A nomogram clinical prediction model for predicting urinary infection stones: development and validation in a retrospective study.

PURPOSE: This study aimed to develop a nomogram prediction model to predict the exact probability of urinary infection stones before surgery in order to better deal with the clinical problems caused by infection stones and take effective treatment measures. METHODS: We retrospectively collected the clinical data of 390 patients who were diagnosed with urinary calculi by imaging examination and underwent postoperative stone analysis between August 2018 and August 2023. The patients were randomly divided into training group (n = 312) and validation group (n = 78) using the "caret" R package. The clinical data of the patients were evaluated. Univariate and multivariate logistic regression analysis were used to screen out the independent influencing factors and construct a nomogram prediction model. The receiver operating characteristic curve (ROC), calibration curves, and decision curve analysis (DCA) and clinical impact curves were used to evaluate the discrimination, accuracy, and clinical application efficacy of the prediction model. RESULTS: Gender, recurrence stones, blood uric acid value, urine pH, and urine bacterial culture (P < 0.05) were independent predictors of infection stones, and a nomogram prediction model ( https://zhaoyshenjh.shinyapps.io/DynNomInfectionStone/ ) was constructed using these five parameters. The area under the ROC curve of the training group was 0.901, 95% confidence interval (CI) (0.865-0.936), and the area under the ROC curve of the validation group was 0.960, 95% CI (0.921-0.998). The results of the calibration curve for the training group showed a mean absolute error of 0.015 and the Hosmer-Lemeshow test P > 0.05. DCA and clinical impact curves showed that when the threshold probability value of the model was between 0.01 and 0.85, it had the maximum net clinical benefit. CONCLUSIONS: The nomogram developed in this study has good clinical predictive value and clinical application efficiency can help with risk assessment and decision-making for infection stones in diagnosing and treating urolithiasis.

A multicenter study to develop a non-invasive radiomic model to identify urinary infection stone in vivo using machine-learning.

Urolithiasis is a common urological disease, and treatment strategy options vary between different stone types. However, accurate detection of stone composition can only be performed in vitro. The management of infection stones is particularly challenging with yet no effective approach to pre-operatively identify infection stones from non-infection stones. Therefore, we aimed to develop a radiomic model for preoperatively identifying infection stones with multicenter validation. In total, 1198 eligible patients with urolithiasis from three centers were divided into a training set, an internal validation set, and two external validation sets. Stone composition was determined by Fourier transform infrared spectroscopy. A total of 1316 radiomic features were extracted from the pre-treatment Computer Tomography images of each patient. Using the least absolute shrinkage and selection operator algorithm, we identified a radiomic signature that achieved favorable discrimination in the training set, which was confirmed in the validation sets. Moreover, we then developed a radiomic model incorporating the radiomic signature, urease-producing bacteria in urine, and urine pH based on multivariate logistic regression analysis. The nomogram showed favorable calibration and discrimination in the training and three validation sets (area under the curve [95% confidence interval], 0.898 [0.840-0.956], 0.832 [0.742-0.923], 0.825 [0.783-0.866], and 0.812 [0.710-0.914], respectively). Decision curve analysis demonstrated the clinical utility of the radiomic model. Thus, our proposed radiomic model can serve as a non-invasive tool to identify urinary infection stones in vivo, which may optimize disease management in urolithiasis and improve patient prognosis.

Do Infection-Associated Stone Subtypes Behave the Same Clinically? A Retrospective Bicenter Study.

Introduction and Objectives: Infection-associated renal stones are commonly composed of magnesium-ammonium-phosphate (MAP) and carbonate apatite (CA). The clinical implications of these two different, but closely related stone types, are unknown. We sought to compare the clinical, microbiologic, and metabolic characteristics of patients with MAP and CA stone types. Methods: We retrospectively reviewed the medical records of patients from two centers (one in the United States and one in Israel) who underwent ureteroscopy or percutaneous nephrolithotomy between 2012 and 2020 and identified patients with a predominant stone analysis component of CA or MAP and clinical data supporting an infection stone. We analyzed and compared demographic data, medical history, postoperative fever, stone and urinary microbiology, and 24-hour urine studies. Results: A total of 79 and 75 patients met the inclusion criteria for the MAP and CA cohorts, respectively. No significant difference was found in patient demographics or comorbidities between the MAP and CA cohort. Female predominance was noted in both. Although there were no significant differences in 24-hour urine parameters between the cohorts, hypercalciuria was common in both cohorts (38% and 32% of patients in the MAP and CA cohorts, respectively). Gram-negative bacteria were more common in the MAP stone cultures. Postoperative fever was significantly more common in the MAP cohort (14.7% vs 3.8%, p < 0.016). Conclusions: MAP and CA stone formers share similar demographic characteristics with a clear female predominance. MAP stones patients appear more likely to develop postoperative fever, possibly related to a higher occurrence of gram-negative bacteria in the stone cultures of the MAP cohort. Although there were no significant differences among metabolic parameters, hypercalciuria was noted in approximately a third of the cohort. The clinical significance of this finding is yet to be determined.

Machine Learning-Assisted Preoperative Diagnosis of Infection Stones in Urolithiasis Patients.

