Abstract
Objectives: To define prognostic factors that affect the success rate after extracorporeal shock‐wave lithotripsy (ESWL) of renal calculi and to estimate the probability of stone‐free status using a regression analysis model.
Material and Methods: Between February 1992 and February 2002, 2954 patients with single or multiple radiopaque renal stones (<30 mm) underwent ESWL monotherapy. The results of treatment were evaluated after 3 months of follow‐up. Treatment success was defined as complete clearance of the stones with no residual fragments. The stone‐free rate was correlated with stone features and patient characteristics using the χ 2 test. Factors found to be significant using the χ 2 test were further analyzed using multivariate regression analysis.
Results: At 3‐month follow‐up, the overall stone‐free rate using ESWL monotherapy was 86.7%. Failure to disintegrate the stones was observed in 7.3% of cases (n = 216) and failure to clear the fragmented stones occurred in 6% (n = 177). Repeat ESWL was needed in 53% of cases. Static steinstrasse occurred in 4.9% of cases (n = 146) and post‐ESWL auxiliary procedures were required in 4% (n = 118). Using the χ 2 test, patient age (p < 0.001), stone size (p < 0.001), location (p < 0.001), number (p < 0.001) and nature (p = 0.003), radiological renal picture (p < 0.001) and congenital renal anomalies (p < 0.001) had a significant impact on the stone‐free rate. Multivariate analysis excluded stone nature from the logistic regression model while other factors maintained their statistically significant effect on success rate, indicating that they were independent predictors. A regression analysis model was designed to estimate the probability of stone‐free status after ESWL. The sensitivity of the model was 83%, the specificity 91% and the overall accuracy 87%.
Conclusion: Patient age, stone size, location and number, radiological renal features and congenital renal anomalies are prognostic factors determining stone clearance after ESWL of renal calculi. Our regression model can predict the probability of the success of ESWL with an accuracy of 87%.