It has been repeatedly shown [1-3] that heart rate variability (HRV) diminishes during anaesthesia, but the physiological background and exact mechanisms behind this phenomenon are not completely understood. The influence of anaesthesia on HRV has been proposed to vary depending on the anaesthetic agent . Kanaya and colleagues demonstrated recently that during the induction phase of anaesthesia, propofol but not sevoflurane decreases the high-frequency (HF) component of HRV . Since the rapid regulation of heart rate (HR) is mainly parasympathetically mediated, the observed reduction in HF HRV can be considered to indicate decreased cardiac vagal activity . However, these earlier studies have not assessed deeper surgical levels of anaesthesia and the concentration-effect response has not been fully characterized.
In this study, we explored the effect of propofol and sevoflurane on HRV during deeper, surgical levels of anaesthesia, with and without nitrous oxide. Electrocardiography (ECG) data were collected from 24 healthy volunteers participating in two positron emission tomography (PET) studies that assessed regional cerebral blood flow and oxygen metabolism during general anaesthesia [6,7]. Our hypothesis was that although propofol and sevoflurane differ markedly in their pharmacological properties, their effects on HRV under comparable depth of anaesthesia and rigorously controlled conditions should be highly similar. We also wanted to further characterize the dose-response effects of these two anaesthetics.
The data for this article were collected during two PET studies, which assessed regional cerebral blood flow and oxygen metabolism during various depths of general anaesthesia [6,7]. The study protocols were approved by the local hospital Ethics Committee. Twenty-four healthy (ASA I) male subjects aged 20-30 yr were recruited after written informed consent. In the first part of the study (Part I), 16 subjects were openly allocated to sevoflurane (the first eight subjects) and propofol (the latter eight) groups, and the stepwise increasing drug concentration levels were targeted to levels 1.0, 1.5 and 2.0 minimal alveolar concentration/effective concentration 50 (MAC/EC50) at 30 min intervals in both groups. In the second part (Part II), a third group (n = 8) participated in an open, five-period study with four anaesthetic regimens at 50 min intervals. The depth of anaesthesia was maintained as close as possible to bispectral index (BIS®) 40 and the anaesthetics were changed three times along the protocol: first from sevoflurane to a mixture of sevoflurane and nitrous oxide (N2O), then to propofol, and at last to a mixture of propofol and N2O.
No premedication was used. In the sevoflurane group of Parts I and II, anaesthesia was induced with mask induction and 8% sevoflurane (Sevorane; Abbot Oy, Espoo, Finland). In the propofol group of Part I, subjects were anaesthetized with intravenous (i.v.) target-controlled infusion of propofol (Diprivan, 20 mg mL−1, AstraZeneca Oy, Masala, Finland) using Harvard 22 syringe pump (Harvard Apparatus, South Natick, MA, USA) and Stanpump software (Steven L. Shafer, Palo Alto, CA, USA), with pharmacokinetic parameters by Marsh and colleagues . The initial target plasma concentration was set to 6 μg mL−1. After the loss of eyelid reflex, 0.6 mg kg−1 rocuronium was administered i.v. and a laryngeal mask was placed (Part I) or subjects were intubated (Part II). Mechanical ventilation (tidal volume) was adjusted so that the individual end-tidal carbon dioxide (etCO2) values remained constant throughout the anaesthesia . The respiratory rate was fixed to 15 breaths min−1 during the anaesthesia. The end-tidal concentration of CO2 was kept strictly at 4.5%. Muscle relaxation was monitored using train-of-four and maintained with bolus doses of rocuronium (5-10 mg).
