Acta Ph.ysiol Scand 1991, 141, 273-278
ADONIS 000167729 100039W
Diurnal variations in central venous pressure B. T. E N G E L and M. I. T A L A N Laboratory of Behavioral Sciences, National Institute on Aging, National Institutes of Health, Gerontology Research Center", Baltimore, MD 21224, U.S.A.
M. I. 1991. Diurnal variations in central venous pressure. Acta ENGEL, B. T . & TALAN, PhjJsiol Scand 141, 273-278. Received 18 January 1990, accepted 12 September 1990. ISSN 0001-6772. National Institute on Aging, NIH, Gerontology Research Center, Baltimore, MD 21224, U.S.A. We analysed the haemodynamic patterns of seven monkeys (Macaca mulatta) during a one-month period during which we monitored beat-to-beat levels of heart rate, stroke volume, intra-arterial pressure and central venous pressure 7 days/week, 18 h/day. We replicated an earlier finding that cardiac output falls and total peripheral resistance rises throughout the night: the fall in cardiac output was mediated by a fall in heart rate since stroke volume did not change; and the rise in peripheral resistance reflected a homeostatic adjustment to the fall in cardiac output since blood pressure was stable throughout the night although it fell early in the evening. In addition, we have shown for the first time that central venous pressure also falls throughout the night; this occurs even in the presence of a- and P-adrenergic sympathetic blockade. Since stroke volume does not change overnight this indicates that the nocturnal fall in cardiac output is not due to a shift of blood volume into the venous compartment. These findings support our hypothesis that there is a nocturnal decline in plasma volume. Key words : blood volume, haemodynamics, nocturnal.
In a previous report we described a stable diurnal pattern of haemodynamic function in chaired monkeys (Engel & Talan 1987). This pattern was characterized by a fall in cardiac output throughout the night, mediated primarily by a fall in heart rate, since stroke volume was relatively unchanged. Peripheral resistance rose in parallel to the fall in cardiac output. A similar pattern has also been described in rats (Smith et al. 1987), except that the decline in cardiac output and related changes in other cardiovascular parameters occurred during the day: a species-specific effect characteristic of nocturnal animals. Also, Miller & Horvath (1976) reported
a night-time fall in stroke volume and cardiac output in normal human subjects, which was unrelated to sleep stage. I n a subsequent study (Talan & Engel 1989) we showed that aadrenergic, /?-adrenergic and complete sympathetic blockade enhanced the nocturnal fall in cardiac output and rise in total peripheral resistance. These results indicated that the sympathetic nervous system is not a major factor in the production of the nocturnal decline in cardiac output and rise in peripheral resistance; rather, it acts to attenuate these changes. Furthermore, this study also ruled out the possibility that changes in vasomotor tone, which might contribute to blood volume distribution, could explain the nocturnal haemodynamic patterns already noted. Nevertheless, there was still the possibility that there was an overnight I
* The Gerontology Research Center is fully accredited by the American Association of Laboratory Animal Care. Correspondence : Dr Bernard T. Engel, Laboratory of Behavioral Sciences, National Institute on Aging, Gerontology Research Center, 4940 Eastern Avenue, Baltimore, MD 21224.
shift
Of
into the
'Om-
Partment which would account for the fall in cardiac output - either through a fall in cardiac contractility or through an expansion of peri-
273
'
B . T. Engel and M. I . Tulun
274
180 j
I
800,
I
m
_*
P
m
300
,~......~~.....~
I
v)
150
f .c m
,---.... #-.-...17 19 21 23 01 03 05 07 09 11 loo TIME
a
TIME
Fig. 1. Average hourly haemodynamic patterns for seven animals. Points within the dashed lines are levels during an interval when the animals were in the dark.
