Arterial baroreflex function is an important short-term regulatory system for maintaining cardiovascular stability and consists of two important limbs, including the sympathetic baroreflex and cardiovagal reflex systems (1,2). Impairment of cardiovagal reflex responses has been associated with some physiological or pathologic conditions, such as aging and hypertension (3,4). More importantly, its clinical application has been highlighted by the increased incidence of cardiac dysrhythmias and decreased survival after myocardial infarction in patients with diminished cardiovagal baroreflex function (5).
The first widely used technique to determine baroreflex control of heart rate (HR) was the Oxford method, by which R-R intervals (RRI) were related to increased or decreased arterial blood pressure (BP) by IV injections of vasoactive drugs (6). This simple method, however, has practical limitations, such as a limited number of observations, possible alterations of baroreceptor transduction by vasoactive drugs, artificially induced pressor and depressor transients, and changes in cardiac filling pressure, which may influence the net gain estimate by loading or unloading cardiopulmonary receptors (7). Other investigators have used spontaneous fluctuations of BP and RRI to calculate spontaneous baroreflex (SBR) gains within naturally occurring BP excursions (8). One of the most frequently used spontaneous indices is a sequence method, which uses spontaneous sequences of three or more consecutive beats in which BP progressively increases and RRI progressively lengthens (up-sequence) or in which BP progressively decreases and RRI progressively shortens (down-sequence). For each sequence, the regression between the BP values and the RRI values of the following cardiac cycle is calculated, and the mean of the slopes of the regression lines is regarded as a gain. The spontaneous sequence gain correlates well with that assessed by the drug-induced technique under resting and various experimental conditions, including cardiac autonomic blockade (2,9). Although many studies on baroreflex function have been performed under general anesthesia in both humans and animals, little is known regarding the effects of general anesthesia on SBR indices. Moreover, dose- or concentration-dependent effects of general anesthetics have never been addressed (10,11).
Potent volatile anesthetics cause concentration-dependent depressions in pharmacological baroreflex gains and continue to exert depressive effects after emergence from general anesthesia in humans (12,13). Because of the significant correlations between pharmacological gains and SBR indices in conscious humans (2,9), we hypothesized that (a) SBR indices would be similarly depressed in a concentration-dependent manner during, and remain depressed after, general anesthesia with a volatile anesthetic such as sevoflurane and that (b) baroreflex gains by various spontaneous methods would correlate with, and be surrogates for, pharmacological gains during sevoflurane anesthesia. In addition, spectral analyses of HR (HR variability; HRV) and BP (BP variability; BPV) were applied in each session of baroreflex determinations in an attempt to interpret the effects of general anesthesia on autonomic modulations of RRI (14).
Ten healthy, nonsmoking volunteers were recruited. All subjects were free of cardiovascular or autonomic disorders and abstained from caffeine-containing beverages and alcohol for at least 24 h before the study. None of the subjects had taken any regular medication during 1 yr before the study. All procedures were approved by the human research committee of Akita University School of Medicine, and written informed consent was obtained from each subject.
All volunteers arrived at the laboratory after a 10-h fast without premedication. They were placed in the supine position, and a 22-gauge IV catheter was inserted, with a local anesthetic, into a peripheral vein for the administration of balanced salt solution containing 5% dextrose at 2 mL · kg−1 · h−1 throughout the experiment. Fluid temperature was maintained at approximately 30°C. Determinations of BP and RRI were made from a radial arterial catheter and an electrocardiography (ECG) lead of the highest signal/noise ratio (Viridia CMS 2000™; Hewlett-Packard, Boeblingen, Germany), respectively. All subjects received whole-body forced-air warming to maintain baseline tympanic temperature, which was monitored throughout the study. The ambient temperature was set to 25°C–30°C to avoid postanesthesia shivering. Subjects were allowed to rest in the supine position for at least 20 min in a quiet environment before the study.
