- Email: [email protected]
Human autonomic responses to blood donation
Autonomic Neuroscience: Basic and Clinical 110 (2004) 114 – 120 www.elsevier.com/locate/autneu
Human autonomic responses to blood donation E´va Zo¨llei a,*, Do´ra Paprika a, Pe´ter Makra b, Zolta´n Gingl b, Kla´ra Vezendi c, La´szlo´ Rudas a a
University of Szeged, Faculty of Medicine, Cardiology Center, Medical Intensive Care Unit, Kora´nyi fasor 7, Szeged 6720, Hungary b University of Szeged, Department of Experimental Physics, Hungary c University of Szeged, Department of Blood Banking, Hungary Received 12 June 2003; received in revised form 27 October 2003; accepted 29 October 2003
Abstract In order to characterize autonomic responses to acute volume loss, supine ECG, blood pressure (BP) and uncalibrated breathing signal (UBS) recordings were taken before and after blood donation in 48 healthy volunteers. Time and frequency domain parameters of RR interval (RRI), BP and UBS variability were determined. Baroreflex gain was calculated by the technique of the spontaneous sequences and crossspectral analysis. The systolic (SAP), diastolic (DAP) and mean BP (MAP) increased after the blood withdrawal. The central frequency of breathing and mean heart rate did not change. RRI variability increased in low frequency band (LF), tended to decrease in high frequency band (HF). Systolic BP variability increased in both frequency bands, but was statistically significant only in the high frequency band. Diastolic BP power increased in both frequencies. From the different baroreflex gain estimates, up sequence BRS and HF alpha index decreased significantly. The phase angle between RRI and systolic blood pressure powers in LF band did not change ( 58 F 24j and 54 F 26j). In the high frequency range, the phase became more negative ( 1 F 29j and 17 F 32j, p = 0.001). The withdrawal of 350 – 400 ml blood in 5 min resulted in sympathetic activation, which was reflected in increased systolic, diastolic and mean BP. The increased BP oscillation was a sensitive marker of the minor volume depletion. This was coupled by increased RRI oscillation via baroreflex mechanisms in the LF band. Changes in the RRI and BP oscillations in the HF band showed no similar coupling. That points to the fact that RRI oscillations in this band should not be explained entirely by baroreflex mechanisms. Vagal withdrawal was reflected in decreased root mean square of successive differences (RMSSD), decreased HF RRI power and decreased up sequence BRS. D 2004 Elsevier B.V. All rights reserved. Keywords: Autonomic nervous system; Baroreflexes; Hypovolemia
1. Introduction Arterial and cardiopulmonary baroreflexes play an essential role in maintaining blood pressure (BP) in acute hypovolemic states. Data from animal studies of volume loading and unloading suggest that cardiopulmonary baroreceptor input can modify arterial baroreflex responses. However, in humans, it is not clear what, if any role play the cardiopulmonary baroreflexes in modulating arterial baroreflex control of the circulation. Regarding the baroreflex control of heart rate, there are opposing data during central volume changes induced by direct volume manipulation or gravitation induced volume shifts.
* Corresponding author. Tel./fax: +36-62-545-689. E-mail address: [email protected] (E´. Zo¨llei). 1566-0702/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2003.10.003
It is well-known that arterial pressure variability increases with hypovolemia. The changes of systolic blood pressure during the respiratory cycle can be described by the systolic pressure variation (SPV) and by its components, the so-called delta up and delta down. These indices are regarded as sensitive indicators of volemic state (Perel et al., 1987; Rooke, 1995; Rooke et al., 1995). Blood pressure fluctuations also can be characterized by frequency domain indices. Using power spectral analysis not just the magnitude, but the frequency distribution of systolic, diastolic and mean blood pressure fluctuations can be assessed. By cross spectral method, the feedback relationship of blood pressure and heart rate modulation can also be investigated. We chose the setting of blood donation to assess autonomic responses to acute volume loss. We characterized autonomic responses by calculating time and frequency domain indices of heart rate and blood pressure variability. To calculate baroreflex gain, we used two non-invasive
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120 Table 1
SAP (mm Hg) DAP (mm Hg) MAP (mm Hg) RRI (ms) SDRR (ms) RMSSD PNN50
Before
After
P-value
124 F 16 74 F 13 91 F 13 816 F 142 41 F 24 33 F 25 3.7 F 0.1
128 F 13 76 F 12 93 F 12 806 F 138 40 F 22 28 F 26 1.0 F 0.2
< 0.001 < 0.001 0.003 NS NS 0.027 NS
The SAP, DAP and MAP, the mean RRI, SDRR, RMSSD and PNN50 before and after blood donation.
methods, the spontaneous sequence and the cross-spectral analysis. These techniques are now widely used and give comparable results to the classical pharmacological and neck-chamber methods, and are regarded more physiological estimates of baroreflex function (Parlow et al., 1995; Watkins et al., 1995; James et al., 1998; Pitzalis et al., 1998; Rudas et al., 1999). The aim of this study was to compare the time and frequency domain parameters of heart rate and blood pressure variability and the baroreflex control of heart rate before and after 350 –400 ml blood was withdrawn from healthy donors, and to get further insight into the short term compensatory mechanisms to acute volume loss.