pheral venous beds. Thus, this study was designed to evaluate these possibilities by measuring central venous pressure as well as cardiac output and total peripheral resistance. M A T E R I A L AND M E T H O D S Subjects
T h e subjects were seven male monkeys (Macucu muluttu). Each weighed about 4 kg at the time of entry into the study. Each animal was chronically chaired and maintained in a primate booth throughout the duration of the study. After 1 week of adaptation to the primate chair a permanent-magnet flow probe ( I n Vzco Metric, model FX-3) was implanted at the root of the aorta. The surgery was performed aseptically, under halothane anaesthesia. A detailed description of the surgery was given in our earlier reports (Engel & Talan 1987, Talan & Engel 1989). After a 3-week recovery, an arterial cannula was implanted aseptically,
under halothane anaesthesia, into the abdominal aorta through the external iliac artery (Engel 1986). The patency of the catheter was maintained by continuous infusion of a solution of 2.0 U/ml of heparin in isotonic saline at the rate of 40 m1/24 h. At the same time that the arterial cannula was implanted, we also implanted a catheter, via the cephalic vein, into the vena cava at the level of the right atrium. The position of the venous catheter was confirmed radiographically at the time of surgery, and it was reconfirmed upon completion of the studies. The diurnal observations reported below were begun not sooner than 3 weeks after catheterization. Throughout the period of study the animals were cleaned daily (7 days/week), and their health was monitored daily. They were also weighed every 2 weeks. Procedure. During diurnal testing the booth door was closed at 1700 h and opened at 1200 h the following day. Feeding time was restricted to the hours between 1200 h and 1700 h (primate laboratory chow was supplemented with fresh fruit), tap water was accessible at all times. The light-dark cycle was
B
-
Central venous pressure
Double-Sympathetic blockade I -Control r _ _ _ _ _ _ _ _ - - _ - _ - -
4
275
3.0 2.5
Fig. 2. Average hourly patterns of central venous pressure: (A) results for all seven animals; (B) results for three animals during control and double sympathetic blockade. Points within the dashed lines are levels during an interval when the animals were in the dark.
12 h with light off at about 2000 h and the ambient temperature was maintained at 2 2 k 1 "C. Data collection was started at 1800 h and completed at 1200 h each day, 7 days/-eek. The computer collected heart period, stroke volume, central venous pressure and systolic blood pressure on a beat-by-beat basis (Engel & Talan 1987). Every 60 s these data were averaged and the means were stored for subsequent analyses. The observation period for each animal was approximately 3 weeks preceded by about 3 4 days of habituation to the procedure: we obtained usable data from three animals for 19 days and from four animals for 20 days. We also tested the effect of sympathetic blockade on central venous pressure in three of the animals (0.06 mg/kg/h, prazosin and 0.4 mg/kg/h, atenolol infused continuously through the arterial catheter for 20 days/animal). The details for administering and testing the adequacy of dosage of these drugs has been reported previously (Talan & Engel 1989). Stutisticul anulyses. Data were analysed separately for the 3 h early evening period (1800-2100 h), for the 10 h period when the animal was in the dark (2100-0700 h) and for the 5 h period during the morning (0700-1200 h). Since the booth lights were turned off at approximately 2000 h and turned back on at about 0800 h, we selected the 10 h interval from 2100 to 0700 h as the dark period. The minute-by-minute means for each of the five primary cardiovascular responses and for two derived indices, cardiac output and total peripheral resistance, were partitioned into 18 hourly time-blocks comprising 60 scores/block. Cardiac output was computed as heart rate x stroke volume. Total peripheral resistance was computed as mean arterial pressure/
cardiac output, where mean arterial pressure was diastolic pressure + 5 pulse pressure -central venous pressure. During the control condition, data were analysed using repeated measures analyses of variance (ANOVA) in which the independent factor was animals, the replication factor was days and the dependent factor was time of day. In addition to the analyses of response levels, i.e. the comparison of the average levels, we also analysed the hour-by-hour trends. Two components of trend were analysed : the linear component that assesses any monotonic trend ; and the quadratic component that assesses any bidirectional change in response, e.g. a fall followed by a rise. The effect of sympathetic blockade on central venous pressure was also analysed using the repeated measures analysis of variance to compare the differences in trend between the non-drug condition and the blockade condition. We are not reporting the effects of blockade on the other haemodynamic parameters since they were similar to those reported earlier (Talan & Engel 1989). Throughout this report, statistical significance is defined as P < 0.05.
RESULTS Figure 1 shows the diurnal patterns for each of the physiological measures except central venous pressure, which is shown in Figure 2A. Table 1 presents the results of the analyses of variance for each of the physiological functions during the early evening, night-time and morning periods, respectively. There is a significant linear trend in the diurnal pattern of heart rate at all times, and there are also significant quadratic trends in the
B . T. Engel and M . I. Talan
276
Table 1. F-ratios from the analyses of the diurnal trends for each of the physiological functions. The degrees of freedom for all analyses were 1/130.
