Multivariance analysis on non-invasive urodynamic parameters for Differentiating DU with BOO in male

Differentiating detrusor underactivity (DU) with bladder outlet obstruction (BOO) can only be done by pressure flow studies. Although a few parameters have been proposed for diagnosing voiding abnormalities, such as flow index1, Delta Q2, peak counting3 and median power frequency4, any single parameter will have limited accuracy and could not exceed the specificity of the simple Qmax<10ml/s to select BOO male patients. We therefore performed multivariate analysis in SPSS on non-invasive urodynamic parameters which have significant statistical difference between BOO and DU groups, including Qmax, Qave, ratio of flow time to voiding time, mean flow rate in rising part, peak counting result, median power frequency parameters and sum of amplitude change in raising slope. Each parameter is assigned with a coefficient according to their diagnosing power, then summed up to create a ‘super parameter’ for achieving the best diagnosing accuracy. From our database of 104 BOO and 93 DU free flow data, the super parameter could achieve 71% sensitivity and 84.6% specificity on differentiating DU from BOO. Moreover, the diagnosing power could be further improved by introducing in additional non-invasive urodynamic parameters.