Background Enhancement Patterns on Screening Breast MRI – Impact on Biopsy Rate and Short Interval Follow Up.

Abstract

Abstract Introduction:Marked background enhancement is one of the perceived limiting factors on breast MRI possibly obscuring suspicious lesions and reducing both the sensitivity and specificity of detection of small cancers. Background enhancement on MRI refers to the normal physiological enhancement of the woman's fibroglandular tissue and varies with age, hormonal status, menstrual cycle and breast glandularity. Patterns of enhancement vary greatly and marked enhancement can occur even when the MRI is performed in week 2 of the menstrual cycle or in postmenopausal women.Aim:To determine what impact background enhancement has on biopsy rate and short interval follow up rate.Method:231 consecutive baseline screening MRIs, in women aged between 16 -79 (average age 48) performed between Jan 2003 and Oct 2004 were selected. Indications for screening included significant family history (46.3%), personal history of contralateral mastectomy (25.1%), personal history of atypia (19.5%), other (3.0%), personal history of mantle radiation (2.2%), personal history of breast cancer > 5 years previously (2.2%), personal history of ovarian cancer (1.7%).The background enhancement in each case was described as minimal (< 25% glandular tissue showing enhancement), mild (25-50% enhancement), moderate (50-75% enhancement) and marked (> 75% enhancement).The BIRADS category, positive predictive value of biopsy and false positive rate was determined and compared for enhancement patterns.Results:Of 231 screening studies 26.6% demonstrated minimal enhancement, 32.8% mild enhancement, 22.7% moderate enhancement and 17.9% marked enhancement. Patients with minimal/mild enhancement had significantly higher rate of normal (BIRADS 1/2) examinations than those with moderate/marked enhancement (47.8% vs. 30.5%, p = 0.008). Patients with minimal/mild enhancement had significantly lower rate of six month followup (BIRADS 3) (38.97% vs. 53.68%, p = 0.02). No significant difference in biopsy rate was noted (11.7% minimal/mild enhancement vs. 14.7% moderate/marked enhancement, p = 0.5). No significant difference in positive predictive value of biopsy between the groups was noted (35.7% for moderate/marked enhancement vs. 25% for minimal/mild enhancement, p = 0.52).Conclusion:No significant association was found between background enhancement pattern and biopsy rate or false positive rate. However, patients with moderate/marked enhancement were significantly more likely to be given a category of BIRADS 3 and assigned to short interval follow up. The increased rate of BIRADS 3 in patients with moderate/marked background enhancement may be related to reduced reader confidence in these patients. Recognition of normal background enhancement patterns is important to limit the number of patients undergoing short interval followup. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4020.