3 h analyses ( 1800-2 100 h)
10 h analyses (2 100-0700 h)
5 h analyses (0700-1200 h)
Linear
Quadratic
Linear
Quadratic
~~~~~~
Response Heart rate (beats min-') Stroke volume (ml beat-') Cardiac output (ml min-') Systolic pressure (mmHg) Diastolic pressure (mmHg) Venous pressure (mmHg) Peripheral resistance (mmHg mi-' min-I)
"
Linear
Quadratic
69.37"
4.82"
190.40"
0.05
156.46"
12.04"
6.14"
6.02"
2.07
3.99"
12.22"
23.73"
60.28"
0.09
46.'20"
1.42
79.58"
21.34"
103.53"
34.37"
1.78
3.30
243.31"
4.36"
116.06"
43.10"
2.06
0.24
222.98"
3.1 1
3.63
2.25
150.15"
0.21
47.32"
8.64"
0.82
10.64"
20.55"
0.78
0.90
25.02"
Significant.
early evening and morning intervals. T h e strokevolume data show both a linear and quadratic trend in the early evening and morning, but only a quadratic trend in the 10 h analysis, indicating that stroke volume did not fall throughout the night. T h e linear trend of change in cardiac output follows the pattern of heart rate. Systolic and diastolic pressures both show linear and quadratic trends during the early evening, and linear trends during the morning intervals, but no trends during the 10 h period. These trend analyses show that blood pressure is maintained throughout the night, and that it rises in the morning to levels which exceed those in the early evening. Central venous pressure shows a quadratic trend in the early evening, a linear trend at night, and both linear and quadratic trends in the morning. T h e quadratic, early evening trend is mediated by the rise in venous pressure between 1900 and 2100 h ; the linear, night-time trend begins at 2100 h and continues until 1200 h the following morning; the quadratic trend in the morning reflects a 'flattening' of the fall during the final 3 morning hours. Figure 2 B shows the pattern of change in central venous pressure during control and sympathetic blockade. T h e level of central venous pressure was always greater during sympathetic blockade irrespective of the time
period. There was never a difference in linear trend between the drug and non-drug condition (F(1,112) = 0.45, 2.68 and 0.88 for the early evening, night-time and early morning intervals, respectively). There was no difference in quadratic trends between the conditions in the early evening (F(1,112) = 0.51); however, there were significant differences in quadratic trends at night (F(1,112) = 6.30) and in the early morning (F(1,112) = 5.57). T h e differences in quadratic trends reflect minor differences in the times at which the peak or trough pressures were attained; however, the overall patterns of trends are relatively similar. I t might also be noted that in the case of these three animals, central venous pressure began returning to early evening levels somewhat earlier than was the case when all seven animals were averaged (Fig. 2B); nevertheless, the pattern of return was similar during the control and blocked conditions.
