Autorhythmometry in hypertension: some methodological aspects and clinical implications.

Autorhythmometry of blood pressure is a technique easy to be performed and well accepted by hypertensive patients. A simple inspection of data self-collected 5 times a day for many days constitute a sufficiently reliable automonitoring of blood pressure both in basal conditions and in relation to the efficacy of some dietary and/or pharmacological treatment. Several examples are reported to show that more sophisticated statistical manipulation of the data collected may give rise to a better understanding of some clinical and physiological aspects. Both analysis of variance, performed on individual subsets of data averaged at each sampling hour, and single cosinor, performed on longitudinal time series, may be used to detect and quantify a circadian rhythm as a systematic daily variation, averaging towards zero the noise superimposed to the actual time series. Serial section analysis of the data, along all the experimental span, is useful to detect a) the reproducibility of the rhythm b) the variation of its parameters induced by changes in the experimental conditions c) the long-term trend. In the first subject the influence on the circadian rhythm of the pulse rate, temperature and blood pressure, due to a 4 h advancing shift in the rest-activity synchronizer, is well documented. In the first hypertensive patient a circadian rhythm is demonstrated also in blood pressure during two non-consecutive months. No difference is detected in both mesor and amplitude of blood pressure and a full resynchronization of the acrophase is achieved when a 1 h delaying shift in the rest activity synchronizer is imposed. In the second patient a well reproducible rhythm of systolic blood pressure and a low noise/signal ratio is documented by the serial section display. In the third patient the 'lability' of hypertention seems mainly due to salt sensitivity, as documented by the significant lowering of the mesor in the second experimental span, when a lowering of only 30 mEq/day in salt intake is imposed. The serial section better documents the salt-sensitivity of this patient, during a span when loading and depressing of salt intake is imposed. In the last patient the prompt effect of therapy in lowering blood pressure within normal range is well documented by serial section with 3 day interval. A possible effect of masking the circadian rhythm of blood pressure, due to therapy, is inferred by the serial section display with an interval of the same length (33 days) as the subspan without therapy. The possibility of prevention in the hypertensive disease is discussed, with the aim of autorhythmometry and statistical methods employed in this paper.