(Hypertension. 1995;26:55-59.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Medicine and Biomedical InstrumentationTNO (K.H.W.) Academic Medical Centre, Amsterdam, Netherlands.
Correspondence to D.P. Veerman, MD, PhD, Room B2-222, Department of Internal Medicine, Academic Medical Centre, Meibergdreef 9, 1105 AZ Amsterdam, Netherlands.
| Abstract |
|---|
|
|
|---|
Key Words: circadian rhythm hemodynamics blood pressure heart rate
| Introduction |
|---|
|
|
|---|
It has been well established that BP and HR are lower during the night than during the day and increase sharply at the moment of arising in the morning.7 8 It has been suggested that the sudden increase in HR and BP might act as a trigger for the rupture of atherosclerotic plaques or for platelet aggregation, thereby causing the acute development of a cardiovascular event.4 Most studies on circadian hemodynamics are solely concerned with changes in BP and HR. However, our understanding of these rhythms would increase by the study of the circadian variation of stroke volume (SV), cardiac output (CO), and total systemic vascular resistance.9
| Methods |
|---|
|
|
|---|
Measurements
Intra-arterial BP was measured through a polytetrafluoroethylene
cannula with a length of 11 cm and internal diameter of 1 mm. After
local anesthesia with 1% lidocaine, the cannula was inserted into the
brachial artery of the nondominant arm with the Seldinger technique.
The cannula was connected via a 70-cm-long polyethylene tube to the
transducer of an Oxford Medilog Mark II system (Romulus Technology
Ltd).11 The average overall resonance frequency of the
combined system was 19 Hz (range 14 to 30 Hz). For storage of the
intrabrachial signal, an instrumentation cassette recorder (HR-10J,
TEAC Corp) was used.
Protocol
Subjects were hospitalized for the 24-hour recording period,
during which several strictly controlled activities were carried out;
sleeping time during the night was also controlled. On the first day of
the registration, arterial cannulation was performed between 9 and 10
AM. The actual protocol started at 1 PM and
lasted until 1 PM of the next day. Controlled activities
comprised a supine period (2:30 to 3:30 PM), during which
the subjects rested on a bed but did not sleep; a 30-minute period of
cycling at 50 W, 50 to 60 rpm on a bicycle ergometer (4:45 to 5:15
PM); and two periods of walking outside the hospital from
9:30 to 11:30 AM with a period of 30 minutes of quiet
sitting in between. During the night the subjects stayed in bed from 10
PM to 7 AM.
Data Analysis
The magnetic tapes on which the 24-hour BP signal was recorded
were replayed. The analog BP signal was digitalized for further
processing by computer at 100 Hz with 0.25 mm Hg resolution. HR, mean
arterial blood pressure (MAP), left ventricular SV, CO, and total
peripheral resistance (TPR) were determined beat-to-beat by a
signal-analysis program, and hourly averages of these parameters
were calculated.
The hemodynamic parameters SV, CO, and TPR were calculated with the pulse contour algorithm of Wesseling (Sprangers et al,12 Wesseling et al,13 and Smith et al14 ), discussed briefly below. All pulse contour methods are, explicitly or implicitly, based on a model. Most models are hemodynamic in nature, relating an arterial pressure or pressure difference to a flow or volume via the impedance through which the flow is driven. If the systemic circulation is considered to be a Windkessel model, SV can be calculated from the pressure as the driving force for flow during the ejection time:
![]() |
where SV is the pulse contour SV of the heart, Asys the area under the systolic portion of the pressure wave, and Zao the characteristic impedance of the aorta.
