Short-interval follow-up mammograms for "probably benign" lesions have low sensitivity

Tuesday, 30 November 1999 00:00 |

Diagnostic short-interval mammograms obtained to follow patients with "probably benign" lesions have a low sensitivity for detecting cancers diagnosed within the following 12 months, according to a paper in the May issue of the American Journal of Roentgenology.

Using data from the Breast Cancer Surveillance Consortium, Dr. Erin J. Aiello Bowles, of the Group Health Center for Health Studies in Seattle, and colleagues examined the accuracy of 45,007 short-interval follow-up mammograms interpreted three to nine months after a probably benign assessment on a screening or diagnostic examination.

The mammograms were linked with patient characteristics and breast cancer diagnoses within 12 months. In addition, the authors linked a subset of 13,907 mammograms with radiologist characteristics.

Breast cancer was diagnosed in 8.0 per 1000 women (0.8 per cent) within six months and 11.3 per 1000 women (1.1 per cent) within 12 months. For cancers diagnosed within 12 months, the sensitivity was 60.5 per cent. Sensitivity was 83.3 per cent for cancers diagnosed within six months. Specificity was 97.2 per cent and 97.3 per cent at six months and 12 months, respectively.

The authors suggest several possible reasons for the low 12-month sensitivity. "First, cancers assessed as probably benign...may not grow as rapidly as cancers that appear more suspicious for malignancy."

Second, the radiologist who interprets the follow-up study may get a false sense of security from the fact that a colleague had interpreted the first study as "probably benign."

A third possibility may be that the standard 12-month auditing definition for follow-up period may not be appropriate.

The bottom line, the investigators conclude, is that "the reasons for this low sensitivity" need to be elucidated.

AJR 2008;190:1200-1208