2. Materials and methods The study population consisted of 48 healthy volunteers (age 35 F 12, 18 –59 years; 23 men, 25 women), who were recruited when they signed up for blood donation. Three of them were first time, the others were regular blood donors. None of the subjects had a previous history of syncope or presyncope. All the subjects were informed and gave their consent for their participation in the study. The study protocol was approved by the Ethical Committee of the University of Szeged. The investigations were carried out according to the Declaration of Helsinki. All the measurements were made in a quiet room at the Blood Bank after 5 min time interval of supine rest. First, 5 min baseline recordings were taken. This was followed by the venipuncture and by the withdrawal of 350– 400 ml blood in approximately 5 min. After the removal of the needle, we started at once with the second recordings. The subjects were not allowed to drink, nor received their caffeinated beverages throughout the investigations. Two of them experienced mild symptoms suggesting vasovagal reaction; they were put to head-down tilt position at the end of the blood withdrawal, according to the protocol of the Blood Bank. By the time of the second recording, they were put back supine. The ECG and blood pressure signals were continuously measured with a Marquette bedside monitor and with the Finapres 2300 non-invasive blood pressure monitor. Signals were recorded and on-line digitalized with 500 Hz by the
115
Dataq/Windaq system. The data were analyzed by using the WinCPRS software (Absolute Aliens Ay, 2000). All recordings were peak detected and checked by one of the investigators. After the confirmation of peak detection, conventional time domain indices of heart rate variability were determined (SDRR: standard deviation of RR intervals, RMSSD: root mean square of successive differences, PNN50: the percentage of cardiac cycles, where the difference was greater than 50 ms between the successive beats), and search for spontaneous sequences and power spectral analysis of RR intervals and systolic arterial pressure (SAP) were conducted. Spontaneous sequences were defined as three or more consecutive cycles of either systolic blood pressure elevation (up sequences) or fall (down sequences) coupled with RR interval changes in the same direction. The blood pressure change had to be at least 1 mm Hg/heartbeat. We made no limit to RR interval changes. When the sequences were selected, linear regression analysis was performed and only sequences with correlation coefficient >0.8 were accepted for further analysis. The phase shift was automatically set by the computer program according to the Yamamoto – Hughson method (Hughson et al., 1993). For all of these sequences, the slope of the function of delta RRI and delta SAP was calculated and averaged separately for up and down sequences. This calculation was performed only if 5 –5 or more sequences were identified in the 5-min recordings. Power spectral analysis was made for RRI and systolic and diastolic BP fluctuations by using fast Fourier transformation (Hanning windowing, triangular smoothing). Traditional low and high frequency range limits were set at 0.05– 0.15 and 0.15 – 0.5 Hz, respectively, and spectral powers were expressed as the integrated areas in both bands. To assess the association between RRI and SAP powers, crossspectral analysis was performed and data with coherence >0.5 were used for alpha index calculation. The cross spectral alpha index was calculated as the square root of ratios of SAP and RRI powers. Phase angles between RRI and SAP fluctuations were determined separately in both frequency ranges. The breathing was recorded by means of a pneumatic rubber belt (pneumobelt), which was applied around the
Table 2 Before 2
RRI LF (ms ) SAP LF (mm Hg2) DAP LF (mm Hg2) RRI HF (ms2) SAP HF (mm Hg2) DAP HF (mm Hg2)
a
221 4.8 F 3.9 2.1 F 1.5 297a 1.2a 0.7 F 0.8
After a
330 5.8 F 3.9 2.7 F 1.8 172a 1.9a 1.0 F 0.8
P-value 0.025 0.071, NS 0.017 0.063, NS < 0.001 0.013
Spectral powers in LF and HF bands of RRI, SAP and DAP before and after blood donation. a Indicates median values for not normally distributed data.
116
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Fig. 1. Time series (the six upper panels) and power spectral densities (PSD) (the six lower panels) of RRI, SAP and DAP. The spectrums indicate increased LF and increased HF peaks in all parameters (in this particular case including RRI) after blood donation.
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
subjects’ abdomen. This device was connected to a pressure transducer and to an analog-digital converter. The central frequency of this uncalibrated breathing signal (UBS) was assessed by the spectral method. Statistical analysis: The measured parameters before and after blood withdrawal were compared by paired t-test and, when the data were not normally distributed, Wilcoxon signed rank test was used. Data are given as mean F S.D. where not otherwise indicated. P-value was set at < 0.05 to be considered statistically significant.
3. Results The systolic, diastolic (DAP) and mean arterial blood pressure (MAP) increased after the blood withdrawal ( p < 0.001 for SAP and DAP and p = 0.003 for MAP). The mean heart rate and the central frequency of the uncalibrated breathing signal did not change (the latter was 0.26 Hz). From the time domain indices of RR interval RMSSD significantly decreased ( p = 0.027), SDRR and PNN50 did not show any change (Table 1). Systolic blood pressure variability increased in both frequency bands, but it reached statistical significance only in the high frequency band after blood withdrawal (LF: p = 0.071, HF: p < 0.001). The diastolic blood pressure variability increased significantly in both frequency ranges (LF: p = 0.017, HF: p = 0.013). Heart rate variability in average increased in LF and tended to decrease in HF band ( p = 0.025 and p = 0.063, respectively); however, the individual responses showed substantial diversion in this range (Table 2). Fig. 1 shows a typical recording, the RRI, SAP and DAP time series and power spectra of one subject before and after blood donation. From the different baroreflex gain estimates, up sequence BRS and HF alpha index decreased significantly due to the intervention ( p = 0.001), the decrease in down sequence BRS did not reach statistical significance ( p = 0.098), and the LF alpha showed no consistent change at all (Table 3). The phase angle between RRI and systolic blood pressure powers in LF did not change due to the volume loss ( 58 F 24j and 54 F 26j). In the HF range, the phase became more negative after blood donation ( 1 F 29j and 17 F 32j, p = 0.001).
Table 3
upBRS (ms/mm Hg) downBRS (ms/mm Hg) LF alpha (ms/mm Hg) HF alpha (ms/mm Hg)
Before
After
P-value
12.0 F 8.6 10.0 F 6.1 9.2a 19.4 F 14.8
9.6 F 7.2 8.0 F 6.1 9.2a 13.2 F 11.1
0.001 0.098, NS NS 0.001
Cardiac vagal baroreflex gain values derived from up sequences (upBRS), down sequences (downBRS), low frequency alpha index (LF alpha) and high frequency alpha index (HF alpha) before and after blood donation. a Indicates median values for not normally distributed data.
117
4. Discussion The main findings of this study are the following. First, the systolic, diastolic and mean blood pressure increased and the blood pressure variability, as it was expected, increased during volume loss. The increase in systolic blood pressure variability mainly occurred in HF band, where the RRI variability tended to decrease. The increase in SAP variability did not reach statistical significance in LF range, where the RRI variability increased. Second, from the different cardiac vagal baroreflex gain indices, the up sequence BRS and HF alpha index decreased significantly. In the LF band, RRI-SAP phase angle remained fairly constant before and after blood donation at around 50j to 60j. This value is compatible with baroreflex mechanism. Phase angles in the HF range significantly decreased from 2j to 17j after the intervention. These values do not support baroreflex transduction. 4.1. Mean values, heart rate and blood pressure variability The mean heart rate did not change due to blood withdrawal. The increase in systolic, diastolic and mean arterial pressures supports the possibility of sympathetic vasoconstriction due to rapid volume loss, so the decreased cardiac output and stroke volume did not result in hypotension. This finding also suggests that the induced volume changes were great enough to stimulate or alter autonomic responses. It is known that arterial pressure fluctuations increase during hypovolemia, especially during mechanical ventilation (Rooke, 1995; Rooke et al., 1995; Perel et al., 1987; Ornstein et al., 1998). Systolic pressure variation, defined as the difference between the maximum and minimum values during a breathing cycle, can be divided into the so-called delta up, the increase, and delta down, the decrease of pressure. These parameters are affected by several factors besides volume status, the type of respiration (spontaneous or mechanical), the lung and chest wall compliance and the cardiac function. The pressure changes are regarded as the result of changing intrathoracic pressure induced variations in venous return, pulmonary circulation, right and left ventricular afterload, and finally stroke volume (Toska and Eriksen, 1993; Rooke, 1995). Why are blood pressure fluctuations more prominent in hypovolemia? On one hand, it is believed that respirationrelated variations in venous return and stroke volume can be increased in hypovolemia. On the other hand, as Taylor demonstrated, the area of the ascending aorta decreases during non-hypotensive hypovolemia. This study suggests that small reduction in blood volume reduce aortic baroreceptor area and arterial baroreflexes can be activated without detectable arterial pressure alterations (Taylor et al., 1995). Barbieri argues that the decreased arterial capacitance itself can result in larger pressure variation (Barbieri et al., 2002).
