Consistency of uroflowmetry analysis in children among observers.

Abstract

The aim of the study is to compare the intra- and inter-observer interpretations of the same uroflowmetry study at two different times. Two-hundred children with a voided volume of 50% above the expected bladder capacity were included. All traces were asked to be evaluated by 11 observers two times in a time span of 1 month. These observers consist of pediatric urologists (n = 2), pediatric urology fellows (n = 2), urology residents (n = 5), and certified urodynamics nurses (n = 2). Each uroflowmetry was asked to be assessed for three domains: voided volume (VV), detrusor sphincter dyssynergia (DSD), and flow curve pattern (FCP). Of the 200 patients with a median age of 10 (4-18) years, 128 (64%) were girls and 72 (36%) boys. The maximum flow rate and the median voided volume were found to be 20 (4-61) mL/s and 232 (116-781) mL. The Fleiss' kappa coefficient of VV, DSD, and FCP in the first assessment was 0.510, 0.501, and 0.346. In the second assessment, κ values were 0.530, 0.422, and 0.373. The best-agreed findings were similar at both times. These were found to be low VV (0.602 and 0.626) and intermittent pattern (0.500 and 0.553). Interpreters were found to have a statistically significant difference in agreement with their own interpretation at different times. Both inter- and intra-observer reliability of the agreement point out the problem in the standardization of uroflowmetry. Inter- and intra-observer reliability of uroflowmetry interpretation can be increased by defining precise numbers and numerical algorithms.