1 Staff Anaesthesiologist, Department of Anaesthesiology University Hospital of Bern, CH-3010 Bern, Switzerland. 2 Department of Anaesthesiology, Kantonsspital Münsterlingen, Switzerland. 3 Research Section, Department of Anaesthesiology University Hospital of Bern, Switzerland. 4 Department of Anaesthesiology and Intensive Care, University Hospital Reykjavik, Iceland *Corresponding author
Accepted for publication: April 16, 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Anaesthesia was induced and maintained with 2% sevoflurane and 50% nitrous oxide in two groups of 10 ASA status I patients. Tetanic stimuli were applied to the flexor muscles of the forearm and the ulnar nerve before and after administration of vecuronium. The change in skin blood flow (laser-Doppler) and pulse wave amplitude (pulse oximetry) after a 5 and 10 s stimulation was measured on the opposite hand. If skin blood flow (laser-Doppler) decreased by more than 10%, a computer-controlled infusion of alfentanil was started and the target plasma concentration was increased in steps until this response was suppressed (<10%). The trachea was intubated and arterial pressure and heart rate responses were recorded. Plasma alfentanil concentration was measured.
Results. When PWR and SVmR were suppressed, the haemodynamic response to tracheal intubation was reduced in 100 and 53% of patients respectively. PWR and SVmR responses decreased with increasing plasma alfentanil concentration. The SVmR response to muscle stimulation was reduced by muscle relaxants. The pulse wave response to both muscle and neural stimulation was reduced by relaxants. The responses to 5 and 10 s stimulations were similar.
Conclusion. An absent SVmR does not predict a blunted arterial pressure or heart rate response to tracheal intubation. The PWR may be a better predictor.
Br J Anaesth 2002: 89: 38997
Keywords: intubation, tracheal; measurement techniques, laser-Doppler flowmetry; measurement techniques, pulse wave amplitude; pain, experimental
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Shimoda and co-workers6 suggested that cutaneous vasoconstriction after electrical tetanic stimulation of the ulnar nerve (skin vasomotor reflex, SVmR) could predict the arterial pressure response to tracheal intubation. After suppressing the skin vasomotor reflex with sevoflurane, they observed a much reduced arterial pressure response to tracheal intubation.6 They used a 3 s, 40 mA, 50 Hz electrical stimulation of the ulnar nerve,6 which was different from the stimulus used by Zbinden and co-workers (10 s, 60 mA, 50 Hz to the forearm muscles).2 The effect of opioids and neuromuscular blockade on these responses was not assessed and the responses to muscle and neural stimulation were not compared.
Measurement of skin blood flow with laser-Doppler flowmetry is expensive and not used routinely in clinical practice. The amplitude of the pulse oximeter wave could be an alternative index of vascular tone7 and correlates with the laser-Doppler signal.8 We considered the possibility that measurement of pulse oximeter amplitude, which is widely available, might be as useful as the laser-Doppler method to indicate vasoconstriction. We measured the pulse wave reflex (PWR) as the minimal pulse wave amplitude after a painful stimulus divided by the mean pulse wave amplitude before the stimulus.