Obviously, this model for the human circulation is too simple. However, sufficient information on human arterial hemodynamics is available today to provide an extension of this model.13 15 MAP is used for the correction of pressure-dependent nonlinear changes in the cross-sectional area of the aorta, and HR is used to correct for reflections from the periphery. (If the time for the pressure wave to travel forward and backward is longer than the ejection time, no correction is needed.) These corrections for pressure and rate are furthermore age dependent. A detailed description can be found in the article by Wesseling at al.13 Basically, the computation can be written as
![]() |
![]() |
where COpc is the pulse contour CO; Pmean is the MAP; and a, b, c, and d are age-dependent parameters. An initial estimate of Zao is computed if age, MAP, and HR are available. The actual characteristic impedance Zao is not known and must be determined for each subject by comparison with an absolute CO estimate such as thermodilution. However, without such calibration one can still determine the relative changes in CO in a subject with the same precision as mentioned below.12 13
CO measured by this method was found to be in good agreement with dye dilution measurements when radial and brachial artery waveforms of young adult subjects were used, when MAP, HR, and CO were varied over wide ranges by pharmacological interventions.14 We have shown previously that when intrabrachial measurements on standing are repeated within a short time period, almost identical responses are obtained for both the arterial pressure measurements and the pulse contour computations.12 On comparison of CO by this method with thermodilution during changes in BP and HR in open-heart surgery, a standard deviation of the difference between the two methods of 10.6% (=0.54 L/min) was found.16
The circulatory circadian profile was analyzed in two ways: (1) To determine the changes of SV, CO, and TPR throughout the whole recording period, we used the entire 24-hour average as the reference level. The percentage of hourly deviations from this reference level were calculated to describe the circadian pattern of these three parameters. (2) The effects of sleep and physical activity were determined by calculating the difference of MAP, HR, SV, CO, and TPR with their daytime averages for the following time periods: the afternoon supine resting period, the nighttime period (10 PM to 7 AM), the walking and cycling periods, and the period of sitting between the two walking periods. For this purpose the daytime average was defined as follows: 1 to 10 PM of the first day combined with 7 AM to 1 PM of the second day. We excluded the nighttime from this reference level as it was one of the periods to be compared. Of course, the height of the chosen reference level may influence the magnitude of the changes in hemodynamic parameters. However, this does not influence the interpretation of the results. Our data consist of relative changes, and the variation of the changes in hemodynamic parameters alters concomitantly with the chosen height of the reference level. Also, the directions of these changes do not change irrespective of any reference level. All results are presented as averages and 95% confidence intervals (95% CI).17 The effects of sleep and physical activity were compared with the daytime average by the Wilcoxon signed rank test. A value of P<.05 was considered significant.
| Results |
|---|
|
|
|---|
|
|
SV showed relatively small changes throughout the entire 24-hour measuring period (Fig 1). Only in the early morning hours after subjects arose did a large increase in SV occur, which lasted until about 11 AM. Compared with the daytime average, nighttime SV showed a decrease of 7% (95% CI, -2% to -12%; P<.05). Because of the large decrease in HR and small change in SV, a large decrease in CO of 29% (95% CI, -24% to -34%; P<.01) was observed during the night (Figs 1 and 2), ranging in individual subjects between -16% and -42%. In the early morning hours CO showed a large increase, with a peak between 10 and 11 AM (Fig 1). The circadian profile of TPR was the mirror image of the BP, HR, and CO profiles (Fig 1). TPR gradually increased when the subjects went to bed and showed a steep decrease when they rose. This pattern was seen in all eight individuals. The average increase in TPR during the night (Figs 1 and 2) was 22% (95% CI, 29% to 16%; P<.01). TPR increased in all subjects during the night (range 15% to 46%). The lowest TPR was reached between 10 and 11 AM (Fig 1).
|
The direction of the changes in hemodynamics during the afternoon supine resting period was comparable to that observed during the night, although the changes were of smaller magnitude (Fig 2, Table). BP and HR dropped as expected, and CO also decreased. TPR increased, although the increase of 8% (95% CI, 15% to 1%; P<.05) was smaller than that seen during the night. During the period of sitting, TPR also increased compared with the daytime average; CO did not differ from the daytime average (Fig 2, Table). During periods of mild physical activity such as walking or cycling, HR and SV increased considerably, resulting in a large increase in CO (Fig 3, Table). As TPR decreased during both periods of physical activity, MAP increased only modestly.
|
| Discussion |
|---|
|
|
|---|
However, reports from animal experiments in primates tend to agree with our findings in humans. In monkeys a hemodynamic circadian rhythm was found that was comparable with our findings in humans.20 21 In these animals CO fell during the night because of a decrease in HR, resulting in a decrease in BP, whereas TPR increased during the night.
