You can respond to this article at http://www.anaesthesiacorrespondence.com
Changes in the photoplethysmogram with tracheal intubation and remifentanil concentration
Version of Record online: 28 SEP 2012
Anaesthesia © 2012 The Association of Anaesthetists of Great Britain and Ireland
Volume 67, Issue 12, pages 1332–1336, December 2012
How to Cite
Wang, X., Chen, X., Ye, S., Feng, Y., Huang, C., Hou, L., Liu, J. and Chen, H. (2012), Changes in the photoplethysmogram with tracheal intubation and remifentanil concentration. Anaesthesia, 67: 1332–1336. doi: 10.1111/anae.12001
- Issue online: 7 NOV 2012
- Version of Record online: 28 SEP 2012
- Accepted: 19 August 2012
Autonomic sympathetic activation, for instance following noxious stimuli, decreases the size and flattens the shape of the photoplethysmographic peripheral pulse waveform. We report a simple measure of the waveform shape, the ratio of mean-to-peak wave amplitude, for measuring nociception level during general anaesthesia. Fifty participants, anaesthetised with propofol and remifentanil, were randomly allocated to one of three different remifentanil effect-site concentrations (1, 3 and 5 ng.ml−1). Tracheal intubation increased the mean (SD) plethysmographic ratio from 0.38 (0.06) to 0.48 (0.04), p = 2.6 × 10−16. The mean (SD) ratios following intubation at remifentanil effect-site concentrations of 1 ng.ml−1, 3 ng.ml−1 and 5 ng.ml−1, were 0.49 (0.03), 0.48 (0.03) and 0.45 (0.04), respectively. Remifentanil therefore suppressed changes in the mean-to-peak ratio caused by tracheal intubation (p = 0.006). The ratio of the mean-to-peak plethysmographic amplitude may represent a simple measure of the balance of autonomic sympathetic and parasympathetic activity under general anaesthesia, and its performance following intubation was significantly different from peak amplitude (p = 0.046).
Several indices of anaesthetic depth, such as bispectral index (BIS), auditory evoked potentials and Entropy, are affected most by the hypnotic component of the anaesthetic triad: hypnosis; analgesia; and relaxation [1–3]. These indices therefore have limited ability to monitor analgesic depth, for which there is no widely accepted objective index. Traditionally, autonomic reactions, such as tachycardia, hypertension, sweating and lacrimation, have been regarded as signs of inadequate analgesia [1, 4].
Nociceptive stimuli activate the autonomic sympathetic system, causing haemodynamic responses like vasoconstriction. The size, shape and frequency of the pulse photoplethysmograph (waveform) reflect the balance of autonomic sympathetic and parasympathetic activities and therefore might be used to measure analgesic depth. Figure 1 illustrates the measurement of four parameters from a representative waveform that respond to noxious stimuli: peak amplitude; notch amplitude; area under the curve; and maximum upward slope [5, 6]. The peak amplitudes have contributed to nociceptive indices, such as the response index of nociception (RN) and surgical stress index (SSI) [7–11]. However, the amplitudes of the pulse waveform can be influenced by the pinch force of the probe and temperature as well as by noxious stimuli [5, 12, 13]. Nociceptive information contained in the shape of the waveform may be less sensitive to artefact than the amplitude of the waveform.
Therefore, the aim of this study was to explore changes in the shape of the waveform, represented by the ratio of mean-to-peak wave amplitude, with tracheal intubation and remifentanil concentrations during general anaesthesia.
The study was approved by the local Research Ethics Committee of the school of medicine, Zhejiang University and written informed consent was obtained from all participants.
We enrolled fifty participants, of ASA physical status 1-2, from the Women’s Hospital and the First Affiliated Hospital, scheduled for laparoscopy under general anaesthesia. We excluded patients with nervous system disease (e.g. neurological disorders, head injury, seizure disorders), chronic use of psychoactive medication or abuse of alcohol or illicit drugs and any clinically significant cardiovascular, renal, hepatic or endocrinological disorder.
We allocated participants randomly to one of three target effect-site remifentanil concentrations (1, 3 or 5 ng.ml−1) using the random number function in ExcelTM (MicrosoftTM Office; Microsoft Corporation, Redmond, WA, USA). One anaesthetist (YF) controlled remifentanil infusions, whereas another (XC) performed all other anaesthetic tasks. A student (XW) collected data.
We induced anaesthesia with target-controlled infusions of propofol and remifentanil [14, 15]. An effect-site concentration of propofol was set at 4 μg.ml−1, adjusted in steps of 0.5 μg.ml−1 every 2 min to maintain BIS between 40 and 60, followed by the allocated infusion of remifentanil. Three minutes after the target concentration of remifentanil was achieved, patients received an infusion of 0.6 mg.kg−1 rocuronium to facilitate tracheal intubation. The anaesthetist could increase the target concentrations of remifentanil as an emergency response to evidence that a patient was inadequately anaesthetised. We excluded study such patients from the analysis. The room temperature was between 22 and 24 °C.
We recorded the photoplethysmograph (pulse waveform) from the index finger of the non-dominant hand, from anaesthetic induction to after intubation (DS-100A Durasensor, OxiMax; Nellcor Puritan Bennett Inc, Pleasanton, CA, USA). We checked and if necessary corrected the automatic measurements of waveform peaks and troughs, especially 1 min before and during intubation (Matlab version 7.1; The Mathworks Inc., Natick, MA, USA). We excluded from the analysis beats with artefacts.