118
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Does only ventilation-related blood pressure fluctuation increase in hypovolemia? Is it accompanied also with changed heart rate variability? In a study of simulated gravity, McKenzie found significant increase in systolic and diastolic pressure LF power, and also an even greater significant increase in HF powers. The RRI power showed no consistent change (McKenzie, 1993). Fortrat found increased high frequency blood pressure variability and no alteration of RRI variability during blood donation (Fortrat et al., 1998). During upright tilt, Cooke found that SAP and DAP spectral power increased in LF and in HF, while the RRI LF power was unaffected and the RRI HF power decreased progressively with the tilt angle (Cooke et al., 1999). In our study, the HF blood pressure variability increased significantly, although LF SAP fluctuations also tended to increase. The RRI power changes were different in LF and HF. While increased HF fluctuations of blood pressure can be the manifestation of greater mechanical effect of greater stroke volume variation, the LF fluctuations can reflect increased sympathetic activity. The inconsistency of the data in literature regarding RRI LF power requires further explanation. It is possible, that the trigger of baroreflex (blood pressure fluctuation) increased, but the gain in heart rate decreased at the same time, and the net effect depends on the magnitude of these opposite influences. In our study, this resulted in increased RRI variation in the low frequency band. The decrease in high frequency power of RR intervals, i.e. the decreased respiratory sinus arrhythmia, can be the effect of sympathetic activation and concomitant vagal withdrawal. So the increased SAP and DAP respiratory fluctuations also can result from the attenuated buffering effect of heart rate (Cooke et al., 1999). 4.2. Cardiac vagal baroreflex gain Data of the influence of volume loading and unloading on human cardiac vagal baroreflex gain are conflicting. The central volume changes can be induced not only by direct blood withdrawal or volume loading, but by different posture, head-up tilt position, by lower body negative pressure or by simulated gravity, so we can parallel our result with such investigations. In an early study Takeshita provoked central venous pressure changes by lower body negative pressure (LBNP) and by leg and lower trunk elevation. The arterial baroreflex gain measured by the phenylephrine and neck suction methods was not changed (Takeshita et al., 1979). Similarly, Eiken found that LBNP induced reduction of central venous pressure did not influence baroreflex control of heart rate, assessed by neck chamber technique over the entire arterial pressure RRI relation (Eiken et al., 1994). In a similar study to ours, Fortrat found that the spontaneous cardiac vagal baroreflex gain did not change after 480-ml blood donation. They concluded that the slight sympathetic activation due to blood withdrawal elicited rapid resetting of the baroreflex (Fortrat et al., 1998). Steptoe and Vo¨gele observed in their
study of postural change from sitting to standing the reduction of the slope of spontaneous sequences (Steptoe and Vo¨gele, 1990). In another study, the spontaneous cardiac baroreflex gain decreased after active standing and head-up tilt position, while the number of sequences increased signing greater baroreflex engagement (BahjaouiBouhaddi et al., 1998). Cooke also demonstrated that baroreflex gain decreased significantly at higher degrees of head-up tilt position (Cooke et al., 1999). In a recent trial, Barbieri concluded that short-term cardiovascular control of heart rate appears to be optimized at mild hypervolemia. They manipulated central blood volume by LBNP and by leg elevation and volume loading. The SAP-RRI baroreflex gain decreased with volume unloading (Barbieri et al., 2002). Why do cardiac vagal baroreflex gain decrease with preload reduction? There are different theoretical solutions for this problem. There can be a cardiopulmonary arterial baroreflex interaction. According to this hypothesis, the unloading of the cardiopulmonary receptors results in sympathetic activation. This sympathetic activation can modify centrally and peripherally the vagal responses. However, in contrast to animals, it is difficult to prove cardiopulmonaryarterial baroreflex interactions in humans, because the influence of the aortic baroreceptors cannot be excluded. As mentioned above, aortic baroreceptors can be activated by alterations in aortic blood volume causing changes in aortic baroreceptor area, even without measurable changes in blood pressure (Taylor et al., 1995). According to Cooke, there is also the possibility that the operational range shifts to the threshold region of the arterial pressure RR interval relationship during central hypovolemia (Cooke et al., 1999). From our results, we cannot answer the question that, if the cardiac vagal baroreflex gain decreases with volume unloading, why was it not reflected by all baroreflex estimates. There are data that the responses to increasing and decreasing blood pressure are not the same (Pickering et al., 1972). This can explain that, though the direction was the same, the magnitude of changes of up sequence and down sequence BRS was not identical. The results of spectral baroreflex gain are more difficult to explain. The phase angles can give further insight into the problem. The phase shift we found in low frequency band is almost exactly the same, as what Cooke found (Cooke et al., 1999). It was in our study: 58 F 24j before and 54 F 26j after blood withdrawal. The phase angle between two waves at a given frequency could be translated into time delays. Baroreflex-mediated time delays are determined by response times, conduction velocities and neurotransmitter kinetics, and therefore are fairly stable in different test conditions. These angles strongly support baroreflex relationship between SAP and RR interval in LF (Cooke et al., 1999; Cevese et al., 2001). The relationship of blood pressure and RR interval fluctuations in the high frequency band could be assessed
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
