BJA Advance Access originally published online on April 9, 2008
British Journal of Anaesthesia 2008 100(6):772-779; doi:10.1093/bja/aen080
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Increased non-stationarity of heart rate during general anaesthesia with sevoflurane or desflurane in children
1 Department of Pediatrics, Hanyang University College of Medicine, Seoul, South Korea
2 Department of Anesthesiology, Seoul National University College of Medicine, #28 Yongon-dong, Jongno-gu, Seoul 110-744, South Korea
* Corresponding author. E-mail: dami0605{at}snu.ac.kr
Accepted for publication February 21, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingReferences |
|---|
Background: During general anaesthesia, the most prominent change in heart rate variability (HRV) is a decrease in the magnitude of heart rate (HR) oscillation in the high- and low-frequency ranges. In children receiving sevoflurane or desflurane, we observed a significant increase in HR non-stationarity, that is, a significant change of mean HR over time. The aim of our study was to describe this increased non-stationarity and compare it with the decrease in the magnitude of HR oscillation.
Methods: Sixty children received sevoflurane (n=30) or desflurane anaesthesia (n=30). The magnitude of HR oscillation and non-stationarity during pre-anaesthesia and anaesthesia were measured by spectral and Hurst analyses using structure function, respectively.
Results: Low- and high-frequency powers decreased significantly and the very-short-term (2
8 s, H
) and short-term Hurst exponent (845 s, Hβ) increased significantly during the anaesthetic period compared with the pre-anaesthetic period, regardless of the anaesthetic agent [sevoflurane: mean (SD) H 0.414 (0.169) vs 0.252 (0.0655), Hβ 0.481 (0.169) vs 0.078 (0.0409); desflurane H 0.336 (0.171) vs 0.261 (0.0614), Hβ 0.471 (0.221) vs 0.0813 (0.049)]. Stepwise discriminant analysis showed that the short-term Hurst exponent was better than the spectral indices at differentiating between the pre-anaesthetic period and anaesthetic period.
Conclusions: During sevoflurane and desflurane anaesthesia in children, there is a significant increase in very-short-term and short-term HR non-stationarity. Furthermore, the greater short-term non-stationarity differentiates better between the pre-anaesthesia and anaesthesia than the decreased magnitude of HR oscillation in the high- and low-frequency ranges.
Keywords: anaesthetics volatile, sevoflurane, desflurane; children; heart, heart rate
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
Characterization of heart rate variability (HRV) is valuable in children and in adults for assessing alterations in heart rate (HR) regulation mechanisms, including autonomic modulation, and for predicting short- and long-term disease outcomes.1–6 Likewise, characterization of HRV during general anaesthesia in children may be useful in the investigation of the influence of anaesthetic agents cardiovascular regulation,7–11 and in evaluating or predicting the anaesthetic or post-anaesthetic outcomes such as anaesthetic depth,12–14 recovery time, and the magnitude of unstable fluctuations in systemic arterial pressure and HR.15–18
During general anaesthesia, the most frequently used method of HRV characterization is quantification of frequency-domain oscillation using spectral analysis of the HR. Low-frequency (0.04–0.15 Hz) and high-frequency (0.15–0.4 Hz) oscillations are markedly suppressed19 20 so that using this method alone may be unhelpful.
We observed markedly increased non-stationarity of HR during general anaesthesia in children. Non-stationarity of HR is defined as a significant change in the mean value of HR with time, that is, a progressive bradycardia or tachycardia. Sometimes oscillation is so greatly suppressed that non-stationarity dominates the HRV. Even when oscillation is less suppressed, it is superimposed on non-stationarity behaviour. Accordingly, for more complete HRV characterization during general anaesthesia, increased non-stationarity needs to be evaluated in addition to oscillation.
The aim of our study was to demonstrate the markedly increased non-stationarity in HR that occurs during general anaesthesia with sevoflurane and desflurane. Another aim was to assess the value of non-stationarity in differentiating between the pre-anaesthetic and anaesthetic states.
