Inter- and intra-observer agreement on evaluating the presence of residual glandular tissue with magnetic resonance tomography following prophylactic mastectomy

BACKGROUND: There are no published international consensus or guideline documents regarding appropriate medical follow-up for women with hereditary increased risk of breast cancer who opt for prophylactic mastectomy. Moreover, it is not known whether breast magnetic resonance imaging (MRI) performed after a prophylactic mastectomy is a reproducible method for evaluating whether clinically relevant amounts of residual glandular tissue remains. PURPOSE: To evaluate the inter- and intra-observer agreement on detecting residual glandular tissue with MRI. MATERIAL AND METHODS: In total, 40 women previously operated with prophylactic mastectomy underwent MRI and two breast radiologists (R1 and R2) independently assessed the presence of residual glandular tissue. Inter- and intra-rater agreements were assessed using Cohen's kappa (k). RESULTS: Residual glandular tissue was found in 69 of 248 quadrants (27.8%) and 32 of 62 breasts (51.6%) by R1 and 77 of 248 quadrants (31.1%) and 35 of 62 breasts (56.5%) by R2. The interrater agreement was observed to be moderate (k = 0.554) and the intra-rater agreement was observed to be substantial (k = 0.623). CONCLUSION: In conclusion, the inter-and intra-rater observer agreement in regard to detection of residual glandular tissue was not excellent, which would be desirable for a method considered reproducible enough to be used as a surveillance tool after the surgical procedure in order to ensure that there is no relevant residual glandular tissue remaining warranting further follow-up. More research is needed, as well as establishment of precise protocols, before using the method in risk assessment of remaining glandular tissue and breast cancer risk.