To develop an individualized nomogram model and explore the feasibility and veracity of nomogram to predict calcium oxalate stones in patients with urinary calculus.

Clinical data of 298 patients with urinary calculus who underwent surgery in the Fifth Affiliated Hospital of Sun Yat-sen University from January 1, 2017 to December 31, 2018 were retrospectively analyzed. The patients were randomly divided into a development group and a validation group by 7∶3 ratio. The least absolute shrinkage and selection operator regression (LASSO) model and multivariable logistic regression analysis were used to select the best prediction characteristics of calcium oxalate stones based on the development group, and a prediction model was constructed in the form of a nomogram according to the best prediction characteristics. Discrimination, calibration, and clinical usefulness of the nomogram were assessed respectively using the C-index, calibration plot, and decision curve analysis, and external validation was assessed based on the validation group.

The best predictive features selected in the LASSO model include stone location, triglycerides (TG), and urine specific gravity (SG). After the gender, age and the best predictive characteristics were used to establish a nomogram model, the C indexes of the development group and the validation group were 0.706 and 0.603, respectively, indicating that the model had good discrimination ability. The standard curve in the calibration curve fit well with the predicted calibration curve, which indicates good calibration. Decision curve analysis showed that the calcium oxalate stones nomogram was clinically useful when intervention was decided at the calcium oxalate stones possibility threshold of 31%.

A nomogram prediction model for the prediction of calcium oxalate stones was established. This model is helpful in screening and early identifying patients who are at high risk of calcium oxalate urinary stones, and is significant to help urologists make clinical treatment decisions.