Which Measure of Stone Burden is the Best Predictor of Interventional Outcomes in Urolithiasis: A Systematic Review and Meta-analysis by the YAU Urolithiasis Working Group and EAU Urolithiasis Guidelines Panel

Background and objective: Stone size has traditionally been measured in one dimension. This is reflected in most of the literature and in the EAU guidelines. However, recent studies have shown that multidimensional measures provide better prediction of outcomes.

Methods: We performed a systematic review and meta-analysis of the prognostic accuracy of measures of stone size (PROSPERO reference CRD42022346967). We considered all studies reporting prognostic accuracy statistics on any intervention for kidney stones (extracorporeal shockwave lithotripsy [ESWL], ureterorenoscopy [URS], or percutaneous nephrolithotomy [PCNL]; Population) using multiplane measurements of stone burden (area in mm² or volume in mm³; Intervention) in comparison to single-plane measurements of stone burden (size in mm; Intervention) for the study-defined stone-free rate (Outcome) in a PICO-framed question. We also assessed complication rates (overall and by Clavien-Dindo grade) and the operative time as secondary outcomes. Searches were made between 1970 and August 2023. We used the DeLong method to compare receiver operating characteristic (ROC) curves.

Key findings and limitations: Of 24 studies included in the review, 12 were eligible for comparative analysis with the DeLong test following meta-analysis of prognostic accuracy. For prediction of stone-free status, the area under the ROC curve (AUC) was significantly higher for stone volume than for stone size (0.71 vs 0.67; p < 0.001). Subanalyses confirmed this for ESWL and URS, but not for PCNL. For URS, the AUC was also significantly higher for stone area than for stone size (0.79 vs 0.77; p < 0.001). Throughout all analyses, there was no difference in AUC between stone area and stone volume. There was high risk of bias for all analyses apart from the URS subanalyses.

Conclusions and clinical implications: According to the limited data currently available, stone-free rates are predicted with significantly higher accuracy using multidimensional measures of stone burden in comparison to a single linear measurement.

Patient summary: We reviewed different ways of measuring the size of stones in the kidney or urinary tract and compared their accuracy in predicting stone-free rates after treatment. We found that measurement of the stone area (2 dimensions) or stone volume (3 dimensions) is better than stone diameter (1 dimension) in predicting stone-free status after treatment.