In Part I, the target end-tidal sevoflurane concentrations were 2% (1 MAC), 3% (1.5 MAC) and 4% (2.0 MAC), and target propofol plasma concentrations were 6 μg mL−1 (1.0 EC50), 9 μg mL−1 (1.5 EC50) and 12 μg mL−1 (2.0 EC50). Effective plasma concentration 50 (EC50) refers to the drug concentration that has been shown to prevent the response to surgical stimulus in 50% of subjects . At the end of each propofol concentration level, an arterial sample was collected for determination of plasma propofol concentrations with high-performance liquid chromatography as previously described. These measured propofol plasma concentrations were somewhat higher, i.e. 7.4 ± 0.7 (mean ± SD), 12.3 ± 2.6 and 18.3 ± 5.0 μg mL−1, respectively . Each of these pseudosteady state steps lasted for 30 min. As expected, both anaesthetics induced a concentration-dependent reduction in mean blood pressure (BP), which was greater in the sevoflurane group (on average, −40 mmHg during the deepest anaesthesia level) than in the propofol group (−26 mmHg) .
In Part II, the end-tidal concentration of sevoflurane was adjusted so that the depth of anaesthesia, measured with BIS® monitor, was as close as possible to value 40. Approximately after 50 min, 70% N2O was added to the anaesthetic regimen and the depth of anaesthesia was maintained by adjusting the sevoflurane vaporizer. Approximately after another 50 min, the administration of sevoflurane and N2O was terminated, oxygen-air mixture restarted and target-controlled infusion of propofol initiated (target plasma concentration 4 μg mL−1). Approximately after 50 min, N2O was added to propofol. During both propofol regimens, the target concentration was adjusted at 0.1-0.2 μg mL−1 steps to maintain a constant anaesthetic depth (value 40 in BIS® monitor). Maintaining BIS® at 40 with sevoflurane alone required end-tidal concentrations of 1.1-1.9%. After the addition of N2O, sevoflurane concentrations could be reduced by 22%, on average, while maintaining a steady BIS® level. Measured plasma concentrations of propofol were 2.6-4.6 μg mL−1, and the introduction of N2O did not allow the propofol concentrations to be reduced. All anaesthetic regimens reduced the mean arterial pressure, on average, by 18% in comparison with awake values .
After the anaesthesia, residual neuromuscular block was reversed with neostigmine/glycopyrrolate combination. The subjects were extubated and they were monitored until the vital functions had been stable for at least an hour.
ECG recordings and analysis of HRV
ECG data were recorded using a Medilog FD-3 digital Holter device (Oxford Medical Ltd, Woking, UK) with a temporal resolution of 1024 Hz. During baseline, the breathing was fixed to 15 min−1 for a period of 5 min with the aid of a sound generator. The Holter ECG recordings were saved in a microcomputer, R-R interval (RRI) data were filtered and beat-by-beat time series of RRIs were generated using Oxford Medilog ECG software (Medilog Cardiology Information System V1.1, Oxford Medical Ltd, Woking, UK), and then analysed using WinCPRS software package for biosignals (Absolute Aliens Oy, Turku, Finland).
The linear time and frequency domain analyses of HRV were based on 200-beat periods registered at the end of each pseudosteady state level or anaesthetic regimen. Nonlinear analysis (approximate entropy, ApEn) was computed at baseline and at the end of each level or regimen from 1000-beat periods of RRI data, in order to improve mathematic reliability . During baseline recordings at the awake state, the segments for the linear analyses were selected from the ectopic-free parts of the recording within the fixed-breathing period, to minimize the effect of mechanical ventilation on the analyses . The mean HR and two-time domain measures were calculated: the percentage of differences between RRIs greater than 50 ms (pNN50) and the square root of the mean squared differences of successive RRIs (RMSSD) . Spectral analysis was performed using fast Fourier transformation. The power spectra were quantified by measuring total power and the areas in the low-frequency (LF: 0.04-0.15 Hz) and high-frequency (HF: 0.15-0.40 Hz) bands . The LF/HF ratio was also calculated. ApEn was analysed using fixed values of the two input variables: number of observations (m) was set to 2, and filter level (r) was set to 20% of the SD of the data sets .
