Salah M. et al., 2025: Performance of 'Triple-D' and 'Quadruple-D' scores compared to a regression-based predictive model for treatment outcomes in extracorporeal shock wave lithotripsy.

Morshed Salah, Maged Al-Ghashmi, Bela Tallai, Mohammed Ibrahim, Tawiz Gul, Maged Alrayashi, Ibrahim Alnadhari, Faisal Ahmed 

Arch Ital Urol Androl. 2025 Oct 22:14265. doi: 10.4081/aiua.2025.14265  FREE ARTICLE

Abstract

Background: Extracorporeal shock wave lithotripsy (ESWL) is a widely utilized, noninvasive treatment for renal and ureteric stones. Accurate prediction of treatment outcomes is essential for improving patient counseling and optimizing clinical management. Established scoring systems, such as the 'Triple-D' score - which incorporates stone Density, Diameter, and skin-to-stone Distance -and the 'Quadruple-D' score - which adds factors like stone location or hydronephrosis status - are used to stratify patients by risk. However, these tools have limitations in predictive accuracy. This study aimed to evaluate and compare the predictive performances of the Triple-D and Quadruple-D scores against a novel regression-based model for ESWL outcomes.

Methods: A retrospective study was conducted on 1,000 adult patients treated with ESWL using the Dornier Compact Delta® III Pro lithotripter from May 2022 to November 2023. Key predictors of ESWL failure were identified using multivariable logistic regression with internal validation. Predictive performances were compared using receiver operating characteristic (ROC) analysis, with statistical differences assessed by DeLong's test. Model calibration and clinical utility were examined through calibration plots and decision curve analysis (DCA).

Results: ESWL treatment success was achieved in 87.5% of patients. Independent predictors of failure included prior urologic intervention (adjusted odds ratio [aOR] 2.64, 95% CI 1.75-3.99), multiple stones (aOR 0.45, 95% CI 0.24-0.77), higher stone density (per 100 Hounsfield Units increase; p<0.001), and increased skin-to-stone distance (per cm; aOR 1.18, 95% CI 1.06-1.30). The regression-based model showed superior discrimination (AUC 0.92) compared to the Quadruple-D (AUC 0.81, p=0.01) and Triple-D (AUC 0.72, p<0.001) scores. Calibration and DCA confirmed the model's improved accuracy and clinical benefit.

Conclusions: The regression-based model outperforms existing Triple-D and Quadruple-D scores in predicting ESWL failure, providing enhanced individualized risk stratification. This may facilitate better patient selection and treatment planning. Prospective validation is warranted.

Comment Hans-Göran Tiselius

The authors of this report focused on prediction of SWL outcome by comparing Triple-D and Quadruple-D scores with an advanced regression-based model. The conclusion was that the regression model was superior to the other two scores.
Although a more precise prediction of SWL was possible with the regression-based model applied by the authors, it is difficult to know if exact prediction really is necessary. Successful SWL is the result of clinical experience by the operator. Any differences between US and fluoroscopic imaging are not mentioned.
The SWL success was 87.5%, but it is not clear whether that number also included patients with residuals up to 4 mm?
Although it is obvious that predictive estimates can be useful from a theoretical point of view, it is unknown how necessary such calculations are for the final clinical outcome of the treatment.

Hans-Göran Tiselius