Ferhat ÇOBAN, Hüseyin KUTLU
Cyprus Journal of Medical Sciences - 2026;11(4):266-272
BACKGROUND/AIMS: Urinary leakage after percutaneous nephrolithotomy (PNL) is a clinically relevant complication that may prolong hospitalization and necessitate additional interventions such as ureteral stenting. However, reliable predictors of clinically significant leakage remain insufficiently defined. MATERIALS AND METHODS: This retrospective study included 305 patients who underwent PNL. The primary outcome was urinary leakage requiring double-J stent placement, defined as persistent drainage >24 hours after nephrostomy tube removal. Patients were divided into leakage (n=46) and non-leakage (n=259) groups. Clinical, radiological, and operative variables were analyzed. Univariate and multivariable analyses were performed using Firth penalized logistic regression. Model performance was evaluated using receiver operating characteristic analysis and calibration testing. RESULTS: Urinary leakage occurred in 15.1% of patients. On multivariable analysis, multiple stones [odds ratio (OR)=55.66, p<0.001] and longer operative time (OR=1.017 per minute, p<0.001) were identified as independent risk factors; nephrostomy tube placement was strongly protective (OR=0.059, p<0.001). Lower calyceal stone involvement also remained significant (OR=0.255, p=0.044). The model demonstrated acceptable discrimination (area under the curve=0.798, 95% confidence interval: 0.729-0.865) and good calibration (Hosmer-Lemeshow p=0.754). CONCLUSION: Clinically significant urinary leakage after PNL is primarily associated with stone complexity and operative factors. Multiple stones and prolonged operative time increase the risk, whereas nephrostomy tube placement has a protective effect. The proposed model may assist in perioperative risk stratification and clinical assessment.