Baseline determinations of HRV, BPV, and baroreflex gains by the pharmacological and SBR methods were made. Then, general anesthesia was induced with 5% sevoflurane (inspiratory) in air (5 L/min) and oxygen (1 L/min), and a laryngeal mask airway was inserted to secure the airway, without any other adjuvant, including muscle relaxant or opioid. The lungs were mechanically ventilated (tidal volume, 7–10 mL/kg at a respiratory rate of 12 breaths/min). Anesthesia was maintained with 0.7%, 1.4%, and 2% end-tidal sevoflurane in air and oxygen (fraction of inspired oxygen, 0.34), and end-tidal CO2 tension was maintained at 35 mm Hg throughout the anesthesia period. Breath-by-breath end-tidal sevoflurane concentration and CO2 tension were measured by a gas analyzer (Capnomac Ultima SV; Datex, Helsinki, Finland), which was calibrated before each use. The order of sevoflurane concentration was randomized to avoid any effect of the testing sequence. To ensure anesthetic equilibration, end-tidal sevoflurane concentration was maintained constant at each level for 20 min by frequently adjusting inspiratory sevoflurane concentrations before baroreflex indices, HRV, and BPV were determined. After approximately 3 h of general anesthesia, sevoflurane was discontinued. After confirming the return of adequate spontaneous respiration and responses to verbal commands, the laryngeal mask airway was removed. The subjects breathed supplemental oxygen 2 L/min via a face mask to ensure oxygen saturation ≥98%. SBR responses, HRV, and BPV were determined at 30, 60, 120, and 180 min after removal of the laryngeal mask airway. At each session, end-tidal sevoflurane concentration was measured through a cannula advanced 2–3 cm into a naris by having subjects take several deep breaths.
To assess pharmacological baroreflex gains and to correlate these with SBR indices, pressor and depressor tests were performed by using IV injections of phenylephrine (150–300 μg) and nitroprusside (150–300 μg) to increase and decrease systolic BP (SBP) by 15–30 mm Hg, respectively. Pressor and depressor tests were performed at conscious baseline and during 0.7% and 2% end-tidal sevoflurane anesthesia. The doses were chosen on the basis of our previous studies with healthy young individuals (13,15). A period of stabilization >10 min between the pressor and depressor tests allowed HR and SBP to return to the pretest values ±5%. All measurements of pharmacological baroreflex gains were made in triplicate and the results were averaged to provide a single data set in each subject. To determine SBR indices, 10-min recordings of RRI and SBP were made. Recordings of ECG and SBP for spectral analyses during the conscious baseline period and on emergence from anesthesia were made while subjects breathed in step with an auditory signal (metronome) at 12 breaths/min. All subjects were trained and familiarized with paced breathing the day before the study. During sevoflurane anesthesia, the respiratory rate was also fixed at 12 breaths/min to avoid confounding effects of respiratory variables on HRV (16).
BP and RRI were determined beat by beat, digitized, stored at a sampling rate of 250 Hz in a computer, and subsequently analyzed offline. A custom program developed to process the digitized data with a 16-bit analog-digital converter (AD7120; ATM Communications, Tokyo, Japan) detected R-waves to determine RRI from ECG signals. The recordings were also observed on an oscilloscope during transfer for elimination of nonsinus or artifactual signals. Digital files were thus generated; each column consisted of SBP, diastolic BP, mean BP generated by electronic integration of the arterial wave form, and RRI values for every cardiac cycle. These files were used for calculations of baroreflex gains and for power spectral analyses.
Pharmacological baroreflex gain was determined by least-square regression analysis between SBP and RRI, when each RRI was plotted as a function of the preceding SBP (6,9,13). Three spontaneous indices were obtained: the spontaneous sequence method, the α-index, and the low-frequency (LF) transfer function index. The sequence method relates sequences of three or more consecutive cardiac cycles in which both SBP and the following RRI either increase (up-sequence) or decrease (down-sequence) simultaneously (8,9). Only sequences in which successive pressure pulses differed by at least 1 mm Hg were selected. A correlation coefficient (R) >0.8 was accepted for each regression analysis with the pharmacological and spontaneous sequence methods. The α-index uses the simple ratio between RRI spectral power and SBP spectral power at LF (0.1 Hz) and high frequency (HF) (0.25 Hz). The square roots of the ratios were derived and averaged for this index (17). The LF transfer function index uses average RRI and SBP cross-spectral magnitude within 0.05–0.15 Hz, where coherence is more than 0.5 (18). All of these spontaneous indices exploit spontaneous fluctuations in BP and RRI and are regarded as less invasive means to assess baroreflex gains.