In order to evaluate the oscillatory and non-stationarity behaviour of HR, we performed power spectral analysis and Hurst analysis using the structure function of HR. This form of Hurst analysis has an inherent advantage for calculation of non-stationarity,21 22 unlike the other Hurst analysis methods such as rescaled range analysis, scaled windowed variance, and dispersional analysis.23 24
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
This study was approved by the Committee of Clinical Investigation of our institution. Sixty children undergoing elective orthopaedic surgery, ASA physical status I and II, aged 2–8 yr were enrolled in the study (Table 1). Thirty were given general anaesthesia with sevoflurane and the other 30 received desflurane. Informed consent was obtained from the parents. Patients with cardiovascular abnormalities, developmental delays, seizures, and metabolic or pulmonary diseases were excluded.
|
The day before surgery, all children visited the monitoring room, and the ECG was recorded in the pre-anaesthetic state for 20 min. This room is beside the operating room and has a quiet environment for patients and their parents. The ECG was recorded and stored into a personal computer by the BioPac MP 100 system (BioPac system Inc., CA, USA) and its companion software AcqKnowledge 3.7.3 (BioPac system Inc.). The sampling rate was 1000 Hz. On the day of surgery, children arrived at the operating room without premedication. ECG, arterial pressure, and SpO2 (Solar 8000M, GE, Milwaukee, WI, USA) were monitored. Anaesthesia was induced with thiopental sodium 6 mg kg–1. After disappearance of the eye-lash reflex, patients were ventilated with 8% of sevoflurane in oxygen 100% and received rocuronium 0.6 mg kg–1 for relaxation. After tracheal intubation, the inhalation agent was changed to desflurane for the patients in the desflurane group, otherwise sevoflurane was continued. End-tidal concentrations of inhalation agent were controlled to 1.0 MAC of each agent with nitrous oxide 66%. During anaesthesia, tidal volume was 8 ml kg–1, and the respiratory rate (RR) was set according to age. At age 1–2 yr, the RR was 24 min–1; at age 3–8 yr, 20 min–1; at age 9–12 yr, 16 min–1; and at age 13 or more, RR was 12 min–1. RRs were then adjusted to achieve an end-tidal CO2 of 4.7–5.3 kPa. The cardiovascular effect of thiopental does not last more than 20 min.25 Therefore, ECG data were recorded for 20 min at a steady end-tidal concentration of volatile agent and at least 20 min after induction. The anaesthetic period was defined as the period when the ECG data were recorded during anaesthesia. No surgical intervention was performed during the recording of ECG and no opiates were administered.
Data preprocessing
All ECG data were reviewed by visual inspection. Segments were discarded in cases of signal loss, significant noise, and the presence of atrial or ventricular extrasystoles. R peak was determined by the algorithm described by Pan and Tompkins.26 27
RR intervals (RRI) were measured and tested for the presence of outliers (HR <30 or >200 beats min–1). Any outliers were removed. All RRI data were linearly interpolated at 1000 Hz to construct a real-time series of RRIs, and then re-sampled at 2 Hz. The 1800 points (15 min of data) were used for calculating HRV indices. All data processing, including visual inspection, spectral analysis, and Hurst analysis, was conducted using Matlab 7.0.1 (Mathwork Inc., MA, USA).
Power spectral analysis
Power spectral analysis was by a fast Fourier transformation.28 The RR interval data were divided into 512 data point sections (4.26 min of data). After being detrended, these sections were fast Fourier transformed and averaged to form a power spectral density function. We calculated total power (TP), very-low-frequency power (VLFP), low-frequency power (LFP), and high-frequency power (HFP) by integrating the power spectral density curve in the 0–0.4, 0–0.04, 0.04–0.15, and 0.15–0.4 Hz ranges, respectively. Normalized LFP (nLFP) and normalized HFP (nHFP) were then calculated by the following formula:29
|
|
|
|
Hurst analysis using structure function21 22 30
The tth RRI was denoted as RRI (t) and we calculated the fluctuations in the difference as the structure function of for different time increments, s. 
y(t) is the magnitude of the change in RRI after s. In order to perform Hurst analysis, this study estimated the average of these local variations, which depended on the scaled time increments shown in the following equation:
|
| (1) |
indicates the presence of the power law behaviour of <y(t)> against |
|
y(t)>, this value was plotted for log2 ()=2...10. The scaling region was then obtained and the slopes of <y(t)> were calculated against at the scaling region, thus yielding the Hurst exponent, H.