DISCUSSION T h e main finding in this study was that central venous pressure falls throughout the night. This finding, in conjunction with the observations that stroke volume and arterial pressure do not change throughout the night, shows that the fall in cardiac output cannot be mediated by a
Central venous pressure reduction in cardiac contractility. Furthermore, the fact that these studies were done under conditions where posture did not change provides further evidence that the nightly fall in cardiac output seen by us and others (Engel 1986, Engel & Talan 1987, Greenleaf 1984, Miller & Horvath 1976, Smith et J.1987, Talan & Engel 1989) is mediated, at least in part, by a fall in plasma volume. W e believe that several mechanisms account for this loss in volume: (1) nocturnal diuresis ; (2) insensible perspiration ; ( 3 ) respiration ; and (4) failure to replace fluid because monkeys d o not drink in the dark although they continue to produce urine (Moore-Ede & Herd 1977). I t is noteworthy that central venous pressure continues to fall in the early morning hours while cardiac output and blood pressure are rising and peripheral resistance shows no linear trend. This haemodynamic pattern probably reflects a shift of plasma volume from the venous to the arterial circulation at a time when total plasma volume is still reduced. This interpretation is consistent with the observation that the fall in central venous pressure is greater during the first 2 or 3 h of the post-dark period and then flattens or rises (Fig. 2A, B) after the animals have begun to drink and to restore plasma volume. I n a study reported earlier (Talan & Engel 1989) we examined the haemodynamic effects of chronic ,B-sympathetic blockade (atenolol), asympathetic blockade (prazosin) and doublesympathetic blockade (atenolol and prazosin). T h e main finding from that study was that partial or total sympathetic blockade not only does not abolish the nocturnal fall in cardiac output and rise in peripheral resistance, it exacerbates it. This finding suggests that the sympathetic nervous system buffers the haemodynamic concomitant typical of sleep. I n the present study, sympathetic blockade was associated with an overall increase in central venous pressure, a well-known consequence of the druginduced vasodilation. However, the diurnal pattern of change in central venous pressure was very similar in the drug and control condition. T h e findings from these two studies provide strong evidence that changes in vasomotor tone d o not mediate the nocturnal fall in central venous pressure. T h e y further suggest that passive filling of venous beds is not responsible for the reduced venous return since (1) under
277
sympathetic blockade the veins would tend to be fully dilated at all times, and (2) the pattern of change was similar in the non-drug and drug conditions. There is considerable evidence that plasma volume falls during sleep. Mice, which are nocturnal animals, have higher haemoglobin and haematocrit levels in the afternoon than they do in the morning (Swoyer 1987); and humans have higher haematocrits in the morning than they do in the afternoon (Cranston & Brown 1983). Greenleaf (1984), in his review of the physiological consequences of bed rest, notes that plasma volume falls by 125 ml (4.3%) after 6 h of head-down bed rest, and by 153-310ml (5.3-9.70/3 after 24 h. I n all of the human studies cited above and reviewed by Greenleaf, the changes in plasma volume were probably mediated by a pressure diuresis following the postural change from upright to recumbency. While this does not militate against the significance of the findings since this is normal human behaviour, it does leave open the question whether the same effect would occur when postural changes are precluded. T h e present study, taken in conjunction with our other findings, shows that in primates that have regular schedules of sleep and wakefulness there are significant diurnal haemodynamic changes, and that these nocturnal changes are probably mediated, at least in part, by a fall in plasma volume. It will be important for future research to develop appropriate methods for measuring plasma volume throughout the night without disturbing the animal’s normal sleep. REFERENCES CRANSTON, W.I. & BROWN,W. 1963. Diurnal variation in plasma volume in normal and hypertensive subjects. Clin Sci Lond 25, 107-114. ENGEL, B.T. 1986.Diurnal variations in cardiovascular integration. A m 3 Physiol 250, (Regulatory Integratiae Comp Physiol 19): R36R40. ENGEL,B.T. & TALAN, M.I. 1987 Diurnal pattern of hemodynamic performance in nonhuman primates. A m 3 Ph,ysiol 253, (Regulatory Integrative Comp Physiol 2 2 ) : R779-R785. GREENLEAF, J.E. 1984. Physiological responses to prolonged bed rest and fluid immersion in humans.
3 Appl Physiol (Respiratory Environmental
Exercise
Physiol) 7, 619-663. MILLER, J.C. & HORVATH, S.M. 1976. Cardiac output during human sleep. Aviat Space Environ Med 41, 104&1051.
278
B. T. Engel and M . I . Talan
MOORE-EDE, M.C. & HERD,J.A. 1975. Renal electrolyte circadian rhythms : independence from feeding and activity patterns. A m 3 Physiol 1 (Renal Fluid Electrolyte Physiol) 2, F128-Fl35. SMITH,T.L., COLEMAN,T.G., STANEK,K.A. & MURPHY,W.R. 1987 A m 3 Physiol 2 (Regulatory Integrative Comp Physiol 22) : R779-785. J., HANS,E. & SACKATT-LUNDEEN, L. 1987. SWOYER,
Circadian reference values from hematologic parameter in several strains of mice. In: J.E. Pauly & L. E. Scheving (eds). Advances in Chronobiology. Liss, New York. TALAN,M.I. & ENGEL,B.T. 1989. Effect of sympathetic blockade on diurnal variation of hemodynamic patterns. Am 3 Ph,ysiol 256 (Regulatory Integrative Comp Physiol) 25, R778-R785.
