# Traumatic spinal cord injury: identifying independent risk factors and predictive model development for symptomatic urinary tract infections

## TL;DR

This study identifies risk factors and creates a predictive model for urinary tract infections in spinal cord injury patients.

## Contribution

The study develops a predictive model for symptomatic UTIs in TSCI patients using independent risk factors.

## Key findings

- The incidence of symptomatic UTIs in TSCI patients was 57.14%, with fever and E. coli infections being common.
- Prolonged hospitalization and cumulative antibiotic exposure were identified as independent risk factors.
- The predictive model showed strong discrimination (AUC = 0.81) and good calibration.

## Abstract

Traumatic spinal cord injury (TSCI) is commonly associated with urinary tract infections (UTIs), with a reported prevalence ranging from 31.7% to 68%. Symptomatic UTIs can result in serious complications, including chronic kidney damage and recurrent infections. The objective of this study was to identify independent risk factors and develop a predictive model for symptomatic UTIs in TSCI patients, thereby providing valuable insights for prevention and management strategies.

A retrospective study was conducted at the China Rehabilitation Research Center, involving 168 TSCI patients admitted between January 1, 2020, and August 1, 2024. Symptomatic UTIs were diagnosed using Delphi consensus criteria, which integrated clinical symptoms, urinalysis, and culture confirmation. Comprehensive clinical data, including demographic characteristics, injury profiles, and laboratory parameters, were systematically extracted from the hospital information system. Potential risk factors were initially screened using univariable logistic regression, with statistically significant variables subsequently analyzed in a multivariable logistic regression model to identify independent predictors. A predictive model for symptomatic UTIs was constructed using the regression coefficients. The model’s performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration with the Hosmer-Lemeshow test, and internal validation through bootstrap resampling.

The incidence of symptomatic UTIs was 57.14%, with the majority presenting with fever (65.07%) and Escherichia coli infections (44.52%). Prolonged hospitalization (OR = 1.005, 95% CI [1.001–1.010]) and cumulative antibiotic exposure (OR = 1.011, 95% CI [1.000–1.022]) were identified as independent risk factors. The predictive model, which incorporated these factors, demonstrated strong discrimination (AUC = 0.81, 95% CI [0.746–0.879]) and good calibration (P = 0.44).

This study presents the incidence of symptomatic UTIs in TSCI patients and identifies two critical predictive factors along with a risk score for early prediction of symptomatic UTIs. The findings provide a foundation for improved clinical practices aimed at preventing and managing symptomatic UTIs in this patient population, potentially reducing healthcare costs and improving patient outcomes.