In Part I, RRI and different measures of HRV were analysed with repeated measures analysis of variance (RM ANOVA) having the drug (sevoflurane or propofol) as a between-factor and the depth of anaesthesia (0, 1.0, 1.5 and 2.0) as a within factor. In Part II, statistical analyses were carried out using RM ANOVA with the anaesthetic regimen as a within factor. RMSSD and the spectral powers (total power, LF, HF, LF/HF) were log-transformed before analyses to meet the assumption of normality. When a significant drug or drug-by-depth interaction effect was detected, the analysis was continued with paired comparisons using linear contrasts within the same model. pNN50 was analysed with non-parametric Friedman's test for overall effect, and with Wilcoxon signed rank sum test for paired comparisons. Statistical analyses were conducted with SAS (Version 8.02, SAS Institute Inc., Cary, NC, USA). A two-sided P-value of <0.05 was considered statistically significant. Data are presented as mean ± SD if not otherwise stated.
The effects of sevoflurane and propofol anaesthesia on the different HRV measures are presented in Table 1. The shift from awake state to 1.0 MAC/EC50 anaesthesia accelerated the average HR by 28 ± 13 beats min−1 in the propofol group (P < 0.001). In deeper anaesthesia, the HR decreased. RMSSD, pNN50, total power and HF power were all markedly and similarly suppressed during both the sevoflurane and propofol anaesthesia, already at the first 1.0 MAC/EC50 step (Table 1). Propofol reduced the LF power less than sevoflurane did, and increased the LF/HF ratio (Table 1). Both anaesthetics decreased ApEn, but at different stages of anaesthesia. Propofol altered HR dynamics already at 1 EC50, while in the sevoflurane group, significant reduction of ApEn was observed only during the deepest (2 MAC) anaesthesia level (Table 1).
In the mean HR, there were no differences between the awake state and the anaesthesia regimens. RMSSD and pNN50 were significantly reduced during both sevoflurane and propofol anaesthesia, similarly with and without adjunct N2O (Table 2, Fig. 1). All anaesthesia regimens suppressed the HF, LF and total powers of RRI variability compared to baseline, but there were no significant differences between the four anaesthetic regimens (Table 2). However, the increase in the LF/HF ratio after propofol alone differed markedly from the effects of the other three regimens (Table 2). ApEn tended to decrease during propofol anaesthesia (P = 0.031, uncorrected).
Vagal tonic activity is believed to be responsible for most of the rapid HRV changes during rest . The measures of rapid HRV - RMSSD, pNN50 and the HF power - are known to reflect cardiac vagal efferent activity , and they are strongly suppressed by parasympathetic blockade . In our data, the detected reductions in these indices during anaesthesia were not associated with significant increases in HR, except at 1 EC50 in the propofol group in Part I. This could indicate that the sympathetic drive was suppressed as well, and the reciprocal sympathovagal mechanism was impaired during anaesthesia. The tachycardic reaction at 1 EC50 propofol may have been a response to insertion of the laryngeal mask, since the analgesic properties of propofol are weak and no additive analgesics were used. The lack of such tachycardic response during propofol in Part II was probably due to the fact that the subjects were intubated at earlier stages of the study.
Our findings indicate a marked reduction of cardiac parasympathetic activity during sevoflurane anaesthesia. In fact, the effect of sevoflurane appeared to be very similar to the effect of propofol at a comparable depth of anaesthesia. The 70-99% average reductions of RMSSD and HF power during sevoflurane anaesthesia seem to indicate a rather prominent vagolytic effect of the drug. For comparison, we have previously demonstrated that supramaximal anticholinergic blockade with glycopyrrolate decreases RMSSD by 97% and HF power by more than 99% in a similar group of healthy subjects . General depression of HRV and HF power during anaesthesia has also been described earlier with another volatile anaesthetic, isoflurane . In PET imaging studies carried out by Pomfrett and Alkire, the attenuation of respiratory sinus arrhythmia (RSA) during isoflurane anaesthesia correlated with the reduction in whole brain glucose metabolic rate, and the level of RSA correlated with functional regional metabolic activity in the right nucleus tractus solitarii of the medulla oblongata [15,16]. Furthermore, our findings with ApEn indicate that the complexity of HRV decreases especially during deeper levels of anaesthesia.