Several studies in humans have measured systemic hemodynamics at various instances during the night. However, none of these studies determined SV, CO, or TPR for the entire 24 hours in ambulatory subjects, and the data from these studies are contradictory. Miller and Horvath22 found a decrease in CO during sleep comparable to our findings and a decrease in SV, with no change in HR. These authors could not calculate vascular resistance, as they did not measure BP. Khatri and Freis23 found that the nightly decrease in CO was mediated by a decrease in HR, with no change in SV compared with the waking hours. TPR increased during the night, although not significantly. The results of these studies contrast with the results of Bristow et al,19 who found no change in CO although BP and TPR decreased during the night.
The differences between these results and our findings can at least in part be explained by the fact that in the above-cited studies in humans hemodynamics during waking time were measured while the subjects were heavily instrumented and resting,19 23 whereas our subjects were truly ambulant during the entire day. Also, in these studies measurements in the waking state were performed a few hours before the sleeping period. However, as Fig 1 clearly shows, in this time period the differences in CO and TPR compared with the sleeping hours were much smaller than during the rest of the daytime.
In our opinion the absence of physical activity during the night is instrumental in the genesis of cardiovascular circadian rhythm. When subjects rest during the daytime instead of being ambulant, the circadian variation in BP and HR is significantly decreased.24 25 Clark et al26 demonstrated that the circadian rhythm of BP depends on physical activity; sleeping was one of the activities in their study. Figs 2 and 3 illustrate that in our subjects the changes in CO and TPR induced by moderate physical activity, as during ordinary everyday life, are exactly opposite to those seen during the night. During quiet supine resting in the daytime, CO and TPR change in the same direction as during nighttime sleeping, albeit to a lesser degree (Fig 2).
Decreased demand for oxygen, especially by skeletal muscles, during nighttime sleep may provide the physiological basis for the relation between the activity pattern and the circadian variation in hemodynamics. The existence of a circadian pattern in oxygen consumption in humans has been established.27 The amount of peripheral flow and thus CO is regulated by local autoregulation in the tissues, in proportion to their need for oxygen and other nutrients.9 28 Muscle blood flow can increase manyfold during exercisefrom 1.2 L/min at rest to 22 L/min during maximal exercise29 and responses of the autoregulation in active skeletal muscle can occur within seconds.28 The heart plays a permissive role, allowing CO to be regulated at any level below its maximum limit.9 Thus, a circadian pattern in activity and in peripheral demand for flow may lead to a circadian pattern in venous return and CO. As the level of blood flow is regulated by peripheral vasoconstriction and vasodilatation, a decreased need for blood flow (during the nighttime) implies an increase in peripheral resistance.9 This reasoning makes our finding of an increased TPR during the nighttime plausible.
Another factor that may contribute to the nightly drop in CO is a reduction in effective circulating blood volume during the night compared with the day. The nightly increase in peripheral resistance will then be explained as a compensatory response to the decrease in CO. The decrease in effective circulating volume may be due to a translocation of blood to the periphery secondary to a fall in venous tone.30 It has been shown that in humans venous tone declines during sleep.23 Also, an absolute reduction in plasma volume may occur during the night because of a loss of water volume in expired air, insensible perspiration, and urine formation that is not replaced, causing a negative water balance. This hypothesis is supported by the finding that in monkeys central venous pressure declines and hematocrit increases during the night.31 32 However, as the reduction in plasma volume develops gradually, this would contradict our finding of more or less instantaneous changes in CO and peripheral resistance at the beginning and end of the sleep period.
The sympathetic nervous system probably plays only a minor role in the
regulation of the circadian pattern of BP and related hemodynamics.
