Desflurane stimulates the sympathetic nervous system in humans, which has not been observed with other volatile anesthetics (1–4). However, whereas heart rate increases with desflurane, arterial blood pressure decreases to similar levels as with isoflurane or sevoflurane during steady-state conditions (1,3,5). This specific effect of desflurane may be caused by an uncoupling of neuroeffector responses by desflurane (6); however, this hypothesis has not been confirmed.
Several studies found that the sympathetic baroreflex sensitivity, measured as the response of muscular sympathetic nerve activity to changes in blood pressure, was better preserved during desflurane compared with isoflurane or sevoflurane (2,5). However, the effect of these anesthetics on blood pressure regulation (being the main target of baroreflex-mediated hemodynamic control) has never been studied in humans.
Therefore, we tested the hypothesis that desflurane has a more pronounced depressive effect on baroreflex-mediated blood pressure control than other volatile anesthetics. We used cyclical stimulation of the carotid baroreceptors by applying sinusoidal neck suction. This technique creates reflex oscillations in hemodynamic signals, such as RR interval, blood pressure, and muscle sympathetic nerve activity with only minor changes in mean values (7). At frequencies more than 0.15 Hz, the effector response of the peripheral baroreflex arc is mediated nearly exclusively by parasympathetic activity, whereas at lower frequencies, parasympathetic and sympathetic activity contribute to the baroreflex response, the absolute and relative contribution of the sympathetic and parasympathetic branch to the baroflex response depends on the general autonomic system status (7,8). Therefore, the application of a cyclic neck suction stimulus at a frequency different from the respiratory frequency enables the evaluation of baroreflex-mediated effector responses independently of the hemodynamic effects of respiration, as well as the assessment of the frequency-dependent characteristics of vagally and sympathetically mediated effector responses (7,9–11). Thus, this method allows one to extend the currently used clinical measure of baroreflex as the simple ratio between changes in blood pressure and RR interval or sympathetic nerve activity by introducing additional elements such as the relationship between baroreceptor stimulation and blood pressure and the delay of the hemodynamic response to baroreceptor stimulation.
With approval of the local Ethical Board and with written informed consent, we studied 40 otherwise healthy patients (ASA physical status I), aged 20–42 yr, who were scheduled for minor surgery for the ear or nose. The patients had no history of cardiopulmonary disease and were not taking any medications.
The patients were randomly allocated to a group receiving either sevoflurane or desflurane for maintenance of anesthesia. On the preoperative afternoon, patients underwent control measurements in the supine position using electrocardiogram monitoring (Sirecust 302D; Siemens, Erlangen, Germany), noninvasive blood pressure (Finapres; Ohmeda, Louisville, CO), and respiration (inductance plethysmograph, Respitrace system; Ambulatory Monitoring, Ardsley, NY). Neck suction was performed using a lead collar (12) by which a pressure was applied that changed sinusoidally between 0 and −30 mm Hg at 0.2 Hz and 0.1 Hz, respectively, while subjects breathed frequency controlled at 0.25 Hz. Measurements were performed after an adequate period of training of the breathing pattern to guarantee a stable breathing frequency and to avoid unintentional hyperventilation. A baseline recording of 3 min without neck suction and 2 3-min recordings during neck suction, with a frequency of 0.1 Hz and 0.2 Hz, respectively, were performed. The measurements were separated by breaks where subjects breathed without frequency control.
