DOI:10.2214/AJR.06.5013
AJR 2006; 186:902
© American Roentgen Ray Society
Computer-Aided Detection (CAD) of Amorphous Calcifications
Ferris M. Hall
Beth Israel Deaconess Medical Center and Harvard Medical School
Boston, MA 02115
Soo et al. [1] make the
important evidence-based conclusion that "CAD sensitivity for malignant
amorphous calcifications is markedly lower than previously reported.... Breast
imaging radiologists who use CAD systems should continue to search diligently
for these difficult-to-detect lesions." These authors note that in their
series these calcifications had a 16% rate of malignancy and were
"frequently overlooked because they are at the threshold of visibility
on routine mammography" and that "identification and biopsy of
all (italics by Dr. Hall) cases of amorphous calcifications are
necessary."
I would like to put these findings and statements in a different context,
suggesting a less rigorous—and more controversial—approach to the
assessment of subtle, lower-risk amorphous calcifications. Ductal carcinoma in
situ (DCIS) is often considered two different diseases: high and low grade.
Amorphous calcifications usually fall into the latter group and, like many
prostate carcinomas, they have been shown to have a relatively long quiescent
stage, measured in years or decades. Most biopsies of amorphous calcifications
are benign, and even when the diagnosis of low-grade DCIS is made, there is
general agreement that relatively few lives are saved by this very early
mammographic diagnosis as compared with finding small invasive carcinomas.
Therefore, in my opinion, there is less urgency about diagnosing low-grade
DCIS, particularly in a population of regularly screened women. The vast
majority of women with early low-grade DCIS would have these lesions picked
up, still at a very early stage, often years later, with little reduction in
mortality.
Mammographers in the United Kingdom and most other affluent countries
ignore or follow these calcifications more often than do their counterparts in
the United States, which undoubtedly contributes to their substantially lower
call-back rates. This presumably reflects population affluence, medical costs,
and medicolegal climates. Of course, the ideal cost-benefit ratio of any
examination is in the eyes of the beholder.
In conclusion, those amorphous breast microcalcifications, which are at the
lower end of suspicion, have not only a low probability of being cancer but,
more important, also have a low, albeit finite, survival benefit from biopsy
and removal even when they turn out to be DCIS. I believe many of us do too
many biopsies for these subtle amorphous calcifications, particularly those
that often cannot be seen well enough to stereotactically biopsy and can only
be wire-localized using magnification mammograms. The positive predictive
value for the biopsy of calcifications is less than for masses but, in my
opinion, it should be above 20%.
References
- Soo MS, Rosen EL, Xia JQ, Ghate S, Baker JA. Computer-aided
detection of amorphous calcifications. AJR2005; 184:887
-892[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
References
