DKI对前列腺癌的鉴别诊断及临床病理分级的定量预测研究

Quantitative research of DKI in the differential diagnosis of prostate cancer and its correlation with pathological grades

目的:探讨扩散峰度成像(Diffusion kurtosis imaging,DKI)相关定量参数对前列腺癌(Prostate cancer,PCa)与前列腺增生(Benign prostatic hyperplasia,BPH)间的鉴别诊断并评估其在临床病理分级中的预测价值。方法:回顾性分析经手术或穿刺活检病理证实的57例前列腺病变,其中PCa 25例,BPH 32例。所有患者均行前列腺常规MRI及DKI扫描,获得病灶的平均扩散峰度(Mean kurtosis,MK)、平均扩散系数(Mean diffusivity,MD)以及各向异性分数(Fractional anisotropy,FA)值,采用受试者工作曲线(Receiver operating characteristic curve,ROC)评估各参数对PCa及BPH的鉴别诊断效能。分析各DKI参数与美国癌症联合会(American joint committee on cancer,AJCC)第8版PCa临床病理分级的相关性,P<0.05认为差异有统计学意义。结果:BPH和PCa组间MK、MD值的差异有统计学意义(P值均为0.000),其中MK值诊断效能最大(AUC=0.99)。AJCC PCa不同临床病理分级的组间MK、FA值差异均有统计学意义(P<0.05),不同Gleason评分组间MD值无统计学意义(P>0.05)。MK值与AJCC临床病理分级之间显著正相关(r值分别为0.860和0.805),FA值与其呈中等负相关(r值分别为-0.463和-0.465)。结论:DKI模型定量分析有助于BPH及PCa的鉴别诊断,其中MK值诊断效能较高,且有助于预测PCa的临床病理分级。

Objective: To investigate the value of the related parameters of diffusion kurtosis imaging(DKI) in differential diagnosis between prostate cancer(PCa) and benign prostatic hyperplasia(BPH), and its relationship with pathological grades.Methods: A total of 57 cases of prostate diseases proved by surgery or biopsy, including 25 PCa and 32 BPH, were collected in the retrospective study. All patients underwent routine prostate MRI and DKI scan. Mean kurtosis(MK), mean diffusivity(MD) and fractional anisotropy(FA) values were obtained. The receiver operating characteristic curve(ROC) was used to evaluate the diagnosis efficiency of these parameters on PCa and BPH. The correlation between DKI parameters and the clinical pathological grade of PCa according to the American Joint Committee on Cancer(AJCC, 8 th edition) was analyzed. P<0.05 was considered statistically significant. Results: The difference of MK and MD value between BPH and PCa were statistically significant(P=0.000). The ROC curve showed that the MK value has the greatest diagnostic efficacy(AUC=0.99). There were significant differences in MK and FA value among different clinical groups and Gleason scores(P<0.05). There was no significant difference in MD value among different Gleason scores(P>0.05). There was a significant positive correlation between the MK value of the DKI model and the AJCC clinical classification and Gleason score(r values were 0.860 and 0.805, respectively) and FA values were moderately negatively correlated(r values were-0.463 and-0.465, respectively). Conclusion: Quantitative analysis of DKI model is helpful for the differential diagnosis of BPH and PCa. The MK value is highly sensitive and specific in diagnosis, which can help predic tthe clinical pathological grade of PCa.