Patients received 20–30 mg of clorazepate orally on the preoperative night. Anesthesia was induced with propofol (2.5 mg/kg) and vecuronium (0.1 mg/kg) IV after prehydration with 500 mL of an isotonic crystalloid solution. Immediately after the anesthesia induction, the vaporizer was set at 3.6% (sevoflurane) or 13.2% (desflurane) using a semiclosed circle system with a fresh gas flow rate of 4 L/min (40% oxygen in air) until an end-tidal concentration of 1.0 minimum alveolar anesthetic concentration (MAC; sevoflurane = 1.8%, desflurane = 6.6%) was achieved (13). Anesthesia was maintained at 1 MAC end-tidal concentration (fresh gas flow rate = 2 L/min; Fio2 = 0.4). The trachea was intubated with a tracheal tube that was lubricated with lidocaine solution and ventilated at 15 breaths/min and an inspiratory/expiratory ratio of 1:1 with tidal volumes adjusted to an end-tidal carbon dioxide tension of 35 mm Hg. End-tidal carbon dioxide and inspiratory and expiratory concentrations of volatile anesthetics were monitored by infrared spectrometry (Capnomac Ultima; Datex Instrumentarium Corp, Helsinki, Finland). After 20 min at steady-state 1.0 MAC, a 3-min baseline recording without neck suction and 2 3-min recordings with neck suction at a frequency of 0.1 Hz and 0.2 Hz, respectively, were performed. Immediately after the recordings, a venous blood sample was drawn for the measurement of the plasma propofol concentration by high-pressure liquid chromatography. The patient’s body was covered to ensure that the core temperature was more than 36.0°C.
Signals were digitized by a 12-bit analog-to-digital converter and sampled at 1000 Hz on a personal computer. R waves and systolic and diastolic arterial blood pressures (Psys; Pdia) were automatically detected. Mean arterial blood pressure was obtained from the area under the pressure curve. Signals were inspected visually and checked for artifacts and heterotopic beats that would have been removed by interpolation using interactive software. Analysis of data was performed in accordance with the suggestions of the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (14). Time series were computed with the pressure of neck suction and the respiratory signal both being registered at the start of the RR interval. Stationary positioning of each period was checked by the reverse arrangement test described by Bendat and Piersol (15). Data were resampled at 4 Hz using a moving 500-ms wide rectangular window. After substraction of the mean value of the sample data, removal of residual linear trends, and application of a cosine function (Hanning window) to avoid distortions of the estimated spectra (15), discrete Fourier analysis was performed for 3 50% overlapping windows, and the results were subsequently averaged. The area under the curve was calculated for the frequency components between 0.075 and 0.125 Hz (frequency component of baroreceptor stimulation at 0.1 Hz) and between 0.175 and 0.225 Hz (frequency component of baroreceptor stimulation at 0.2 Hz). The relationship between 2 signals was analyzed by computing the coherence function and complex transfer function using 6 50% overlapping windows. A squared coherence >0.53 was interpreted as a measure of a statistically significant linear relationship between 2 signals (16), and the relationship between these signals was expressed by gain and phase factors.
Statistical analysis was performed using commercially available software (SPSS for Windows, version 10.0; SPSS Inc, Chicago, IL). Because the data of spectral power were left skewed, logarithmic transformation was required to obtain normal distribution. Data are presented as mean ± sd. The mean values of RR interval and blood pressure, obtained during each registration, as well as the results of power spectral analysis, were compared between the groups using a general linear model procedure (repeated-measures analysis of variance) and the Student’s t-test for unpaired data with adjustment of the α-error, as appropriate. Under the condition that an effect size of approximately 0.7 was regarded as sufficient, the β-error was determined to be approximately 0.3 for the comparison of hemodynamic data and the results of power spectral analysis between groups. Assuming that a difference in phase shifts of at least 0.5 was of interest, the β-error for the comparison of phases was determined to be 0.1. The number of subjects with significant coherence between tested variables was compared between groups using the χ2 test. A P value <0.05 was regarded as statistically significant.
The sevoflurane group consisted of 13 men and seven women (age, 32.1 ± 6.3 yr; weight, 77.7 ± 17 kg; height, 177 ± 8 cm) and the desflurane group consisted of 13 men and seven women (age, 33.5 ± 6.5 yr; weight, 78 ± 12 kg; height, 176 ± 10 cm; no significant differences between groups). Seven subjects of each group reported a history of cigarette smoking.
