Prostate cancer is clinically understaged in 50% of cases. It was improved with the use of random biopsies, PSA analysis (balanced and unbalanced) and Gleason grade on biopsies. Traditional imaging techniques such as endorectal ultrasonography and CT Scan are known to be of no interest. MRI is a non-evasive technique which allows the three dimensional study of an organ with a high quality of tissue contrast. Moreover, the positioning of surface antennae or "phased array" allowed improvement in spatial resolution.
Objective: To determine whether MRI allows an accurate staging of clinically localised adenocarcinoma by the direct estimation of capsular effraction, measure of glandular and tumor volumes and finally by obtaining reliable PSAd.
Material and methods: A prospective study was done from March 1995 to November 1996 on 55 patients with a clinically localised adenocarcinoma. MRI 1 Tesla Imager (SIEMENS), pelvic phased array antennae. FSE sequences (T1 and T2), axial and frontal. Evaluation of volumes by on-screen. Standford protocol for histo-pathologic reading on the axial cuts. Direct planimetry on a millimetre paper.
Results: Reliable estimation of glandular volume (average difference 0.7 cc). Sub-evaluation of tumoral volume (average difference 1.04 cc) with a 30% margin of error, accuracy 87%. Tumoral stage exact in almost 90% of the cases with a PPV of 94%.
Conclusion: Prostatic MRI, using surface antennae, allows to complete the conventional stage by stage development by correctly diagnosing 75% of locally evolved stages. Its limits are linked to the nature of the signal of the tumorous tissue as to the localisation of neoplasic lesions.