Our subjects were mechanically ventilated. Since the HF power of HRV is strongly influenced by the breathing pattern , we standardized the breathing frequency according to current guidelines . It should be acknowledged, however, that mechanical ventilation produces an unphysiological condition, in which pulmonary pressures are reversed and normal respiratory sinus arrhythmia may be disturbed. Thus, the influence of mechanical ventilation on our current HRV results cannot be entirely excluded. The effects of mechanical ventilation and sevoflurane anaesthesia on HRV have recently been studied by Nakatsuka and colleagues , who also discovered a reduction of HF power during sevoflurane anaesthesia. At 2% sevoflurane, they registered slightly higher values of HF power during mechanical ventilation than during apnoea, but in deeper anaesthesia (defined by electroencephalogram suppression), this difference was abolished. Therefore, our results are not likely to be explained by mechanical ventilation, but rather by the anaesthesia itself. It should also be noted that estimation of the influence of artificial ventilation on HRV per se is very difficult because mechanically ventilated subjects are usually anaesthetized, sedated or critically ill. Furthermore, Pöyhönen and colleagues  have recently shown that high carbon dioxide concentrations increase HF and LF components of HRV, but this reactivity is abolished during anaesthesia. This notion lends support to the assumption that the influence of respiratory control mechanisms on HRV during anaesthesia is rather limited.
It has been demonstrated that during induction of anaesthesia, spontaneous breathing is more often preserved with sevoflurane than propofol . This phenomenon may also, in part, explain the differences between sevoflurane and propofol in the study by Kanaya and colleagues  in which the subjects breathed spontaneously during constantly deepening anaesthesia. In the present investigation, we wanted to improve the assessment of rapid, parasympathetically mediated HRV by computing RMSSD and pNN50 also, which are not so easily affected by alterations of the breathing pattern . In our procedure, each change in the anaesthetic regimen was followed by a 20-min stabilization period before the measurements. This approach was supposed to provide more stable conditions for the HRV assessments, i.e. to resemble steady state. Of note, Keyl and colleagues  have previously suggested that propofol may destabilize haemodynamic control mechanisms, leading to irregular dynamic behaviour of the cardiovascular system.
Although sevoflurane and propofol both depressed short-term HRV in a very similar manner in our study, there were also differences between the anaesthetics. During light anaesthesia, the LF power was more suppressed by sevoflurane than by propofol, and, in consequence, propofol markedly increased the LF/HF ratio. The physiological background of LF power is somewhat controversial; it is considered to be influenced by both sympathetic and parasympathetic innervations of the heart in an interactive manner . However, it is possible that the LF reduction that we observed during deeper anaesthesia could reflect the attenuation of sympathetic regulation, because during a profound anaesthesia-associated parasympathetic blockade, the vagal control of HR should be at its minimum. Accordingly, the more prominent depression of LF power during sevoflurane anaesthesia could result from a stronger impact of the drug on the sympathetic nervous system. Indeed, the relative preservation of LF variability during propofol anaesthesia has been described earlier , and it is presumed to indicate a weaker sympathetic depression property of the drug. Unfortunately, there is no HRV index that specifically reflects the cardiac sympathetic activity. Ledowski and colleagues have recently studied stress response during volatile and i.v. anaesthesia titrated to BIS 45-55. Interestingly, the stress hormone levels during surgery did not seem to correlate with spectral HRV indices, and HRV stayed reduced even during surgical stimulus .
