[Radical prostatectomy in prostate adenocarcinoma. Clinical factors influencing the pathological stage. Diagnostic model].
MATERIAL AND METHODS: Study on the efficacy of stage diagnosis, how to support it based on clinical objective data and description of a prognostic model. Analysis of 160 patients diagnosed with localized prostate adenocarcinoma undergoing radical prostatectomy in the Clínica Universitaria de Navarra between 1988-1997. The statistical study used Fisher's or Pearson's tests for the comparison of qualitative variables. A logistic regression multivariate analysis was run to avoid confounding factors in the pathological stage. RESULTS: 85/160 (53%) were correctly staged. Incorrect staging occurred in patients with higher clinical stage (T1-T2a: 25%; T2bc: 65%). The univariate study shows that the pathological stage is significantly correlated to: a) serum PSA levels (15 ng/mL in P2 vr. 25 ng/mL in P3-4), the most suitable cutoff value being 15 ng/mL. b) digital rectal examination and Gleason. Negatively influencing factors in the multivariate study were: PSA greater than 15 ng/mL, Gleason greater than 5 and a T2bc clinical stage. Risk groups: 4 risk groups are established based on the above factors (inclusion in group 1 involves an 8% risk of having P3, 30% in group 2, 56% in group 3 and 84% in group 4). CONCLUSIONS: The clinical factors with influence in the pathological stage are PSA, Gleason and clinical stage. The reliability of the risk groups established based on these factors is remarkable.