Most Doubtful Mammographic Findings Are Benign

Hearing or reading about subtle breast cancer findings that were missed on screening mammography should not prompt the radiologist to change his/her recall threshold unless they have a really valid reason.

In radiology practices that perform mammography, reducing the rate of false-positives is an important area of concern. One way to reduce false-positive rates is by reducing the recall rate for screening mammography.

One approach to reducing the recall rate is by resisting the temptation to change the recall threshold because of new information you may have heard or read about a subtle breast cancer that was missed.

Although the notion of a subtle missed breast cancer is frightening, you do not need to begin recalling everything. Instead, you need to learn what things really need recalling.

Doubtful Mammographic Findings
Those findings identified at screening that may represent precursor lesions, but were judged highly likely to be benign (no recall needed) are classified as “doubtful mammographic findings.”

A study published by Wolverton and Sickles (AJR Am J Roentgenol 1996;167,, 1041-1045) addressed the following problem: for what percentage of exams on which we can see a cancer in retrospect could we have seen the cancer prospectively?

In about 50% of breast cancers that are visible at screening, some precursor finding can be seen in retrospect on the screening exam just prior to the most recent one. We asked, “Can we increase our true-positive rate by changing our threshold for recall so that we actually find these cancers one or two years earlier?” The mean follow-up was 2.5 years.

Of findings evaluated, 75% were unchanged (presumed benign), 20% went away or decreased (benign), 3% were slightly more prominent, but so slight that they were not recalled on the second exam (presumed benign), and 1% were read as abnormal because they increased on the second exam (underwent workup).

Three workups were benign. Three workups led to biopsy, with one low-grade ductal carcinoma in situ (DCIS). This particular low-grade DCIS was completely excised, and the woman remains disease-free at 15 years.

Study Conclusions
Virtually all doubtful findings were benign. Of some 500 cases, no woman was harmed by not recalling for these doubtful mammographic findings. Had we recalled all of these cases, our recall rate would have been sky high, and it would not have helped us find many more cancers.

The bottom line is that you should not change your recall threshold unless you have very good reason to do so.

Author: Bokran Won, MD

Reference:
Edward Sickles, MD. Most Doubtful Mammographic Findings Do Not Need Recalling.