Objectives:To assess the adherence of academic radiologists in a university center to BI-RADS lexicon (BLA) and to evaluate the structural completeness of breast MRI reports.Materials and Methods:Breast MRI reports made during 2012 in a single academic center by six readers were scored for formal completeness (FS) including recording the MRI protocol, making relevant clinical correlation, and describing background enhancement; BLA including mass rather than lesion, describing lesion outline, enhancement characteristics, and dynamic curve; and also expressing the final conclusion using BLA, resulting in a maximal total score of 8. FS and BLA were correlated with reader characteristics including breast imaging background, years of academic experience, and number of breast MRIs reported yearly. Tests used for statistical analysis were the Mann–Whitney U test and analysis of variance (ANOVA).Results:Overall BLA was 38.9%. This percentage was 60.1% and 3.7% in radiologists with and without breast imaging background, respectively (P = 0.000). Mean FS among all readers was 3.81 ± 1.75. This score was 2.54 ± 1.1 for readers without breast imaging background and 4.6 ± 1.6 for the readers regularly involved in breast imaging (P = 0.000).Conclusions:Higher degree of BLA and higher mean FS were associated with radiologists regularly involved in breast imaging. No association was found with years of academic experience or number of breast MRIs interpreted yearly.