A novel nomogram and a simple scoring system for urinary leakage after percutaneous nephrolithotomy

Abstract

Introduction: The present study aimed to investigate the factors of prolonged urinary leakage (PUL) after percutaneous nephrolithotomy (PCNL) and develop a new and simple scoring system to predict it.

Patients and methods: We retrospectively reviewed patients with renal stones who underwent PCNL at the University of Health Sciences Izmir Bozyaka Training and Research Hospital between April 2011 and January 2020. The patients were divided into two groups according to the presence of PUL, and their preoperative and perioperative data were compared. A multivariate regression analysis was applied to examine the relationship between perioperative descriptors and PUL, and a nomogram was developed using significant predictors. Then, the individual components of the nomogram were assigned points to form a scoring system.

Results: There were 92 and 840 patients in the groups with and without PUL, respectively. The results of the univariate logistic regression analysis showed that hydronephrosis grade, parenchymal thickness, duration of nephroscopy, and duration of nephrostomy catheter were significantly associated with PUL. Subsequently, a multivariate regression analysis was carried out with these four factors as possible independent risk factors of PUL after PCNL. Based on the results of this analysis, a nomogram prediction model was developed with an area under the curve value of 0.811, which was consequently used to develop a new simple score system consisting of three characteristics: parenchymal thickness (1-5 points), duration of nephroscopy (1-3 points), and hydronephrosis grade (1-3 points).

Conclusion: A novel scoring system is a useful tool for predicting PUL in patients who have undergone percutaneous nephrolithotomy.

Keywords: Kidney Calculi; Nephrolithotomy, Percutaneous; Urinary Incontinence, Stress.

Figure 1. Nomogram and scoring system predicting urinary leakage after PCNL. The scoring system is based on radiological (parenchymal thickness, hydronephrosis grade) and surgical parameters (nephroscopy time). Parenchymal thickness (1–5 points), nephroscopy time (1–3 points), and hydronephrosis grade (1–3 points) are summed to provide a total score ranging from 3 to 11 points.

Figure 2. Receiving operator characteristic curve for predicting urinary leakage based on the nomogram and the scoring system. (a) The area under the curve (AUC) value of the nomogram was 0.811 (95% CI: 0.767-0.855) with an optimal cut-off value of 14.96%, at which it had a sensitivity of 77.2% and specificity of 74.2%. (b) The AUC value of the scoring system was 0.793 (0.745-0.841) with an optimal cut-off value of 6.5, at which it had 76.1% sensitivity and 71.0% specificity.

Figure 3. Stacked bar graph of the classification of urinary leakage risk as low, moderate, and high based on our score. Based on novel scores, the patients were divided into the risk groups of low (3-6), moderate (7-9), and high (10-11), which were found to have the PUL rates of 3.6%, 19.4%, and 80%, respectively.