by taking several factors into consideration. The presence of a spectral peak in both signals with identical central frequency may imply a cause – effect relationship. Such a relationship seems to be indicated by the high level of coherence between signals. The intertwined peaks nevertheless do exist following baroreflex denervation and the level of coherence remains high (Di Rienzo et al., 1997). In our study, the phase angles in HF were variable, unstable and basically incompatible with arterial baroreflex time delays. According to the phase angles, SAP and RRI changes happened nearly at the same time, suggesting that the spectral peaks should therefore be related to a third factor, which is respiration itself. Non-baroreflex mechanisms, such as pulmonary stretch receptor reflexes, could be operational (Taha et al., 1995). After all, it was surprising that HF alpha index behaved as it is expected from a baroreflex gain, it decreased. A recent study of Barbieri demonstrated that preload manipulations in humans bring on parallel modulations in baroreflex gain and respiratory sinus arrhythmia (Barbieri et al., 2002). This parallelism sufficiently explains the observed phenomenon. In the opposite, LF alpha index, which is regarded as a measure of baroreflex gain, even increased in 17 subjects, resulting in overall not significant change in average. 4.3. Limitations France found in his study of blood donors that those who had previous presyncope – syncope exhibited higher up sequence BRS to painful stimuli than those who had not. Unfortunately, these investigations were not carried out during blood donation (France, 1995). In our study, we calculated the baroreflex gain for the whole group; we did not separate the results of the individuals who had some kind of vasovagal reactions. However, these reactions were rare and no full blown syncope was observed. We did not measure central venous pressure, stroke volume or intrathoracic blood volume, but in previous studies this quantity of blood withdrawal was found great enough to induce central blood volume changes. The observed elevation of blood pressure also supports this assumption. Finally, and most importantly, we did not measure muscle sympathetic nerve activity, so we can just speculate on sympathetic responses and on their effects on other parameters. In summary, in our study, the withdrawal of 350 –400 ml blood in 5 min resulted in sympathetic activation, which was reflected in increased systolic, diastolic and mean arterial pressure and increased LF BP variability. This sympathetic response also resulted in vagal withdrawal, as was shown by decreased RMSSD, decreased HF RRI variability. Increased HF blood pressure variability could have been also related to decreased heart rate buffering. The decreased vagal responsiveness resulted in smaller up sequence baroreflex gain. The phase angles in
119
high frequency range between SAP and RRI do not support baroreflex mechanism, so the high frequency alpha index is not recommended to use for baroreflex gain estimation. References Bahjaoui-Bouhaddi, M., Henriet, M.T., Cappelle, S., Dumoulin, G., Regnard, J., 1998. Active standing and passive tilting similarly reduce the slope of spontaneous baroreflex in healthy subjects. Physiol. Res. 47, 227 – 235. Barbieri, R., Triedman, J.K., Saul, J.P., 2002. Heart rate control and mechanical cardiopulmonary coupling to assess central volume: a systems analysis. Am. J. Physiol. 283, R1210 – R1220. Cevese, A., Gulli, G., Polati, E., Gottin, L., Grasso, R., 2001. Baroreflex and oscillation of heart period at 0.1 Hz studied by alpha-blockade and cross-spectral analysis in healthy humans. J. Physiol. 531, 235 – 244. Cooke, W.H., Hoag, J.B., Crossman, A., Kuusela, T.A., Tahvanainen, K.U.O., Eckberg, D.L., 1999. Human responses to upright tilt: a window on central autonomic integration. J. Physiol. 517, 617 – 628. Di Rienzo, M., Castiglioni, P., Mancia, G., Parati, G., Pedotti, A., 1997. Critical appraisal of indices for the assessment of baroreflex sensitivity. Methods Inf. Med. 36, 246 – 249. Eiken, O., Sun, J.C.L., Mekjavic, I.B., 1994. Effects of blood-volume distribution on the characteristics of the carotid baroreflex in humans at rest and during exercise. Acta Physiol. Scand. 150, 89 – 94. Fortrat, J.O., Nasr, O., Duvareille, M., Gharib, C., 1998. Human cardiovascular variability, baroreflex and hormonal adaptations to a blood donation. Clin. Sci. 95, 269 – 275. France, C., 1995. Baroreflex sensitivity during noxious stimulation in vasovagal reactors to blood donation. Int. J. Psychophysiol. 19, 13 – 22. Hughson, R.L., Quintin, L., Annat, G., Yamamoto, Y., Gharib, C., 1993. Spontaneous baroreflex by sequence and power spectral methods in humans. Clin. Physiol. 13, 663 – 676. James, M.A., Panerai, R.B., Potter, J.F., 1998. Applicability of new techniques in the assessment of arterial baroreflex sensitivity in the elderly: a comparison with established pharmacological methods. Clin. Sci. 94, 245 – 253. McKenzie, I., 1993. Heart rate and blood pressure variability in subjects exposed to simulated increases in gravity. Exp. Physiol. 78, 825 – 834. Ornstein, E., Eidelman, L.A.E., Drenger, B., Elami, A., Pizov, R., 1998. Systolic pressure variation predicts the response to acute blood loss. J. Clin. Anesth. 10, 137 – 140. Parlow, J., Viale, J.-P., Annat, G., Hughson, R., Quintin, L., 1995. Spontaneous cardiac baroreflex in humans. Comparison with drug-induced responses. Hypertension 25, 1058 – 1068. Perel, A., Pizov, R., Cotev, S., 1987. Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded haemorrhage. Anesthesiology 67, 498 – 502. Pickering, T.G., Gribbin, B., Sleight, P., 1972. Comparison of the reflex heart rate response to rising and falling arterial pressure in man. Cardiovasc. Res. 6, 277 – 283. Pitzalis, M.V., Mastropasqua, F., Passantino, A., et al., 1998. Comparison between noninvasive indices of baroreceptor sensitivity and the phenylephrine method in post-myocardial infarction patients. Circulation 97, 1362 – 1367. Rooke, G.A., 1995. Systolic pressure variation as an indicator of hypovolemia. Curr. Opin. Anaesth. 8, 511 – 515. Rooke, G.A., Schwid, H.A., Shapira, Y., 1995. The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation. Anesth. Analg. 80, 925 – 932. Rudas, L., Crossman, A., Morillo, C.A., et al., 1999. Human sympathetic and vagal baroreflex responses to sequential nitroprusside and phenylephrine. Am. J. Physiol. 276, H1691 – H1698.
120
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Steptoe, A., Vo¨gele, C., 1990. Cardiac baroreflex function during postural change assessed using non-invasive spontaneous sequence analysis in young man. Cardiovasc. Res. 24, 627 – 632. Taha, B.H., Simon, P.M., Dempsey, J.A., Skatrud, J.B., Iber, C., 1995. Respiratory sinus arrhythmia in humans: an obligatory role for vagal feedback from the lungs. J. Appl. Physiol. 78, 638 – 645. Takeshita, A., Mark, A.L., Eckberg, D.L., Abboud, F.M., 1979. Effect of central venous pressure on arterial baroreflex control of heart rate. Am. J. Physiol. 236, H42 – H47.
Taylor, J.A., Halliwill, J.R., Brown, T.E., Hayano, J., Eckberg, D.L., 1995. Non-hypotensive hypovolaemia reduces ascending aortic dimensions in humans. J. Physiol. 483, 289 – 298. Toska, K., Eriksen, M., 1993. Respiration—synchronous fluctuations in stroke volume, heart rate and arterial pressure in humans. J. Physiol. 472, 501 – 512. Watkins, L.L., Fainman, C., Dimsdale, J., Ziegler, M.G., 1995. Assessment of baroreflex control from beat-to-beat blood pressure and heart rate changes: a validation study. Psychophysiology 32, 411 – 414.
