Abstract
Background: Considering the non-invasive nature of magnetic resonance spectroscopy (MRS) and its ability to detect prostate lesions, the present study aimed to investigate the accuracy of MRS techniques in distinguishing between prostate cancer (PCa) and prostatitis.
Methods: Thirty-three patients (18 patients with PCa and 15 patients with prostatitis) were recruited for this study. Magnetic resonance imaging (MRI) and MRS were performed using 1.5-T system GE- modle Optima 450 Discovery (GE Medical Systems, US). The (Cho+Cr)/Cit ratio of hypointense T2 areas were calculated. The diagnostic accuracy including sensitivity and specificity indices, with 0.95 confidence interval as well as PPV and NPV were calculated for each variable. The receiver operating characteristic (ROC) area under the curve (AUC) was outlined and investigated. The mean quantitative values between the two groups (PCa and Prostatitis) were compared using independent t test.
Results: The mean ratios of Cho+Cr/Cit in PCa was 1.54 ± 0.63 and 0.83 ± 0.48 for PCa and prostatitis, respectively, indicating a significant statistical difference (P = 0.00). A reduction in citrate was seen in both PCa and prostatitis tissue. Significant elevation in choline peak was shown for PCa. Moreover, creatinine level was low in both normal tissue and PCa without significant difference. Sensitivity, specificity, accuracy, PPV and NPV of MRS were 94.4% (95% CI, 74.2–99), 80% (95% CI, 54.8–93), 96%, 85% and 92.4%, respectively.
Conclusion: The results of this study indicate an acceptable level of sensitivity, specificity and accuracy of MRS in the differential diagnosis of PCa and prostatitis.