Diffusion tensor imaging of prostate at 3.0 Tesla

Abstract

Background

The feasibility of diffusion tensor imaging (DTI) of prostate has been confirmed by several studies. Apparent diffusion coefficient (ADC) and fractional anisotropy (FA) values of prostate with healthy or pathological changes remain controversial.

Purpose

To investigate the efficacy of DTI at 3.0 T in diagnosing prostate cancer in the peripheral zone and to determine the normative ADC and FA values of the normal prostate.

Material and Methods

T2-weighted imaging (T2WI) and DTI of the prostate was performed in 33 cases with diagnosis confirmed by biopsy at 3.0 T MR. ADC and FA values of cancerous and non-cancerous sextants were compared. Receiver-operating characteristic (ROC) curves were used to compare the ability of ADC, FA and the two values combined in diagnosing prostate cancer in the peripheral zone. DTI data of 20 healthy young volunteers were also acquired. ADC and FA values of the normal central gland and the peripheral zone were compared.

Results

ADC value (1.10 ± 0.09 × 10⁻³ mm²/s) was lower while FA value (0.37 ± 0.05) was higher in normal central gland, compared with that in normal peripheral zone (1.63 ± 0.15 × 10⁻³ mm²/s, 0.21 ± 0.07, respectively) (P < 0.0001, P < 0.0001, respectively). Sixteen of 33 cases were confirmed to be malignant while 17 of 33 cases were benign. All the 198 sextants were confirmed by biopsy, including 136 non-cancerous sextants and 62 cancerous sextants. There were significant differences between cancerous sextants (1.02 ± 0.16 × 10⁻³ mm²/s, 0.38 ± 0.09) and non-cancerous sextants (1.22 ± 0.14 × 10⁻³ mm²/s, 0.31 ± 0.06) for both ADC and FA values (P <0.0001, P <0.0001, respectively). Significant differences were noted between the AUC of DTI and FA alone (0.86 vs. 0.76, P = 0.0009), but no differences between the AUC of DTI and ADC alone (0.86 vs. 0.84, P = 0.1595).

Conclusion

ADC and FA values of normal prostate may be compatible with the microstructural organization of prostate. Furthermore, DTI may be a potential tool in diagnosing prostate cancer in the peripheral zone.