Human autonomic responses to blood donation E´va Zo¨llei a,*, Do´ra Paprika a, Pe´ter Makra b, Zolta´n Gingl b, Kla´ra Vezendi c, La´szlo´ Rudas a a
University of Szeged, Faculty of Medicine, Cardiology Center, Medical Intensive Care Unit, Kora´nyi fasor 7, Szeged 6720, Hungary b University of Szeged, Department of Experimental Physics, Hungary c University of Szeged, Department of Blood Banking, Hungary Received 12 June 2003; received in revised form 27 October 2003; accepted 29 October 2003
Abstract In order to characterize autonomic responses to acute volume loss, supine ECG, blood pressure (BP) and uncalibrated breathing signal (UBS) recordings were taken before and after blood donation in 48 healthy volunteers. Time and frequency domain parameters of RR interval (RRI), BP and UBS variability were determined. Baroreflex gain was calculated by the technique of the spontaneous sequences and crossspectral analysis. The systolic (SAP), diastolic (DAP) and mean BP (MAP) increased after the blood withdrawal. The central frequency of breathing and mean heart rate did not change. RRI variability increased in low frequency band (LF), tended to decrease in high frequency band (HF). Systolic BP variability increased in both frequency bands, but was statistically significant only in the high frequency band. Diastolic BP power increased in both frequencies. From the different baroreflex gain estimates, up sequence BRS and HF alpha index decreased significantly. The phase angle between RRI and systolic blood pressure powers in LF band did not change ( 58 F 24j and 54 F 26j). In the high frequency range, the phase became more negative ( 1 F 29j and 17 F 32j, p = 0.001). The withdrawal of 350 – 400 ml blood in 5 min resulted in sympathetic activation, which was reflected in increased systolic, diastolic and mean BP. The increased BP oscillation was a sensitive marker of the minor volume depletion. This was coupled by increased RRI oscillation via baroreflex mechanisms in the LF band. Changes in the RRI and BP oscillations in the HF band showed no similar coupling. That points to the fact that RRI oscillations in this band should not be explained entirely by baroreflex mechanisms. Vagal withdrawal was reflected in decreased root mean square of successive differences (RMSSD), decreased HF RRI power and decreased up sequence BRS. D 2004 Elsevier B.V. All rights reserved. Keywords: Autonomic nervous system; Baroreflexes; Hypovolemia
1. Introduction Arterial and cardiopulmonary baroreflexes play an essential role in maintaining blood pressure (BP) in acute hypovolemic states. Data from animal studies of volume loading and unloading suggest that cardiopulmonary baroreceptor input can modify arterial baroreflex responses. However, in humans, it is not clear what, if any role play the cardiopulmonary baroreflexes in modulating arterial baroreflex control of the circulation. Regarding the baroreflex control of heart rate, there are opposing data during central volume changes induced by direct volume manipulation or gravitation induced volume shifts.
* Corresponding author. Tel./fax: +36-62-545-689. E-mail address: [email protected] (E´. Zo¨llei). 1566-0702/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2003.10.003
It is well-known that arterial pressure variability increases with hypovolemia. The changes of systolic blood pressure during the respiratory cycle can be described by the systolic pressure variation (SPV) and by its components, the so-called delta up and delta down. These indices are regarded as sensitive indicators of volemic state (Perel et al., 1987; Rooke, 1995; Rooke et al., 1995). Blood pressure fluctuations also can be characterized by frequency domain indices. Using power spectral analysis not just the magnitude, but the frequency distribution of systolic, diastolic and mean blood pressure fluctuations can be assessed. By cross spectral method, the feedback relationship of blood pressure and heart rate modulation can also be investigated. We chose the setting of blood donation to assess autonomic responses to acute volume loss. We characterized autonomic responses by calculating time and frequency domain indices of heart rate and blood pressure variability. To calculate baroreflex gain, we used two non-invasive
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120 Table 1
SAP (mm Hg) DAP (mm Hg) MAP (mm Hg) RRI (ms) SDRR (ms) RMSSD PNN50
Before
After
P-value
124 F 16 74 F 13 91 F 13 816 F 142 41 F 24 33 F 25 3.7 F 0.1
128 F 13 76 F 12 93 F 12 806 F 138 40 F 22 28 F 26 1.0 F 0.2
< 0.001 < 0.001 0.003 NS NS 0.027 NS
The SAP, DAP and MAP, the mean RRI, SDRR, RMSSD and PNN50 before and after blood donation.
methods, the spontaneous sequence and the cross-spectral analysis. These techniques are now widely used and give comparable results to the classical pharmacological and neck-chamber methods, and are regarded more physiological estimates of baroreflex function (Parlow et al., 1995; Watkins et al., 1995; James et al., 1998; Pitzalis et al., 1998; Rudas et al., 1999). The aim of this study was to compare the time and frequency domain parameters of heart rate and blood pressure variability and the baroreflex control of heart rate before and after 350 –400 ml blood was withdrawn from healthy donors, and to get further insight into the short term compensatory mechanisms to acute volume loss.