Hemodynamic data and spectral power were not differentbetween groups in the awake state and are comprehensively reported in Table 1 and Table 2. During anesthesia, heart rate decreased in the sevoflurane group and increased in the desflurane group, whereas blood pressure decreased to similar values in both groups. With neck suction, heart rate decreased significantly in the awake state and during sevoflurane as well as desflurane anesthesia, whereas blood pressure remained stable.
The number of patients with significant coherence between neck suction and hemodynamic signals was similar in the awake state and during anesthesia as well as during sevoflurane and desflurane anesthesia (Table 3).
Spectral power of the RR interval around 0.2 Hz during neck suction at 0.2 Hz was significantly diminished by sevoflurane and desflurane without differences between groups (Table 2). Likewise, desflurane and sevoflurane decreased the gain of the relationship between neck suction and RR interval at 0.2 Hz in a similar way (Fig. 1).
Blood pressure did not change during neck suction at 0.2 Hz. Neck suction at 0.2 Hz created, during awake conditions and during anesthesia, only minimal blood pressure responses (Table 2). The phase relationship between neck suction and oscillation in the RR interval and Psys, respectively, remained unchanged during anesthesia (Table 3;Fig. 2). The phase shift between neck suction and mean arterial blood pressure and Pdia, respectively, increased significantly during anesthesia in both groups (Table 3). In the awake state, Pdia fluctuated in opposite phase to the RR interval. This relationship was absent during anesthesia.
In contrast to baroreceptor stimulation at 0.2 Hz, neck suction at 0.1 Hz induced a marked response in blood pressure in the awake state as well as during anesthesia (Table 2). The power spectral density around 0.1 Hz was more diminished during sevoflurane than during desflurane, with statistically significant differences between groups (Table 2). Sevoflurane, as well as desflurane, markedly depressed the gain of the transfer function between neck suction and fluctuations in blood pressure at 0.1 Hz (Fig. 3). The phase relationship between 0.1 Hz neck suction and oscillation in the RR interval and blood pressure changed significantly during anesthesia without differences between sevoflurane and desflurane (Table 3). Thus, the time delay between neck suction and RR interval and blood pressure response, respectively, was prolonged for approximately 1 s during anesthesia in both the sevoflurane and desflurane group (Figs. 2 and 4).
Plasma propofol concentration, measured at the end of the measurements, was similar in the sevoflurane group and in the desflurane group (383 ± 235 ng/mL and 344 ± 199 ng/mL, respectively).
The major findings of this study are:
- Sevoflurane and desflurane did not disturb the linear coupling between baroreceptor stimulation and heart rate or blood pressure response.
- The vagal-mediated heart rate response to baroreceptor stimulation was similarly suppressed by sevoflurane and desflurane.
- Sevoflurane and desflurane impaired the baroreflex-mediated blood pressure regulation to the same extent and increased the time delay of the blood pressure response to baroreceptor stimulation.
Neck suction increased the mean RR interval in the awake state as well as during anesthesia. This phenomenon occurred independently of the fact that desflurane decreased, whereas sevoflurane increased, the mean RR interval. This effect indicates an increase in mean vagal activity because of the intermittent baroreceptor activation (9,11). The RR interval response to neck suction, measured as spectral power in the specific frequency component and as the gain of the transfer function between neck suction and RR interval oscillation, was depressed by sevoflurane and desflurane in a comparable manner. This observation is in accordance with previous studies that reported a similar reduction of cardiac baroreflex sensitivity by sevoflurane and desflurane (2).
The data obtained during neck suction at 0.2 Hz mainly reflect the vagal-mediated cardiac baroreflex caused by the sluggish sympathically mediated effector response (17). The number of subjects with significant coherence between neck suction and oscillation in the RR interval was not significantly different between the awake state and anesthesia. The latency between stimulation of baroreceptors and changes in the RR interval (Fig. 2) is typical for the vagal-mediated cardiac effector response (18,19) and re- mained unchanged at 1 MAC of sevoflurane or desflurane. Thus, our results demonstrate a diminished but preserved reactivity of the parasympatheticallymediated cardiac baroreflex and indicate the absence of major disturbances in the linear coupling of baroreceptor stimulation and vagal-mediated heart rate response.
