One-year mammographic follow-up recommendations fall short

May 28, 2008

Short-interval follow-up is most often suggested for women whose diagnostic mammograms are flagged as probably benign. But this second look at 12 months has a low sensitivity for detecting cancer, according to a study in the American Journal of Roentgenology.

Short-interval follow-up is most often suggested for women whose diagnostic mammograms are flagged as probably benign. But this second look at 12 months has a low sensitivity for detecting cancer, according to a study in the American Journal of Roentgenology.

Erin Aiello Bowles and colleagues at several U.S. institutions evaluated the accuracy of short-interval follow-up using data from the Breast Cancer Surveillance Consortium. Bowles is a research associate at the Group Health Center for Health Studies in Seattle. Her coauthors included Bonnie Yankaskas, Ph.D., a professor of radiology and principal investigator for the Carolina Mammography Registry at the University of North Carolina, Chapel Hill.

Their data set included 45,007 women aged 40 to 59 who underwent short-interval follow-up between 1994 and 2004. The majority of the women were asymptomatic. Based on their analysis, the authors found that the average abnormal interpretation rate of short-term follow-up was 3.4%. Breast cancer was diagnosed in 11.3 per 1000 women within 12 months of the short-interval follow-up.

The six-month sensitivity for short-term follow-up was 83.3%. The 12-month sensitivity of all short-term exams was 60.5%. Sensitivity was lower in women with extremely dense breasts, according to Bowles' group. Women who underwent unilateral mammograms had increased sensitivity versus those with bilateral studies.

The average specificity at six and 12 months was 97.3%. Women with dense breasts had significantly lower specificity than women with fatty breasts, the authors said.

Finally, they noted that radiologists who spent more time in breast imaging had increased sensitivity and specificity in comparison with those who devoted less than 10 hours a week to mammography.

Bowles and colleagues offered several possible reasons why 12-month short-term follow-up turned in such low sensitivity.

"Cancers assessed as probably benign (BI-RADS category 3) may not grow as rapidly as cancers that appear more suspicious for malignancy," they said. "[R]adiologists interpreting (these exams) might be reassured by the previous radiologist's probably benign interpretation and thus have a higher threshold for calling the initial examination suspicious."

There may be a discrepancy between the recommended time period for short-term follow-up intervals and what actually happens in clinical practice (AJR 2008;190:1200-1208). Bowles' group called for additional studies on the differences in sensitivity between six-month and 12-month follow-up periods.

In related news, results from the American College of Radiology Imaging Network (ACRIN) 6666 trial on screening with breast ultrasound did find the modality particularly adept in high-risk women with dense breasts (JAMA 2008;299:2151-2163).

Lead investigator Dr. Wendie Berg told Diagnostic Imaging that ultrasound could not, however, be used as an alternative to short-term interval mammography.

"Short-interval follow-up mammography is used for particular findings seen on the mammogram, such as punctate calcifications, circumscribed masses, or focal asymmetries," Berg said. "Ultrasound is usually performed to characterize masses seen on mammography. If a mass is found to be a cyst on initial ultrasound, then we don't need short-interval follow-up. Such workup with ultrasound has typically already been performed before deciding on follow-up: We are following lesions on mammography that we cannot see at all or well enough with ultrasound to characterize as definitely benign."

For more information from the Diagnostic Imaging archives:

Copayments turn older women off to mammography

Large screening mammography study restores faith in CAD

Breast ultrasound adjusts to get past false positives