2. Materials and methods The study population consisted of 48 healthy volunteers (age 35 F 12, 18 –59 years; 23 men, 25 women), who were recruited when they signed up for blood donation. Three of them were first time, the others were regular blood donors. None of the subjects had a previous history of syncope or presyncope. All the subjects were informed and gave their consent for their participation in the study. The study protocol was approved by the Ethical Committee of the University of Szeged. The investigations were carried out according to the Declaration of Helsinki. All the measurements were made in a quiet room at the Blood Bank after 5 min time interval of supine rest. First, 5 min baseline recordings were taken. This was followed by the venipuncture and by the withdrawal of 350– 400 ml blood in approximately 5 min. After the removal of the needle, we started at once with the second recordings. The subjects were not allowed to drink, nor received their caffeinated beverages throughout the investigations. Two of them experienced mild symptoms suggesting vasovagal reaction; they were put to head-down tilt position at the end of the blood withdrawal, according to the protocol of the Blood Bank. By the time of the second recording, they were put back supine. The ECG and blood pressure signals were continuously measured with a Marquette bedside monitor and with the Finapres 2300 non-invasive blood pressure monitor. Signals were recorded and on-line digitalized with 500 Hz by the
115
Dataq/Windaq system. The data were analyzed by using the WinCPRS software (Absolute Aliens Ay, 2000). All recordings were peak detected and checked by one of the investigators. After the confirmation of peak detection, conventional time domain indices of heart rate variability were determined (SDRR: standard deviation of RR intervals, RMSSD: root mean square of successive differences, PNN50: the percentage of cardiac cycles, where the difference was greater than 50 ms between the successive beats), and search for spontaneous sequences and power spectral analysis of RR intervals and systolic arterial pressure (SAP) were conducted. Spontaneous sequences were defined as three or more consecutive cycles of either systolic blood pressure elevation (up sequences) or fall (down sequences) coupled with RR interval changes in the same direction. The blood pressure change had to be at least 1 mm Hg/heartbeat. We made no limit to RR interval changes. When the sequences were selected, linear regression analysis was performed and only sequences with correlation coefficient >0.8 were accepted for further analysis. The phase shift was automatically set by the computer program according to the Yamamoto – Hughson method (Hughson et al., 1993). For all of these sequences, the slope of the function of delta RRI and delta SAP was calculated and averaged separately for up and down sequences. This calculation was performed only if 5 –5 or more sequences were identified in the 5-min recordings. Power spectral analysis was made for RRI and systolic and diastolic BP fluctuations by using fast Fourier transformation (Hanning windowing, triangular smoothing). Traditional low and high frequency range limits were set at 0.05– 0.15 and 0.15 – 0.5 Hz, respectively, and spectral powers were expressed as the integrated areas in both bands. To assess the association between RRI and SAP powers, crossspectral analysis was performed and data with coherence >0.5 were used for alpha index calculation. The cross spectral alpha index was calculated as the square root of ratios of SAP and RRI powers. Phase angles between RRI and SAP fluctuations were determined separately in both frequency ranges. The breathing was recorded by means of a pneumatic rubber belt (pneumobelt), which was applied around the
Table 2 Before 2
RRI LF (ms ) SAP LF (mm Hg2) DAP LF (mm Hg2) RRI HF (ms2) SAP HF (mm Hg2) DAP HF (mm Hg2)
a
221 4.8 F 3.9 2.1 F 1.5 297a 1.2a 0.7 F 0.8
After a
330 5.8 F 3.9 2.7 F 1.8 172a 1.9a 1.0 F 0.8
P-value 0.025 0.071, NS 0.017 0.063, NS < 0.001 0.013
Spectral powers in LF and HF bands of RRI, SAP and DAP before and after blood donation. a Indicates median values for not normally distributed data.
116
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Fig. 1. Time series (the six upper panels) and power spectral densities (PSD) (the six lower panels) of RRI, SAP and DAP. The spectrums indicate increased LF and increased HF peaks in all parameters (in this particular case including RRI) after blood donation.
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
subjects’ abdomen. This device was connected to a pressure transducer and to an analog-digital converter. The central frequency of this uncalibrated breathing signal (UBS) was assessed by the spectral method. Statistical analysis: The measured parameters before and after blood withdrawal were compared by paired t-test and, when the data were not normally distributed, Wilcoxon signed rank test was used. Data are given as mean F S.D. where not otherwise indicated. P-value was set at < 0.05 to be considered statistically significant.
3. Results The systolic, diastolic (DAP) and mean arterial blood pressure (MAP) increased after the blood withdrawal ( p < 0.001 for SAP and DAP and p = 0.003 for MAP). The mean heart rate and the central frequency of the uncalibrated breathing signal did not change (the latter was 0.26 Hz). From the time domain indices of RR interval RMSSD significantly decreased ( p = 0.027), SDRR and PNN50 did not show any change (Table 1). Systolic blood pressure variability increased in both frequency bands, but it reached statistical significance only in the high frequency band after blood withdrawal (LF: p = 0.071, HF: p < 0.001). The diastolic blood pressure variability increased significantly in both frequency ranges (LF: p = 0.017, HF: p = 0.013). Heart rate variability in average increased in LF and tended to decrease in HF band ( p = 0.025 and p = 0.063, respectively); however, the individual responses showed substantial diversion in this range (Table 2). Fig. 1 shows a typical recording, the RRI, SAP and DAP time series and power spectra of one subject before and after blood donation. From the different baroreflex gain estimates, up sequence BRS and HF alpha index decreased significantly due to the intervention ( p = 0.001), the decrease in down sequence BRS did not reach statistical significance ( p = 0.098), and the LF alpha showed no consistent change at all (Table 3). The phase angle between RRI and systolic blood pressure powers in LF did not change due to the volume loss ( 58 F 24j and 54 F 26j). In the HF range, the phase became more negative after blood donation ( 1 F 29j and 17 F 32j, p = 0.001).
Table 3
upBRS (ms/mm Hg) downBRS (ms/mm Hg) LF alpha (ms/mm Hg) HF alpha (ms/mm Hg)
Before
After
P-value
12.0 F 8.6 10.0 F 6.1 9.2a 19.4 F 14.8
9.6 F 7.2 8.0 F 6.1 9.2a 13.2 F 11.1
0.001 0.098, NS NS 0.001
Cardiac vagal baroreflex gain values derived from up sequences (upBRS), down sequences (downBRS), low frequency alpha index (LF alpha) and high frequency alpha index (HF alpha) before and after blood donation. a Indicates median values for not normally distributed data.
117
4. Discussion The main findings of this study are the following. First, the systolic, diastolic and mean blood pressure increased and the blood pressure variability, as it was expected, increased during volume loss. The increase in systolic blood pressure variability mainly occurred in HF band, where the RRI variability tended to decrease. The increase in SAP variability did not reach statistical significance in LF range, where the RRI variability increased. Second, from the different cardiac vagal baroreflex gain indices, the up sequence BRS and HF alpha index decreased significantly. In the LF band, RRI-SAP phase angle remained fairly constant before and after blood donation at around 50j to 60j. This value is compatible with baroreflex mechanism. Phase angles in the HF range significantly decreased from 2j to 17j after the intervention. These values do not support baroreflex transduction. 4.1. Mean values, heart rate and blood pressure variability The mean heart rate did not change due to blood withdrawal. The increase in systolic, diastolic and mean arterial pressures supports the possibility of sympathetic vasoconstriction due to rapid volume loss, so the decreased cardiac output and stroke volume did not result in hypotension. This finding also suggests that the induced volume changes were great enough to stimulate or alter autonomic responses. It is known that arterial pressure fluctuations increase during hypovolemia, especially during mechanical ventilation (Rooke, 1995; Rooke et al., 1995; Perel et al., 1987; Ornstein et al., 1998). Systolic pressure variation, defined as the difference between the maximum and minimum values during a breathing cycle, can be divided into the so-called delta up, the increase, and delta down, the decrease of pressure. These parameters are affected by several factors besides volume status, the type of respiration (spontaneous or mechanical), the lung and chest wall compliance and the cardiac function. The pressure changes are regarded as the result of changing intrathoracic pressure induced variations in venous return, pulmonary circulation, right and left ventricular afterload, and finally stroke volume (Toska and Eriksen, 1993; Rooke, 1995). Why are blood pressure fluctuations more prominent in hypovolemia? On one hand, it is believed that respirationrelated variations in venous return and stroke volume can be increased in hypovolemia. On the other hand, as Taylor demonstrated, the area of the ascending aorta decreases during non-hypotensive hypovolemia. This study suggests that small reduction in blood volume reduce aortic baroreceptor area and arterial baroreflexes can be activated without detectable arterial pressure alterations (Taylor et al., 1995). Barbieri argues that the decreased arterial capacitance itself can result in larger pressure variation (Barbieri et al., 2002).
