Utility Of Multi-parametric MRI For Differentiating Prostatitis From Prostate Cancer In Peripheral Zone

Part ? Utility of multi-parametric MRI for DiscriminatingProstatitis from Prostate Cancer in Peripheral ZoneObjective To evaluate the value of multi-parametric MRI(Mp MRI)in the differential diagnosis of prostate cancer(PCa)and prostatitis in peripheral zone(PZ).Materials and methods A total of 77 patients with PCa in PZ and 29 with prostatitis in PZ were retrospectively analyzed,who underwent multiparametric MRI(T2WI,DWI,and DCE-MRI)examination and all patients were confirmed by biopsy.Two radiologists assessed cases for multi-parametric MRI features(T2-morphology,boundary of transition zone and peripheral zone,lesion shape)and DWI(slightly high signal intensity or very high intensity on high b-value images,and ADC values).Then,perfusion type were evaluated.Furthermore,All MRI features were initially screened by chi-square test,then indexes with significant difference were entered into multivariate logistic regression analysis.The statistical indicators and the receiver operating characteristic(ROC)curve analysis of each method were compared to histopathological findings obtained by transrectal prostate biopsy.Results Pca was present in 73%;chronic prostatitis was present in 37%.Prostatitis more frequently showed bilateral diffuse inhomogeneous low signal intensity,swelling of the prostate peripheral zone,unclear boundaries of transition zone and peripheral zone.slightly hyperintensity on DWI and type ?.The distribution of curve types for Pca was type?2.6%(2/77),type?18.2%(14/77)and type ? 79.2%(61/77).While the numbers for prostatitis was type ? 13.8%(4/29),type ? 82.8%(24/29)and type ? 27.0%(1/29)respectively.The mean ADC values in cancerous and prostatitis regions were(0.80±0.20)×10-3mm2/s and(1.17±0.24)×10-3mm2/s,respectively(P<0.001).An ADC ?0.893×10-3mm2/s achieved the highest predictive value for prostatitis(AUC = 0.882,sensitivity,93.1%;specificity,70.1%).Multivariate analysis revealed that occurring bilaterally,mass effect,unclear boundary between transition zone and peripheral zone was the risk factors of peripheral zone prostatitis on T2 WI and lesion shape,heterogeneous intensity,slightly signal intensity on DWl and perfusion type were risk factors of peripheral zone prostatitis on Mp-MRI.In the ROC analysis,the correlative areas under the ROC curves(AUC)were 0.960,0.845,and 0.869 for T2 WI,DWI,and DCE,respectively,and 0.982 for Mp-MRI.The combination of T2WI+DWI+DCE achieved the highest predictive value for prostatitis(AUC=0.982).Conclusion The multi-parametric MR imaging features improve the power of MR imaging in discriminating prostatitis from PCa.The diagnostic accuracy of prostate cancer is increased through a combination of the three techniques.Part ? Utility of 1H MR Spectroscopy for DifferentiatingProstatitis from Prostate Cancer in Peripheral ZoneObjective To defferentiate prostatitis and prostate cancer of peripheral zone of prostate gland with MR spectroscopy(MRS).Methods 25 patients with PZPCa(peripheral zone prostate cancer)and 25 patients with PZ prostatitis were recruited,MRI and MRS characteristic were analyzed retrospectively,the values of [C(choline)+C(creatine)]/C(citrate)of low signal intensities area were calculated.The CC/C value of prostate cancer and prostatitis were compared by t-test.Referring the MRS curve of contrary side normal tissue as the standard,The citrate/choline-level were campared between prostate cancer and prostatitis of peripheral zone.Results The CC/C value of prostate cancer was 3.0±2.48,that of prostatitis was5.2±7.08,without significant statistic difference(P= 0.306);Both prostate cancer and prostatitis showed Cit peak decreased obviously,whereas prostate cancer showed Cho peak increased sharply,Cho peak of prostatitis is normal.Conclusion The CC/C value can not be used to defferentiate prostate cancer from prostatitis,whereas the change of Cho peak play an important role,the Cho peak higher,the greater possibility of PCa.Part ? Differentiation of prostatitis and prostate cancer inPeripheral Zone using the Prostate Imaging-Reporting and DataSystem(PI-RADS)Objective To determine whether prostate cancer(PCa)and prostatitis in the peripheral region can be identified by using PI-RADS.Methods 3T MR images of 77 patients with PCa and 29 chronic prostatitis were analyzed.The findings were correlated with histopathology.T2 WI,DWI,DCE,and MR-Spectroscopy(MRS)were acquired.Every lesion was given a single PI-RADS score for each parameter,as well as a sum score and a PI-RADS v2 score.Results The single PI-RADS score for T2 WI,DWI,DCE,as well as the aggregated score including and not including MRS,and the PI-RADS v2-score were all significantly higher for PCa than for prostatitis(p<0.001).Of these parameters,the aggregated score including MRS achieved the highest predictive value for the presence of prostatitis with an AUC of 0.937.Conclusion Prostatitis can be differentiated from PCa using PI-RADS,Prostatitis usually presented borderline pathological PI-RADS scores.