The autonomic regulation of heart rate at 0.1 Hz is mediated by combined vagal and sympathetic activity (20,21). In the awake state, the RR interval response to baroreceptor stimulation is more marked at 0.1 Hz than at 0.2 Hz (7) according to the phenomenon by which hemodynamic fluctuations are reinforced at 0.1 Hz (22). This characteristic behavior was less obvious during anesthesia thus indicating a depressive effect of sevoflurane and desflurane on both the sympathetic and parasympathetic arch of autonomic cardiac regulation. However, we observed a slightly increased cardiac baroreflex sensitivity during anesthesia at 0.1 Hz compared with 0.2 Hz, with the difference being significant in the sevoflurane group (Fig. 1).
The delay between baroreceptor stimulation and RR interval response at 0.1 Hz indicates a mainly vagal-mediated cardiac effector response during baroreceptor stimulation at 0.1 Hz in the awake state (Fig. 2). During anesthesia, this latency was markedly prolonged. This phenomenon may be interpreted as a result of increased sympathetic contribution to the heart rate response at 0.1 Hz during anesthesia. Sundblad et al. (23) observed a similar phenomenon in awake subjects using a different experimental design and attributed their finding to an increase in sympathetic and decrease in parasympathetic control of the sinus node.
The minor blood pressure response to neck suction at 0.2 Hz, compared with the response at 0.1 Hz, can be regarded as the expression of the suppressed sympathetically mediated baroreflex effector response at higher frequencies (7,17). During neck suction at 0.2 Hz, diastolic blood pressure fluctuated nearly in opposite phase with the RR interval. This relationship has been explained by the fact that the length of the RR interval has a direct influence on the diastolic runoff and thus on diastolic blood pressure (24). Consequently, the suppression of the vagal-mediated heart rate response during anesthesia was associated with a loss of this coupling.
The blood pressure response to cyclic baroreceptor stimulation at 0.1 Hz is related to fluctuations in muscular sympathetic activity (7) and can be regarded as an expression of the neurally mediated short-term control of vascular tone (25).
Our results indicate that the blood pressure response to baroreceptor stimulation at 0.1 Hz is similarly suppressed by desflurane and sevoflurane (Fig. 3). Thus, our study does not support the hypothesis that desflurane has a more depressant effect on the baroreflex-mediated blood pressure response than sevoflurane (Table 2). The data of spectral power of Psys in the 0.1-Hz component suggest that the blood pressure response may be even better preserved with desflurane than with sevoflurane. This interpretation is in accordance with the previous observation that the sympathetic baroslope was less impaired with 1 MAC of desflurane compared with sevoflurane or isoflurane (2,5).
The delay between neck suction and blood pressure response registered in the awake state (Fig. 4) is identical with the findings of previous studies investigating the autonomic short-term regulation of blood pressure and is typical for a sympathetically mediated effector response (19,26–28). The faster response of Pdia to baroreceptor stimulation compared with Psys has been explained by the above-mentioned direct effect of the RR interval on Pdia(24). Because of the same reason, the time delay of the blood pressure response may have been prolonged during anesthesia, and the increased latency of the RR interval to baroreceptor stimulation may have contributed to the more marked delay in blood pressure response. This phenomenon may be interpreted as an expression of the feed-forward activity of the RR interval on blood pressure regulation (29) that reinforces (physiologically meaningful) the autonomic blood pressure control. However, an increased latency of the sympathically mediated effector response by sevoflurane and desflurane cannot be excluded by our study because the method provides information about the complete baroreflex arch and does not distinguish between neural and mechanical components. Obviously, the increased latency of blood pressure regulation had no destabilizing effect on hemodynamic regulation as observed under propofol in one study (30). This can be explained by the fact that the phase shift between baroreflex stimulation and blood pressure response at 0.1 Hz was less than half a period (i.e., 5 seconds) thus avoiding the amplification of under-dampened oscillations that would have been associated with an increase in power around 0.1 Hz (22).