There are at least two possible reasons why the parasympathetically mediated short-term HRV diminishes profoundly during anaesthesia. First, anaesthetic agents cause vasodilatation and influence the contractility of the heart, thus decreasing arterial pressure, which, in turn, can reflectorily decrease vagal efferent activity on the sinus node. Although anaesthesia is known to impair the function of baroreflex as well [23-25], it could still be operative during light-to-moderate anaesthesia, and explain the profound depression of parasympathetic tone at 1 MAC/EC50. Second, anaesthetics may also depress function of the brainstem nuclei responsible for haemodynamic control. There is growing evidence that the LF fluctuations of HR are mainly of central origin [26-28]. Therefore, the reduction of LF power that we registered during deeper anaesthesia could be due to the central effects of the anaesthetic drugs.
ApEn quantifies the regularity and complexity of time series data: lower values of ApEn indicate a more regular (less complex) signal, while higher values indicate more irregular/complex data . Quite interestingly, our observation of the loss of complexity in HR dynamics (decreased ApEn) during anaesthesia seems to be the opposite to what takes place in physiologic sleep [29,30]. The physiological background for ApEn of HRV is not fully understood, but evidently both sympathetic and parasympathetic regulation play an important role in the genesis of complex RRI fluctuations . Decreased ApEn of HRV has been described in various cardiovascular diseases and pathologic conditions, such as preceding the spontaneous onset of arrhythmias . In our earlier published data , supramaximal anticholinergic blockade with glycopyrrolate decreased ApEn by 40% in a similar group of healthy male subjects. ApEn has been shown to correlate with the HF spectral power of HRV  which, as already depicted, is believed to reflect mainly cardiac parasympathetic regulation . In our propofol group, the HF power and ApEn were both strongly suppressed already at the lightest anaesthesia level (1.0 MAC/EC50). In the sevoflurane group, the HF power was also markedly decreased at the first anaesthesia level, but ApEn continued to decrease further during deepening anaesthesia. This could partly be due to propofol-induced tachycardia at the 1 EC50 level of Part I, as ApEn is known to be sensitive to baseline fluctuations .
The comparability of MAC and EC50 (Part I) and the use of BIS® 40 (Part II) as a reference to equivalent anaesthesia can be criticized. As discussed earlier , propofol induced deeper anaesthesia than sevoflurane, partly because target propofol concentrations were exceeded and also because our 1 MAC sevoflurane dose (2%) was a bit low (for the current age group MAC is actually 2.6% approximately) . This shortcoming probably underestimated the true difference between the drugs in lowering the LF power. In contrast, the observed differences in ApEn can probably be explained by the fact that the concentration levels were not equipotent. Because of these problems, we wanted to compare the anaesthetics also at an equal hypnotic depth of anaesthesia in Part II using the BIS® monitor. The main weakness of BIS® is its varying behaviour during N2O . Maintaining BIS® 40 through all regimens allowed us to reduce sevoflurane but not propofol during N2O. The non-randomized design of Part II can also be criticized, because anaesthesia regimens followed each other at 50 min intervals, which may have been somewhat short to abolish the residual effects of preceding drug administration. Furthermore, the limited number of study subjects also carries the risk for type II error. Nevertheless, our study design allowed us to compare several anaesthesia settings during both surgical and hypnotic anaesthesia levels. Although the standardized conditions during our study made it possible to minimize effects other than anaesthesia on HRV, we still find it important to study further the behaviour of HRV during stressful conditions like hypoxemia, hypercarbia or surgical stimulus.