Methods for spectral analyses of HR and BP have been explained previously in detail (19,20). For each study period, a time series of 512 consecutive RRI and SBP data free of artifacts were selected. A fast Fourier transformation was applied to the tachogram to calculate the amplitude of variations of RRI and SBP as a function of frequency (power spectral density: RRI and SBP powers [units of milliseconds squared and millimeters of mercury squared, respectively] as a function of frequency [hertz]). The RRI and SBP powers were the integrated area under the power spectral density plots as an index of the frequency-specific degree of RRI and SBP variability. Spectral power was determined over the LF (0.04–0.15 Hz) and HF (0.15–0.4 Hz) ranges. The LF/HF ratio was calculated for HRV (21).
All the statistical analyses were performed by an investigator blinded to the treatment of subjects. Power analysis based on previous baroreflex studies revealed that at least 9 subjects would provide a power more than 0.8 (P = 0.05) for a 40% difference in temporal changes in baroreflex gains (13,15). Hemodynamic and baroreflex data were first analyzed by one-way analysis of variance for repeated measurements. If a significant difference was detected with respect to time, this was followed by a paired Student’s t-test with Bonferroni’s correction as a post hoc test to compare hemodynamic and baroreflex data with baseline values and between sessions. Log transformation was used before performing one-way analysis of variance if data were not distributed normally, such as HF and LF power. Correlations and agreements between pharmacological and SBR indices were analyzed by Pearson’s correlation coefficient and Bland-Altman plots, respectively (22). All data are presented as mean ± sd, and P < 0.05 was considered statistically significant.
One subject was excluded because of difficulty in placing the laryngeal mask airway. The mean age, weight, and height of the remaining nine volunteers were 25 ± 3 yr, 62.4 ± 5.2 kg, and 170 ± 3 cm, respectively. All of them were male. Compared with the conscious baseline value, SBP decreased significantly in a concentration-dependent manner during sevoflurane anesthesia and returned to the baseline value after emergence from anesthesia (Table 1). A significant increase in HR was seen 30 min after emergence from anesthesia compared with the awake baseline value. Compared with HR during 0.7% sevoflurane, significantly greater HR values were seen at 1.4% and 2.0% sevoflurane. The tympanic temperature in each individual was maintained within 0.2°C higher or lower than the conscious baseline value throughout the study.
Baroreflex gains assessed by the sequence method demonstrated concentration-dependent, significant decreases during sevoflurane anesthesia and remained depressed for 30 min after emergence (Fig. 1). Significant depressions of α-index gains were also noted during and for 30 min after sevoflurane anesthesia compared with the baseline value, but a concentration-dependent effect was not evident (Fig. 2). In contrast, the LF transfer function index did not show any significant alteration during and after sevoflurane anesthesia.
Power spectral analysis of RRI revealed that both HF and LF were concentration-dependently depressed during sevoflurane anesthesia (Fig. 3). During the recovery period, HF power remained significantly depressed at 30 min and returned to the conscious baseline value from 60 to 180 min after emergence. LF power was similarly depressed in a concentration-dependent manner during sevoflurane anesthesia. After anesthesia, however, LF power rapidly recovered and was significantly more than the baseline value at the 120- and 180-min recovery periods. As a result, the LF/HF ratio remained unaltered during and for 60 min after sevoflurane anesthesia but was significantly more than the awake value from 120 to 180 min after emergence from anesthesia (data not shown). Conversely, the HF power of BPV was significantly increased during sevoflurane anesthesia without a concentration-dependent effect, whereas the LF power of BPV was significantly depressed by sevoflurane. Both HF and LF powers recovered rapidly after emergence from anesthesia (Fig. 4).
