Recurrence in patients submitted to radical prostatectomy. Statistical analysis of the predictive parameters

After some years of radical surgical treatment of prostatic carcinoma, we retrospectively evaluated our case series and analysed the predictive parameters, which were deducible by means of an analytical and statistical comparison between clinical data, pathological data and disease recurrence. From February 1990 to December 1996, 205 patients were submitted to radical prostatectomy at our Institution. The observation periods of this consecutive series vary between 96 and 14 months. All patients were submitted to clinical staging by means of: rectal exploration, transrectal prostatic ultrasound, total PSA, bioptic grade, clinical stage, pathological stage, lymph nodal positivity, grade of the operative specimen, tumor volume, relationship between tumor volume and glandular volume. 188 of the 205 patients are alive. 3 died of disease progression; 14 of unrelated causes. There were 2 serious early complications which required reoperation. There were 19 late complications. The statistically significant correlated parameters were: grade, because we discovered a very high correspondence (87.7%) between the grade of the biopsy sample and that of the subsequent prostatectomy; stage, because we found a correspondence (52.1%) between clinical and pathological stage; tumor volume, which was highly significant (p= 0.0001) in the correlation between pathological stage and lymphnodal state. During follow up, we found histological confirmation of local recurrence in 11 patients. At the time that recurrence was diagnosed, all these patients bad PSA values in progression. In patients submitted to prostatic mapping the correspondence between clinical and pathological stage increased by 18.6%. Therefore, it is confirmed as the only instrument capable of improving clinical stage. The statistical correlation between positive lymph nodes and pathological stage is confirmed (p: 0.002). Tumor volume is shown to be an important predictive parameter, correlating with all the other prognostic factors.