Magnetic resonance imaging of the prostate with spectroscopic imaging using a surface coil. Initial clinical...

The purpose of this study was to evaluate the diagnostic ability of proton magnetic resonance spectroscopic imaging (MRSI) in the detection and localisation of prostate cancer, prospectively compared with histopathologic findings.

Magnetic resonance imaging (MRI) and MRSI were performed on 39 patients with prostate-specific antigen (PSA) levels greater than 4 ng/ml and suspicious findings at trans-rectal ultrasound (TRUS). All patients underwent a TRUS ten-core biopsy within 30 days according to a subdivision of the prostate into octants. All studies were interpreted by a dedicated radiologist who reported the areas of interest as normal, equivocal or suspicious on MRI. At MRSI, cancer was defined as possible if the ratio of choline plus creatine to citrate exceeded mean normal peripheral zone values by two standard deviations (SD) or as definite if that ratio exceeded the normal value by three SD. MRI and MRSI findings were spatially correlated with findings obtained from individual biopsy sites.

MRI and MRSI alone had sensitivity, specificity, positive and negative predictive values and diagnostic accuracy in the detection of prostate cancer equal to 85%, 75%; 53%, 89%; 65%, 88%; 77%, 74%; and 69%, 79%, respectively. These values were 70%, 89%, 88%, 74% and 79% when MRI and MRSI were combined. Site-by-site analysis of MRI and MRSI findings and biopsy results yielded no significant correlation.

The combination of MRSI and MRI provides a significantly higher specificity in the detection of tumours as compared with MRI alone and can be recommended as a problem-solving modality before biopsy in patients with high PSA levels and suspicious TRUS.