In general, there were highly significant positive correlations among various baroreflex indices during sevoflurane anesthesia, with the exception of the LF transfer function index (Table 2). The least-square regressions of these spontaneous indices against the pharmacological gains are shown in Figure 5. Because of the significant relations, the agreement was further analyzed by Bland-Altman plots (22), as shown in Figure 6. In each Bland-Altman plot, however, the limits of agreement (mean ± 1.96 sd) were as large as or exceeded the baroreflex gain itself, and 95% confidence intervals of the limits showed wide variations (Table 3), thus indicating weak agreement.
A major finding of our study is that spontaneous sequence gain is depressed by sevoflurane in a concentration-dependent manner and remains depressed during the immediate postanesthesia period (Fig. 1). A few previous investigations reported depressions of SBR indices by general anesthesia in humans. Van Vlymen and Parlow (11) found depressions of spontaneous sequence gains during propofol/fentanyl anesthesia, whereas Huang et al. (10) documented decreases in sequence gains by various concentrations of sevoflurane during minor surgery. Furthermore, depressions of spontaneous sequence gain and HF gain of cross-spectral analysis have been reported (23). However, none of those studies controlled depth of anesthesia, determined the dose-response relationship between anesthetics and SBR indices, or reported relation/agreement between pharmacological gains and spontaneous indices. The concentration-dependent depressions in spontaneous sequence gains and their recovery characteristics by sevoflurane in our study are remarkably similar to changes in pharmacological gains by volatile anesthetics, including sevoflurane (13). These results imply that the depth of anesthesia should be strictly controlled when a spontaneous sequence method is used to quantify cardiovagal baroreflex function.
Except for the LF transfer function index, our results are in accordance with several previous reports (2,9) in which highly significant correlations were demonstrated between spontaneous indices and pharmacological baroreflex gains (Table 2). However, no previous study has further analyzed agreement between any spontaneous index and pharmacological gain. A more recent study by Lipman et al. (24) demonstrated poor agreement between pharmacological and spontaneous sequence gains in conscious humans. They also showed that carotid distensibility, a critical determinant of baroreflex sensitivity, was directly related to the pharmacological gain, but not to spontaneous indices. Their results, together with the present results, suggest that spontaneous indices may simply reflect beat-to-beat synchronous fluctuations of BP and RRI associated with respiratory-related changes in the intrathoracic pressure, but not arterial baroreflex function. Therefore, the lack of agreement shown by the Bland-Altman analysis in our study (Fig. 6, Table 3) does not support the utility of any spontaneous index as a surrogate for pharmacological gain during general anesthesia. Rather, spontaneous indices may be useful simply for qualitative assessments of beat-to-beat vagal modulation of HR.
We have simultaneously performed frequency-domain analysis of RRI as one of the autonomic indices, because fluctuations of RRI reflect the beat-to-beat dynamic response of the autonomic nervous system to various physiological perturbations (21). The HF component of the HRV spectrum is considered to represent predominantly vagal modulation of the cardiac cycle, LF is thought to be under the influence of both the parasympathetic and sympathetic nervous systems, and the LF/HF ratio may reflect sympathetic predominance (21,25). Consistent with a previous report in which isoflurane concentration-dependently reduced the powers of both the HF and LF components without altering the LF/HF ratio (26), our results suggest that efferent autonomic nervous activity was diminished concentration dependently, but the relative balance of cardiac sympathetic and parasympathetic activities appeared to be maintained during sevoflurane anesthesia (Fig. 3). With respect to recovery characteristics, our results also agree with a report by Donchin et al. (27) in which the HF component remained depressed for 20–30 minutes after 2 minimum alveolar anesthetic concentration of isoflurane and 70% nitrous oxide anesthesia. The recovery profile of spontaneous sequence baroreflex response in our study is similar to that of HF power after emergence from anesthesia. These results, together with previous reports, suggest that depressed SBR response and HF power during and after sevoflurane anesthesia are closely linked with impaired beat-to-beat vagal modulation of the cardiac cycle.
