The Prediction Of Postoperative Renal Function From Renal Cortex Volumetry With Preoperative Multidetector Computed Tomography

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA INTRODUCTION: To predict the postoperative total renal function after renal surgery, we measured preserved renal cortex volume and examined the estimation of postoperative eGFR of from preoperative multidetector CT. MATERIALS AND METHODS: From 2011 to 2013, 39 patients with preoperative contrast enhanced multi detector CT were underwent renal removal surgery at Teikyo University Hospital. Preoperative and postoperative renal function were assessed by estimating glomerular filtration rate (eGFR). The renal cortex volume (RCV) was calculated automatically by imaging volumetry computer software. Addition of normal side RCV and disease side RCV that was subtracted by tumor region determined as total RCV. The postoperative RCV was defined as preserved renal cortex volume (RCV preservation), which is normal side RCV. The factors that affected the postoperative renal function were examined by univariate and multivariate analysis. For developing the models to estimate postoperative eGFR, we used stepwise (backward elimination) method for identifying the predictors. To assess of the relative quality of these statistical models, both Akaike information criterion (AIC) and Root mean square error of the cross validation (RMSE CV) were calculated. Then tour model was examined as the external validation with 10 patients in Jyuntendo University Hospital. RESULTS: The postoperative eGFR value was associated with age (P <0.01), preserved Renal Parenchymal Volume, and preserved Renal Cortex Volume (P <0.01, P =0.03, respectively). With univariate analysis, %GFR alternation significant correlated with % RCV preservation (r = 0.58, p <0.01) and with % RPV preservation (r = 0.54, p <0.01). Age and preoperative eGFR show poor statistically significant correlations. The alteration of renal cortex volume is the strongest predictor for the postoperative eGFR alteration rate.From the results, we identified the independent predictors of eGFR. AIC and RMSE CV values were 198.7 and 60.4 respectively. Our developed regression model as follows: postoperative eGFR=57.87 -0.55(Age)-15.01(BSA)+0.30(preoperative eGFR)+52.92(% RCV preservation). A strong correlation was seen between postoperative eGFR and estimation model (R=0.805, p <0.001 and RMSE value was 7.16). The results of external validation cohort, residual sum of squares of our equation was calculated as 647.77, which show better predictive value than the previous reported estimation models. CONCLUSIONS: Preoperative MDCT based RCV measurements are associated with postoperative renal function in patients undergoing renal surgery. Computer aided image analysis system makes this procedure simple and reproducible method. Combining CT renal volumetry and patient data might be the important prediction tool for postoperative renal function and risk assessment tool for the CKD after renal surgery. Citation Format: Shuji Isotani, Hiroshi Shimoyama, Yasuhiro Noma, Keisuke Saito, Satoru Muto, Hisamitsu Ide, Raizo Yamaguchi, Shigeo Horie. The prediction of postoperative renal function from renal cortex volumetry with preoperative multidetector computed tomography. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 114. doi:10.1158/1538-7445.AM2014-114