We set out to confirm that SVmR could predict a reduced haemodynamic response to tracheal intubation. We used an opioid instead of sevoflurane and assessed the PWR as a predictor of the haemodynamic response to intubation. We also assessed the effect of plasma alfentanil concentration and compared the effects of muscle relaxation and the effects of muscle and ulnar nerve stimulation on the two reflex responses.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Study plan
Patients were given omeprazole 40 mg (Antra®; Astra-Zeneca, Zug, Switzerland) on the evening before the study. In the operating theatre, standard monitoring devices were applied (Datex AS3 Compact, Datex-Ohmeda; Instrument arium Corporation, Helsinki, Finland), a venous cannula was inserted in the non-dominant arm, and patency of the ulnar artery of both arms was assessed with Allens test. On the non-dominant hand, an angled laser-Doppler flowmeter probe (PF 404; Perimed, Järfälla, Sweden) was placed on the fourth finger, a pulse-oximeter probe on the third finger and a skin temperature probe on the palm. The patient was then covered with a warming blanket (Bair Hugger; Augustine Medical, MN, USA). When the laser-Doppler signal was stable, the patients were asked to perform a brisk inspiration (inspiratory gasp), which should cause a short-lasting skin vasoconstriction.9 Patients without this reaction were excluded from the study as this indicates abnormal vasomotor function.9
First study
Anaesthesia was induced in the first group of 10 patients by vital capacity inhalation of 7 vol% sevoflurane in nitrous oxide/oxygen (ratio 1:1), and the end-tidal sevoflurane and nitrous oxide concentrations were kept at 2 and 50 vol% respectively. The patients were ventilated by facemask and the end-tidal carbon dioxide concentration was maintained between 35 and 45 mm Hg throughout the study. A 20-G radial arterial cannula was inserted in the non-dominant arm. A temperature probe was placed in the rectum.
Skin electrodes for electrical stimulation were placed along a line between a point 1 cm ulnar to the mid-point of the cubital fold (point A) and a point 1.5 cm ulnar to the mid-point of the first wrist fold (point B). The muscle electrodes were attached 5 and 13 cm distal to A2 and the ulnar nerve electrodes 15 and 23 cm distal to A.6 A 5 s, 60 mA, 50 Hz, 0.25 ms square-wave electrical stimulus10 was applied by a Digitimer DS 7 constant current stimulator with a Digitimer DG2 trigger generator (Digitimer, Welwyn Garden City, Hertfordshire, UK) and a timer device (constructed in our laboratory). The positive pole of the stimulator was attached proximally. The tetanic stimulation was always applied first to the muscles and then to the ulnar nerve. After 10 min of equilibration, the first tetanic stimulation was applied. The second stimulation followed 35 min after vecuronium 0.1 mg kg1 i.v., when at least three of the four twitches of the train-of-four were suppressed. Muscle relaxation was maintained at this level by further doses of vecuronium (0.02 mg i.v.) until the study was completed. If a positive laser-Doppler skin vasomotor reflex SVmR was present on ulnar nerve stimulation, an infusion of alfentanil was started to give a target plasma concentration of 25 ng ml1. A Harvard 22 infusion pump (Harvard Apparatus, South Natick, MA, USA) was controlled by an IBM-compatible laptop computer running the Stanpump program (S. L. Shafer, Anesthesiology Service (112A), PAVAMC, Palo Alto, CA, USA) using pharmacokinetic parameters for alfentanil.11 After each positive SVmR, the alfentanil target concentration was increased by 25 ng ml1 and the stimulations were repeated. As soon as the laser-Doppler SVmR was negative (for definition see below), tracheal intubation was performed by an experienced anaesthetist. The haemodynamic response to tracheal intubation was determined off-line from the heart rate computed from the ECG and the arterial pressure values (Datex AS3 compact, Datex-Ohmeda) recorded on a personal computer every 10 s (Labview version 5.0; National Instruments Corporation, Austin, TX, USA). An arterial pressure increase of more than 25 mm Hg and a heart rate increase of more than 25 beats min1 were considered relevant. The mean arterial pressure and heart rate in the 120 s before intubation were used as reference. The duration of the procedure from the insertion of the laryngoscope to the inflation of the cuff and the view of the vocal cords according to Cormack and Lehane12 were recorded.
Before each stimulation and before intubation, arterial blood samples were taken to measure plasma alfentanil concentration. Arterial blood gas measurements were made before the first tetanic stimulation and before intubation. Ten minutes after tracheal intubation, data acquisition was stopped and anaesthesia continued according to clinical routine.
Second study
The second group of 10 patients took part in this study. A 10 s, 60 mA, 50 Hz, 0.25 ms square-wave electrical impulse was used. The target concentration of alfentanil was initially 50 ng ml1 and was increased in steps of 50 ng ml1. In all other aspects the plan was as in the first study.
