In the study of Geerlings et al. (1), one of five type 1 diabetic women had asymptomatic bacteriuria (ASB). In the few studies of diabetic children a low prevalence of ∼1% was found (2–4). Our clinical experience suggested a much higher prevalence; therefore, we decided to estimate the prevalence and possible risk factors of ASB in type 1 diabetic children.
There were 178 (86 male) type 1 diabetic children and young adults (age 15.1 ± 5.9 years) with diabetes duration of 6.2 (3.0–10.1) [median (interquartile range)] years who participated in this study.
The control group consisted of 194 (103 male) school children/medical students (14.4 ± 5.1 years). After careful cleaning, midstream voiding morning urine samples were collected and immediately cultured on 2 consecutive days.
ASB was defined as the presence of ≥105 colony-forming units/ml of one and the same bacterial species in both samples without symptoms of urinary tract infection (UTI). In 140 diabetic patients and 191 control subjects, we had the possibility to evaluate the presence of leukocyturia by dip-slide method and/or microscopically (>5 cells/high-power field).
Student’s t test, Mann-Whitney test, χ2 test, and Fisher’s exact test were used to assess statistically significant differences.
The prevalence of ASB was 10.1% (95% CI 5.7–14.5%), which was significantly higher than in the control group (2.6%, 0.35–4.8%) (P = 0.003) and tended to increase with age (P = 0.064). We did not find any difference in prevalence of ASB between diabetic male (9.3%) and female (10.9%) subjects (P = 0.73). The age, duration of diabetes, BMI, morning and mean daily blood glucose levels, urinary glucose excretion, HbA1c, and albumin excretion rate (in the normal range) were similar in diabetic patients with and without ASB.
Leukocyturia tended to be more frequent in diabetic patients without ASB than in control subjects (14.4 vs. 7.6%; P = 0.052). Almost half (46.7%) of the diabetic children with ASB, but only 14.4% of those without ASB, had leukocyturia (P = 0.002).
In the 18 diabetic patients with ASB S. agalactiae (n = 6), Enterococcus sp. (n = 5), E. coli (n = 4), and K. pneumoniae (n = 3) were cultured. The proportion of leukocyturia in patients with Gram-positive and Gram-negative bacteria was 2/11 to 6/7 (P = 0.041), respectively. In control subjects, E. coli (n = 2), S. agalactiae, K. pneumoniae, and P. vulgaris (n = 1:1) were cultured.
In contrast to pediatric studies performed several decades ago (2–4), but in agreement with more recent adult observations (1) and textbook data (5), ASB in diabetic children occured with a higher frequency. Samples were collected by diabetes nurse specialists, and care was taken to avoid contamination. The increased prevalence of ASB is not readily explainable, particularly because our study could not demonstrate a relationship with ASB and hyperglycemia. Other possible reasons for the higher prevalence may include increased residual urine volume or impairment of several aspects of host defense mechanisms (e.g., cytokine secretion in the urinary tract), factors that are currently being investigated in our patients.
In UTI, the most common bacterium is E. coli, which was isolated in ∼80–90% of positive cultures (6). In our diabetic patients with ASB, E. coli was found in only about a quarter of cases. Similarly low rates were found in adult type 1 diabetic women with ASB (1). It therefore seems that the spectrum of pathogenic bacteria causing ASB and UTI is different. The reason for this may be that the virulence factors are different in ASB and UTI.
In conclusion, we have found that the prevalence of ASB and leukocyturia was higher in diabetic children and young adults than in control subjects and that the spectrum of bacteria in ASB was different from the usual spectrum of UTI. As the treatment of ASB in adult type 1 diabetic women did not appear to prevent UTI (7) and may promote the invasion of more virulent pathogens (8), a careful follow-up of these pediatric patients is warranted before antibiotic therapy can be considered. This conclusion is underlined by a recent observation (9) that showed that women with type 1 diabetes and ASB had a tendency toward a decline in renal function.
