Prevalence of asymptomatic bacteriuria in subjects with NIDDM in San Luis Valley of Colorado.

This study examined whether non-insulin-dependent diabetic (NIDDM) subjects have an increased prevalence of asymptomatic bacteriuria compared with subjects with normal glucose tolerance. Diabetic (n = 206) and normal (n = 418) subjects were identified from a defined geographic area in the San Luis Valley of southern Colorado. Presence of asymptomatic bacteriuria was determined by testing the subjects' urine with a reagent-strip test for nitrite and leukocyte esterase (Chemstrip LN). The ability of the Chemstrip LN to detect bacteriuria was evaluated by comparing its results with those from urine culture on a subsample of subjects. There were 7 control and 12 diabetic subjects with bacteriuria as measured by the Chemstrip LN. The prevalence of urinary tract colonization among diabetic compared with control subjects was increased 3.5-fold (95% confidence interval 1.4-8.6). Adjustment for confounding by age, sex, ethnicity, and county of residence resulted in an adjusted prevalence ratio of 4.4 (95% confidence interval 1.1-17.4). Among diabetic subjects, prevalence of bacteriuria increased with longer disease duration but was not affected by measures of glucose control. We conclude that NIDDM increases the prevalence of bacterial colonization of the urine and, therefore, probably also increases the risk of symptomatic urinary tract infection.

Duke Scholars

Author Lyman Barth Reller Pathology