Risk factors for contamination of catheterized urine specimens in febrile children.

BACKGROUND

Urinary tract infections are the most common serious bacterial infection in febrile infants. Bladder catheterization is the preferred method of obtaining urine for culture in young children. Contamination of urine can be recognized when nonpathogens or multiple pathogens are isolated; preliminary culture results may lead to unnecessary antibiotics pending final identification. Some low-colony count (<50,000 colony-forming units per milliliter) cultures may represent contamination or asymptomatic bacteriuria.

OBJECTIVE

Identify clinical factors that lead to contamination of catheterized urine specimens.

METHODS

Physicians and nurses in an urban pediatric emergency department completed a survey after performing bladder catheterization of febrile children 36 months or younger. Contamination was defined by multiple pathogens, nonpathogens, or colony counts less than 10,000 colony-forming units per milliliter.

RESULTS

: One hundred eighty-five children were studied. The median age was 8.4 months (interquartile range, 2.4-14.4 months). Sixty-eight percent were girls. Forty-six percent of boys were circumcised. Of the 185 children, 18 (10%) had true UTI. Fourteen percent of cultures were contaminated. Univariate analysis of potential predictors identified age younger than 6 months (odds ratio [OR], 6.8; 95% confidence interval [CI], 2.6-17.9), difficult catheterization (OR, 3.6; 95% CI, 1.5-8.6), and uncircumcised boys (OR, 5.7; 95% CI, 1.2-29.4). The contamination rate in uncircumcised boys younger than 6 months was 43% (95% CI, 26-66). Volume of urine, sex, and catheter size were not predictive of contamination.

CONCLUSIONS

Children younger than 6 months and uncircumcised boys are at increased risk of contaminated specimens from bladder catheterization. Suprapubic aspiration or use of a fresh, sterile catheter with each repeated attempt at catheterization may lead to less contamination in these patients.