Sample size calculation was performed using G*Power software (version 3.1, Franz Faul, Universitat Kiel, Germany) . To detect a difference of 0.02 of the mean-to-peak ratios between groups, we calculated that we would need to recruit 12 patients to each group, assuming α = 0.05, β = 0.2, and a SD of the ratio within each group of 0.03. We used the paired-samples t-test to compare average measurements from the minute before intubation with the average during the peak 10-s period during intubation (Fig. 2). The plethysmographic measurements we analysed from these times, were the mean-to-peak ratio and the peak amplitude. We used one-way ANOVA to compare the distributions of mean-to-peak ratios among the three remifentanil groups , Levene’s test for homogeneity of variance and Bonferroni’s method for multiple comparisons. All the tests were calculated using SPSS (version 16.0; SPSS Inc., Chicago, IL, USA). We assessed the probability that the mean-to-peak ratio of the waveform could predict remifentanil effect-site concentrations using the jack-knife method . A prediction probability of 1 means perfect prediction, whereas a value of 0.5 means that the ratio has no predictive power.
One patient was not studied from analysis due to incomplete data. The characteristics of the remaining 49 participants are shown in Table 1.
|1 ng.ml−1(n = 15)||3 ng.ml−1(n = 20)||5 ng.ml−1(n = 14)||All(n = 49)|
|Age; years||37 (10)||43 (11)||43 (9)||41 (10)|
|Weight; kg||59.0 (8.2)||60.7 (9.8)||62.7 (9.6)||60.7 (9.2)|
|Height; cm||162.6 (5.9)||162.6 (6.8)||167.0 (8.0)||163.9 (7.0)|
Figure 3 shows a representative waveform. The amplitude of the waveform decreases during intubation whilst the shape flattens, with an increase in the ratio of mean-to-peak amplitudes. The ratios of mean-to-peak amplitude were similar before intubation. During intubation, the ratio increased least in the 5 ng.ml−1 remifentanil infusion group (Table 2). The peak waveform amplitude decreased during intubation, but without differences among remifentanil groups (Table 3). The ratio or peak amplitude before or during intubation were poor predictors of their respective remifentanil concentration, with the predictive probability ranging between 0.51 and 0.63.
|1 ng.ml−1(n = 15)||3 ng.ml−1(n = 20)||5 ng.ml−1(n = 14)||All(n = 49)||p value|
|Before||0.39 (0.05)||0.38 (0.06)||0.36 (0.07)||0.38 (0.06)||0.350|
|During||0.49 (0.03)†||0.48 (0.03)‡||0.45 (0.04)||0.48 (0.04)*||0.006|
|1 ng.ml−1(n = 15)||3 ng.ml−1(n = 20)||5 ng.ml−1(n = 14)||All(n = 49)||p value|
|Before||954 (519)||685 (321)||676 (372)||765 (416)||0.100|
|During||322 (187)||265 (169)||328 (246)||300 (197)*||0.590|
Both the ratio of mean-to-peak amplitude and the peak amplitude changed significantly during intubation, whereas during intubation the remifentanil dose affected the ratio of mean-to-peak amplitude, but not the peak waveform amplitude. The shape of the plethysmographic waveform therefore could help monitor the balance of autonomic sympathetic and parasympathetic activity.
The amplitude of the plethysmographic waveform does not have any absolute units of measurement. This means that the value of the amplitude can only be interpreted in the context of a trend. The incorporation of the peak waveform amplitude into the SSI required a complex method to normalise the signal amplitude . A study reported a delay of ∼10 s from a change in waveform amplitude to the change in the index . Normalisation might have contributed most to this delay. However, a parameter based on waveform shape does not require normalisation and therefore would not generate such a delay.
Researchers have used two methods to calculate waveform changes following a stimulus. One method has been to calculate a mean value for a period [6, 7, 20–22], the other method has been to calculate the peak value within a period . In our study, tracheal intubation took different times for different patients, hence it was not suitable to calculate a mean value for an arbitrary period (e.g. 120 s) for all patients. On the other hand, a peak value within a period might represent an artefact. Therefore, we chose to calculate a mean value for a short period (10 s) for every patient.
There were some aspects of our study that could be improved. First, we only used tracheal intubation as a stimulus, and the intensity of stimulation probably varied between participants. The response of the ratio of mean-to-peak amplitude to other stimuli should be studied. Second, different remifentanil concentrations might generate different results . Third, we only studied 49 patients, so to determine how useful the waveform shape might be when monitoring analgesic depth more patients should be studied.
In conclusion, the shape of the plethysmographic waveform has some advantages over the amplitude of the waveform as a measure of autonomic activity to noxious stimuli.
This study was funded by the National Natural Science Foundation of China (grant no. 81070885), Foundation of the Education Department of Zhejiang Province of China (grant no. 20100699), Qianjiang Talent Plan of Zhejiang Province of China (grant no. 2011R10041), and Foundation of the Science and Technology Department of Zhejiang Province of China (Grant no. 2011c33035).
No other funding and no competing interests declared.