118
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Does only ventilation-related blood pressure fluctuation increase in hypovolemia? Is it accompanied also with changed heart rate variability? In a study of simulated gravity, McKenzie found significant increase in systolic and diastolic pressure LF power, and also an even greater significant increase in HF powers. The RRI power showed no consistent change (McKenzie, 1993). Fortrat found increased high frequency blood pressure variability and no alteration of RRI variability during blood donation (Fortrat et al., 1998). During upright tilt, Cooke found that SAP and DAP spectral power increased in LF and in HF, while the RRI LF power was unaffected and the RRI HF power decreased progressively with the tilt angle (Cooke et al., 1999). In our study, the HF blood pressure variability increased significantly, although LF SAP fluctuations also tended to increase. The RRI power changes were different in LF and HF. While increased HF fluctuations of blood pressure can be the manifestation of greater mechanical effect of greater stroke volume variation, the LF fluctuations can reflect increased sympathetic activity. The inconsistency of the data in literature regarding RRI LF power requires further explanation. It is possible, that the trigger of baroreflex (blood pressure fluctuation) increased, but the gain in heart rate decreased at the same time, and the net effect depends on the magnitude of these opposite influences. In our study, this resulted in increased RRI variation in the low frequency band. The decrease in high frequency power of RR intervals, i.e. the decreased respiratory sinus arrhythmia, can be the effect of sympathetic activation and concomitant vagal withdrawal. So the increased SAP and DAP respiratory fluctuations also can result from the attenuated buffering effect of heart rate (Cooke et al., 1999). 4.2. Cardiac vagal baroreflex gain Data of the influence of volume loading and unloading on human cardiac vagal baroreflex gain are conflicting. The central volume changes can be induced not only by direct blood withdrawal or volume loading, but by different posture, head-up tilt position, by lower body negative pressure or by simulated gravity, so we can parallel our result with such investigations. In an early study Takeshita provoked central venous pressure changes by lower body negative pressure (LBNP) and by leg and lower trunk elevation. The arterial baroreflex gain measured by the phenylephrine and neck suction methods was not changed (Takeshita et al., 1979). Similarly, Eiken found that LBNP induced reduction of central venous pressure did not influence baroreflex control of heart rate, assessed by neck chamber technique over the entire arterial pressure RRI relation (Eiken et al., 1994). In a similar study to ours, Fortrat found that the spontaneous cardiac vagal baroreflex gain did not change after 480-ml blood donation. They concluded that the slight sympathetic activation due to blood withdrawal elicited rapid resetting of the baroreflex (Fortrat et al., 1998). Steptoe and Vo¨gele observed in their
study of postural change from sitting to standing the reduction of the slope of spontaneous sequences (Steptoe and Vo¨gele, 1990). In another study, the spontaneous cardiac baroreflex gain decreased after active standing and head-up tilt position, while the number of sequences increased signing greater baroreflex engagement (BahjaouiBouhaddi et al., 1998). Cooke also demonstrated that baroreflex gain decreased significantly at higher degrees of head-up tilt position (Cooke et al., 1999). In a recent trial, Barbieri concluded that short-term cardiovascular control of heart rate appears to be optimized at mild hypervolemia. They manipulated central blood volume by LBNP and by leg elevation and volume loading. The SAP-RRI baroreflex gain decreased with volume unloading (Barbieri et al., 2002). Why do cardiac vagal baroreflex gain decrease with preload reduction? There are different theoretical solutions for this problem. There can be a cardiopulmonary arterial baroreflex interaction. According to this hypothesis, the unloading of the cardiopulmonary receptors results in sympathetic activation. This sympathetic activation can modify centrally and peripherally the vagal responses. However, in contrast to animals, it is difficult to prove cardiopulmonaryarterial baroreflex interactions in humans, because the influence of the aortic baroreceptors cannot be excluded. As mentioned above, aortic baroreceptors can be activated by alterations in aortic blood volume causing changes in aortic baroreceptor area, even without measurable changes in blood pressure (Taylor et al., 1995). According to Cooke, there is also the possibility that the operational range shifts to the threshold region of the arterial pressure RR interval relationship during central hypovolemia (Cooke et al., 1999). From our results, we cannot answer the question that, if the cardiac vagal baroreflex gain decreases with volume unloading, why was it not reflected by all baroreflex estimates. There are data that the responses to increasing and decreasing blood pressure are not the same (Pickering et al., 1972). This can explain that, though the direction was the same, the magnitude of changes of up sequence and down sequence BRS was not identical. The results of spectral baroreflex gain are more difficult to explain. The phase angles can give further insight into the problem. The phase shift we found in low frequency band is almost exactly the same, as what Cooke found (Cooke et al., 1999). It was in our study: 58 F 24j before and 54 F 26j after blood withdrawal. The phase angle between two waves at a given frequency could be translated into time delays. Baroreflex-mediated time delays are determined by response times, conduction velocities and neurotransmitter kinetics, and therefore are fairly stable in different test conditions. These angles strongly support baroreflex relationship between SAP and RR interval in LF (Cooke et al., 1999; Cevese et al., 2001). The relationship of blood pressure and RR interval fluctuations in the high frequency band could be assessed
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
