This study describes the diurnal blood pressure (BP) profile and identifies its correlates in older patients with isolated systolic hypertension (ISH). The ambulatory BP readings of 408 patients, aged > or = 60 years, with ISH on clinic measurement, enrolled in the placebo run-in phase of the Syst-Eur Trial were examined. The time-weighted 24 h BP, daytime and night-time BP and the cusum-derived crest and trough BP were computed to express the BP level. The daily alteration between the high and low BP span was estimated from the day-night BP difference, the cusum derived circadian alteration magnitude and plot height, as well as the amplitude of the Fourier curve. The 24 h SBP and DBP tended to be higher in men (150 +/- 15/82 +/- 9 mm Hg) than in women (147 +/- 17/79 +/- 10 mm Hg), but the sex difference was only significant for DBP. In multiple regression analysis, the 24 h SBP increased (P < 0.05) by 3 mm Hg for each 10 year increment in age and was also 10 mm Hg higher (P < 0.001) in smokers than in non-smokers; the 24 h DBP was 2 mm Hg higher (P < 0.05) in men than in women and decreased (P < 0.05) by 1.5 mm Hg for each 10 year increment in age. The day-night difference in SBP increased with 2 mm Hg for each 10 mm Hg increase in the conventional pressure, decreased with 5 mm Hg for each 10 year increment in age and was 6 mm Hg higher in smokers than in non-smokers; the day-night difference in diastolic pressure was 2 mm Hg greater in women than in men. We conclude that the main determinants of the diurnal BP variation in older patients with isolated systolic hypertension were sex, age, smoking habits and the level of pressure on conventional measurement.