Predictors of Urinary Abnormalities in Children Hospitalised for Their First Urinary Tract Infection

Abstract

We aimed to investigate if children with their first UTI and a concomitant positive blood culture have a higher risk of abnormalities. We performed a retrospective study of children younger than 18 years of age with their first UTI. Multivariate logistic regression and receiver operating characteristic (ROC) curves were used to evaluate if positive blood cultures are associated with urinary abnormalities. After the screening process, we considered the enrolled 161 children with UTIs. The median age was three months, and 83 were females (43.2%). In multivariate analysis, age (p = 0.001, 95% CI 1.005-1.020), the presence of Pseudomonas aeruginosa or unusual germs in urine cultures (p = 0.002, 95% CI 2.18-30.36) and the positivity of blood cultures (p = 0.001, 95% CI 2.23-18.98) were significantly associated with urinary abnormalities. A model based on these parameters has an AUC of 0.7168 to predict urinary malformations (p = 0.0315). Conclusions include how greater age, a positive blood culture and the presence of Pseudomonas aeruginosa or unusual germs in urine culture in children hospitalised for their first episode of a UTI are factors associated with a significantly higher risk of urinary abnormalities. These data can guide the implementation of more personalized strategies to screen for urinary abnormalities that may be included in future guidelines.

Keywords: children; urinary abnormalities; urinary tract infections.

Figure 1

Correlation between age and urinary anomaly frequency.

Figure 2

Relationship between age and different bacteria retrieved in urine cultures in predicting a urinary abnormality.

Figure 3

AUC of the model used to predict urinary abnormalities. Inclusion of positivity on blood culture increases the AUC from 0.6349 to 0.7168.