Bladder shape impact on the accuracy of ultrasonic estimation of bladder volume

Objective: Variations in bladder shape can lead to errors in ultrasonic est imation of the bladder volume. The purposes of this study were to compare t he accuracies of various formulas to estimate bladder volume from sonograph ic measurements and to assess the impact of bladder shape on the accuracy o f bladder volume estimation. Study design: Twenty-four healthy volunteers (13 men, 11 women) and 55 spin al cord injury patients (48 men, 7 women) underwent ultrasonographic measur ements of dimensions before and after voiding. Bladder shape was classified as cuboid, ellipsoid, or triangular prism. Ten formulas from the literatur e were applied to estimate bladder volume, using the volume voided or cathe terized as the standard, and then linear regression was used to obtain opti mal correction coefficients for the whole data set as well as each of the t hree bladder shapes. Setting: Rehabilitation hospital affiliated with a medical college. Results: The most accurate of the 10 formulas tested was height (H) X trans verse depth (Dt) X weight (W) X 0.7 (mean error 17.4% +/- 11.6%). Linear re gression analysis yield optimal correction coefficients of .72 for the whol e data set and .89, .81, and .66 for cuboidal, ellipsoid, and triangular pr ism-shaped bladders, respectively. The mean error for the estimation of bla dder volume using .72 as the correction coefficient was 16.9% +/- 11.9% and decreased to 12.7% +/- 10.1% (p < .0001, paired t test) when the bladder s hape was taken into account. Conclusion: Using bladder shape for reference and applying the correspondin g correction coefficient to volume calculations will improve the accuracy o f the estimation. (C) 1998 by the American Congress of Rehabilitation Medic ine and the American Academy of Physical Medicine and Rehabilitation.