Reproducibility of renal volume measurement in adults using 3-dimensional sonography

The purpose of this study was to determine the interobserver reproducibility of renal volume measurement performed by 3-dimensional (3D) sonography.

We conducted an exploratory, quantitative, observational, and descriptive cross-sectional study. We calculated ranges, means, and standard deviations of the renal volumes obtained with 3D sonography by 2 different examiners from a sample of 30 patients. The interobserver reproducibility was assessed by calculating intraclass correlation coefficients (ICCs) and by construction of Bland-Altman plots.

The ICC between the right 3D renal volumes obtained by examiners 1 and 2 was 0.8552 (95% confidence interval, 0.7153-0.9292), showing excellent reproducibility (P < .0001). For the left 3D renal volumes obtained by examiners 1 and 2, the ICC was 0.6411 (95% confidence interval, 0.3652-0.8135), showing average to good reproducibility (P = .0001). The Bland-Altman plots showed mean differences ± SD of -4.8% ± 23.9% between the right renal volumes obtained by examiners 1 and 2 and -2.2% ± 38% between the left renal volumes. The 95% limits of agreement between the right renal volumes obtained by examiners 1 and 2 were -28.7% to 19%, whereas those between the left renal volumes were -40.2% to 35.7%.

Three-dimensional sonography showed good interobserver reproducibility, which was better for the right kidney. It appears necessary to establish standardized techniques for acquisition of sonographic renal volumes.