| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Viewpoint |
1 From the Department of Radiology, University of California Medical Center, 2330 Post St, Rm 180, San Francisco, CA 94115. Received February 9, 1999; revision requested February 26; revision received March 12; accepted March 24. Address reprint requests to the author.
Index terms: Breast neoplasms, diagnosis, 00.30 • Breast radiography, utilization, 00.30
There is much in Dr Rubin's Viewpoint (1) with which I agree. Clearly, many radiologists use 6-month follow-up mammography inappropriately, either as a (poor) substitute for a full imaging work-up or by ignoring the several recommendations described in my own Viewpoint (exclude palpable lesions, compare with previous mammograms, give follow-up recommendations to patients in person in the radiology department, biopsy lesions that progress during surveillance) (2). In addition, I agree with eliminating 6-month follow-up mammography for multiple bilateral probably benign lesions and have already been doing this for several years. Finally, I also agree that there is little if any justification for 6-month follow-up mammography over the entire period of surveillance, nor have I ever endorsed this practice.
However, Dr Rubin goes too far by proposing the complete elimination of 6-month follow-up mammography. She makes two questionable arguments that deserve further discussion: (a) The added cost and anxiety resulting from 6-month follow-up mammography that affects all women with probably benign lesions does not justify the benefit of earlier detection of relatively few cancers; and (b) since some probably benign lesions may have a slightly higher likelihood of malignancy, these cases might better be managed with prompt tissue diagnosis. This combination of arguments proposes two extreme positions against the middle ground that I prefer to occupy.
First, let us consider the efficacy of Dr Rubin's proposal to select a subset of probably benign lesions for prompt tissue diagnosis, on the basis of patient age, lesion size, and the imaging features depicted by modern ultrasonography (US). Simply stated, there is no convincing scientific evidence that these criteria are efficacious. Rather, my own published experience, involving 1,403 probably benign masses, strongly argues against the efficacy of age and lesion size criteria (3). Even use of the most successful age or size thresholds would result in the biopsy of at least 60 benign lesions for every cancer found, a ratio that should be just as unacceptable to radiologists as it overwhelmingly is to surgeons. Furthermore, modern US does not permit identification of substantially more cancers among probably benign masses, thereby leading to immediate tissue diagnosis. At the University of California at San Francisco, the observed positive predictive value of malignancy for probably benign masses has been 1.4% with use of modern US (calculated for the past 2 years and for the past 5 years), just as it was 1.4% with use of older US from 1978 to 1990 (3). I believe it unwise to select a subset of probably benign masses for prompt tissue diagnosis and to skip 6-month follow-up mammography for the rest until it has been convincingly demonstrated that for the remaining probably benign masses, this approach results in a positive predictive value that is substantially less than 1.4% and is preferably closer to the 0.2% level that has caused me to abandon 6-month follow-up for multiple bilateral probably benign lesions.
Next, let us consider the issue of cost. The 11.6% frequency of probably benign assessment observed in my initial published series (4), on the basis of case accrual from 1978 to 1987, has no bearing on my current practice (or that of others), which now involves a great preponderance of screening examinations. In the screening setting, I now observe only a 1.2% frequency of probably benign assessment, therefore inducing a 6-month follow-up "cost" that is negligible compared to the cost of screening itself. Furthermore, since the cost of percutaneous biopsy is approximately eight times that of 6-month follow-up (5), and the cost of surgical biopsy is approximately two to three times that of percutaneous biopsy (6), substitution of tissue diagnosis for 6-month follow-up for even a minority of probably benign lesions, as suggested by Dr Rubin, disproportionately weakens the argument for reduction in cost.
Concerning the issue of anxiety, as discussed in my Viewpoint (2), 6-month follow-up mammography induces significantly less anxiety than does percutaneous biopsy (7), even though percutaneous biopsy almost always results in a prompt diagnosis of benignity compared to the same benign diagnosis that will come only after several years of surveillance. However, there is increasing evidence that the low level of anxiety imparted by 6-month follow-up may have more benefit than harm, since such not-completely-normal assessments encourage women to comply with mammography screening guidelines (2). I fear that eliminating 6-month follow-up will substantially decrease compliance with what Dr Rubin and I agree are the most important (1-year and 2-year) follow-up examinations in the surveillance protocol.
Finally, does the benefit of 6-month follow-up mammography (ie, earlier detection of some cancers) justify the added cost and anxiety? As discussed previously, I believe that Dr Rubin overestimates the effects of added cost and anxiety and underestimates the value of earlier diagnosis by overlooking the aggressive nature of the cancers that can be found. Table 3 in my Viewpoint shows that the cancers actually detected at 6-month follow-up grow more rapidly and carry a poorer prognosis than the majority of cancers found among probably benign lesions (2). The data also show that cancers not detected because 6-month follow-up was omitted carry an even poorer prognosis. Even though 6-month follow-up will enable detection of cancer earlier in only a few cases (eight of 7,484 [0.1%]), this represents a detection rate not that much smaller than what we consider acceptable for incidence screening mammography, for cancers that grow more rapidly than those usually detected at screening. I believe that the benefit of earlier detection does exceed the associated harms; it makes good sense to assess more frequently in order to detect the most aggressive cancers.
Footnotes
See also the articles by Sickles (pp 11–14 ) and Rubin (p 15–18 ).
References
Related Articles
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