Laser-Doppler SVmR
Blood flow in the finger was measured continuously with a laser-Doppler flow system (Periflux 4001; Perimed). The time constant of the flowmeter output amplifier was set to 0.02 s. The raw signal of the laser-Doppler was digitized at 128 Hz (A/D conversion card; National Instruments Corporation) and stored on a computer, and was also displayed during the study (arbitrary perfusion units according to the manufacturers specifications). The laser-Doppler system was calibrated with a standard suspension of 2 mm latex spheres provided by the manufacturer (Perimed).
The mean laser-Doppler flow (perfusion units) was calculated at intervals of 1 s from the digitized raw signal. The skin vasomotor reflex amplitude (SVmR) was defined according to the equation
SVmR=1(b/a)(1)
where a is the mean laser-Doppler flow during the 120 s before stimulation and b is the minimal laser-Doppler flow after stimulation plus the standard deviation of a. The reflex was calculated for every stimulation according to the monitor readings and was considered absent if SVmR was 0.1 or less (i.e. the blood flow decreased by 10% or less; modified from Shimoda and colleagues).6
Pulse wave reflex
The pulse wave of the pulse oximeter (Datex AS3 Compact, Datex-Ohmeda) was scaled in arbitrary units from 100 to +100 and digitized at 128 Hz, like the laser-Doppler signal. The beat-to-beat amplitude of the pulse wave was measured off-line using software developed in our laboratory. The PWR was computed in the same way using equation (1), with the laser-Doppler flow replaced by the pulse wave amplitude.
Alfentanil plasma concentrations and performance of the target-controlled infusion
Blood samples for alfentanil measurement were immediately centrifuged at 3500 g for 30 min and the plasma was stored at 26°C. Alfentanil plasma concentration was measured by gas chromatographymass spectrometry13 with a detection limit of 0.04 ng ml1. We found coefficients of variation of 2.2, 2.5 and 4.7% at concentrations of 200, 50 and 5 ng ml1 respectively for measurements made on the same day. Between days we found coefficients of variation of 4.2, 6.8 and 7.5% respectively. Fentanyl was used as internal standard.
Data analysis and statistics
The characteristics of the two groups were compared using Fishers exact test and the Wilcoxon rank test as appropriate. Values are given as medians (first and third quartiles) or as mean (SD) if normally distributed.
Anaesthetic drug concentrations
From the measured alfentanil plasma concentrations of each subject, the prediction error of the pharmacokinetic model was calculated:
where Cm is the measured concentration and Cp is the predicted plasma alfentanil concentration. The median prediction error (MDPE; a measure of bias) and the median absolute prediction error (MDAPE=median |PE|; a measure of accuracy) were computed for each individual.14 Subsequently, the means of the MDPE (MDPE) and the MDAPE (MDAPE) were calculated for the groups.
The mean end-tidal sevoflurane and nitrous oxide concentrations for the period from 60 s before to 60 s after each stimulation were determined for each subject. Subsequently, the mean coefficients of variation of the sevoflurane and nitrous oxide concentrations were determined for the study population.
Prediction of response to tracheal intubation
The responses to the last ulnar nerve stimulation before tracheal intubation were analysed to estimate the probability of reduced arterial pressure and heart rate response to tracheal intubation if the laser-Doppler SVmR and the PWR were absent, i.e. the negative predictive value of these tests.
Effect of stimulation pattern on the reflex responses
The responses to stimulation before alfentanil administration in the first and second studies were compared (MannWhitney rank sum test) to detect a difference between the 5 and 10 s stimulations, considering muscle and ulnar nerve stimulation separately. As there was no difference between these responses, the results of the two studies were pooled for further analysis.
