Epidemiology of Kidney Stones: The incidence of nephrolithiasis is rising globally, with peaks in middle age. Factors include geography, diet, and associated conditions like obesity and diabetes. Regions with higher temperatures and certain occupations with heat exposure show increased kidney stone prevalence. Surgical Treatments: Procedures such as Shock Wave Lithotripsy, Percutaneous Nephrolithotomy, and Retrograde Intrarenal Surgery (RIRS) are preferred based on stone size and location. RIRS is often recommended for stones up to 2 cm in the largest diameter. Sepsis Risk Post-RIRS: Sepsis is a severe complication after RIRS, with rates varying between 0.5% to 11.1%. Factors contributing to higher risk include diabetes, recent urinary infections, stent placement, and longer surgical times. These findings underline the importance of identifying high-risk patients and providing tailored preventive care. The Infection post Flexible UreteroreNoscopy (I-FUN) Study: This multicenter and prospective study across 16 centers focused on predicting sepsis post-RIRS. The study included 1552 kidney stone adult patients and aimed to build a machine learning model to predict the individual risk of sepsis within 30 days of surgery. The second aim was to assess the influence of a positive stone culture obtained by laser lithotripsy on the occurrence of postoperative sepsis. Machine Learning Predictive Model: A Random Forest algorithm was developed to predict sepsis risk after RIRS. The model includes factors such as age, stone volume, and surgical time and achieves high accuracy. A web-based interface was also created for clinical use to help preemptively identify high-risk patients (https://emabal.pythonanywhere.com/). Stone culture: Patients with a positive SC have a significantly higher incidence of sepsis following RIRS. A poor pathogen concordance between preoperative urine and stone culture highlights the need to perform the latter in all patients whenever feasible. Conclusion: The research emphasizes routine stone cultures when possible, given their role in predicting post-operative infections, and highlights that the machine-learning model can significantly aid clinicians in identifying patients at risk for sepsis, potentially improving surgical outcomes and patient care.
Epidemiologia dei calcoli renali: L'incidenza della nefrolitiasi è in aumento a livello globale, con picchi nella mezza età. I fattori includono geografia, dieta e condizioni associate come obesità e diabete. Le regioni con temperature più elevate e alcune occupazioni esposte al calore mostrano una maggiore prevalenza di calcoli renali. Trattamenti chirurgici: Procedure come la litotrissia extracorporea a onde d'urto, la nefrolitotomia percutanea e la chirurgia intrarenale retrograda (RIRS) vengono scelte in base alla dimensione e alla posizione del calcolo. La RIRS è spesso consigliata per calcoli fino a 2 cm di diametro massimo. Rischio di sepsi post-RIRS: La sepsi è una complicanza grave dopo la RIRS, con tassi che variano tra lo 0,5% e l'11,1%. Fattori che aumentano il rischio includono diabete, infezioni urinarie recenti, posizionamento di stent e tempi chirurgici più lunghi. Questi dati sottolineano l'importanza di identificare i pazienti ad alto rischio e fornire cure preventive personalizzate. Studio I-FUN (Infection post Flexible UreteroreNoscopy): Questo studio multicentrico e prospettico condotto in 16 centri si è concentrato sulla previsione della sepsi post-RIRS. Ha incluso 1552 pazienti adulti con calcoli renali ed è stato finalizzato a sviluppare un modello di apprendimento automatico per prevedere il rischio individuale di sepsi entro 30 giorni dall'intervento chirurgico. Un secondo obiettivo era valutare l'influenza di una coltura del calcolo positiva, ottenuta tramite litotrissia laser, sull'insorgenza di sepsi post-operatoria. Modello predittivo basato sull'apprendimento automatico: È stato sviluppato un algoritmo Random Forest per prevedere il rischio di sepsi dopo la RIRS. Il modello include fattori come età, volume del calcolo e durata dell'intervento chirurgico, e offre un'elevata precisione. È stata inoltre creata un'interfaccia web per l'uso clinico, utile per identificare preventivamente i pazienti ad alto rischio (https://emabal.pythonanywhere.com/). Coltura del calcolo: I pazienti con una coltura del calcolo positiva presentano un'incidenza significativamente più alta di sepsi dopo la RIRS. La scarsa concordanza tra le colture delle urine preoperatorie e quelle del calcolo sottolinea la necessità di effettuare quest'ultima ogni volta che sia possibile. Conclusione: La ricerca evidenzia l'importanza delle colture dei calcoli quando possibile, dato il loro ruolo nella previsione delle infezioni post-operatorie, e sottolinea che il modello di apprendimento automatico può aiutare significativamente i medici nell'identificazione dei pazienti a rischio di sepsi, migliorando potenzialmente i risultati chirurgici e la cura del paziente.
How to predict the risk of sepsis after flexible ureteroscopy for kidney stone disease. Results from a large, prospective, multicenter study: The Infection post Flexible UreteroreNoscopy (I-FUN) study / Castellani, Daniele. - (2025 Mar 31).
How to predict the risk of sepsis after flexible ureteroscopy for kidney stone disease. Results from a large, prospective, multicenter study: The Infection post Flexible UreteroreNoscopy (I-FUN) study.
CASTELLANI, DANIELE
2025-03-31
Abstract
Epidemiology of Kidney Stones: The incidence of nephrolithiasis is rising globally, with peaks in middle age. Factors include geography, diet, and associated conditions like obesity and diabetes. Regions with higher temperatures and certain occupations with heat exposure show increased kidney stone prevalence. Surgical Treatments: Procedures such as Shock Wave Lithotripsy, Percutaneous Nephrolithotomy, and Retrograde Intrarenal Surgery (RIRS) are preferred based on stone size and location. RIRS is often recommended for stones up to 2 cm in the largest diameter. Sepsis Risk Post-RIRS: Sepsis is a severe complication after RIRS, with rates varying between 0.5% to 11.1%. Factors contributing to higher risk include diabetes, recent urinary infections, stent placement, and longer surgical times. These findings underline the importance of identifying high-risk patients and providing tailored preventive care. The Infection post Flexible UreteroreNoscopy (I-FUN) Study: This multicenter and prospective study across 16 centers focused on predicting sepsis post-RIRS. The study included 1552 kidney stone adult patients and aimed to build a machine learning model to predict the individual risk of sepsis within 30 days of surgery. The second aim was to assess the influence of a positive stone culture obtained by laser lithotripsy on the occurrence of postoperative sepsis. Machine Learning Predictive Model: A Random Forest algorithm was developed to predict sepsis risk after RIRS. The model includes factors such as age, stone volume, and surgical time and achieves high accuracy. A web-based interface was also created for clinical use to help preemptively identify high-risk patients (https://emabal.pythonanywhere.com/). Stone culture: Patients with a positive SC have a significantly higher incidence of sepsis following RIRS. A poor pathogen concordance between preoperative urine and stone culture highlights the need to perform the latter in all patients whenever feasible. Conclusion: The research emphasizes routine stone cultures when possible, given their role in predicting post-operative infections, and highlights that the machine-learning model can significantly aid clinicians in identifying patients at risk for sepsis, potentially improving surgical outcomes and patient care.File | Dimensione | Formato | |
---|---|---|---|
tesi_Castellani.pdf
embargo fino al 01/10/2025
Descrizione: tesi_Castellani
Tipologia:
Tesi di dottorato
Licenza d'uso:
Creative commons
Dimensione
416.58 kB
Formato
Adobe PDF
|
416.58 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.