ADONIS 000167729 100039W
Diurnal variations in central venous pressure B. T. E N G E L and M. I. T A L A N Laboratory of Behavioral Sciences, National Institute on Aging, National Institutes of Health, Gerontology Research Center", Baltimore, MD 21224, U.S.A.
M. I. 1991. Diurnal variations in central venous pressure. Acta ENGEL, B. T . & TALAN, PhjJsiol Scand 141, 273-278. Received 18 January 1990, accepted 12 September 1990. ISSN 0001-6772. National Institute on Aging, NIH, Gerontology Research Center, Baltimore, MD 21224, U.S.A. We analysed the haemodynamic patterns of seven monkeys (Macaca mulatta) during a one-month period during which we monitored beat-to-beat levels of heart rate, stroke volume, intra-arterial pressure and central venous pressure 7 days/week, 18 h/day. We replicated an earlier finding that cardiac output falls and total peripheral resistance rises throughout the night: the fall in cardiac output was mediated by a fall in heart rate since stroke volume did not change; and the rise in peripheral resistance reflected a homeostatic adjustment to the fall in cardiac output since blood pressure was stable throughout the night although it fell early in the evening. In addition, we have shown for the first time that central venous pressure also falls throughout the night; this occurs even in the presence of a- and P-adrenergic sympathetic blockade. Since stroke volume does not change overnight this indicates that the nocturnal fall in cardiac output is not due to a shift of blood volume into the venous compartment. These findings support our hypothesis that there is a nocturnal decline in plasma volume. Key words : blood volume, haemodynamics, nocturnal.
In a previous report we described a stable diurnal pattern of haemodynamic function in chaired monkeys (Engel & Talan 1987). This pattern was characterized by a fall in cardiac output throughout the night, mediated primarily by a fall in heart rate, since stroke volume was relatively unchanged. Peripheral resistance rose in parallel to the fall in cardiac output. A similar pattern has also been described in rats (Smith et al. 1987), except that the decline in cardiac output and related changes in other cardiovascular parameters occurred during the day: a species-specific effect characteristic of nocturnal animals. Also, Miller & Horvath (1976) reported
a night-time fall in stroke volume and cardiac output in normal human subjects, which was unrelated to sleep stage. I n a subsequent study (Talan & Engel 1989) we showed that aadrenergic, /?-adrenergic and complete sympathetic blockade enhanced the nocturnal fall in cardiac output and rise in total peripheral resistance. These results indicated that the sympathetic nervous system is not a major factor in the production of the nocturnal decline in cardiac output and rise in peripheral resistance; rather, it acts to attenuate these changes. Furthermore, this study also ruled out the possibility that changes in vasomotor tone, which might contribute to blood volume distribution, could explain the nocturnal haemodynamic patterns already noted. Nevertheless, there was still the possibility that there was an overnight I
* The Gerontology Research Center is fully accredited by the American Association of Laboratory Animal Care. Correspondence : Dr Bernard T. Engel, Laboratory of Behavioral Sciences, National Institute on Aging, Gerontology Research Center, 4940 Eastern Avenue, Baltimore, MD 21224.
shift
Of
into the
'Om-
Partment which would account for the fall in cardiac output - either through a fall in cardiac contractility or through an expansion of peri-
273
'
B . T. Engel and M. I . Tulun
274
180 j
I
800,
I
m
_*
P
m
300
,~......~~.....~
I
v)
150
f .c m
,---.... #-.-...17 19 21 23 01 03 05 07 09 11 loo TIME
a
TIME
Fig. 1. Average hourly haemodynamic patterns for seven animals. Points within the dashed lines are levels during an interval when the animals were in the dark.
