[Nyctohemeral rhythm of arterial pressure].

It has long been known that blood pressure (BP) varies over time. Several methods for monitoring BP over the 24-hour cycle are available. The Oxford intra-arterial method is the most accurate but is too invasive to be suitable for routine use. Noninvasive monitors that measure BP intermittently have gained acceptance in clinical practice. The Finapress device provides noninvasive beat-by-beat BP recordings that are useful for studying BP variability. The reproducibility of successive recordings remains controversial. BP recordings with measurements at intervals of up to 15 minutes yield acceptable results provided the duration of the recording is at least 8-12 hours. BP decreases at night. Whether or not BP varies according to a circadian rhythm is unsettled; if a circadian BP cycle does exist, it may be dependent on the sympathetic nervous system. The method of expression of ambulatory BP monitoring data is not well-standardized: the 24-hour mean, daytime and nighttime means, histograms, and the BP load (based on values greater than an arbitrary norm) are the most widely used parameters in clinical practice. Mathematical methods have been developed for processing outlying data (ten methods have been proposed), identifying low-BP and high-BP periods (the "cumulative sum" and "square wave" methods), and modeling the 24-hours BP profile (a four-harmonic Fourier model seems to be the most appropriate model). Variability can be divided into short-term and long-term variability. Many factors influence the BP curve, including genetic factors, age, physical activity, and to a lesser extent psychosensory factors (the "office effect"). Apart from the difference in BP levels, the appearance of the BP curve is similar in normotensive subjects and in subjects with borderline or established hypertension. Ambulatory BP data suggest that the WHO definitions for hypertension may not be appropriate, and the normal values obtained in the meta-analysis by Staessen are now widely accepted. Ambulatory BP data provide better prognostic information that office BP measurements. In drug trials, use of ambulatory BP monitoring may abolish the placebo effect reduce required sample sizes, and provide new information on the pharmacology of experimental drugs.