Prevalence of hypertension and microalbuminuria in adult type 1 (insulin-dependent) diabetic patients without renal failure in Italy. I. Validation of screening techniques to detect microalbuminuria

Abstract

The prevalence of microalbuminuria and arterial hypertension among type 1 (insulin-dependent) diabetic patients is poorly known in Italy. In the preliminary phase of a large outpatient screening programme, we addressed the possibility of using non-time urine samples to predit the chance of detecting albumin excretion rate (AER) in the range of microalbuminuria. We therefore measured urinary albumin and creatinine concentration in timed overnight collections from 641 type 1 diabetic patients with serum creatinine levels lower than 133 μmol/l. AER was strongly and comparably predicted both by urinary albumin concentration (UAlb;r2=0.754) and by the urinary albumin to creatinine concentration ratio (A/C;r2=0.773). After exploring several independent cut-off levels for UAlb and A/C, AER in the range 20–200 μg/min (n=91) was found to be predicted with 90% sensitivity and specificity either by UAlb≧20 mg/l or by A/C≧2.0 mg/mmol. UAlb was negatively associated with diuresis, and false negative outcomes were explained by polyuria when screening by this variable. A/C was positively associated with female gender among normoalbuminuric patients, in line with the lower urinary excretion of creatinine in women (7.2±0.25 vs 10.2±0.35 μmol/min,P<0.00001). A significant excess of false positive outcomes in women compared with men was found when screening by any A/C cut-off level equal to or less than 2.5 mg/mmol. Simplified screening techniques seem to remain, however, a practicable option for the detection of microalbuminuria both in epidemiology and in clinical practice.