Extracorporeal Shock Wave Lithotripsy (ESWL) is the recommended treatment for renal stones ≤20 mm or ureteric stones ≤10 mm, according to the European Association of Urology guideline. Predicting the number of ESWL sessions required to achieve treatment success is valuable for patient counselling, resource planning, and improving cost-effectiveness. We aim to develop a novel prediction model for the number of ESWL sessions needed to treat urolithiasis.

Eligible patients aged ≥16 years who underwent ESWL within the Oxford University Hospitals Trust from January 2013 to June 2021 were included in this retrospective cohort study. Patient demographics, stone characteristics, and ESWL variables were extracted at the first ESWL session for each stone episode. Univariate and multivariate analyses were conducted using Stata BE (version 18.0).

A total of 5,711 stone episodes were included, 1,201 (21%) of which achieved treatment success with ESWL. 67% (n = 808) of patients required only one ESWL session. Univariate analysis using Poisson regression identified age, stone size, stone location, number of shocks, Storz Medical Lithotripsy Index (SMLI), and focus of the shockwave as statistically significant predictors of the number of ESWL sessions. These variables were subsequently included in a multivariate analysis, with results shown in Table 1.

The results showed that size and location of stone (kidney/ ureter), SMLI, focus of the shockwave (precise, extended or combined) are all statistically significant predictors of the number of ESWL session required to achieve complete stone clearance. These predictors can be obtained during the first ESWL session, and calculate the need for repeat ESWL treatment. Patients who might benefit from ureteroscopy and laser lithotripsy could be identified early to avoid unnecessary ESWL session.

The ability to predict the number of ESWL sessions required is a valuable tool for shared decision-making and identifying the most suitable treatment for urolithiasis. To date, this is the first study to attempt such prediction. The next phase involves developing a probability calculator for clinicians to integrate into daily practice.