When the values of log2 <y(t)> vs log2 were plotted, two linear scaling relations could be observed approximately at 28 and 845 s. Therefore, very-short-term and short-term generalized Hurst exponents (H, 28, and Hβ, 845 s, respectively) could be obtained from equation (1). The very-short-term Hurst exponent was defined as H and the short-term Hurst exponent was defined as Hβ. Since H is the slope of <y(t)> against in the log-log scale and H>0, this value measures the rate of increase in the mean RRI difference over time . In other words, the exponent measures how greatly the mean RRI changes as time advances. Accordingly, unlike the other methods of Hurst analysis, this measurement is well suited for measuring non-stationarity.
Statistical methods
Repeated-measures ANOVA was performed to compare the differences in HRV indices during pre-anaesthetic and anaesthetic periods. P-values <0.05 were considered significant. All values shown are mean (SD). To find the best discriminator among all of the variables between the pre-anaesthetic and anaesthetic states, we used stepwise discriminant analysis. The variable that produces the largest F value is the best discriminating variable selected, provided F is statistically significant. All statistical analyses were conducted using SPSS 12.0 (SPSS Inc., Chicago, IL, USA).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
Haemodynamic variables are shown in Table 1. HRV indices and comparisons between the indices during the pre-anaesthetic period and the anaesthetic period are presented in Table 2. Comparisons between those of the sevoflurane group and those of the desflurane group during the pre-anaesthetic and anaesthetic period are also presented in Table 2.
|
P<0.05, During sevoflurane anaesthesia, the mean and standard deviation of RRI did not change significantly. The TP, LFP, HFP, nLFP, and nHFP significantly decreased when compared with the pre-anaesthetic period (Table 2). During desflurane anaesthesia, mean, standard deviation, TP, LFP, HFP, nLFP, and nHFP decreased significantly when compared with the pre-anaesthetic values. H
, Hβ, and Hβ/H were significantly greater during sevoflurane and desflurane anaesthesia than during the pre-anaesthetic period. The values of TP, LFP, HFP, and nHFP of RRI during desflurane anaesthesia were significantly lower than those during the period of sevoflurane anaesthesia. However, all Hurst indices during the desflurane anaesthesia were not significantly different to those during sevoflurane anaesthesia. Stepwise discriminant analysis showed that the short-term Hurst exponent, Hβ, was best at differentiating between pre-anaesthetic and anaesthetic states for both sevoflurane (F value: 15.5, P=0.001, Table 2) and desflurane (F value: 65.9, P=0.000, Table 2).
Figure 1A and B displays HR tachograms during the pre-anaesthetic and anaesthetic, respectively, period that were calculated from the RRI sequences in Figure 1C and D, respectively. Figure 1C and D shows two representative 15 min RRI sequences recorded before and during, respectively, desflurane anaesthesia. The most marked change in HR behaviour during desflurane anaesthesia is the decreased magnitude of oscillation, which is further corroborated by the markedly decreased low- and high-frequency powers (345.4 and 92.5 ms2, respectively, in Fig. 2A, vs 1.1 and 2.0 ms2, respectively, in Fig. 2B). However, another prominent difference in HR behaviour is that during anaesthesia the mean of HR changes significantly with time, that is, the HR is significantly non-stationary (Fig. 1B, arrows), which is not the case in the pre-anaesthetic state (Fig. 1A). The increased non-stationarity of very short-term and short-term scales is further confirmed by higher H (0.2877 vs 0.4243) and markedly higher Hβ (0.0165 vs 0.8395).
