# Predictive factors for spontaneous dislodgement of percutaneous nephrostomies for malignant ureteral obstruction

## TL;DR

This study identifies factors like muscle thickness and patient health status that predict when kidney catheters might dislodge on their own.

## Contribution

The study introduces new predictive markers for spontaneous catheter dislodgement based on muscle and renal tissue characteristics.

## Key findings

- Renal parenchymal and abdominal wall muscle thickness are significant predictors of catheter dislodgement.
- ECOG performance status and psoas muscle index are significantly associated with dislodgement risk.
- More frequent catheter replacements correlate with higher dislodgement risk.

## Abstract

To investigate the etiology of spontaneous percutaneous nephrostomy (PCN) catheter dislodgements and evaluate factors potentially associated with these dislodgements, including muscle-to-fat composition and tissue characteristics of catheter traces.

Data from 92 patients (63 males, 29 females; mean age 63.9 ± 11.4 years, range 28–88) undergoing 151 PCN catheter replacements between January 2016 and June 2021 were analyzed. Patients were divided into Group 1 (prophylactic replacements every 3 months, n = 41) and Group 2 (at least one spontaneous dislodgement, n = 51). Associations were evaluated for factors including intraabdominal visceral adipose tissue index (IAVATI), subcutaneous adipose tissue index (SATI), and abdominal perimeter. Other variables assessed were Eastern Cooperative Oncology Group (ECOG) performance status scores, psoas muscle index (PMI), renal size, renal parenchymal thickness, renal cortex-to-skin distance, posterolateral abdominal wall muscle thickness, and PCN replacement frequency.

No significant differences were identified between Group 1 and Group 2 in IAVATI, SATI, or abdominal perimeter values (p = 0.210–0.412). A significant difference in ECOG performance status scores (p = 0.0001), PMI (p = 0.04) and lower renal size, renal parenchymal thickness, renal cortex-to-skin distance, and posterolateral abdominal muscle thickness (p = 0.0001–0.039) were observed in Group 2. PCN replacements were significantly more frequent in Group 2 (p = 0.0001). Multivariate regression identified renal parenchymal thickness and abdominal wall muscle thickness as significant independent predictors (p = 0.0001, p = 0.02). ROC analysis yielded an AUC of 0.843 (95% CI: 0.769–0.917) for renal parenchymal thickness and 0.694 (95% CI: 0.610–0.778) for abdominal wall muscle thickness. Sensitivity and specificity rates were 73.1% and 96.4% for a 16 mm cutoff in renal parenchymal thickness, and 50.7% and 79.8% for an 8 mm cutoff in abdominal wall muscle thickness.

A significant association was identified between spontaneous PCN catheter dislodgement and both the psoas muscle index and ECOG performance status scores, while no notable relationship was observed with abdominal visceral or subcutaneous fat tissue volumes or abdominal perimeter. The risk of dislodgement was found to increase with reduced renal parenchymal and abdominal wall muscle thickness, as well as with more frequent nephrostomy replacements, suggesting these parameters may serve as useful markers for identifying patients at higher risk.