Optimal bacterial colony counts for the diagnosis of upper urinary tract infections in infants.

Abstract

There are no consensus criteria for diagnosing upper urinary tract infections (UTI). Therefore, we conducted a study to assess whether bacterial colony counts of ≥ 103CFU/ml are optimal for diagnosing upper UTIs among infants. This retrospective observational study included 673 patients (<4months of age) with urine samples obtained by catheterization for bacterial cultures. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were obtained when cutoff values of 103, 104, and 105CFU/ml were used for diagnosing upper UTIs. Upper UTI patients were divided based on cutoff values: Group A (103CFU/ml), Group B (104CFU/ml), and Group C (≥ 105CFU/ml). Of the 197 positive (≥ 103CFU/ml) patients, 92 were diagnosed with an upper UTI. These patients were divided into Group A (n = 23), Group B (n = 16), and Group C (n = 53). No significant differences were detected in terms of clinical findings, including the incidence of vesicoureteral reflex. When cutoff values of 103, 104, and 105CFU/ml were used for diagnosing upper UTIs, the sensitivity/specificity percentages were 100/81.3, 75.0/95.9, and 57.6/97.5, and the PPVs/NPVs were 46.7/100, 75.0/95.9, and 79.1/93.4. Using ≥ 105CFU/ml as a diagnostic threshold leads to approximately 40% of positive cases being missed. In contrast when ≥ 103CFU/ml is used, all upper UTIs were identified. Therefore, bacterial colony counts of ≥ 103CFU/ml should be considered the cutoff value for the diagnosis of upper UTIs in infants (< 4months of age).