pheral venous beds. Thus, this study was designed to evaluate these possibilities by measuring central venous pressure as well as cardiac output and total peripheral resistance. M A T E R I A L AND M E T H O D S Subjects
T h e subjects were seven male monkeys (Macucu muluttu). Each weighed about 4 kg at the time of entry into the study. Each animal was chronically chaired and maintained in a primate booth throughout the duration of the study. After 1 week of adaptation to the primate chair a permanent-magnet flow probe ( I n Vzco Metric, model FX-3) was implanted at the root of the aorta. The surgery was performed aseptically, under halothane anaesthesia. A detailed description of the surgery was given in our earlier reports (Engel & Talan 1987, Talan & Engel 1989). After a 3-week recovery, an arterial cannula was implanted aseptically,
under halothane anaesthesia, into the abdominal aorta through the external iliac artery (Engel 1986). The patency of the catheter was maintained by continuous infusion of a solution of 2.0 U/ml of heparin in isotonic saline at the rate of 40 m1/24 h. At the same time that the arterial cannula was implanted, we also implanted a catheter, via the cephalic vein, into the vena cava at the level of the right atrium. The position of the venous catheter was confirmed radiographically at the time of surgery, and it was reconfirmed upon completion of the studies. The diurnal observations reported below were begun not sooner than 3 weeks after catheterization. Throughout the period of study the animals were cleaned daily (7 days/week), and their health was monitored daily. They were also weighed every 2 weeks. Procedure. During diurnal testing the booth door was closed at 1700 h and opened at 1200 h the following day. Feeding time was restricted to the hours between 1200 h and 1700 h (primate laboratory chow was supplemented with fresh fruit), tap water was accessible at all times. The light-dark cycle was
B
-
Central venous pressure
Double-Sympathetic blockade I -Control r _ _ _ _ _ _ _ _ - - _ - _ - -
4
275
3.0 2.5
Fig. 2. Average hourly patterns of central venous pressure: (A) results for all seven animals; (B) results for three animals during control and double sympathetic blockade. Points within the dashed lines are levels during an interval when the animals were in the dark.
12 h with light off at about 2000 h and the ambient temperature was maintained at 2 2 k 1 "C. Data collection was started at 1800 h and completed at 1200 h each day, 7 days/-eek. The computer collected heart period, stroke volume, central venous pressure and systolic blood pressure on a beat-by-beat basis (Engel & Talan 1987). Every 60 s these data were averaged and the means were stored for subsequent analyses. The observation period for each animal was approximately 3 weeks preceded by about 3 4 days of habituation to the procedure: we obtained usable data from three animals for 19 days and from four animals for 20 days. We also tested the effect of sympathetic blockade on central venous pressure in three of the animals (0.06 mg/kg/h, prazosin and 0.4 mg/kg/h, atenolol infused continuously through the arterial catheter for 20 days/animal). The details for administering and testing the adequacy of dosage of these drugs has been reported previously (Talan & Engel 1989). Stutisticul anulyses. Data were analysed separately for the 3 h early evening period (1800-2100 h), for the 10 h period when the animal was in the dark (2100-0700 h) and for the 5 h period during the morning (0700-1200 h). Since the booth lights were turned off at approximately 2000 h and turned back on at about 0800 h, we selected the 10 h interval from 2100 to 0700 h as the dark period. The minute-by-minute means for each of the five primary cardiovascular responses and for two derived indices, cardiac output and total peripheral resistance, were partitioned into 18 hourly time-blocks comprising 60 scores/block. Cardiac output was computed as heart rate x stroke volume. Total peripheral resistance was computed as mean arterial pressure/
cardiac output, where mean arterial pressure was diastolic pressure + 5 pulse pressure -central venous pressure. During the control condition, data were analysed using repeated measures analyses of variance (ANOVA) in which the independent factor was animals, the replication factor was days and the dependent factor was time of day. In addition to the analyses of response levels, i.e. the comparison of the average levels, we also analysed the hour-by-hour trends. Two components of trend were analysed : the linear component that assesses any monotonic trend ; and the quadratic component that assesses any bidirectional change in response, e.g. a fall followed by a rise. The effect of sympathetic blockade on central venous pressure was also analysed using the repeated measures analysis of variance to compare the differences in trend between the non-drug condition and the blockade condition. We are not reporting the effects of blockade on the other haemodynamic parameters since they were similar to those reported earlier (Talan & Engel 1989). Throughout this report, statistical significance is defined as P < 0.05.
RESULTS Figure 1 shows the diurnal patterns for each of the physiological measures except central venous pressure, which is shown in Figure 2A. Table 1 presents the results of the analyses of variance for each of the physiological functions during the early evening, night-time and morning periods, respectively. There is a significant linear trend in the diurnal pattern of heart rate at all times, and there are also significant quadratic trends in the
B . T. Engel and M . I. Talan
276
Table 1. F-ratios from the analyses of the diurnal trends for each of the physiological functions. The degrees of freedom for all analyses were 1/130.