In summary, the suppression of HRV was very similar during all anaesthetic regimens measured with classical linear methods, which indicates a profound cardiac parasympatholytic effect and also suggests considerable sympathetic suppression as well. Further research is needed to discover the clinical significance and interpretation of the decreasing complexity (ApEn) of RRI fluctuations during anaesthesia. Pomfrett  has previously presented a hypothesis of a ‘brainstem component’ of anaesthesia, which denotes a general depression of autonomic nervous system nuclei that could be a common mechanism of action of a wide range of anaesthetic agents and regimens. Even though it is quite probable that peripheral mechanisms also markedly contribute to the anaesthesia-associated changes in HRV, it should be kept in mind that various pathological conditions, especially those directly affecting heart and nervous system (e.g. hypertension, heart failure, ischaemic heart disease, diabetes and polyneuropathia), exert strong influences on HRV . Furthermore, various drugs can affect HRV by interfering with the cardiac autonomic regulation. In consequence, it is quite unlikely that any single HRV index could become a general indicator of anaesthetic depth, but it is possible that the analysis of HRV could be exploited in future monitoring devices for anaesthesia.
1. Galletly DC, Buckley DH, Robinson BJ, Corfiatis T. Heart rate variability during propofol anaesthesia. Br J Anaesth
1994; 72: 219-220.
2. Sleigh JW, Donovan J. Comparison of bispectral index, 95% spectral edge frequency and approximate entropy of the EEG, with changes in heart rate variability during induction of general anaesthesia. Br J Anaesth
1999; 82: 666-671.
3. O'Flaherty D. Heart rate variability and anaesthesia. Eur J Anaesthesiol
1993; 10: 419-432.
4. Kanaya N, Hirata N, Kurosawa S et al.
Differential effects of propofol and sevoflurane on heart rate variability. Anesthesiology
2003; 98: 34-40.
5. Penttilä J, Helminen A, Jartti T et al.
Time domain, geometrical and frequency domain analysis of cardiac vagal outflow: effects of various respiratory patterns. Clin Physiol
2001; 21: 365-376.
6. Kaisti KK, Metsähonkala L, Teräs M et al.
Effects of surgical levels of propofol and sevoflurane anesthesia on cerebral blood flow in healthy subjects studied with positron emission tomography. Anesthesiology
2002; 96: 1358-1370.
7. Kaisti KK, Långsjö JW, Aalto S et al.
Effects of sevoflurane, propofol, and adjunct nitrous oxide on regional cerebral blood flow, oxygen consumption, and blood volume in humans. Anesthesiology
2003; 99: 603-613.
8. Marsh B, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth
1991; 67: 41-48.
9. Davidson JA, Macleod AD, Howie JC et al.
Effective concentration 50 for propofol with and without 67% nitrous oxide. Acta Anaesthesiol Scand
1993; 37: 458-464.
10. Pincus SM. Approximate entropy as a measure of system complexity. Proc Natl Acad Sci USA
1991; 88: 2297-2301.
[11. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Eur Heart J
1996; 17: 354-381.
12. Kleiger RE, Stein PK, Bosner MS, Rottman JN. Time domain measurements of heart rate variability. Cardiol Clin
1992; 10: 487-498.
13. Hayano J, Sakakibara Y, Yamada A et al.
Accuracy of assessment of cardiac vagal tone by heart rate variability in normal subjects. Am J Cardiol
1991; 67: 199-204.
14. Kato M, Komatsu T, Kimura T et al.
Spectral analysis of heart rate variability during isoflurane anesthesia. Anesthesiology
1992; 77: 669-674.
15. Alkire MT, Pomfrett CJD. Toward a monitor of depth: bispectral index (BIS) and respiratorory sinus arrhytmia (RSA) both monitor cerebral metabolic reduction during isoflurane anesthesia. Anesthesiology
1997; V87: A421.
16. Pomfrett CJD, Alkire MT. Respiratory sinus arrhythmia as an index of anaesthetic depth in man: evidence from functional imaging studies. J Ph
1999; 518P: 180P.
17. Nakatsuka I, Ochiai R, Takeda J. Changes in heart rate variability in sevoflurane and nitrous oxide anesthesia: effects of respiration and depth of anesthesia. J Clin Anesth
2002; 14: 196-200.
18. Pöyhönen M, Syväoja S, Hartikainen J et al.
The effect of carbon dioxide, respiratory rate and tidal volume on human heart rate variability. Acta Anaesthesiol Scand
2004; 48: 93-101.
