Original articleNew-generation nephrometry systems: head-to-head comparison of tumor contact surface area and resected and ischemic volume

ObjectivesWe propose a calculus-based formula to calculate contact surface area (CSA). We examined the correlation of CSA and renal volume loss and the predictability for renal function after partial nephrectomy (PN).Materials and methodsWe conducted a retrospective study in patients who underwent PN between January 2012 and December 2014. Based on abdominopelvic computed tomography and magnetic resonance imaging, we calculated the CSA with the formula: 2 × π × r × d; where r = radius and d = depth, while resected and ischemic volume (RAIV) was determined by the equation [2w2 + 3w(r + d) + 6rd] × w × π⁄3, where w = width of parenchymal ischemia and resection, r = radius, and d = depth. We evaluated the correlation between CSA, RAIV, and perioperative parameters. We compared the ability of CSA and RAIV to predict the reduction in renal function.ResultsThere were 35, 26, and 45 patients receiving open, laparoscopic, or robot-assisted PN, respectively. The mean ± standard deviation CSA was 30.7 ± 26.1 cm2, and the mean ± standard deviation RAIV was 19.1 ± 14.4 cm3. In Spearman correlation analysis, we found that CSA and RAIV were highly correlated (coefficient: 0.99, p < 0.001). In univariate analysis, body mass index (p = 0.02), estimated blood loss (p = 0.001), RAIV (p < 0.001), and CSA (p < 0.001) significantly affected postoperative renal function (PRF). In receiver operating characteristic curve analysis, both CSA and RAIV had good ability to predict >10% change in estimated glomerular filtration rate (area under the curve: 0.86 vs. 0.87). There was no significant difference in area under the curve between CSA and RAIV. The area difference in PCE10 was 0.002 (p = 0.51).ConclusionCSA and RAIV were correlated with several perioperative outcomes and affected PRF. The ability to predict PRF between CSA and RAIV was nearly identical. CSA was simpler to use, and may possess less interobserver variability in comparison with RAIV. Therefore, we believe that CSA can represent renal parenchymal loss.