Effect of plasma alfentanil concentration on the reflex responses
The data were fitted to a non-linear regression model relating the laser-Doppler SVmR and PWR values to the corresponding measured plasma alfentanil concentrations according to the equation
y=y0xeaxx(3)
where y is the reflex (either PWR or SVmR) at plasma alfentanil concentration x, y0 is the reflex at zero alfentanil and a is the regression coefficient. The inter-individual variation of the estimated parameters (y0 and a) was modelled with an additive error model.
Pi=PTV+i(4)
where Pi is the parameter of the ith individual, PTV is the typical parameter value of the population, and i is the random inter-individual variation of the parameter (with mean 0 and variance
2).
The residual error of the predicted effect including intra-individual variation15 was also assumed to be additive:
Eij=ETVi+ij(5)
where Eij is the effect in the ith individual at the jth measurement, ETVi the typical effect value for the ith individual and j the residual variation of the effect at the jth measurement (with mean 0 and variance
2).
This non-linear mixed-effects model was implemented in Fortran pseudocode for use with the non-linear mixed-effects modelling program NONMEM.16 The NONMEM objective function was minus twice the logarithm of the likelihood (2LL). The objective function was minimized to obtain the best estimation of the parameters of the structural model, y0 and a, and the related parameters of the error models, and
. The two parameters of the structural model were considered significant if the 95% confidence interval of the estimates did not include zero. The significance of each of the structural and error parameters was tested with the likelihood ratio test. A difference in the minimal value of the objective function between the reduced and the full model exceeding 3.841 was considered significant (P<0.05). If the minimal value of the objective function did not differ significantly between the full and the reduced model, a parameter was eliminated from the model.
Effect of muscle relaxation and stimulation site on the reflex responses
A signed rank test was used to test for differences between neural and muscle stimulation (using the responses to stimuli before neuromuscular block) and to compare the responses to neural and muscle stimuli before and after injection of the muscle relaxant.
The statistical analysis, except for NONMEM, was performed with Sigmastat version 2.0 (Jandel Scientific Corporation, San Rafael, CA, USA). P<0.05 was taken as significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
The MDPE, representing the bias of the alfentanil target-controlled infusion, and the MDAPE, representing its accuracy, were 7.6 and 17.7% respectively. Thus, the capacity of the selected pharmacokinetic parameters to predict plasma concentrations in our study population was comparable to that in a previous study (Fig. 1).11
|
|
Effects of stimulus pattern, alfentanil concentration, muscle relaxation and site of stimulation
The SVmR and PWR responses after 5 and 10 s stimuli given before alfentanil were similar (Fig. 3). The systolic arterial pressure increases after 5 and 10 s stimuli were 13.3 (6.6) mm Hg (range 023) and 19.2 (12.1) mm Hg (range 5.239) respectively (not significant).
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A test of analgesia in the anaesthetized patient may help prevent unexpected haemodynamic reactions to severe stimuli. It would supplement anaesthetic depth monitors, such as the bispectral index and auditory evoked potential monitors,17 18 which indicate the hypnotic state. Although both SVmR and PWR were suppressed by alfentanil, the prediction of no response to tracheal intubation was poor with SVmR and good with PWR. The PWR, which could be easily measured clinically, could be a good alternative to the expensive and rarely available laser-Doppler (SVmR) technique. The cut-off value of 0.1 that has been suggested for SVmR,6 and which was also used for PWR in the present study, needs to be validated in a larger study.
From knowledge of the value for MAC intubation of sevoflurane in adults19 and the relationship between MAC and MAC bar,5 the responses after tracheal intubation found in patients not given alfentanil are unsurprising. They support the fact that haemodynamic responses to painful stimulation are not suppressed by volatile anaesthetics alone.3
Because the response to tracheal intubation was badly predicted by laser-Doppler SVmR, we decided to increase the stimulus applied after the first 10 patients had been studied. The stronger stimulation did not affect the predictive value of the SVmR, however, and the responses (SVmR and PWR) were similar for the two stimulation patterns. Although the two patient groups were not randomized, they did not differ except for slightly greater age in the second group. If the same concentrations of sevoflurane and nitrous oxide are given to subjects of different age, the expected effect would be greater in the older subjects.20 In fact, the haemodynamic responses to tracheal intubation, but not the SVmR or PWR response, were significantly greater in the older subjects.