3 h analyses ( 1800-2 100 h)
10 h analyses (2 100-0700 h)
5 h analyses (0700-1200 h)
Linear
Quadratic
Linear
Quadratic
~~~~~~
Response Heart rate (beats min-') Stroke volume (ml beat-') Cardiac output (ml min-') Systolic pressure (mmHg) Diastolic pressure (mmHg) Venous pressure (mmHg) Peripheral resistance (mmHg mi-' min-I)
"
Linear
Quadratic
69.37"
4.82"
190.40"
0.05
156.46"
12.04"
6.14"
6.02"
2.07
3.99"
12.22"
23.73"
60.28"
0.09
46.'20"
1.42
79.58"
21.34"
103.53"
34.37"
1.78
3.30
243.31"
4.36"
116.06"
43.10"
2.06
0.24
222.98"
3.1 1
3.63
2.25
150.15"
0.21
47.32"
8.64"
0.82
10.64"
20.55"
0.78
0.90
25.02"
Significant.
early evening and morning intervals. T h e strokevolume data show both a linear and quadratic trend in the early evening and morning, but only a quadratic trend in the 10 h analysis, indicating that stroke volume did not fall throughout the night. T h e linear trend of change in cardiac output follows the pattern of heart rate. Systolic and diastolic pressures both show linear and quadratic trends during the early evening, and linear trends during the morning intervals, but no trends during the 10 h period. These trend analyses show that blood pressure is maintained throughout the night, and that it rises in the morning to levels which exceed those in the early evening. Central venous pressure shows a quadratic trend in the early evening, a linear trend at night, and both linear and quadratic trends in the morning. T h e quadratic, early evening trend is mediated by the rise in venous pressure between 1900 and 2100 h ; the linear, night-time trend begins at 2100 h and continues until 1200 h the following morning; the quadratic trend in the morning reflects a 'flattening' of the fall during the final 3 morning hours. Figure 2 B shows the pattern of change in central venous pressure during control and sympathetic blockade. T h e level of central venous pressure was always greater during sympathetic blockade irrespective of the time
period. There was never a difference in linear trend between the drug and non-drug condition (F(1,112) = 0.45, 2.68 and 0.88 for the early evening, night-time and early morning intervals, respectively). There was no difference in quadratic trends between the conditions in the early evening (F(1,112) = 0.51); however, there were significant differences in quadratic trends at night (F(1,112) = 6.30) and in the early morning (F(1,112) = 5.57). T h e differences in quadratic trends reflect minor differences in the times at which the peak or trough pressures were attained; however, the overall patterns of trends are relatively similar. I t might also be noted that in the case of these three animals, central venous pressure began returning to early evening levels somewhat earlier than was the case when all seven animals were averaged (Fig. 2B); nevertheless, the pattern of return was similar during the control and blocked conditions.
