Commonly used daytime measurements in previous investigations on community noise and arterial hypertension (AH) may be a source of exposure bias, as urban residents spend most of their daytime hours out of the home on workdays. For this reason, we focused on the relation of nighttime noise and AH. A cross-sectional study was performed on a sample of 2,503 (995 men and 1,508 women) adult residents of a downtown Belgrade municipality. The inclusion criteria were a period of residence longer than 10 years and a bedroom oriented toward the street. The exclusion criteria were a high level of noise annoyance at work and diseases related to AH. Noise measurements were performed in all 70 streets of the municipality. The streets were grouped into noisy areas (equivalent noise level [Leq]>45 dB(A)) and quiet areas (Leq< or =45 dB(A)). The residents were interviewed in regard to antihypertensive therapy. Subjects who responded that they had not received such therapy were contacted for blood pressure measurements with mercury sphygmomanometer. Possible confounding factors: family history of AH, age, body mass index, smoking habits, physical activity and alcohol consumption were controlled for. The proportions of men with AH in the noisy and quiet areas were 23.6% and 17.5%, respectively. The adjusted odds ratio (OR) for AH was 1.58; the 95% confidence interval (CI) ranged from 1.03-2.42; and the probability value was 0.038, when men living in quiet streets were taken as a reference category. This relation was statistically insignificant for women: adjusted OR: 0.90; 95% CI: 0.59-1.38; p: 0.644. This cross-sectional study showed that nighttime urban road-traffic noise might be related to occurrence of AH in men.