropriate preimaging clinical breast examination in patients with a discernable mass. Of particular concern are patients with invasive lobular carcinoma, which can often be a diagnostic dilemma. Patients will commonly present with subtle thickening at physical examination and a negative mammogram. Butler et al (2) reported that 26% of patients with invasive lobular carcinoma and a negative mammogram had sonographically invisible tumors. Findings from a second study (3) revealed that in 9% of patients with a palpable lesion and negative mammographic and US findings, carcinoma was identified at surgical excision.
Again, we congratulate the authors on their fine work; however, we hope that readers of this article do not take the message that a negative radiologic work-up is a substitute for a thorough clinical breast examination, particularly when a clinically suspicious abnormality exists.
REFERENCES
Sally Jobe Breast Centers, Radiology Imaging Associates, 8200 East Belleview Avenue, Suite 102, Englewood, CO 80111 e-mail: mark.dennis@riaco.com; Surgical Consultants of Aurora, Colo
We appreciate the concerns and gracious commentary offered by Drs Burak and Lerner in regard to our article (1). The roles of imaging and surgical subspecialists in the evaluation of palpable abnormalities of the breast are indeed complementary in most practice settings today. Surgical referral directly from primary care has been the standard for years. First-line triage by means of breast imaging, as suggested in our article, is a relatively new concept, and a breast treatment paradigm shift may eventually occur as experience with breast US increases. However, we agree that during this transitional time of early experience with US triage, caution must be exercised. Appropriate didactic training in breast US and a supervised palpation-guided real-time US experience are required and are indispensable. Breast radiologists must be prepared to accept the responsibility of patient triage to biopsy.
Although there may be debate as to what constitutes a lump versus a thickening, the bottom line is that our primary care colleagues in the "trenches" frequently are not comfortable in the categorizing or triaging of palpable breast abnormalities: There is simply something in the breast that they and/or their patient can palpate that just does not seem right. Therefore, in the interest of good patient care and decreased liability exposure, these patients are referred for consultation and further evaluation. It is at this juncture that the philosophy of many surgeons diverges from our own approach as imagers. This is perhaps the heart of the commentary by Drs Burak and Lerner.
In our model of initial imaging triage, a woman with a palpable abnormality benefits from an expeditious and complete diagnostic pathway that includes (a) another physical examination (by a breast radiologist), (b) complete diagnostic imaging, and, if necessary, (c) minimally invasive percutaneous breast biopsy, all performed in one sitting. In contrast to primary triage performed by surgeons in the traditional manner, the patient who may not yet have undergone mammography after her primary care appointment makes an appointment with a surgeon. The surgeon examines the patient and then refers her for imaging evaluation (mammography and US). If a surgeon declines to order US when a lump is present, a reasonably high accuracy of the surgeon’s physical examination is implied. After imaging evaluation, the woman returns to the surgeon. The surgeon then decides if the patient requires biopsy and arranges for imaging-guided percutaneous biopsy or surgical lumpectomy at another date. This regime can add weeks to diagnostic evaluation and patient anxiety.
Additionally, even if it were logistically possible to refer every patient with a palpable abnormality to a breast surgeon, there continues to be the very real specter of subjective variability in the interpretation of the physical examination. Also, there is less than ideal specificity of physical examination for breast cancer when it is performed by primary care physicians or surgeons (2,3).
With the assumption that it were possible for all physicians examining the breasts to consistently and reliably distinguish a situation of purportedly lower concern (thickening) from a more worrisome finding (discrete lump), the distinction may not be clinically relevant. For example, Drs Burak and Lerner express concerns about invasive lobular carcinoma and describe subtle thickening, which may be the presenting symptom. Therefore, if only a preselected group of women with discreet lumps found at surgical consultation were allowed to undergo US, a number of cancers would be potentially overlooked. Our practice of not discriminating between a discrete mass and a thickening may be responsible for successful use of our management algorithm for more than 10 years, with excellent negative predictive results.
We apologize if the manner in which we reported some of the data in the article is misleading. Although we state that nearly half of the palpable findings were discovered by the patients, these patients were subsequently examined, and findings were corroborated by their primary care physician. We assume that this would be considered "appropriate breast examination" prior to imaging referral. Self-referral for lumps in our system is very unusual. We also cannot state how many of the study patients ultimately visited a surgeon for a second opinion after they were examined with US. What we can say, however, is that despite the possible multiplicity of referrals or lack thereof, these patients did not develop cancer in the area of the original palpable abnormality over a mean 43-month follow-up.
As long as there is a human element involved in patient care, there will be an opportunity for mistakes. It is our charge as conscientious physician-imagers to minimize mistakes as much as possible with continuing education, as well as judicious use of consultants in other subspecialties. As the experience of breast imagers across the country increases, we believe that efficient and possibly more objective consultation (palpation combined with real-time visual appraisal with US) at initial imaging will become more commonplace and that surgical consultation will be increasingly dedicated to therapy. For now, we agree that the job of triage must be placed in the hands of those who have the most experience, greatest accuracy, and the willingness to accept final responsibility. Depending on the locality, these doctors may be either breast surgeons or breast radiologists.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
