3.0T MR DTI技术对外周带前列腺癌的定量分析

Quantitative Analysis of Prostate Cancer in Peripheral Zone with 3.0T MR Diffusion Tensor Imaging

目的:探讨扩散张量成像(DTI)对外周带前列腺癌(PCa)的诊断价值。方法:回顾性分析行常规MRI、DWI及DTI扫描,后经穿刺病理证实的25例外周带PCa患者(恶性组),40例良性前列腺增生(BPH)和/或慢性前列腺炎(CP)患者(良性组)影像学资料,采用ROC的曲线下面积(AUC)值比较ADC、FA值及两者联合诊断外周带PCa的诊断效能,并初步确定ADC、FA的诊断阈值。结果:前列腺外周带癌区与非癌区ADC值和FA值分别为(0.89±0.19)×10^(-3)mm^2/s,0.25±0.05和(1.46±0.23)×10^(-3)mm^2/s,0.17±0.04,两者差异均有统计学意义(t值分别为10.414和-7.789,P值均<0.05)。ADC阈值为1.135×10^(-3)mm^2/s,敏感度、特异度分别为96.0%和95.0%;FA阈值为0.196,敏感度、特异度分别为96.0%和85.0%。ROC曲线上,ADC值、FA值及两者联合的AUC分别为0.974,0.907和0.990,95%置信区间分别为0.900~0.998,0.809~0.965和0.926~1.000。AUC的两两比较,ADC值与FA值、ADC值与两者联合,FA值与两者联合,z值分别为1.498、1.312、2.151,P值分别为0.134、0.190、0.032。结论:DTI评价参数ADC、FA值均可为外周带PCa的诊断及鉴别诊断提供有价值的信息,ADC值较FA值对外周带PCa的诊断效能高,两者联合的诊断效能较ADC值无统计学差异,较FA值差异有统计学意义。

Purpose： To explore the diagnostic values of diffusion tensor imaging （DTI） in prostate cancer in peripheral zone. Methods： Twenty-five patients with prostate cancer in peripheral zone （malignant group） and 40 with benign prostate hyperplasia （BPH） and/or chronic prostatitis （CP） （benign group）, who were all confirmed by biopsy results, were undergone conventional MRI, diffusion-weighted Imaging （DWI） and diffusion tensor imaging （DTI）. Receiver operating characteristic （ROC） curves were used to compare the diagnostic efficiency of ADC value, FA value and the combination of them. The cutoff threshold of ADC value and FA value were determined. Results： The mean ADC value and FA value of the cancerous and noncancerous regions were （0.89-4-0.19）×10^-3mm2/s, 0.25±0.05 and （1.46±0.23）×10-3mm^2/s, 0.17±0.04, respectively. There were statistical significant differences of ADC value and FA value between the cancerous and noncancerous regions （t=10.414, -7.789, P〈0.05）. Using the ADC value of 1.135×10^-3mm^2/s and the FA value of 0.196 as the cutoff threshold, the sensitivity and specificity were 96.0%, 95.0% and 96.0%, 85.0%, respectively. When ADC value, FA value and combination of them were used as criteria, the area under curve （AUC） of ROC were 0.974, 0.907 and 0.990, 95% CI were 0.900-0.998, 0.809-0.965 and 0.926- 1.000, respectively. For the comparison of the differences of AUC for the ADC and FA value, the ADC value and the combination, the FA value and the combination, their z values and P values were 1.498, 1.312, 2.151, and 0.134, 0.190, 0.032, respectively. Conclusion： Both the ADC value and FA value provide valuable information for the diagnosis and differential diagnosis of PCa. The ADC value has higher diagnostic efficiency than the FA value. The difference of diagnostic efficacy between the combination criteria and the ADC value was with no statistical significant, and the difference between the combination criteria and the FA value was with statistical significant.