Breast birads 4 and above lesions: Magnetic resonance imaging findings and pathologic correlation

Aim: In this study, we retrospectively evaluated breast lesions diagnosed BIRADS 4 and above with MRI and investigated diagnostic reliability of the morphologic findings and contrast enhancement patterns and kinetics in correlation of pathologic diagnosis. Material and Method: In our study, patients with BIRADS 4 and above lesions diagnosed on MRI were retrospectively evaluated between August 2013 – January 2015. Breast MR examination of all patients were obtained with 1,5 T MR and 3T MR Device. Conventional sequences were taken and after that dinamik contrast enhanced MR examinations were obtained. Lesions diagnosed with malignancy were classified according to morphological features, distribution, enhancement patterns, kinetic types, and histopathologic diagnosis. Results: 114 patients were investigated, mean age of all patients was 48. Most of malignancy cases were between 40-60 yrs old. Almost half of BIRADS 4 diagnosed lesions were benign and half were malign, all of the BIRADS 5 and 6 patients were malign. 89 patients had breast ductolobular unit Malignancies as 66 lesions (%74.2)IDC (invasive ductal carcinoma), 14 lesions (%15.7)ILC (invasive lobular karsinoma), 6 lesions (%6.7)DCIS (ductal karsinoma insitu) ve 3 lesions (%3.4) LCIS (lobular carcinoma insitu). They were classified be morphologic findings, internal enhancement and enhancement kinetics. Most of ducktal malignancy cases were mass like, lobular malignancies lesions had mass like and nan mass like appearance in the same rate. Most of 68 malign mass lesions had irregular shape and irregular and spikulated margins, there was no significant statistic difference between shape form but there was significant statistic difference between margins and irregular and spikulated margins sound for malignancies. Most of mass like IDC (%70.5) & ILK (%55.5) cases had irregular shape, and spikulated and irregular margins. Most of invasive & non invasive mass like carcinomas showed heterogeneous enhancement. Most non mass like lesions showed segmental and multisegmental distribution( for both ductal & lobular pathologies). According to internal enhancement most cases (%66.7) had clumped enhancement. Clumped nodular enhancement was significantly higher in malignancies and there was significant statistic difference. But internal enhancement was not so helpful to distinguish breast malignancy sub types. In dynamic Contrast-enhanced series most of 89 ductolobuler malignancies showed type 2 (plato) and type 3 (wash out) kinetic curves. In opposition type 1 kinetic curves had higher prevalence in the benign lesions. According to pathologic sub type, there was higher prevalence of type 2 (plato) kinetic curves in both ductal and lobular sub types. Conclusion: Mass like appearance was significantly higher in IDC cases, lobular pathologies and DCIS can show both mass like and non mass like appearance. Lobular pathologies showed more 9 multi-centricity. Most of non mass like malignancies had segmental & multisegmental distribution and clumped nodular internal enhancement. In our study ,in Contrast-enhanced dynamic series the prevalence of type 2 curves in malignancies was higher than type 3, but most of type 3 curves was seen in IDC cases. In breast magnetic resonance imaging there are many morphologic parameters for malignancy such as shape , margins, distribution pattern, enhancement pattern and also kinetic parameters. Morphologic findings has the most important role especially margins in determining malignancy, but each of this criterias are not enough to diagnosis alone and must be considered all together in correlation. It is not possible to differentiate pathologic sub types by this criterias but these criteria can suggest probability of specific sub type. Considering all morphologic, enhancement, kinetic patterns would increase diagnostic reliability