Purpose: The decision-making of how to treat urinary infection stones was complicated by the difficulty in preoperative diagnosis of these stones. Hence, we developed machine learning (ML) models that can be leveraged to discriminate between infection and noninfection stones in urolithiasis patients before treatment. Materials and Methods: We enrolled 462 patients with urinary stones and randomly stratified them into training (80%) and testing sets (20%). ML models were constructed using five algorithms (decision tree, random forest classifier [RFC], extreme gradient boosting, categorical boosting, and adaptive boosting) and 15 preoperative variables and were compared with conventional logistic regression (LR) analysis. Performance measurement was the area under the receiver operating characteristic curve (AUC) in the testing set. We also analyzed the importance of 15 features on the prediction of infection stones in each ML model. Results: Sixty-two (13.4%) patients with infection stones were included in the study. On the testing set, all the five ML models demonstrated strong discrimination (AUC: 0.892-0.951). The RFC model was chosen as the final model [AUC: 0.951 (95% confidence interval, CI, 0.934-0.968); sensitivity: 0.906; specificity: 0.924], significantly outperforming the traditional LR model [AUC: 0.873 (95% CI 0.843-0.904)]. Gender, urine white blood cell counts, and urine pH level were the top 3 important features. Conclusion: Our RFC model was the first model for the preoperative identification of infection stones with superior predictive performance. This novel model could be useful for risk assessment and decision support for infection stones.

A Multi-Institutional Study of Struvite Stones: Patterns of Infection and Colonization.

OBJECTIVE: To examine urine and stone bacteriology of struvite stone formers in a large cohort of patients undergoing percutaneous nephrolithotomy (PCNL). MATERIALS AND METHODS: A total of 1191 patients, with stone and urine cultures, treated with PCNL for renal calculi were included in the study. Statistical differences were assessed using Mann-Whitney U and T-tests. RESULTS: Stone cultures were positive in 72% of patients with struvite stones. Urea-splitting organisms accounted for only half of the positive stone cultures. Enterococcus (9/50, 18%), Proteus (9/50, 18%), and Escherichia coli (6/50, 12%) were the most commonly identified organisms. Notably, two-thirds of struvite formers with negative stone culture had at least one positive culture for a urea-splitting organism on urine culture going back 1 year from the time of surgery. A majority (67%) of struvite stone cultures were found to be resistant to first- and second-generation cephalosporins. CONCLUSIONS: The bacteriology of struvite stones has shifted away from traditional urea-splitting organisms and antibiotic coverage must be expanded to include organisms such as Enterococcus that do not respond to cephalosporins. Causative organisms may be found by going back in time to identify the initial organism that could have induced struvite stone formation to inform preventative therapy.

Metabolic Work-up of Patients with Urolithiasis: Indications and Diagnostic Algorithm.

CONTEXT: The incidence of urinary tract stone disease is increasing and the risk of recurrent stone formation is high. Appropriate therapeutic procedures with the aim of counteracting the progress of stone formation are highly desirable. Metabolic work-up is considered essential as a base for optimal design and follow-up of effective recurrence prevention. OBJECTIVE: To scrutinize the current literature with regard to principles of metabolic work-up for this heterogeneous group of patients. EVIDENCE ACQUISITION: Relevant articles in PubMed, guideline documents, consensus reports, and the Cochrane Library published during the past 20 yr were consulted. EVIDENCE SYNTHESIS: Grades of recommendation were used according to the principles applied in the European Association of Urology and American Urological Association guidelines. Medical efforts to prevent recurrent stone formation should be part of the care of patients with urinary tract stone disease (grade of recommendation A). A careful medical history and imaging together with analysis of stone composition, blood, and urine provide the basis for appropriate measures, but the treatment has to be individualized (grade of recommendation D). Whenever possible, stone analysis should be carried out at least once for every patient or each time when a long time has elapsed between two stone episodes because the risk factors explaining stone formation may have changed (clinical principle). The medical history, including information on dietary and drinking habits as well as lifestyle, is necessary for appropriate advice (grade of recommendation C). The medical history, together with imaging and stone composition, is used to estimate the severity of the disease (clinical principle). Identification of specific medical conditions should be supported by blood and/or urine analysis (grade of recommendation B). Pharmacological agents associated with an increased risk of stone formation should be identified (grade of recommendation C). Patients who have formed noncalcium stones should always be given recurrence preventive treatment. Analysis of urine composition for these patients is optional, but might be of value in the follow-up to support decisions on appropriate dosage regimens (grade of recommendation C). For patients with idiopathic calcium stone disease information from 24-h urine samples should be used, although the number of samples to be taken is debated (grade of recommendation C). Information from 24-h urine analysis should be used for selective dietary and drinking advice as well as for selection of the most appropriate pharmacological agent (grade of recommendation B). The treatment effects on the risk of stone formation can be followed by estimates of supersaturation based on urine composition (grade of recommendation C). CONCLUSIONS: It is clear that the metabolic work-up of patients with urinary tract stone disease should be individualized according to stone type and severity of the disease, and that the different therapeutic approaches are closely associated with the availability of therapeutic tools and motivation by the patient. PATIENT SUMMARY: Effective prevention of recurrent stone formation is determined by several factors such as the current and previous stone episodes and surgical procedures, stone composition, medical history, dietary and drinking habits, lifestyle, and ongoing pharmacological therapy. Analysis of blood and urine is an important part of the metabolic evaluation, but how extensive the risk evaluation should be is determined by the type of stone and the severity of the disease.