OBJECTIVE Objective parameters in computed tomography (CT) scans that could predict calyceal access during percutaneous nephrolithotomy have not been evaluated. These parameters could improve access planning for percutaneous nephrolithotomy. We aimed to determine which parameters extracted from a preoperative multiplanar reconstructed CT could predict renal calyceal access during a percutaneous nephrolithotomy. METHODS From January 2009 through April 2011, 230 patients underwent 284 percutaneous nephrolithotomies at our institution. Sixteen patients presented with complete staghorn calculi, and 11 patients (13 renal units) were analyzed. Five parameters were extracted from a preoperative reconstructed CT and compared with the surgical results of percutaneous nephrolithotomy. RESULTS Fifty-eight calyces were studied, with an average of 4.4 calyces per procedure. A rigid nephroscope was used to access a particular calyx, and a univariate analysis showed that the entrance calyx had a smaller length (2.7 vs. 3.98 cm, p=0.018). The particular calyx to be accessed should have a smaller length (2.22 vs. 3.19 cm, p=0.012), larger angles (117.6 vs. 67.96, p<0.001) and larger infundibula (0.86 vs. 0.61 cm, p=0.002). In the multivariate analysis, the only independent predictive factor for accessing a particular calyx was the angle between the entrance calyx and the calyx to be reached (OR 1.15, 95% confidence interval [CI], 1.053-1.256, p=0.002). CONCLUSIONS The angle between calyces obtained by multiplanar CT reconstruction is the only predictor of calyx access.