The aim of the present study was to examine the relationships between ambulatory blood pressure (ABPM) and urinary albumin excretion (UAE) in diabetic (non-insulin dependent [NIDDM] and insulin-dependent [IDDM]) hypertensives at baseline and after treatment by an angiotensin converting enzyme (ACE) inhibitor. After a 3-week placebo period, patients were treated for 16 weeks with trandolapril, 2 to 4 mg/day. The UAE and blood pressure (mercury sphygmomanometer and 24-h ABPM) were measured at baseline and repeated on trandolapril. Predictive factors of abnormal UAE (24-h UAE > or = 30 mg) were determined using univariate and multivariate analysis (logistic regression). Predictors of UAE decrease were also searched. One hundred seventy-one patients entered the analysis. Baseline office BP was 164+/-14 / 97+/-6 mm Hg and 24-h BP was 142+/-17 / 83+/-10 mm Hg. Seventy-four patients (43%) had UAE > or = 30 mg. Independent risk factors for abnormal UAE were nighttime diastolic BP (odds ratio [OR] = 4.1, confidence interval [CI] = 2.0 to 8.6, P = .0001), diabetes duration (OR = 2.4, CI = 1.1 to 5.0, P = .025), and presence of retinopathy (OR = 3.2, CI = 1.0 to 10.0, P = .047). Conversely, office BP level was not significantly related to UAE. On treatment, office BP levels decreased to 143+/-13 / 82+/-8 mm Hg (P < .0001) and 24-h BP levels to 134+/-17 / 78+/-9 mm Hg (P < .0001). In the abnormal UAE group, UAE significantly decreased from 76 to 50 mg/day (P = .006). After treatment, independent predictive factors of abnormal UAE were: on-drug fasting plasma glucose (OR = 3.5, CI = 1.7 to 7.4, P = .0009) and on-drug nighttime diastolic BP (OR = 3.5, CI = 1.7 to 7.4, P = .001). The only predictor of UAE decrease was a 24-h systolic BP decrease (OR = 2.3, CI = 1.3 to 4.3, P = .007). We conclude that in diabetic hypertensives with abnormal UAE, trandolapril exhibited a sustained 24-h antihypertensive effect and provided a consistent reduction of microalbuminuria. This study confirmed the superiority of ABPM over clinical BP to predict target organ damage.