The ulnar nerve is the preferred site of stimulation to cause vasoconstriction. During neuromuscular blockade, PWR was inversely correlated with the measured alfentanil concentration regardless of the site of stimulation. Neuromuscular blockade did not suppress the PWR response a great deal and did not reduce its capacity to predict the response to tracheal intubation.
Differences between our study and the previous study6 might partly explain the lower predictive value of laser-Doppler SVmR in our study. Shimoda and co-workers6 carried out laryngoscopy and intubation 35 min after i.v. induction of anaesthesia in premedicated patients, but we stimulated our patients during pharmacological steady-state conditions. We applied the first stimulus a mean (range) of 22 (2129) min after inhalation induction and intubated 50 min (3083) after induction. To prevent hypothermia, our patients were actively heated during the study, which increased skin temperature and induced vasodilation, shown by an increase in laser-Doppler flow and pulse wave amplitude before stimulation. Despite this vasodilation, tracheal intubation decreased pulse wave amplitude and finger blood flow (laser-Doppler). The decrease in pulse wave amplitude caused by tracheal intubation was correlated with the analgesic drug concentration, but the change in skin blood flow was not (data not shown). Because the SVmR in response to nerve stimulation was negative in nine patients, they were intubated without alfentanil administration first. The increase in arterial pressure after tracheal intubation in these patients could have increased the superficial skin blood flow in the fingers and thus reduced SVmR, because the vasodilation induced by sevoflurane and skin warming may have prevented the transient skin vasoconstriction induced by intubation. In the 11 patients given alfentanil, however, the SVmR was not suppressed and the decrease in skin blood flow was still found. Considering the whole study population, neither PWR nor SVmR was correlated with the skin temperature, so heating and sevoflurane may have suppressed skin vasomotor reactions (laser-Doppler) only in some patients. PWR was not affected by temperature, which suggests it is a better measure than SVmR.
The laser-Doppler probe measures the microvascular blood flow in a small area of the fingertip to a depth of a few millimetres. The pulse oximeter probe, in contrast, measures the changing blood volume in the whole fingertip. Finger blood flow is regulated mainly by local arteriovenous shunts.21 Vasoconstriction to a cold stimulus in awake humans is more pronounced in deep finger vessels than in superficial ones.22 Vasoconstriction to pain in deep and superficial finger vessels has not yet been compared. Our data might suggest vasoconstriction in deep finger vessels is more pronounced than in superficial ones and is thus less suppressed by active warming of the body and volatile anaesthetics. This would explain how the PWRin contrast to the laser-Doppler SVmRwas not attenuated in our study. Our results, in which acute sympathetic activation elicited by pain caused a transient vasoconstriction and thereby a change in the PWR, support previous data on angiotensin-induced vasoconstriction.23 Vasoconstriction with angiotensin slightly decreased the maximal photoplethysmography wave amplitude, by reducing the size of the early systolic component of the pulse wave, while the pulse pressure and the late systolic arterial pressure in the ascending aorta increased because the reflection wave increased.23
We conclude that a negative SVmR does not predict a reduced response of arterial pressure or heart rate to tracheal intubation. The PWR to a 5 s, 60 mA, 50 Hz tetanic impulse to the ulnar nerve could be a better predictor and warrants further study to confirm that the selected cut-off limit is appropriate and to determine the efficiency of PWR in predicting responses in patients of different ages and medical conditions.