|
|
Since Hβ best discriminated between anaesthetic and pre-anaesthetic periods, and Hβ is calculated from <
y> (845 s), we show how y and <y> differ between the pre-anaesthetic and anaesthetic states when increases from 8 to 45 s. Figure 1E and G displays the sequences of y=8s and y=45s calculated from the RRIs during the pre-anaesthetic state shown in Figure 1C. Each value of y=8s and y=45s is the magnitude of RRI change at 8 and 45 s after a given time point. The mean of y=45s (<y=45s>=32.4 ms, the horizontal line in Fig. 1E) is not significantly different from the mean of y=8s (<y=8s>=32.0 ms, the horizontal line in Fig. 1G). In other words, in the pre-anaesthetic state on average, the magnitude of change of RRI at 45 s after a certain instant does not differ from that at 8 s after the instant, which implies that the mean of the RRI is unchanged as time evolves from 8 to 45 s after the instant. Figure 1F and H displays the sequences of y=8s and y=45s calculated from the RRIs during the anaesthetic period shown in Figure 1D. The mean of the y=45s (<y=45s>=23.3 ms, the horizontal line in Fig. 1H) is significantly larger than the mean y=8s (<y=8s>=6.4 ms, the horizontal line in Fig. 1F). In other words, during anaesthesia, the mean change in the RRI at 45 s after a given time is significantly larger than that at 8 s. This indicates a large non-stationarity with passage of time. Though we have demonstrated only the time evolution from 8 to 45 s during anaesthesia, the change in mean RRI can be found in any short-term time period within the scale, such as that from 9 to 11 s. This is due to the fractal property of the scale invariance, that is, linear scaling behaviour, and a similar slope from 8 to 45 s (Fig. 2C and D).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
The major new finding of the present study was that, in children, the short-term and very-short-term Hurst exponents increased significantly during both sevoflurane and desflurane anaesthesia, compared with the pre-anaesthetic period. Another new finding was that for both sevoflurane and desflurane anaesthesia, the short-term Hurst exponent was better than other HRV indices, including spectral indices, in discriminating between pre-anaesthetic and anaesthetic periods.
This indicates that the HRV characteristic that differentiates between pre-anaesthesia and anaesthesia most effectively in children may be increased short-term non-stationarity rather than decreased low- and high-frequency oscillations. Since two different general anaesthetics both produced similar prominent changes, these changes are probably a non-specific response to general anaesthesia. Thus, we suggest that when characterizing HRV during general anaesthesia in children evaluation of short-term non-stationarity by short-term Hurst exponent and the evaluation of oscillation by spectral indices should be considered.
In the pre-anaesthetic period in the sevoflurane group and the desflurane group, the mean values of the short-term Hurst exponent were 0.078 and 0.081, respectively, that is, close to zero. During sevoflurane and desflurane anaesthesia, these mean values were 0.481 and 0.471, respectively, that is, close to 0.5. When the Hurst exponent value equals zero,21 22 the time series is completely anti-persistent, and when the time series increases then the Hurst exponent always decreases, and vice versa. When the Hurst exponent equals 0.5, the time series is random; when the time series increases, then an increase or decrease is equally probable. Accordingly, in the pre-anaesthetic period if HR increases at a certain time, then at 8–45 s after that time HR will usually decrease, and vice versa for initial decrease in HR. Consequently, the mean of HR on the short-term scale changes little, and is almost completely stationary. During anaesthesia in contrast, if HR increases, then 8–45 s later, there is an equal probability of a further HR increase or decrease. Accordingly, the mean of HR can change significantly after a short-term period and can be non-stationary on the short-term scale.
Although we have presented the results of the increased non-stationarity during general anaesthesia on a scale of less than 45 s, the increase also occurs on a greater scale. Figure 3B shows a HR tachogram recorded during sevoflurane anaesthesia. To highlight the non-stationarity for a larger scale of 5 min, we displayed a sequence of average values for the three successive, non-overlapping, 5 min long HR segments in the HR tachogram during the pre-anaesthetic period that is depicted in Figure 3A (C) and the anaesthetic period that is shown in Figure 3B (D). The mean HR continuously decreased during anaesthesia (D) but did not change in the pre-anaesthetic state. However, we could not present the results of increased non-stationarity on larger scales. This is because at the larger scales, the range of scales at which increased non-stationarity occurs differs greatly among patients, and no scale suitable for all patients could be found.
|
The Hurst exponent quantifies the magnitude of non-stationarity but does not capture the direction (or polarity) of the non-stationarity. Negative polarity and positive polarity indicate progressive bradycardia (Fig. 1B, broken arrows) and progressive tachycardia (Fig. 1B, arrow), respectively. Zero polarity indicates stationarity—no progressive change in mean HR. It could well be that a progressively more prominent bradycardia develops as there is a deepening of depth of anaesthesia in the absence of surgical stimuli. In that case, the non-stationarity is not an indication of stable anaesthesia, but of a deepening of anaesthesia when the steady state is not yet reached. This warrants further study into whether the non-stationarity of HR can be used as a probe for anaesthetic depth or a change in anaesthetic depth. Modification of the algorithm for the Hurst exponent to capture not only the magnitude of the non-stationarity but also the polarity would be needed.