DISCUSSION T h e main finding in this study was that central venous pressure falls throughout the night. This finding, in conjunction with the observations that stroke volume and arterial pressure do not change throughout the night, shows that the fall in cardiac output cannot be mediated by a
Central venous pressure reduction in cardiac contractility. Furthermore, the fact that these studies were done under conditions where posture did not change provides further evidence that the nightly fall in cardiac output seen by us and others (Engel 1986, Engel & Talan 1987, Greenleaf 1984, Miller & Horvath 1976, Smith et J.1987, Talan & Engel 1989) is mediated, at least in part, by a fall in plasma volume. W e believe that several mechanisms account for this loss in volume: (1) nocturnal diuresis ; (2) insensible perspiration ; ( 3 ) respiration ; and (4) failure to replace fluid because monkeys d o not drink in the dark although they continue to produce urine (Moore-Ede & Herd 1977). I t is noteworthy that central venous pressure continues to fall in the early morning hours while cardiac output and blood pressure are rising and peripheral resistance shows no linear trend. This haemodynamic pattern probably reflects a shift of plasma volume from the venous to the arterial circulation at a time when total plasma volume is still reduced. This interpretation is consistent with the observation that the fall in central venous pressure is greater during the first 2 or 3 h of the post-dark period and then flattens or rises (Fig. 2A, B) after the animals have begun to drink and to restore plasma volume. I n a study reported earlier (Talan & Engel 1989) we examined the haemodynamic effects of chronic ,B-sympathetic blockade (atenolol), asympathetic blockade (prazosin) and doublesympathetic blockade (atenolol and prazosin). T h e main finding from that study was that partial or total sympathetic blockade not only does not abolish the nocturnal fall in cardiac output and rise in peripheral resistance, it exacerbates it. This finding suggests that the sympathetic nervous system buffers the haemodynamic concomitant typical of sleep. I n the present study, sympathetic blockade was associated with an overall increase in central venous pressure, a well-known consequence of the druginduced vasodilation. However, the diurnal pattern of change in central venous pressure was very similar in the drug and control condition. T h e findings from these two studies provide strong evidence that changes in vasomotor tone d o not mediate the nocturnal fall in central venous pressure. T h e y further suggest that passive filling of venous beds is not responsible for the reduced venous return since (1) under
277
sympathetic blockade the veins would tend to be fully dilated at all times, and (2) the pattern of change was similar in the non-drug and drug conditions. There is considerable evidence that plasma volume falls during sleep. Mice, which are nocturnal animals, have higher haemoglobin and haematocrit levels in the afternoon than they do in the morning (Swoyer 1987); and humans have higher haematocrits in the morning than they do in the afternoon (Cranston & Brown 1983). Greenleaf (1984), in his review of the physiological consequences of bed rest, notes that plasma volume falls by 125 ml (4.3%) after 6 h of head-down bed rest, and by 153-310ml (5.3-9.70/3 after 24 h. I n all of the human studies cited above and reviewed by Greenleaf, the changes in plasma volume were probably mediated by a pressure diuresis following the postural change from upright to recumbency. While this does not militate against the significance of the findings since this is normal human behaviour, it does leave open the question whether the same effect would occur when postural changes are precluded. T h e present study, taken in conjunction with our other findings, shows that in primates that have regular schedules of sleep and wakefulness there are significant diurnal haemodynamic changes, and that these nocturnal changes are probably mediated, at least in part, by a fall in plasma volume. It will be important for future research to develop appropriate methods for measuring plasma volume throughout the night without disturbing the animal’s normal sleep. REFERENCES CRANSTON, W.I. & BROWN,W. 1963. Diurnal variation in plasma volume in normal and hypertensive subjects. Clin Sci Lond 25, 107-114. ENGEL, B.T. 1986.Diurnal variations in cardiovascular integration. A m 3 Physiol 250, (Regulatory Integratiae Comp Physiol 19): R36R40. ENGEL,B.T. & TALAN, M.I. 1987 Diurnal pattern of hemodynamic performance in nonhuman primates. A m 3 Ph,ysiol 253, (Regulatory Integrative Comp Physiol 2 2 ) : R779-R785. GREENLEAF, J.E. 1984. Physiological responses to prolonged bed rest and fluid immersion in humans.
3 Appl Physiol (Respiratory Environmental
Exercise
Physiol) 7, 619-663. MILLER, J.C. & HORVATH, S.M. 1976. Cardiac output during human sleep. Aviat Space Environ Med 41, 104&1051.
278
B. T. Engel and M . I . Talan
MOORE-EDE, M.C. & HERD,J.A. 1975. Renal electrolyte circadian rhythms : independence from feeding and activity patterns. A m 3 Physiol 1 (Renal Fluid Electrolyte Physiol) 2, F128-Fl35. SMITH,T.L., COLEMAN,T.G., STANEK,K.A. & MURPHY,W.R. 1987 A m 3 Physiol 2 (Regulatory Integrative Comp Physiol 22) : R779-785. J., HANS,E. & SACKATT-LUNDEEN, L. 1987. SWOYER,
Circadian reference values from hematologic parameter in several strains of mice. In: J.E. Pauly & L. E. Scheving (eds). Advances in Chronobiology. Liss, New York. TALAN,M.I. & ENGEL,B.T. 1989. Effect of sympathetic blockade on diurnal variation of hemodynamic patterns. Am 3 Ph,ysiol 256 (Regulatory Integrative Comp Physiol) 25, R778-R785.
