[Multifactorial urodynamic study of 115 patients with dysuria and benign prostatic hypertrophy]

Discriminant analyses based on preoperative urodynamic parameters have proved inaccurate in predicting functional results after prostatectomy. To investigate the cause of this failure, we studied a group of 115 patients consecutively referred for prostatism and selected for prostatic surgery only on clinical criteria and flow rate. A preoperative urodynamic work-up was performed, including cystometry and urethrometry. The patients were reexamined 2 months postoperatively and underwent control uroflowmetry. All preoperative urodynamic data were computed simultaneously using principal component analysis and canonical discriminant analysis. The significance of the diagrams obtained with these multifactorial analyses was more closely examined than in previous studies. The results of classification by the discriminant function demonstrating the best combination of preoperative urodynamic variables resulted in 44% of patients being correctly rated. Examination of the diagrams showed that the major cause of failure in previous studies was the great variety of urodynamic conditions in men with prostatism, which indicates a need for multifactorial interpretation of the results of urodynamic explorations.