In general, spectral analysis is based on a hypothesis of stationary signal. If the signal is non-stationary, the calculated spectral power contains energy originating not only from oscillation, but also from the non-stationarity. Since HR is highly non-stationary during general anaesthesia, the value of the HR spectral powers during general anaesthesia is influenced by the non-stationarity31 and by cardiac autonomic nervous activity. Furthermore, during general anaesthesia, autonomic activity is strongly suppressed, and the value is much more influenced by non-stationarity. Accordingly, when evaluating autonomic activity by a power spectral analysis of the HR, simultaneous evaluation of the degree of non-stationarity seems to be mandatory.
To identify increased non-stationarity as a phenomenon that occurs universally during general anaesthesia, future studies are needed in adults and to see whether i.v. anaesthetics have similar effects. Furthermore, the magnitude of the parasympathetic modulation of high-frequency oscillation (HFP) in middle-age adults and the autonomic modulation of low frequency oscillation (LFP) in the young adults are significantly smaller and larger, respectively, than in childhood.32 A future study is also required on whether spectral indices are worse than the short-term Hurst exponent in discriminating between pre-anaesthetic and anaesthetic states in adults.
Spectral analysis shows the difference in cardiac autonomic modulation between sevoflurane and desflurane in children. In adults,33–35 desflurane but not sevoflurane can cause neurocirculatory excitation manifest as increased HR, arterial pressure, and sympathetic nerve activity during induction and the transitional period when desflurane concentrations are rapidly increasing. Ebert and Muzi33 also showed that during steady-state anaesthesia desflurane results in a progressive increase in sympathetic nerve activity whereas sevoflurane does not. Our study was performed under steady-state sevoflurane or desflurane anaesthesia at 1 MAC concentration, a similar concentration to that used by Ebert. It seems from our study that desflurane-associated sympathetic excitation does not occur in children because the LFP and nLFP during desflurane anaesthesia were not increased but significantly and markedly decreased and because LFP and nLFP during desflurane anaesthesia were not higher than that during sevoflurane anaesthesia. When results of other spectral indices are analysed together, the effect on autonomic modulation between the two anaesthetics in children seems to be different from that in adults. In children, desflurane suppresses autonomic modulation more strongly than sevoflurane does. Unlike sevoflurane, desflurane suppresses parasympathetic modulation more than sympathetic modulation, which is supported by a significant decrease in RRI during the desflurane anaesthetic period (Table 2). We did not further investigate why the sympathetic activation that was observed in the adults during desflurane anaesthesia in Ebert's study was not observed in the children during desflurane anaesthesia in the current study. However, our result of the spectral analyses showed that the behaviour of the sympathetic nervous system during desflurane anaesthesia in children may be different from that in adults.
In this study, ventilatory rate was set according to age. Different rates may influence HRV, especially the HFP which is affected by respiratory-related vagal HR modulation.36 HFP values vary with tidal volume and RR, if RR is equal to or less than 8 bpm.37 Since in our study, ventilatory rate remained at 12–24 bpm in both groups, HFP was unaffected by ventilatory rate.
Non-stationarity markedly increased, despite the strong suppression of autonomic nerve activity by general anaesthesia. Thus, the non-stationarity may be related to the non-neural modulation of HR which may be due to stress hormones, including circulating catecholamines and cortisol, both of which are known to increase during sevoflurane anaesthesia.38 39 With marked suppression of the neural modulation of the HR, the non-neural modulation by stress hormones can dominate HR regulation and produce a markedly increased non-stationarity during the anaesthetic period.
If non-stationarity is related to non-neural HR regulation, both spectral and non-stationarity analysis will provide information about both neural and non-neural regulation of HR. Even if future studies fail to prove the relationship between non-stationarity and stress hormones, the short-term non-stationarity exponent provides another way of characterizing HRV in children during general anaesthesia. Future studies should therefore address the question of whether non-stationarity or spectral indices measured during general anaesthesia in children can be used to assess anaesthetic depth or to predict post-anaesthetic outcomes such as cardiovascular instability.
In conclusion, there is a significant increase in HR non-stationarity during sevoflurane or desflurane anaesthesia in children on the short- and very-short-term scales. The increase in short-term non-stationarity is more prominent than the decrease in HR oscillation. We suggest that short-term non-stationarity should be evaluated when characterizing HRV during sevoflurane or desflurane anaesthesia in children.