![]() |
Acknowledgement |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Zbinden AM, Maggiorini M, Petersen-Felix S, Lauber R, Thomson DA, Minder CE. Anesthetic depth defined using multiple noxious stimuli during isoflurane/oxygen anesthesia. I. Motor reactions. Anesthesiology 1994; 80: 25360[ISI][Medline]
3 Zbinden AM, Petersen-Felix S, Thomson DA. Anesthetic depth defined using multiple noxious stimuli during isoflurane/oxygen anesthesia. II. Hemodynamic reactions. Anesthesiology 1994; 80: 2617[ISI][Medline]
4 Inomata S, Watanabe S, Taguchi M, Okada M. End-tidal sevoflurane concentration for tracheal intubation and minimum alveolar concentration in pediatric patients. Anesthesiology 1994; 80: 936[ISI][Medline]
5 Roizen MF, Horrigan RW, Frazer BM. Anesthetic doses blocking adrenergic (stress) and cardiovascular responses to incisionMAC BAR. Anesthesiology 1981; 54: 3908[ISI][Medline]
6 Shimoda O, Ikuta Y, Sakamoto M, Terasaki H. Skin vasomotor reflex predicts circulatory responses to laryngoscopy and intubation. Anesthesiology 1998; 88: 297304[ISI][Medline]
7 Shelley KH, Murray WB, Chang D. Arterial-pulse oximetry loops: a new method of monitoring vascular tone. J Clin Monit 1997; 13: 2238[ISI][Medline]
8 Shelley KH, Stout B, Silverman D. The pulse oximeter waveform compared to the laser Doppler waveform. J Clin Monit Comput 1999; 15: 2478
9 Low PA, Neumann C, Dyck PJ, Fealey RD, Tuck RR. Evaluation of skin vasomotor reflexes by using laser Doppler velocimetry. Mayo Clin Proc 1983; 58: 58392[ISI][Medline]
10 Petersen-Felix S, Zbinden AM, Fischer M, Thomson DA. Isoflurane minimum alveolar concentration decreases during anesthesia and surgery. Anesthesiology 1993; 79: 95965[ISI][Medline]
11 Raemer DB, Buschman A, Varvel JR, et al. The prospective use of population pharmacokinetics in a computer-driven infusion system for alfentanil. Anesthesiology 1990; 73: 6672[ISI][Medline]
12 Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia 1984; 39: 110511[ISI][Medline]
13 Kingsbury DP, Makowski GS, Stone JA. Quantitative analysis of fentanyl in pharmaceutical preparations by gas chromatography-mass spectrometry. J Anal Toxicol 1995; 19: 2730[ISI][Medline]
14 Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992; 20: 6394[ISI][Medline]
15 Karlsson MO, Sheiner LB. The importance of modeling interoccasion variability in population pharmacokinetic analyses. J Pharmacokinet Biopharm 1993; 21: 73550[ISI][Medline]
16 Sheiner LB, Beal SL. NONMEM Users Guide 1994. San Francisco (CA): University of California San Francisco, 1994
17 Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology 1998; 89: 9801002[ISI][Medline]
18 Dutton RC, Smith WD, Rampil IJ, Chortkoff BS, Eger EI. Forty-hertz midlatency auditory evoked potential activity predicts wakeful response during desflurane and propofol anesthesia in volunteers. Anesthesiology 1999; 91: 120920[ISI][Medline]
19 Katoh T, Nakajima Y, Moriwaki G, et al. Sevoflurane requirements for tracheal intubation with and without fentanyl. Br J Anaesth 1999; 82: 5615
20 Mapleson WW. Effect of age on MAC in humans: a meta-analysis. Br J Anaesth 1996; 76: 17985
21 Sessler DI, McGuire J, Hynson J, Moayeri A, Heier T. Thermoregulatory vasoconstriction during isoflurane anesthesia minimally decreases cutaneous heat loss. Anesthesiology 1992; 76: 6705[ISI][Medline]
22 Richardson D, Tyra J, McCray A. Attenuation of the cutaneous vasoconstrictor response to cold in elderly men. J Gerontol 1992; 47: M211M214[ISI][Medline]
23 Takazawa K, Tanaka N, Fujita M, et al. Assessment of vasoactive agents and vascular aging by the second derivative of photoplethysmogram waveform. Hypertension 1998; 32: 36570