Predictors of clinically significant residual stones for renal stones after percutaneous nephrolithotomy

The aim of this study is to analyze the predictors of clinically significant residual stones (CSRS) after percutaneous nephrolithotomy (PCNL). Between March 2006 and January 2008, all consecutive patients with renal stones undergoing percutaneous creation and dilatation of renal access by radiologists for subsequent scheduled PCNL were reviewed. Patients who have completed the whole procedure of PCNL were included in this study. We reviewed clinical characteristics and the follow-up plain radiograph of the kidney, ureter and bladder for assessing CSRS in each patient. The relationships of clinical characteristics and CSRS were analyzed by univariate analysis and multivariate analysis. Univariate analysis showed larger stone size, less body mass index, presence of prior open nephrolithotomy and staghorn stone were significantly associated with CSRS. Multivariate analysis showed that stone size, prior open nephrolithotomy and their interaction were significant predictors of CSRS. In patients with prior open nephrolithotomy, the probability of CSRS is as high as 62.5% even in small (< 3 cm) renal stones. On the other hand, the probability of CSRS in patients without prior open nephrolithotomy increases rapidly with increasing stone size despite a much lower baseline probability. When the stone sizes are larger than 6 cm, the CSRS probability in patients without prior open nephrolithotomy becomes as high as 92.3%, which is much higher than that (50.0%) of patients with prior open nephrolithotomy. In conclusion, stone size, prior open nephrolithotomy and their interaction should be taken into account for selecting patients undergoing PCNL.