The Diagnostic Value Of MRI Dynamic Contrast-enhanced Angiography Combined With DWI In Breast Diseases

Objective:Combining Fischer’s score and diffusion-weighted imaging,the study discussed how to subclassify four types of BI-RADS-MRI lesions and improve the accuracy of BI-RADS classification.Method:Cases of 301 patients who underwent breast MRI examination in our hospital from June 2014 to April 2017 were collected.Among them,150 cases were confirmed by puncture biopsy and operation pathology.7cases were excluded because of the heavy artifacts of DWI,and finally 143 cases（150 lesions）were included in the study.All patients underwent breast MRI plain scan,multi-phase dynamic contrast-enhanced scan and DWI examination before operation.After scanning,the lesions were analyzed morphologically,including morphology,edge,size,enhancement mode,etc.Dynamic enhancement curve,early maximum enhancement rate and ADC value were measured and recorded.According to Fischer’s score（F score）system,Fischer’s+ADC score（F+ADC score）was used to draw the ROC curve of ADC value to differentiate the benign from the malignant,and to determine the best diagnostic threshold of ADC value.Meanwhile,the ROC curves of ADC value,F score and F+ADC score were drawn to differentiate the benign lesions from the malignant ones,and the best diagnostic threshold of ADC value was determined.The area under ROC curves of three diagnostic methods was compared by Z test.The sensitivity,specificity,positive predictive value,negative predictive value and diagnostic accuracy of various diagnostic methods were counted and calculated with pathological results as the gold standard of diagnosis.Kappa was used to analyze the consistency of ADC value,F score and F+ADC score with pathological results.Results:Through ROC curve analysis,the ADC value=1.08×10^-3mm²/s was used as the diagnostic threshold for breast benign and malignant lesions.The area under ROC curve（AUC）was 0.855（95%CI:0.789-0.907）.The area under ROC curve of F score method and F+ADC score method was 0.875 and 0.949,respectively.There was no significant difference between F score and AUC of DWI（Z=0.526,P=0.599）;AUC of F+ADC score was significantly higher than that of DWI and F score（Z=3.359,3.644,P=0.0008,0.0003）.Fischer’s score combined with DWI was better than F score and DWI in diagnosing pathological findings.The F+ADC score was in good agreement with the pathological results（Kappa=0.800）.F score and DWI were moderately consistent with pathological findings（Kappa=0.583,Kappa=0.621）.There were 62 cases of BI-RADS-MRI,including 12 cases of benign lesions and 1 case of malignant lesions in 4A,10 cases of benign lesions,11 cases of malignant lesions in 4B,25 cases of malignant lesions and 3 cases of benign lesions in 4C.The positive predictive values of 4A,4B and4C lesions were 7.7%,52.4%and 89.3%respectively.Conclusion:Fischer’s score combined with DWI can reasonably classify the four types of BI-RADS lesions,and 4A lesions can be used as the diagnostic threshold to judge benign and malignant lesions,which can maintain high diagnostic sensitivity of breast MRI,and significantly improve its diagnostic specificity and accuracy.Objective: Combined with dynamic contrast-enhanced mammography and F+ADC score,the study discussed how to improve the diagnostic accuracy of BI-RADS-MRI,reduce missed diagnosis and misdiagnosis,and lower the clinical puncture biopsy rate.Method: Cases of 301 patients who underwent breast MRI examination in our hospital from June 2014 to April 2017 were collected.Among them,150 cases were confirmed by puncture biopsy and operation pathology.7 cases were excluded because of the heavy artifacts of DWI,and 143 cases（150 lesions）were included in the study.All patients underwent breast MRI plain scan,multi-phase dynamic contrast-enhanced scan and DWI examination before operation.The optimum critical value of ADC is 1.08×10-3 mm2/s calculated in the first part.Four types of lesions were subclassified by F + ADC system.BI-RADS-MRI4 A,4B and 4C were classified according to whether there were increased blood supply asymmetry and adjacent vascular signs.In the absence of vascular changes,it is defined as BI-RADS-MRI4A（-）,4B（-）,4C（-）;in the presence of increased vascular asymmetry or proximal vascular signs,it is defined as BI-RADS-MRI4A（+）,4B（+）,4C（+）.The positive predictive values of BI-RADS-MRI 4A,4B,4C and 4A（-）,4A（+）,4B（-）,4B（+）,4C（-）,4C（+）were calculated.The difference was statistically significant with P< 0.05.Results: There were 62 cases of BI-RADS-MRI.Among them,there were 13 cases of 4A,including 12 cases of benign lesions and 1 case of malignant lesions;21 cases of 4B,including 10 cases of benign lesions and 11 cases of malignant lesions;28 cases of 4C,including 25 cases of malignant lesions and 3 cases of benign lesions.The positive predictive values of 4A,4B and 4C lesions were 7.7%,52.4% and 89.3% respectively（see table 4）.There were 11 cases of benign lesions in 4A（-）;2 cases of 4A（+）,1 case of benign lesions and 1 case of malignant lesions;9 cases of 4B（-）,6 cases of benign lesions and 3 cases of malignant lesions;12 cases of 4B（+）,4 cases of benign lesions and 8 cases of malignant lesions;9 cases of 4C（-）,3 cases of benign lesions and 6 cases of malignant lesions;19 cases of 4C（+）were all malignant lesions.The positive predictive values of 4A（-）,4A（+）,4B（-）,4B（+）,4C（-）,4C（+）were 0,50%,33.3%,66.7%,66.7%,100%,respectively（see table 4）.Conclusion: Asymmetric increase of breast blood supply and adjacent vascular signs can be effective complement to the comprehensive diagnosis of breast cancer.Combined with dynamic angiography and Fischer’s score plus ADC,the classification of BI-RADS4 lesions can be more reasonable,and the rate of biopsy can be effectively lowered.