19. Joo HS, Perks WJ. Sevoflurane versus propofol for anesthetic induction: a meta-analysis. Anesth Analg
2000; 91: 213-219.
20. Keyl C, Schneider A, Dambacher M et al.
Dynamic cardiocirculatory control during propofol anesthesia in mechanically ventilated patients. Anesth Analg
2000; 91: 1188-1195.
21. Eckberg DL. Sympathovagal balance: a critical appraisal. Circulation
1997; 96: 3224-3232.
22. Ledowski T, Bein B, Hanss R et al.
Neuroendocrine stress response and heart rate variability: a comparison of total intravenous versus balanced anesthesia. Anesth Analg
2005; 101: 1700-1705.
23. Sato M, Tanaka M, Umehara S, Nishikawa T. Baroreflex control of heart rate during and after propofol infusion in humans. Br J Anaesth
2005; 94: 577-581.
24. 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: 1127-1131.
25. Ebert TJ, Muzi M. Propofol and autonomic reflex function in humans. Anesth Analg
1994; 78: 369-375.
26. Cooley RL, Montano N, Cogliati C et al.
Evidence for a central origin of the low-frequency oscillation in RR-interval variability. Circulation
1998; 98: 556-561.
27. Montano N, Porta A, Malliani A. Evidence for central organization of cardiovascular rhythms. Ann N Y Acad Sci
2001; 940: 299-306.
28. Bernardi L, Leuzzi S, Radaelli A et al.
Low-frequency spontaneous fluctuations of R-R interval and blood pressure in conscious humans: a baroreceptor or central phenomenon? Clin Sci (Lond)
1994; 87: 649-654.
29. Vanoli E, Adamson PB, Ba L et al.
Heart rate variability during specific sleep stages. A comparison of healthy subjects with patients after myocardial infarction. Circulation
1995; 91: 1918-1922.
30. Pikkujämsä SM, Mäkikallio TH, Sourander LB et al.
Cardiac interbeat interval dynamics from childhood to senescence: comparison of conventional and new measures based on fractals and chaos theory. Circulation
1999; 100: 393-399.
31. Singer DH, Baumgarten CM, Ten Eick RE. Cellular electrophysiology of ventricular and other dysrhythmias: studies on diseased and ischemic heart. Prog Cardiovasc Dis
1981; 24: 97-156.
32. Vikman S, Mäkikallio TH, Yli-Mäyry S et al.
Altered complexity and correlation properties of R-R interval dynamics before the spontaneous onset of paroxysmal atrial fibrillation. Circulation
1999; 100: 2079-2084.
33. Penttilä J, Helminen A, Jartti T et al.
Effect of cardiac vagal outflow on complexity and fractal correlation properties of heart rate dynamics. Auton Autacoid Pharmacol
2003; 23: 173-179.
34. Yeregani VK, Srinivasan K, Vempati S et al.
Fractal dimension of heart rate time series: an effective measure of autonomic function. J Appl Physiol
1993; 75: 2429-2438.
35. Pomeranz B, Macaulay RJ, Caudill MA et al.
Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol
1985; 248: H151-H153.
36. Fragen RJ, Dunn KL. The minimum alveolar concentration (MAC) of sevoflurane with and without nitrous oxide in elderly versus young adults. J Clin Anesth
1996; 8: 352-356.
37. Puri GD. Paradoxical changes in bispectral index during nitrous oxide administration. Br J Anaesth
2001; 86: 141-142.
38. Pomfrett CJ. Heart rate variability, BIS and ‘depth of anaesthesia’. Br J Anaesth
1999; 82: 659-662.
39. Goldberger AL, Amaral LA, Hausdorff JM et al.
Fractal dynamics in physiology: alterations with disease and aging. Proc Natl Acad Sci USA
2002; 99(Suppl 1): 2466-2472.