During nighttime hours the activity of the sympathetic nervous system
decreases,33 which has been considered to induce a
decrease in peripheral resistance during sleep,18 in
contrast to our findings. However, in monkeys neither
- nor
ß-blockade nor a combined
- and ß-blockade could eliminate the
circadian pattern of decreased MAP and HR and increased TPR during the
night.21 Also, in patients with continuously high levels
of circulating catecholamines due to pheochromocytoma, the circadian
drop in BP is still present.34
We found a large increase in BP and HR in the morning after subjects had risen. This increase in BP was related to an increase in SV and CO larger than the decrease in TPR (Fig 1). The extent of these hemodynamic changes coincided with a brisk, 90-minute walk outside the hospital. This probably amplified the changes in hemodynamics that normally occur at that time, as shown in Fig 3. During a period of quiet sitting in the morning, MAP, HR, SV, CO, and TPR levels did not differ from their daytime averages, confirming that circadian hemodynamics are mainly determined by the intensity of physical activity and not by the clock hour.26
The change in TPR and CO we observed after subjects had risen is different from that found in the cardiovascular laboratory after they had stood up from a supine position. During quiet standing, CO is decreased and TPR increased compared with values during the supine position.35 However, this only holds when subjects remain standing quietly without moving, whereas our subjects were engaged in normal daily activities after arising. The physical activity explains these differences in hemodynamics.
The large and rather steep increase in BP and CO after arising results in a sudden increase of the workload of the heart. These factors may contribute to the onset of acute cardiovascular events in the morning, as they increase the oxygen demand of the myocardium at that time of day.36
The large hemodynamic differences between day and night may have consequences for the treatment of hypertension. Our subjects were all young and normotensive, and as yet we have no similarly detailed information on the circadian profile of CO and TPR in hypertensive subjects. However, as the circadian profile of BP and HR differs only quantitatively between normotension and hypertension,8 the results of this study suggest that different antihypertensive drugs for day and night, with differing hemodynamic mechanisms, may be needed to ensure adequate BP control for the entire 24 hours.
Received September 12, 1994; first decision November 1, 1994; accepted February 10, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. van Dijk, M. C. Boer, T. De Santo, N. Grovale, A. J.J. Aerts, L. Boersma, and W. Wieling Daily, weekly, monthly, and seasonal patterns in the occurrence of vasovagal syncope in an older population Europace, September 1, 2007; 9(9): 823 - 828. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Sayk, C. Becker, C. Teckentrup, H.-L. Fehm, J. Struck, J. P. Wellhoener, and C. Dodt To Dip or Not to Dip: On the Physiology of Blood Pressure Decrease During Nocturnal Sleep in Healthy Humans Hypertension, May 1, 2007; 49(5): 1070 - 1076. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Durgan, N. A. Trexler, O. Egbejimi, T. A. McElfresh, H. Y. Suk, L. E. Petterson, C. A. Shaw, P. E. Hardin, M. S. Bray, M. P. Chandler, et al. The Circadian Clock within the Cardiomyocyte Is Essential for Responsiveness of the Heart to Fatty Acids J. Biol. Chem., August 25, 2006; 281(34): 24254 - 24269. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Young The circadian clock within the heart: potential influence on myocardial gene expression, metabolism, and function Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H1 - H16. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Kawano, T. Motoyama, H. Yasue, N. Hirai, H. M. Waly, K. Kugiyama, and H. Ogawa Endothelial function fluctuates with diurnal variation in the frequency of ischemic episodes in patients with variant angina J. Am. Coll. Cardiol., July 17, 2002; 40(2): 266 - 270. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Yang, E. Scalbert, P. Delagrange, P. M. Vanhoutte, and S. T. O'Rourke Melatonin potentiates contractile responses to serotonin in isolated porcine coronary arteries Am J Physiol Heart Circ Physiol, January 1, 2001; 280(1): H76 - H82. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Deguchi and C. J Mathias Continuous haemodynamic monitoring in an unusual case of swallow induced syncope J. Neurol. Neurosurg. Psychiatry, August 1, 1999; 67(2): 220 - 222. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |