Application Value Studying Of Dynamic Contrast-enhanced MR Imaging And~1H Magnetic Resonance Spectroscopy To The Diagnosis And Differential Diagnosis Of Breast Cancer

Objective1. To evaluate the diagnostic effectiveness and limitation of3.0T dynamiccontrast-enhanced MR imaging(DCE-MRI）and single-voxel1H magnetic resonancespectroscopy(1H MRS) in patients with breast diseases.2. To investigate if it can improve the diagnostic effectiveness by DCE-MRIassociated with1H MRS.Materials and MethodsSixty-three patients with78lesions were examined with Phlips Acheiva3.0Tmagnetic resonance system which equipped with16-channel breast coil.MR scanincluded T1-weighted imaging(T1WI) and T2-weighted imaging(T2WI),and fatsuppressed T2-weighed imaging;DCE-MRI using Vdyn-eTHRIVE SENSEsequence;and1H MRS by PRESS technology.DCE-MRI is tested on the total78lesions,and1H MRS is tested on51lesions.Images analysis is used Pilips special workstation and software.(1)enhancedmorphological performance were include shape,margin,border,and enhancedform.The irregular,lobulated or speculated margins,and heterogeneous enhancementare considered as three types:typeⅠis steady enhancement,typeⅡis plateau,andtypeⅢ is washout.The TIC typeⅡ and typeⅢ are considered as malignantlesions.(3)Prophase enhancement rate is the enhancement rate2minutes after injectedcontrast agent,and≥50%is diagnose malignant lesions.(4) the MRI BI-RADScategory increases:shape of regular (round,oval) is0point,irregular is1point;margin type of well-defined is0point,ill-defined is1point;enhancement pattern ofhomogeneous is0point,heterogeneous is1point,rim is2points;dynamics of initialenhancement of﹤50%is0point,50～100%is1point,and﹥100%is2points;TIC oftypeⅠis0point,typeⅡis1point,type Ⅲ is2points.The range of total possible scoresis0～8points. BI-RADS categories Ⅲ,Ⅳ,andⅤcorrespond to scores of0～3,4,and5～8,respectively.To improve the sensitivity,we consider the scores≥4points ismalignant disease.(5)1H MRS:Lesions presenting a detectable Choline signal in thespectra at3.2ppm were classified as malignant.(6) Combination of MRI and MRS:iftwo methods is both positive,the combination diagnose is maligance;if one diagnoseof the two methods is negative,the combination diagnose is benign.Statistical analysis use software SPSS13.0. Different imaging technologies areused to emphasize on sensitivity,specificity, positive and negative predictive value andaccuracy.Two independent samples t test is used for measyrement data.The x2test isused for the numeration data.Statistics showed bilateral P value＜0.05that thedifference was statistically significant.Results1. Clinic and Pathology results:The63patients ages is19～62years old，meta-age is42years old.78lesions were confirmed by pathology,including52malignant lesions,37of Infiltrating Duct Carcinoma,8of Infiltrating lobular carcinoma,2ofmedullary carcinoma,2of mucinous adenocarcinoma,3of ductal carcinoma insitu;The minimus diameter is1.6cm,the average diameteris4.1cm;the age is28～62years old,the average age is49.2years old.26benign lesions,fibroadenoma(n=13),puberty enormous mammary glands(n=2),Plasmacellmastitis(n=1),hyperplasia lobular(n=2),papilloma(n=4),lymph nodes in mammaryglands (n=1),chronic granuloma(n=1),fibrocystic mastopathy(n=2);The minimusdiameter is0.9cm, the average diameter is2.7cm; the age is19～53years olds,theaverage age is32.3years old. There is no significant difference between the ages ofbenign and malignant lesions. The benign lesions are smaller than the malignantlesions(p=0.013).2. DCE-MRI:(1) morphologic features:benign lesions had the characteristic ofround,smooth edge,clear boundary,homogeneous enhancement;whereas malignantlesions had the characteristic of more irregular,lobulated or speculated margins,borderunclear,and heterogeneous enhancement.The sensitivity is78.85%,specificity is61.54%, positive predictive value is80.39%, negative predictive value is59.26%, accuracy is73.08%, which was based on the morphologic feature.(2) prophaseenhancement rate:In the52maliganant lesiongs,18lesions(34.61%) is＜50%,15lesions(28.85%) is50%～100%,19lesions(36.54%) is＞100%.In the26benignlesiongs,there are18lesions(69.23%) is＜50%,6lesions(23.08%) is50%～100%,2lesions(7.69%) is＞100%. The sensitivity is65.38%,specificity is69.23%, positivepredictive value is80.95%, negative predictive value is50%, accuracy is66.67%.(3)time-signal intensity curve(TIC):there are15lesions are typeⅠ,12arebenign lesions;the typeⅡhave17lesions,and10are malignant lesions;type Ⅲ have46lesions,and39are malignant.The sensitivity is94.23%,specificity is46.15%,positive predictive value is77.78%, negative predictive value is80%,accuracy is78.21%,supposed the typeⅡand type Ⅲ curve were the characteristic of malignantlesions.(4)MRI BI-RADS:there are23lesions are0～3points,15lesion are4points,40lesions are5～8points.The score≥4points is malignant lesions.Thesensitivity specificity, positive predictive value, negative predictive value, accuracyare92.31%,73.08%,87.27%,82.61%,85.90%.3.1H MRS:51lesions that was to be tested,the choline peak appears3.2ppmdiagnostic indicators of benign and malignant lesions, the sensitivity,specificity,positive predictive value, negative predictive value, accuracy are88.57%,81.25%,91.17%，76.47%，86.27%.4. Combination of MRI BI-RADS and MRS：51lesions take both the DCE-MRIand MRS,the combination diagnose sensitivity is77.14%,specificity is93.75%,positive predictive value is96.43%, negative predictive value is65.22%, accuracy is82.35%.5. The sensitivity of the six method have significant difference（p=0.000）.Thesensitivity of the prophase enhancement rate is the lowest,the sensitivity of MRIBI-RADS connected with MRS is higher than morphologic features and prophaseenhancement rate,but lower than TIC and MRI BI-RADS.The specificity of TIC islower than MRS.And there is no significant difference among the specificity of themorphologic features, prophaseenhancement rate,MRI BI-RADS and MRS.Thespecificity of combination of MRI BI-RADS and MRS is higher than morphologicfeatures, prophas enhancement rate,TIC and DCE-MRI.The specificity of theconnected method is also higher than1H MRS,but there is no significant difference inprinciple of statistics.The accuracy of the six method also have significant difference,accuracy of MRI BI-RADS and MRS are high,but there is no significant difference between them,and the accuracy of combination method have no significantdeduce.The positive predictive value of combination method is higher than other fivemethod.Conclusion1. The morphologic features is helpful for diagnosis and differential diagnosis ofthe breast cancer,but both of the sensitivity and specificity are low.2. The diagnosis sensitivity by TIC is very high,but its specificity is low,theoverlapping of the benign and malignant lesions is high.3. The sensitivity of the prophase enhancement rate is the lowest among all of thesix methods,so its application value in clinical is not high.4. MRI BI-RADS is a diagnose method that connect the morphology andhemodynamics of DCE-MRI,its sensitivity and specificity are both high,and itsapplication value is very high.5. Diagnosing breast cancer only by using Choline peak has low application valuedue to the limitation of the technological and lesions factors.The combination of1HMRS and DCE-MRI can improve the specificity and positive predictive value,and theaccuracy is also high.