BROAD-BAND SPECTRAL-ANALYSIS OF BLOOD-PRESSURE AND HEART-RATE-VARIABILITY IN VERY ELDERLY SUBJECTS

Systolic blood pressure (SEP) variability is increased and R-R interva l variability is reduced in the elderly. Little is known, however, abo ut how SEP and R-R interval variabilities change in the very elderly. More important, however, it is not known which frequency components of SEP and R-R interval variability are affected significantly. We addre ssed this issue in subjects older than 70 years by broadband spectral analysis, which allows all variability components from the lowest to t he highest frequency to be considered. In 20 very elderly normotensive subjects (mean+/-SD age, 78.1+/-6.8 years) and 28 normotensive adult subjects (36.1+/-7.1 years), noninvasive finger blood pressure and R-R intervals were recorded continuously for 30 minutes in the supine pos ition and 15 minutes in the upright position. SEP and R-R interval pow er spectral densities were computed over the entire frequency region b etween 0.005 Hz (0.007 Hz in the upright position) and 0.5 Hz. Overall SEP variability (SD) was greater and overall R-R interval variability was less in very old subjects than in adult subjects. All spectral R- R interval powers were reduced significantly in very elderly individua ls. The spectral SEP powers were greater in the very elderly group tha n in the adult group only in the very-low-frequency range (<0.04 Hz). This was true in the supine and the standing positions. With subjects in the standing position, the shape of the broadband spectra differed in the very old and adult subjects because in the former group the inc rease in SEP and R-R interval power around 0.1 Hz that was seen in the latter was blunted. Therefore, in very elderly subjects a reduction i n overall R-R interval variability is accounted for by a reduction in all of its frequency components. The accompanying increase in overall BP variability, however, results from a nonhomogeneous behavior of its frequency components, which consists of an increase in the very low f requency and a concomitant reduction in the higher frequency powers. T he mechanisms responsible for these changes may be complex, but at lea st they may in part reflect the baroreflex impairment and autonomic dy sfunction that characterize aging.