In contrast, the HF power of BPV showed neither a depression nor a concentration-dependent effect by sevoflurane and returned to the baseline value immediately after discontinuance of sevoflurane and positive pressure ventilation (Fig. 4). Although vagally mediated changes in cardiac output may play a role in determining the HF power of BPV, the HF power of BPV is not substantially modified in patients with a denervated, transplanted heart (28). Therefore, the absence of concentration-dependent alterations of HF power of BPV in our study is more likely to reflect mechanical effects of positive pressure ventilation on pressure gradients of large thoracic vessels and functions of the heart rather than vagal influences.
Our study also demonstrated that the immediate recovery phase ≤30 minutes after emergence from sevoflurane anesthesia was characterized by the diminished HF component of HRV and cardiovagal baroreflex response with an unaltered LF power and LF/HF ratio. Conversely, the late recovery phase ≥120 minutes after emergence was characterized by augmented LF power and LF/HF ratio with restored HF power and baroreflex response (Figs. 1–3). Little is known regarding the effects of surgical/anesthetic factors on the recovery characteristics of autonomic indices. Our previous investigations and the present results may suggest a differential recovery profile of the cardiac autonomic nervous system after general anesthesia, i.e., more rapid recovery of sympathetic than parasympathetic nervous activity during the immediate postanesthesia period and overactivity of the sympathetic nervous activity thereafter (13,15,29). However, to confirm the recovery characteristics of sympathetic nervous system activity more precisely, determinations of norepinephrine spillover from the heart and/or examination of microneurography would be mandatory (25).
The results of our study should be interpreted with some caution. First, a significant difference was not detected in up-sequence indices between end-tidal sevoflurane 0.7% and 1.4% and down-sequence indices between end-tidal sevoflurane 1.4% and 2.0%. In addition, a concentration-dependent depression in the α-index was not demonstrated. A possible explanation is that the step size of the sevoflurane concentration was not adequate to differentiate depressive effects at two anesthetic levels, although the lack of statistical power, i.e., a small sample size, cannot be excluded. Second, application of our results in actual surgical patients may be limited, because we have examined only healthy individuals not undergoing surgery. Changes in baroreflex responses during surgery and the recovery period would be affected by not only coexisting disorders, but also multiple anesthetic and nonanesthetic drugs, including cholinergic and anticholinergic drugs, which could affect autonomic nervous system activity (11,30).
In conclusion, our results demonstrated that the spontaneous sequence baroreflex response was depressed in a concentration-dependent manner during, and remained depressed for 30 minutes after, sevoflurane anesthesia in humans. Spontaneous sequence gains and the α-index correlated well, but did not agree, with pharmacological baroreflex gains during anesthesia. Our results extend the findings in conscious humans that the utility of a spontaneous index as a surrogate for pharmacological gain is limited during general anesthesia.
1. Ogoh S, Fadel PJ, Monteiro F, et al. Haemodynamic changes during neck pressure and suction in seated and supine positions. J Physiol 2002;540:707–16.
2. Rudas L, Crossman AA, Morillo CA, et al. Human sympathetic and vagal baroreflex responses to sequential nitroprusside and phenylephrine. Am J Physiol 1999;276:H1691–8.
3. Goldstein DS. Arterial baroreflex sensitivity, plasma catecholamines, and pressor responsiveness in essential hypertension. Circulation 1983;68:234–40.
4. Laitinen T, Hartikainen J, Vanninen E, et al. Age and gender dependency of baroreflex sensitivity in healthy subjects. J Appl Physiol 1998;84:576–83.
5. La Rovere MT, Bigger JT Jr, Marcus FI, et al. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. Lancet 1998;351:478–84.
6. Smyth HS, Sleight P, Pickering GW. Reflex regulation of arterial pressure during sleep in man. Circ Res 1969;24:109–21.
7. Pawelczyk JA, Raven PB. Reductions in central venous pressure improve carotid baroreflex responses in conscious men. Am J Physiol 1989;257:H1389–95.