Different methods for evaluating baroreflex sensitivity are comparable only in a limited way. Cyclic stimulation of baroreceptors enables the assessment of dynamic baroreflex activity but offers no information about static blood pressure regulation. Nevertheless, the typical baroreceptor response is mainly dynamic rather than static. Furthermore, the effect of the isolated stimulation of carotid baroreceptors may be counteracted by baroreceptors located in other areas. However, it has been shown that these effects are mainly additive and can be described as a simple linear relationship (31).
The anesthetics used for the induction may have influenced our results. In one study, we observed that propofol had a marked effect on autonomic hemodynamic regulation, which seems to be much more pronounced than the effects demonstrated in the present study (30). Although the plasma concentration of propofol was much less than that measured in the former study, an influence of propofol on the results cannot be excluded. However, because most studies that investigated the influence of volatile anesthetics on autonomic hemodynamic regulation used a similar induction regimen, the comparability of studies may be preserved.
The modulation of autonomic activity depends on a variety of physiological variables (such as respiratory depth and frequency; spontaneous breathing versus mechanical ventilation) (32–34). All studies investigating autonomic hemodynamic control during anesthesia face this problem, and although the technique applied in our study separates influences of ventilation and of neck suction, we measured the net effect of baroreflex stimulation on heart rate and arterial blood pressure.
In conclusion, our study reveals that sevoflurane and desflurane decrease dynamic cardiac baroreflex response in a similar manner without affecting the time delay between baroreceptor stimulation and the vagal-mediated heart rate response. There is no indication that desflurane exerts a more depressant effect on autonomic short-term blood pressure control than sevoflurane. Both anesthetics increase the delay of the blood pressure response to baroreceptor stimulation without loss of significant coupling between baroreceptor stimulation and effector response.
The authors thank Michael Gruber, PhD, Department of Anesthesiology, University of Regensburg, for technical assistance.
1. Ebert TJ, Muzi M. Sympathetic hyperactivity during desflurane anesthesia in healthy volunteers: a comparison with isoflurane. Anesthesiology 1993; 79: 444–53.
2. Ebert TJ, Harkin CP, Muzi M. Cardiovascular responses to sevoflurane: a review. Anesth Analg 1995; 81: S11–22.
3. Ebert TJ, Muzi M, Lopatka CW. Neurocirculatory responses to sevoflurane in humans: a comparison to desflurane. Anesthesiology 1995; 83: 88–95.
4. Ebert TJ, Trotier TS, Arain SR, et al. High concentrations of isoflurane do not block the sympathetic nervous system activation from desflurane. Can J Anaesth 2001; 48: 133–8.
5. Ebert TJ, Perez F, Uhrich TD, Deshur MA. Desflurane-mediated sympathetic activation occurs in humans despite preventing hypotension and baroreceptor unloading. Anesthesiology 1998; 88: 1227–32.
6. Ebert TJ, Schmid PGI. Inhalation anesthesia. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical anesthesia. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2001: 377–417.
7. Bernardi L, Hayoz D, Wenzel R, et al. Synchronous and baroceptor-sensitive oscillations in skin microcirculation: evidence for central autonomic control. Am J Physiol 1997; 273: H1867–78.
8. Sunagawa K, Ikeda Y, Kawada T, et al. Dynamic control of arterial blood pressure by the sympathetic baroreflex. Fundam Clin Pharmacol 1998; 12: S23–8.
9. Cross SJ, Cowie MR, Rawless JM. Mutual interactions of respiratory sinus arrhythmia and the carotid baroreceptor-heart rate reflex. Clin Sci 1992; 82: 139–45.
10. Keyl C, Dambacher M, Schneider A, et al. Cardiocirculatory coupling during sinusoidal stimulation and fixed-frequency breathing. Clin Sci 2000; 99: 113–24.
11. Keyl C, Schneider A, Dambacher M, Bernardi L. Time delay of vagally mediated cardiac baroreflex response varies with autonomic cardiovascular control. J Appl Physiol 2001; 91: 283–9.