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
This work was supported by the research fund of Hanyang University (HY-2006-I).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingReferences |
|---|
1 Ohuchi H, Takasugi H, Ohashi H, et al. Stratification of pediatric heart failure on the basis of neurohormonal and cardiac autonomic nervous activities in patients with congenital heart disease. Circulation (2003) 108:2368–76.
2 Harrington C, Kirjavainen T, Teng A, et al. Altered autonomic function and reduced arousability in apparent life-threatening event infants with obstructive sleep apnea. Am J Respir Crit Care Med (2002) 165:1048–54.
3 Evrengul H, Tavli V, Evrengul H, et al. Spectral and time-domain analyses of heart-rate variability during head-upright tilt-table testing in children with neurally mediated syncope. Pediatr Cardiol (2006) 27:670–8.[CrossRef][Web of Science][Medline]
4 Franco P, Scaillet S, Groswasser J, et al. Increased cardiac autonomic responses to auditory challenges in swaddled infants. Sleep (2004) 27:1527–32.[Web of Science][Medline]
5 Ohuchi H, Hasegawa S, Yasuda K, et al. Severely impaired cardiac autonomic nervous activity after the Fontan operation. Circulation (2001) 104:1513–8.
6 Franco P, Seret N, Van Hees JN, et al. Decreased arousals among healthy infants after short-term sleep deprivation. Pediatrics (2004) 114:e192–7.
7 Kanaya N, Hirata N, Kurosawa S, et al. Differential effects of propofol and sevoflurane on heart rate variability. Anesthesiology (2003) 98:34–40.[CrossRef][Web of Science][Medline]
8 Constant I, Abbas M, Boucheseiche S, et al. Non-invasive assessment of cardiovascular autonomic activity induced by brief exposure to 50% nitrous oxide in children. Br J Anaesth (2002) 88:637–43.
9 Luginbühl M, Yppärilä-Wolters H, Rüfenacht M, et al. Heart rate variability does not discriminate between different levels of haemodynamic responsiveness during surgical anaesthesia. Br J Anaesth (2007) 98:728–36.
10 Tanaka M, Kimura T, Goyagi T, et al. Gender differences in baroreflex response and heart rate variability in anaesthetized humans. Br J Anaesth (2004) 92:831–5.
11 Licker M, Spiliopoulos A, Tschopp JM. Influence of thoracic epidural analgesia on cardiovascular autonomic control after thoracic surgery. Br J Anaesth (2003) 91:525–31.
12 Tsuchiya S, Kanaya N, Hirata N, et al. Effects of thiopental on bispectral index and heart rate variability. Eur J Anaesthesiol (2006) 23:454–9.[CrossRef][Web of Science][Medline]
13 Weber F, Bein T, Hobbhahn J, et al. Evaluation of the alaris auditory evoked potential index as an indicator of anesthetic depth in preschool children during induction of anesthesia with sevoflurane and remifentanil. Anesthesiology (2004) 101:294–8.[CrossRef][Web of Science][Medline]
14 Hans P, Dewandre PY, Brichant JF, et al. Comparative effects of ketamine on bispectral index and spectral entropy of the electroencephalogram under sevoflurane anaesthesia. Br J Anaesth (2005) 94:336–40.
15 Kim HS, Kim SD, Kim CS, et al. Prediction of the oculocardiac reflex from pre-operative linear and nonlinear heart rate dynamics in children. Anaesthesia (2000) 55:847–52.[CrossRef][Web of Science][Medline]
16 Fujiwara Y, Kurokawa S, Asakura Y, et al. Correlation between heart rate variability and haemodynamic fluctuation during induction of general anaesthesia: comparison between linear and non-linear analysis. Anaesthesia (2007) 62:117–21.[CrossRef][Web of Science][Medline]
17 Yum MK, Kim HS. Prediction of severity of mean arterial pressure elevation after tracheal intubation in hypertensive patients by preanesthetic recurrence quantification analysis of heart rates. Acta Anaesthesiol Scand (2003) 47:475–81.[CrossRef][Web of Science][Medline]
18 Yum MK, Oh AY, Lee HM, et al. Identification of patients with childhood moyamoya diseases showing temporary hypertension after anesthesia by preoperative multifractal Hurst analysis of heart rate variability. J Neurosurg Anesthesiol (2006) 18:223–9.[CrossRef][Web of Science][Medline]
19 Kato M, Komatsu T, Kimura T, Sugiyama F, Nakashima K, Shimada Y. Spectral analysis of heart rate variability during isoflurane anesthesia. Anesthesiology (1992) 77:669–74.[CrossRef][Web of Science][Medline]
20 Galletly DC, Westenberg AM, Robinson BJ, Corfiatis T. Effect of halothane, isoflurane and fentanyl on spectral components of heart rate variability. Br J Anaesth (1994) 72:177–80.
