Department of Radiology, The Affiliated Wuxi People's Hospital of Nanjing Medical University, No.299,Qingyang Road, Wuxi, 214000, Jiangsu Province, China.
Department of ophthalmology, The Affiliated Wuxi People's Hospital of Nanjing Medical University, No.299, Qingyang Road, Wuxi, 214000, Jiangsu Province, China.
Eur J Radiol. 2019 Sep;118:138-146. doi: 10.1016/j.ejrad.2019.07.009. Epub 2019 Jul 10.
To prospectively assess the feasibility of diffusional kurtosis (DK) imaging for distinguishing prostate cancer(PCa) from benign prostate hyperplasia (BPH) in comparison with standard diffusion-weighted (DW) imaging, as well as low-from high-grade malignant regions.
147 consecutive patients with suspected PCa underwent multi-parametric 1.5-TMR. Diffusion kurtosis imaging was acquired with with 5 b values (0,600,800,1600,and 2400sec/mm2).Region of interest (ROI)-based measurements were performed on ADC, D, and K map by two radiologists. Data were analyzed by using mixed-model analysis of variance and receiver operating characteristic curves. Correlations among the three parameters (ADC,D and K) in all patients, and correlations between three parameters with the tumor Gleason score (GS) in PCa group were analyzed using Pearson's correlation coefficient in peripheral zone(PZ) and transiton zone(TZ).
58 patients were proved with PCa (9 GS 3 + 3[PZ/TZ = 4/5], 49 GS ≥ 7 [PZ/TZ = 26/23]), and 89 patients were with BPH. ADC,D and K were able to distinguish benignance from tumor tissue both in PZ and TZ(P＜0.01), but performed poorly in neither differentiating low-(GS 3 + 3) from high-grade (GS≥3 + 4) disease, nor GS(3 + 4) from GS(4 + 3).There was a weak correlation between the GS and ADC, D (PZ:ADC r=-0.113, D r=-0.139; TZ:ADC r=-0.104,D r=-0.103), while a moderate correlation between the GS and K(PZ:K r = 0.492; TZ:K r = 0.433, P＜0.01).K had significantly greater area under the curve for differentiating PCa from BHP than ADC both in PZ and TZ.
DK model may add value in PCa detection and diagnosis, but none can differentiate low-from high-grade PCas (including GS=3+4 from GS=4+3).
前瞻性评估扩散峰度（DK）成像在区分前列腺癌（PCa）和良性前列腺增生（BPH）方面的可行性，并与标准扩散加权（DW）成像以及低级别和高级别恶性区域进行比较。
147 例疑似 PCa 的患者连续接受了 1.5TMR 多参数检查。使用 5 个 b 值（0、600、800、1600 和 2400sec/mm2）采集扩散峰度成像。两位放射科医生在 ADC、D 和 K 图上对感兴趣区域（ROI）进行了测量。使用混合模型方差分析和受试者工作特征曲线对数据进行了分析。分析了所有患者的三个参数（ADC、D 和 K）之间的相关性，以及 PCa 组中三个参数与肿瘤 Gleason 评分（GS）之间的相关性，使用外周区（PZ）和移行区（TZ）中的 Pearson 相关系数进行分析。
58 例患者被证实患有 PCa（9 例 GS 3+3[PZ/TZ=4/5]，49 例 GS≥7[PZ/TZ=26/23]），89 例患者患有 BPH。ADC、D 和 K 均可区分 PZ 和 TZ 中的良性和肿瘤组织（P＜0.01），但无法区分低级别（GS 3+3）和高级别（GS≥3+4）疾病，也无法区分 GS（3+4）和 GS（4+3）。GS 与 ADC、D 之间存在弱相关性（PZ：ADC r=-0.113，D r=-0.139；TZ：ADC r=-0.104，D r=-0.103），而 GS 与 K 之间存在中度相关性（PZ：K r=0.492；TZ：K r=0.433，P＜0.01）。K 在 PZ 和 TZ 中区分 PCa 和 BPH 的曲线下面积均显著大于 ADC。
DK 模型在 PCa 检测和诊断中可能具有一定价值，但无法区分低级别和高级别 PCa（包括 GS=3+4 和 GS=4+3）。