12. Eckberg DL, Cavanagh MS, Mark AL, Abboud FM. A simplified neck suction device for activation of carotid baroreceptors. J Lab Clin Med 1975; 85: 167–73.
13. Mapleson WW. Effect of age on MAC in humans: a meta-analysis. Br J Anaesth 1996; 76: 179–85.
14. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability, standards of measurement, physiological interpretation, and clinical use. Circulation 1996; 93: 1043–65.
15. Bendat JS, Piersol AG. Random data: analysis and measurement procedures, 2nd ed. New York: Wiley, 1986.
16. Taylor JA, Carr DL, Myers CW, Eckberg DL. Mechanisms underlying very-low-frequency RR-interval oscillations in humans. Circulation 1998; 98: 547–55.
17. Ikeda Y, Kawada T, Sugimachi M, et al. Neural arc of baroreflex optimizes dynamic pressure regulation in achieving both stability and quickness. Am J Physiol 1996; 271: H882–90.
18. Pickering TG, Davies J. Estimation of the conduction time of the baroreceptor-cardiac reflex in man. Cardiovasc Res 1973; 7: 213–9.
19. Borst C, Karemaker JM. Time delays in the human baroreceptor reflex. J Auton Nerv Syst 1983; 9: 399–409.
20. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res 1986; 59: 178–93.
21. Saul JP, Berger RD, Albrecht P, et al. Transfer function analysis of the circulation: unique insights into cardiovascular regulation. Am J Physiol 1991; 261: H1231–45.
22. Toska K, Eriksen M, Walløe L. Short-term control of cardiovascular function: estimation of control parameters in healthy humans. Am J Physiol 1996; 270: H651–60.
23. Sundblad P, Haruna Y, Tedner B, Linnarson D. Short-term cardiovascular responses to rapid whole-body tilting during exercise. Eur J Appl Physiol 2000; 81: 259–70.
24. de Boer RW, Karemaker JM, Strackee J. Hemodynamic fluctuations and baroreflex sensitivity in humans: a beat-to-beat model. Am J Physiol 1987; 253: H680–9.
25. Furlan R, Porta A, Costa F, et al. Oscillatory patterns in sympathetic neural discharge and cardiovascular variables during orthostatic stimulus. Circulation 2000; 101: 886–92.
26. Eckberg DL. Nonlinearities of the human carotid baroreceptor-cardiac reflex. Circ Res 1980; 47: 208–16.
27. Levenhagen D, Evans JM, Wang M, Knapp CF. Cardiovascular regulation in humans in response to oscillatory lower body negative pressure. Am J Physiol 1994; 267: H593–604.
28. Toska K, Eriksen M, Walløe L. Short-term cardiovascular responses to a step decrease in peripheral conductance in humans. Am J Physiol 1994; 266: H199–211.
29. Parati G, Di Rienzo M, Castaglioni P, et al. Cardiovascular regulation and analysis of blood pressure-heart rate variability interactions. Fundam Clin Pharmacol 1998; 12: S48–56.
30. Keyl C, Schneider A, Dambacher M, et al. Dynamic cardiocirculatory control during propofol anesthesia in mechanically ventilated patients. Anesth Analg 2000; 91: 1188–95.
31. Kawada T, Inagaki M, Takaki H, et al. Counteraction of aortic baroreflex to carotid sinus baroreflex in a neck suction model. J Appl Physiol 2000; 89: 1979–84.
32. Melcher A. Carotid baroreflex heart rate control during the active and the assisted breathing cycle in man. Acta Physiol Scand 1980; 108: 165–71.
33. Novak V, Novak P, de Champlain J, et al. Influence of respiration on heart rate and blood pressure fluctuations. J Appl Physiol 1993; 74: 617–26.
© 2002 International Anesthesia Research Society
34. Haggenmiller C, Baumert JH, Adt M, Frey AW. Modulation of respiratory sinus arrhythmia by breathing pattern. Clin Sci 1996; 91: 40–2.