To investigate the relationships among diurnal blood pressure (BP) variations and autonomic nervous system dysfunction, we assessed heart rate variability (HRV) using power spectral analysis of the 24-hour RR interval in 51 asymptomatic elderly hypertensive patients with various patterns of nocturnal BP fall. The extreme-dippers with marked nocturnal BP fall (n=16) had lower asleep low-frequency power (LF)/high-frequency power (HF) ratios (a relative index of sympathetic nervous system activity), while the nondippers without nocturnal BP fall (n=18) had lower awake LF/HF ratios and asleep/awake ratio for HF (an index of parasympathetic nervous activity), when compared with dippers with appropriate nocturnal BP fall (n=17). The incidence of multiple lacunar infarction detected by brain magnetic resonance imaging was 56% in the extreme-dippers and 38% in the nondippers, and both were markedly higher than that (6.3%) in the dippers (both P<.01). There was no significant relationship between the BP level and any HRV parameter for either the daytime or nighttime period. The asleep/awake ratio for systolic BP was significantly correlated with the asleep/awake ratio for HF (r= -.363, P<.01) and with the asleep/awake ratio for the LF/HF ratio (r=.540, P<.001), regardless of whether multiple lacunar infarction was present. In conclusion, the autonomic nervous system activity is not related to high BP level per se, rather its diurnal variation is more important as a determinant of the diurnal BP patterns, regardless of the presence or absence of cerebrovascular disease.