不同计算化高扩散敏感系数对外周带前列腺癌第二版前列腺影像和数据报告系统中弥散加权成像评分的影响

Effects of different computed high b-values on diffusion weighted imaging scores in Prostate ImagingReporting and Data System version 2 of prostate cancer in peripheral zone

目的探讨不同计算化高扩散敏感系数（b值）对外周带前列腺癌第二版前列腺影像和数据报告系统（PI-RADSv2）中弥散加权成像（DWI）评分的影响。方法回顾性分析2012年1月至2015年12月在同济大学附属同济医院经直肠超声引导下前列腺饱和穿刺活检或前列腺根治术后病理证实外周带前列腺癌104例患者的临床资料。所有患者均采用SiemensVerio3．0TMRI扫描，成像序列包括横断位、矢状位高分辨T2加权成像（T2WI），b值=0、50、1000s／mm2横断位弥散加权成像（aDWI）及动态对比增强（DCE）扫描，通过Matlab后处理计算化合成b=1000、1400、2000s／mm2的DWI图像（eDWI）。依据PI．RADSv2评分标准对b=1000、1400、2000s／mm2的三组cDWI资料进行评分，并对各组病例的病灶信号强度比（SIR）进行独立样本t检验及单因素方差分析。结果三组cDWI中DWI评分2、5分的病例数无差别，27例b=1000s／mm2DWI评分3分的病例中有9例（33．3％）在b=1400s／mm2及b=2000s／mm2时DWI评分均升高到4分。升高至DWI评分4分的病例与仍为DWI评分3分的病例间的病灶SIR在b=1400s／mm2及b=2000s／mm2时的比值分别为1．86±0．21比1．61±0．27、2．18±0．26比1．75±0．30，两组问SIR差异均有统计学意义（t=2．486、3．671，均P〈0．05）。升高至OWI评分4分的9例病例的病灶SIR在b=1000、I400、2000s／mm2时两两之间差异均有统计学意义（F=10．907、33．768、8．043，均尸〈0．05），且病灶SIR均随b值的升高而增大。结论b≥1400s／mm2的高b值主要影响DWI评分3分的病例，但b=2000s／mm2不改变b=1400s／mm2时的DWI评分，b=1400s／mm2可能更适用于外周带前列腺癌的PIRADSv2中DWI评分。

Objective To determine the effects of different computed high b-value on diffusion weighted imaging （DWI） scores in Prostate Imaging Reporting and Data System version 2 （PI-RADS v2） of prostate cancer in peripheral zone. Methods A retrospective study of 104 cases of prostate cancer in peripheral zone was conducted, all of the patients were histopathologically confirmed by transrectal ultrasound guided saturation biopsy or radical prostatectomy in Tongji Hospital of Tongji University from January 2012 to December 2015. All MRI imaging examinations were performed by using a 3.0T Siemens Verio MRI scanner. The imaging protocol consisted of high-resolution axial and sagittal T2 weighted imaging （T2WI） , axial acquired diffusion weighted imaging （aDWI） with b = 0, 50, 1 000 s/mm2 and dynamic contrast-enhanced （DCE） scans. Computed diffusion weighted imaging （cDWI） images with b = 1 000, 1 400, 2 000 s/mm2 were processed by Matlab. These three groups of cDWI images were analyzed according to the PI-RADS v2 criteria, and signal intensity of ratio （SIR） of lesions were analyzed by independent t testand one-way ANOVA in each group. Results The numbers of cases with a DWI score of 2 and 5 were similar among three groups. Nine cases （33.3%） in all 27 cases with a DWI score of 3 on b = 1 000 s/ram2 upgraded to score 4 when b-value rose to 1 400 and 2 000 s/ram2. The ratios of SIR of lesions in cases upgraded from DWI score 3 to 4 to those unchanged cases on b = 1 400 and 2 000 s/mm2 were 1, 86 ±0. 21 to 1.61 ± 0. 27 and 2. 18 ± 0. 26 to 1.75 ± 0. 30, respectively （ t = 2. 486, t = 3. 671, both P 〈 0. 05 ）. In these 9 cases who upgraded to DWI score 4, SIRs of the lesion were significantly different between groups when b = 1 000, 1 400 and 2 000 s/ram2 （F = 10. 907, 33. 768, 8. 043, all P 〈 0. 05）, and their SIRs increased with the rising of b-value. Conclusions The computed high b-value （b 1〉 1 400 s/mm2 ） mainly affects cases with a DWI score of 3, but DWI scores would not change neither in b = 1 400 s/ram2 nor in 2 000 s/ram2 cases. For DWI score in PI-RADS v2 of prostate cancer in peripheral zone, b = l 400 s/ram2 is probably of more scoring value than b = l 000 or 2 000 s/ram2.