Question

• In comparison to suprapubic bladder aspiration (gold standard), how "bad" (false-positives) is a cathed urine sample in differentiating significant bacteriuria from contamination in children?

Clinical Bottom Lines

1. If the criteria for significant bacteriuria is >1000cuf/ml, the second portion of the urine sample obtained by catheter (C₂) has a greater diagnostic correlation (97.5%) with suprapubic specimens, than the first portion of the urine sample (92.8%).¹
2. If the criteria for significant bacteriuria is >10,000cuf/ml, the second portion of the urine sample obtained by catheter (C₂) has a 100% diagnostic correlation with suprapubic specimens.  The first portion of the urine sample has a 97.6% diagnostic correlation with suprapubic specimens.
3. Continue to use cathed urine samples.

Summary of Key Evidence

1. In this study, 42 children who were free from clinical evidence of infection of the urinary tract were selected.  After surgery children underwent:  Suprapubic bladder aspiration- sample sent for colony counts and culture, Urethral Catheterization- first 2-5ml (C₁) sent for colony counts and culture, second 2-5ml (C₂) sent for colony counts and culture, cath tip sent for culture.
2. If cut off of significant bacteriuria >1000 cfu/ml; SPA: 0/42=0% false positives or 100% diagnostic correlation with clinical impression, C₁: 3/42=7.2% false positives or 92.8% diagnostic correlation with clinical impression, C₂: 1/41=2.5% false positives or 97.5% diagnostic correlation with clinical impression.
3. If cut off of significant bacteriuria > 10,000 cfu/ml;  SPA: 0/42=0% false positives or 100% diagnostic correlation with clinical impression, C₁: 1/42=2.4% false positives or 97.6% diagnostic correlation with clinical impression, C₂: 0/42=0% false positives or 100% diagnostic correlation with clinical impression.
4. Gram stains of all specimens were negative for organisms.
5. All patients in the study were free from infection of the urinary tract on follow-up, even up to 4-6 months after catheterization.  Thus, while organisms were introduced into the urine of children during catheterization, these organisms were not responsible for the initiation of a urinary tract infection. 

Additional Comments

• This study was done in children who were free of active infection.  How well the two tests compare for children with UTIs is unclear.
• In Hansson et al, 7.4% of 366 infants with symptomatic urinary tract infections had <10,000 cuf/ml in urine obtained by suprapubic aspiration.  Thus, we see that the determination of positive and negative cultures can vary depending on the cut-off.²

Citation

1. Pryles CV, et al.  Comparative bacteriologic study of urine obtained from children by percutaneous suprapubic aspiration of the bladder and by catheter.  Pediatrics 1959;24:983-991.
2. Hansson S, et al.  Low bacterial counts in infants with urinary tract infection.  J Pediatrics 1998;132:180-182.
3. Pollack VC, et al.  Suprapubic bladder aspiration versus urethral catheterization in ill infants; success, efficacy, and complication rates.  Ann Emerg Med 1994;23:225-230.

CAT Author: Ray Castro, MD

CAT Appraisers:

Date appraised: January 11, 1999