by taking several factors into consideration. The presence of a spectral peak in both signals with identical central frequency may imply a cause – effect relationship. Such a relationship seems to be indicated by the high level of coherence between signals. The intertwined peaks nevertheless do exist following baroreflex denervation and the level of coherence remains high (Di Rienzo et al., 1997). In our study, the phase angles in HF were variable, unstable and basically incompatible with arterial baroreflex time delays. According to the phase angles, SAP and RRI changes happened nearly at the same time, suggesting that the spectral peaks should therefore be related to a third factor, which is respiration itself. Non-baroreflex mechanisms, such as pulmonary stretch receptor reflexes, could be operational (Taha et al., 1995). After all, it was surprising that HF alpha index behaved as it is expected from a baroreflex gain, it decreased. A recent study of Barbieri demonstrated that preload manipulations in humans bring on parallel modulations in baroreflex gain and respiratory sinus arrhythmia (Barbieri et al., 2002). This parallelism sufficiently explains the observed phenomenon. In the opposite, LF alpha index, which is regarded as a measure of baroreflex gain, even increased in 17 subjects, resulting in overall not significant change in average. 4.3. Limitations France found in his study of blood donors that those who had previous presyncope – syncope exhibited higher up sequence BRS to painful stimuli than those who had not. Unfortunately, these investigations were not carried out during blood donation (France, 1995). In our study, we calculated the baroreflex gain for the whole group; we did not separate the results of the individuals who had some kind of vasovagal reactions. However, these reactions were rare and no full blown syncope was observed. We did not measure central venous pressure, stroke volume or intrathoracic blood volume, but in previous studies this quantity of blood withdrawal was found great enough to induce central blood volume changes. The observed elevation of blood pressure also supports this assumption. Finally, and most importantly, we did not measure muscle sympathetic nerve activity, so we can just speculate on sympathetic responses and on their effects on other parameters. In summary, in our study, the withdrawal of 350 –400 ml blood in 5 min resulted in sympathetic activation, which was reflected in increased systolic, diastolic and mean arterial pressure and increased LF BP variability. This sympathetic response also resulted in vagal withdrawal, as was shown by decreased RMSSD, decreased HF RRI variability. Increased HF blood pressure variability could have been also related to decreased heart rate buffering. The decreased vagal responsiveness resulted in smaller up sequence baroreflex gain. The phase angles in
119
high frequency range between SAP and RRI do not support baroreflex mechanism, so the high frequency alpha index is not recommended to use for baroreflex gain estimation. References Bahjaoui-Bouhaddi, M., Henriet, M.T., Cappelle, S., Dumoulin, G., Regnard, J., 1998. Active standing and passive tilting similarly reduce the slope of spontaneous baroreflex in healthy subjects. Physiol. Res. 47, 227 – 235. Barbieri, R., Triedman, J.K., Saul, J.P., 2002. Heart rate control and mechanical cardiopulmonary coupling to assess central volume: a systems analysis. Am. J. Physiol. 283, R1210 – R1220. Cevese, A., Gulli, G., Polati, E., Gottin, L., Grasso, R., 2001. Baroreflex and oscillation of heart period at 0.1 Hz studied by alpha-blockade and cross-spectral analysis in healthy humans. J. Physiol. 531, 235 – 244. Cooke, W.H., Hoag, J.B., Crossman, A., Kuusela, T.A., Tahvanainen, K.U.O., Eckberg, D.L., 1999. Human responses to upright tilt: a window on central autonomic integration. J. Physiol. 517, 617 – 628. Di Rienzo, M., Castiglioni, P., Mancia, G., Parati, G., Pedotti, A., 1997. Critical appraisal of indices for the assessment of baroreflex sensitivity. Methods Inf. Med. 36, 246 – 249. Eiken, O., Sun, J.C.L., Mekjavic, I.B., 1994. Effects of blood-volume distribution on the characteristics of the carotid baroreflex in humans at rest and during exercise. Acta Physiol. Scand. 150, 89 – 94. Fortrat, J.O., Nasr, O., Duvareille, M., Gharib, C., 1998. Human cardiovascular variability, baroreflex and hormonal adaptations to a blood donation. Clin. Sci. 95, 269 – 275. France, C., 1995. Baroreflex sensitivity during noxious stimulation in vasovagal reactors to blood donation. Int. J. Psychophysiol. 19, 13 – 22. Hughson, R.L., Quintin, L., Annat, G., Yamamoto, Y., Gharib, C., 1993. Spontaneous baroreflex by sequence and power spectral methods in humans. Clin. Physiol. 13, 663 – 676. James, M.A., Panerai, R.B., Potter, J.F., 1998. Applicability of new techniques in the assessment of arterial baroreflex sensitivity in the elderly: a comparison with established pharmacological methods. Clin. Sci. 94, 245 – 253. McKenzie, I., 1993. Heart rate and blood pressure variability in subjects exposed to simulated increases in gravity. Exp. Physiol. 78, 825 – 834. Ornstein, E., Eidelman, L.A.E., Drenger, B., Elami, A., Pizov, R., 1998. Systolic pressure variation predicts the response to acute blood loss. J. Clin. Anesth. 10, 137 – 140. Parlow, J., Viale, J.-P., Annat, G., Hughson, R., Quintin, L., 1995. Spontaneous cardiac baroreflex in humans. Comparison with drug-induced responses. Hypertension 25, 1058 – 1068. Perel, A., Pizov, R., Cotev, S., 1987. Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded haemorrhage. Anesthesiology 67, 498 – 502. Pickering, T.G., Gribbin, B., Sleight, P., 1972. Comparison of the reflex heart rate response to rising and falling arterial pressure in man. Cardiovasc. Res. 6, 277 – 283. Pitzalis, M.V., Mastropasqua, F., Passantino, A., et al., 1998. Comparison between noninvasive indices of baroreceptor sensitivity and the phenylephrine method in post-myocardial infarction patients. Circulation 97, 1362 – 1367. Rooke, G.A., 1995. Systolic pressure variation as an indicator of hypovolemia. Curr. Opin. Anaesth. 8, 511 – 515. Rooke, G.A., Schwid, H.A., Shapira, Y., 1995. The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation. Anesth. Analg. 80, 925 – 932. Rudas, L., Crossman, A., Morillo, C.A., et al., 1999. Human sympathetic and vagal baroreflex responses to sequential nitroprusside and phenylephrine. Am. J. Physiol. 276, H1691 – H1698.
120
E´. Zo¨llei et al. / Autonomic Neuroscience: Basic and Clinical 110 (2004) 114–120
Steptoe, A., Vo¨gele, C., 1990. Cardiac baroreflex function during postural change assessed using non-invasive spontaneous sequence analysis in young man. Cardiovasc. Res. 24, 627 – 632. Taha, B.H., Simon, P.M., Dempsey, J.A., Skatrud, J.B., Iber, C., 1995. Respiratory sinus arrhythmia in humans: an obligatory role for vagal feedback from the lungs. J. Appl. Physiol. 78, 638 – 645. Takeshita, A., Mark, A.L., Eckberg, D.L., Abboud, F.M., 1979. Effect of central venous pressure on arterial baroreflex control of heart rate. Am. J. Physiol. 236, H42 – H47.
Taylor, J.A., Halliwill, J.R., Brown, T.E., Hayano, J., Eckberg, D.L., 1995. Non-hypotensive hypovolaemia reduces ascending aortic dimensions in humans. J. Physiol. 483, 289 – 298. Toska, K., Eriksen, M., 1993. Respiration—synchronous fluctuations in stroke volume, heart rate and arterial pressure in humans. J. Physiol. 472, 501 – 512. Watkins, L.L., Fainman, C., Dimsdale, J., Ziegler, M.G., 1995. Assessment of baroreflex control from beat-to-beat blood pressure and heart rate changes: a validation study. Psychophysiology 32, 411 – 414.