вход по аккаунту



код для вставкиСкачать
Malignant Breast Masses Detected Only by
Ultrasound: A Retrospective Review
octors Gordon and Goldenberg should be complemented on their
prodigious work using breast ultrasound, which they reported in the
August 15 edition of Cancer.’ Unfortunately, the authors’ suggestion that
“a survey scan of the ipsilateral breast be performed whenever diagnostic
ultrasound is indicated for a palpable or mammographic finding” could
be misconstrued in our litigious environment. It is important to make it
clear that this is not the standard of care, and that the standard of care
should not be based on a single report. As the authors have nicely summarized, the entire world’s experience prior to this report, including their
own first 15 patients, includes only 46 cancers that have been reported
as having been detected by ultrasound alone. Their most recent report
is retrospective, and leaves many unanswered questions.
The authors provide insufficient information regarding the type, size,
and stage of the cancers that they detected so that the benefit of finding
these lesions cannot even be inferred. A randomized, controlled trial
would be needed to actually prove the efficacy of screening using ultrasound. The authors calculated their results based on 44 lesions, but, as
they indicate, these calculations actually involved only 30 women. Fifteen
of the women had breast cancer that was already evident by either clinical
examination or mammography. They speculate on the potential impact
of finding the additional foci of cancer in these women, but they do not
actually evaluate what the impact was in their 15 patients. In at least one
of the patients (shown in Fig. Z), the mammography was performed six
months prior to the detection of the two cancers so that it is not clear
that the second cancer was actually mammographically occult.
As they point out, the authors actually are reporting on only 15
women who apparently had an occult cancer that was detected by ultrasound alone. These 15 patients were among 12,706 women who were
scanned over the 5-year period (0.1%). The details of detection, the time
involved in finding the cancers, and the scanning techniques used are
not provided.
As the authors point out, it should be emphasized that since this was
not a prospective, blinded study, the true benefit of screening a breast
using ultrasound cannot be deduced. Surveying a breast using a handheld transducer is extremely operator-dependent. It is very easy to skip
over portions of the breast. Although they state that the scanning takes
only “a short time,” they have not documented the exact length of the
study. Breast ultrasound is sufficiently complex that it probably should
be performed by a physician. It appears that the authors evaluated approximately 3000 women each year using ultrasound. Conservatively assuming a ten-minute examination for each woman, this would involve at
least two hours each day of radiologist time just performing the ultrasound studies. As they suggest, the costlbenefit ratio of this technique
cannot be ignored.
This article provides an interesting, anecdotal experience, but the
implications are sufficiently important that the work must be reproduced
0 1996 American Cancer Society
in a prospective fashion with a clear description of specific patient selection criteria, scanning techniques, and
more detailed results. Until the results are corroborated,
and the other fi3ctors, such as actual benefit as well as
cost/benefit ratios, evaluated, ultrasound should not be
used to routinely scan a breast in search of occult lesions.
Gordon PB, Goldenberg SL. Malignant breast masses detected only by ultrasound: a retrospective review. Cancer
Daniel B. Kopans, M.D., F.A.C.R.
Stephen A. Feig, M.D.
Edward A. Sickles, M.D.
Department of Radiology
Massachusetts General Hospital
Haruard Medical School
Boston, MA
Author Reply
e agree with Dr. Kopans et al. that survey scans during breast ultrasound are not yet the standard of
care. We also agree that the actual benefit of finding these
“ultrasound-only” cancers cannot be estimated in terms
of mortality reduction. However, if we are all in
agreement that our goal is early detection, then clearly
these cancers were detected earlier with ultrasonography
than they would have been by screening mammography
or clinical breast examination. With what is known about
the sample sizes required to prove mortality reduction
with mammographic screening, it is unlikely that such a
study could ever be performed for breast ultrasonograPhY.
Our colleagues correctly stressed the fact that ultrasonography is operator-dependent. However, their comment that it is easy to skip areas while scanning can be
avoided by using a transducer with a wide field of view
and overlapping the scanning rows. This is exactly the
technique used in our study. The 7.5 megahertz linear
array probe described in the article has a field of view
that is 6 cm in width. Therefore, even when overlapping
the scans, it would take only two to three minutes of
scanning time per breast, assuming that the findings are
negative. Clearly, when a mass is found, it requires a
greater amount of time to study and measure the mass.
We are in agreement that breast sonography should be
performed by the radiologist, but since the radiologist is
already performing the scan in patients who are referred
because of palpable findings or mammographic findings,
the additional two to three minutes should not be significant.
The cost of the increased scanning time varies c o ~ i siderably by center. In British Columbia, the fee for uiltrasonography (including both technical and professional
components) is $46.51 Canadian dollars per patienl, regardless of whether one or both of the breasts is scanned.
Given the fact that magnetic resonance imaging (MRI) is
now being investigated for the very indications suggested
in our paper, ultrasonography can certainly be regarded
as the “poor man’s MRI.” Hopefully, there will soon be
data comparing the effectiveness of MRI not only with
palpation and mammography but also with ultrasonograPhY.
Perhaps the greatest reason that our colleagues are
reluctant to embrace breast ultrasonography is the fact
that there is no permanent global record of the study.
Ultrasound is unique in this regard not only for breast
examination but examination elsewhere in the body. All
other radiologic techniques include images that can be
reviewed in retrospect. The finding of an abnormali~yon
ultrasonography is based on the perception of the operator, and if no abnormality is detected, then no number
of normal images elswhere in the breast can prove, i n
retrospect, that no abnormality was present. This should
not deter us from using the technology, however. We can
regard the ultrasound examination as being more analogous to physical examination or other clinical techniques,
such as endoscopy.
With respect to the medical legal implications, if a
patient has a negative sonogram and then six months
later is diagnosed with breast cancer, it is impossible to
say whether the cancer has become detectable since the
normal scan or whether it was present at the time of the
scan but not perceived. This would be similar to a patient
who has a normal physical examination and them six
months later finds a breast lump herself, or a patient who
has a negative cystoscopy and then six months later is
diagnosed with a bladder tumor. The weaknesses of ultrasonography can be acknowledged without abandoning
its potential usefulness.
Finally, we have heard from numerous collea,gues
who have similar experience in finding cancers only with
ultrasound, and we await publication of further research
to corroborate our findings.
Paula B. Gordon, M.D.
S. Larry Goldenberg, M.D.
Department of Radiology
The University of British Columbia
Vnncoiriw, HC
Без категории
Размер файла
187 Кб
Пожаловаться на содержимое документа