To compare freehand scripting and semiautomated renal parenchymal volume measurements on preoperative or postoperative computed tomography scans and assess relationships between parenchymal volume loss and functional changes within the operated kidney after partial nephrectomy (PN).

Fifty patients (16 solitary kidneys, 34 bilateral kidneys) with renal tumors managed by PN with necessary studies for analysis were included. Freehand scripting and semiautomated threshold-based analysis were performed before and 4-12 months after PN to obtain preoperative normal parenchymal volumes, projected residual parenchymal volumes, and actual postoperative volumes. Glomerular filtration rate was determined by the Modification of Diet in Renal Disease 2 equation along with nuclear renal scan to provide split function for patients with 2 kidneys. Limits of agreement and Bland-Altman analyses were performed. The relationship between the amount of vascularized parenchyma preserved and renal function saved was correlated for each measurement method using Pearson correlation.

The semiautomated method yielded estimates that were higher than freehand scripting by a mean of 14 cm³ for estimation of preoperative normal parenchymal volume, 21 cm³ for projected residual parenchymal volume, and 9 cm³ for actual postoperative parenchymal volume. For the semiautomated method, correlation between the amount of normal parenchyma preserved and renal function saved was 0.52 (95% confidence interval [CI], 0.28-0.69; P <.001), and for the scripting method, correlation was 0.60 (95% CI, 0.39-0.76; P <.001).

Semiautomated and freehand scripting measurements of parenchymal volumes were relatively consistent before and after PN, although the semiautomated approach tended to yield volumes that were approximately 5%-10% higher on average. Measurement of parenchymal volume changes by both approaches correlated significantly with functional changes after PN.