Quantitative iodine-based material decomposition images with spectral CT imaging for differentiating prostatic carcinoma from benign prostatic hyperplasia.

RATIONALE AND OBJECTIVES: To investigate the value of iodine-based material decomposition images produced via spectral computed tomography (CT) in differentiating prostate cancer (PCa) from benign prostate hyperplasia (BPH).

MATERIALS AND METHODS: Fifty-six male patients underwent CT examination with spectral imaging during arterial phase (AP), venous phase (VP), and parenchymal phase (PP) of enhancement. Iodine concentrations of lesions were measured and normalized to that of the obturator internus muscle. Lesion CT values at 75 keV (corresponding to the energy of polychromatic images at 120 kVp) were measured and also normalized; their differences between AP and VP, VP and PP, and PP and AP were also obtained. The two-sample t-test was performed for comparisons. A receiver operating characteristic curve was generated to establish the threshold for normalized iodine concentration (NIC).

RESULTS: Fifty-two peripheral lesions were found, which were confirmed by biopsy as 28 cases of PCa and 24 BPHs. The NICs of prostate cancers significantly differed from those of the BPHs: 2.38 ± 1.72 compared with 1.21 ± 0.72 in AP, respectively, and 2.67 ± 0.61 compared with 2.27 ± 0.77 in VP. Receiver operating characteristic analysis indicated that an NIC of 1.24 in the AP provided a sensitivity of 88% and a specificity of 71% for differentiating PCa from BPH.

CONCLUSIONS: Spectral CT imaging enabled quantitative depiction of contrast medium uptake in prostatic lesions and improved sensitivity and specificity for differentiating PCa from BPH.