21 Yu CX, Gilmore M, Peebles WA, et al. Structure function analysis of long-range correlations in plasma turbulence. Phys Plasma (2003) 10:2772–9.[CrossRef]
22 Hurst HE. Long term storage of reservoirs. Trans Am Soc Civil Eng (1951) 116:770–99.
23 Bassingthwaighte B, Raymond GM. Evaluation of the dispersional analysis method for fractal time series. Ann Biomed Eng (1995) 23:491–505.[Web of Science][Medline]
24 Mandelbrot BB, Wallis JR. Robustness of the rescaled range R/S in the measurement of noncyclic long run statistical dependence. Water Resour Res (1969) 5:967–88.[CrossRef]
25 Mather LE, Duke CC, Ladd LA, et al. Direct cardiac effects of coronary site-directed thiopental and its enantiomers: a comparison to propofol in conscious sheep. Anesthesiology (2004) 101:354–64.[CrossRef][Web of Science][Medline]
26 Pan J, Tompkins WJ. A real-time QRS detection algorithm. IEEE Trans Biomed Eng (1985) 32:230–6.[Web of Science][Medline]
27 Tompkins WJ. Biomedical Digital Signal Processing (1993) Ch. 12. New Jersey: Prentice Hall International Inc. 236–64.
28 Akselrod S, Gordon D, Ubel FA, et al. Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat to beat cardiovascular control. Science (1981) 213:220–2.
29 Hanss R, Bein B, Turowski P, et al. The influence of xenon on regulation of the autonomic nervous system in patient at high risk of perioperative cardiac complications. Br J Anaesth (2006) 96:427–36.
30 Yum MK, Kim JH. A very-short-term intermittency of fetal heart rates and developmental milestone. Pediatr Res (2003) 53:915–9.[CrossRef][Web of Science][Medline]
31 Haaksma J, Brouwer J, Dijk WA, Mulder LJM, Crijns HJGM, Lie KI. Heart rate dependent changes in spectral analysis. Proc Comp Cardiol (IEEE) (1994) 20:45–8.
32 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) 100:393–9.
33 Ebert TJ, Muzi M. Sympathetic hyperactivity during desflurane anesthesia in healthy volunteers: a comparison with isoflurane. Anesthesiology (1993) 79:444–53.[CrossRef][Web of Science][Medline]
34 Weiskopf RB, Moore MA, Eger EI 2nd, et al. Rapid increase in desflurane concentration is associated with greater transient cardiovascular stimulation than with rapid increase in isoflurane concentration in humans. Anesthesiology (1994) 80:1035–45.[Web of Science][Medline]
35 Ebert TJ, Muzi M, Lopatka CW. Neurocirculatory responses to sevoflurane in humans: a comparison to desflurane. Anesthesiology (1995) 83:88–95.[CrossRef][Web of Science][Medline]
36 Yodlowski EH, Introna RPS, Montano N, Crumrine RS. Heart rate variability and neural control of the circulation: implications for anesthesiologists. Adv Anesth (1998) 15:33–74.
37 Pomeranz B, Macaulay RJ, Caudill MA, et al. Assessment of autonomic function in human by HR spectral analysis. Am J Physiol (1985) 248:H151–3.[Web of Science][Medline]
38 Shono A, Sakura S, Saito Y, et al. Comparison of 1% and 2% lidocaine epidural anaesthesia combined with sevoflurane general anaesthesia utilizing a constant bispectral index. Br J Anaesth (2003) 91:825–9.
39 Yoo K, Hwang J, Jeong S, et al. Anesthetic requirements and stress hormone responses in spinal cord-injured patients undergoing surgery below the level of injury. Anesth Analg (2006) 102:1223–8.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||