8. Bertinieri G, di Rienzo M, Cavallazzi A, et al. A new approach to analysis of the arterial baroreflex. J Hypertens 1985;3:S79–81.
9. Parlow J, Viale JP, Annat G, et al. Spontaneous cardiac baroreflex in humans: comparison with drug-induced responses. Hypertension 1995;25:1058–68.
10. Huang CL, Huang HH, Chao A, et al. Rapid recovery of spontaneous baroreflex after sevoflurane anesthesia in ambulatory surgery. Acta Anaesthesiol Sin 2001;39:23–6.
11. Van Vlymen JM, Parlow JL. The effects of reversal of neuromuscular blockade on autonomic control in the perioperative period. Anesth Analg 1997;84:148–54.
12. Kotrly KJ, Ebert TJ, Vucins E, et al. Baroreceptor reflex control of heart rate during isoflurane anesthesia in humans. Anesthesiology 1984;60:173–9.
13. Tanaka M, Nagasaki G, Nishikawa T. Moderate hypothermia depresses arterial baroreflex control of heart rate during, and delays its recovery after, general anesthesia in humans. Anesthesiology 2001;95:51–5.
14. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure variations as a marker of sympatho-vagal interaction in man and conscious dogs. Circ Res 1986;59:178–93.
15. Tanaka M, Nishikawa T. Sevoflurane speeds recovery of baroreflex control of heart rate after minor surgical procedures compared with isoflurane. Anesth Analg 1999;89:284–9.
16. Brown TE, Beightol LA, Koh J, Eckberg DL. Important influence of respiration on human R-R interval power spectra is largely ignored. J Appl Physiol 1993;23:2310–7.
17. Pagani M, Somers V, Furlan R, et al. Changes in autonomic regulation induced by physical training in mild hypertension. Hypertension 1988;12:600–10.
18. DeBoer RW, Karemaker JM, Strackee J. Hemodynamic fluctuations and baroreflex sensitivity in humans: a beat-to-beat model. Am J Physiol 1987;253:H680–9.
19. Macor F, Fagard R, Vanhaecke J, Amery A. Respiratory-related blood pressure variability in patients after heart transplantation. J Appl Physiol 1994;76:1961–2.
20. Yamamoto Y, Hughson RL, Peterson JC. Autonomic control of heart rate during exercise studied by heart rate variability spectral analysis. J Appl Physiol 1991;71:1136–42.
21. 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.
22. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10.
23. Wang YP, Shih RL, Huang CL, et al. Differential changes in cardiac baroreflex sensitivity estimated by sequence and spectral analysis during etomidate anesthesia. Clin Auton Res 2000;10:117–21.
24. Lipman RD, Slisbury JK, Taylor JA. Spontaneous indices are inconsistent with arterial baroreflex gain. Hypertension 2003;42:481–7.
25. Kingwell BA, Thompson JM, Kaye DM, et al. Heart rate spectral analysis, cardiac norepinephrine spillover, and muscle sympathetic nerve activity during human sympathetic nervous activation and failure. Circulation 1994;90:234–40.
26. Kato M, Komatsu T, Kimura T, et al. Spectral analysis of heart rate variability during isoflurane anesthesia. Anesthesiology 1992;77:669–74.
27. Donchin Y, Feld JM, Porges SW. Respiratory sinus arrhythmia during recovery from isoflurane-nitrous oxide anesthesia. Anesth Analg 1985;64:811–5.
28. Van de Borne P, Schintgen M, Niset G, et al. Does cardiac denervation affect the short-term blood pressure variability in humans? J Hypertens 1994;12:1395–403.
29. Nagasaki G, Tanaka M, Nishikawa T. The recovery profile of baroreflex control of heart rate after isoflurane or sevoflurane anesthesia in humans. Anesth Analg 2001;93:1127–31.
© 2005 International Anesthesia Research Society
30. Parlow JL, van Vlymen JM, Odell MJ. The duration of impairment of autonomic control after anticholinergic drug administration in humans. Anesth Analg 1997;84:155–9.