RESUMO
<p><b>BACKGROUND</b>Transurethral resection of prostate (TURP) has been widely used as a golden standard therapy of benign prostatic hyperplasia for over 40 years. However, not all patients achieved favorable outcome postoperatively. Since the level of bladder outlet obstruction and the dysfunction of detrusor (overactive and underactive) were both found to affect surgical efficacy, urodynamics was recommended as routine preoperative examination in selecting proper surgical candidates by International Continence Society in spite of its invasiveness and high cost. The aim of this research was to compare the predictive value between ultrasonography and urodynamics for TURP efficacy and determine if preoperative urodynamic test could be replaced by ultrasonography.</p><p><b>METHODS</b>Two hundred and seventy-one patients took part in the retrospective analysis. All the subjects had preoperative evaluation of symptoms, life quality, and combined examination of ultrasonography and urodynamics. Surgical efficacy was measured according to the recovery of international prostate symptom score, quality of life score, and maximal flow rate 6 months after TURP. Fisher's linear discriminant analysis was applied to establish the predictive models of surgical efficacy by choosing parameters from ultrasonography or urodynamics as independent factors. Receiver's operating characteristic curve was then plotted to compare the values between the models.</p><p><b>RESULTS</b>Sensitivity, specificity, positive and negative predictive value of models consisting of parameters from both ultrasonography and urodynamics were favorable. Corresponding models of ultrasonography and urodynamics were found to have non-significant difference in area under curve (P > 0.05).</p><p><b>CONCLUSIONS</b>Preoperative ultrasonography has as strong value as urodynamics does in predicting surgical outcome of patients undergone TURP and might take the place of urodynamics in selecting surgical candidates. Further prospective analysis with larger popularity and longer period of follow up should be launched to verify the result of this research.</p>