Determining residual urine volumes using a portable ultrasonographic device

Residual urine volumes of 24 men with neurogenic bladder dysfunction were repetitively assessed 400 times with the BVI-2000 BladderScan portable ultrasonographic device prior to 100 episodes of intermittent catheterization. By comparing each examiner's first ultrasonographic measurement of urine volume with the catheterized urine volume, the mean error of the ultrasonographic measurements was -26mL (-11%) and the mean absolute error was 44mL (22%). The ultrasonographic measurements detected the presence of residual urine volumes of > or = 100mL with a sensitivity of 90% and a specificity of 81%. In the subset of catheterization episodes in which the catheterized urine volumes were < or = 200mL, the mean error of the ultrasonographic measurements was -15mL (-9%), the mean absolute error was 37 mL (28%), and the sensitivity and specificity were 77% and 81%, respectively. There was no clear advantage in using the average or maximum of two repeated ultrasonographic measurements over using each examiner's first ultrasonographic measurement alone. Increased examiner experience did not significantly decrease the errors encountered.