The effect of remifentanil on propofol ... - Wiley Online Library

Anaesthesia 2017, 72, 479–487

doi:10.1111/anae.13781

Original Article The effect of remifentanil on propofol requirements to achieve loss of response to command vs. loss of response to pain H. B. Scott,1 S. W. Choi,2 G. T. C. Wong3 and M. G. Irwin4 1 Clinical Associate, 2 Postdoctoral Fellow, 3 Associate Professor, 4 Professor and Head of Department, Department of Anaesthesiology, University of Hong Kong, Hong Kong

Summary When providing total intravenous anaesthesia, careful selection of end-points is required in titrating dose to effect during induction. Although propofol and remifentanil have predominantly different pharmacodynamic effects, they are seen to interact in achieving loss of consciousness and analgesia. To highlight these differences, we performed a double-blind, randomised controlled trial, comparing one group of patients receiving propofol alone (n = 42) with another group receiving remifentanil plus propofol (n = 46) as a target-controlled infusion of remifentanil (Minto; 3 ng.ml 1). Propofol was also titrated using a target-controlled infusion (Marsh effect model) to produce loss of response to tactile and vocal stimuli, and subsequently to loss of response to pain. The effect-site concentration of propofol at which 50% of patients lost tactile/verbal response was 2.9 lg.ml 1 in the propofol only group and 2.4 lg.ml 1 in the remifentanil with propofol group. In contrast, loss of pain response occurred at 4.4 lg.ml 1 in the propofol group, and 2.7 lg.ml 1 in the remifentanil with propofol group, with correspondingly lower bispectral index values. Judicious use of analgesia in total intravenous anaesthesia can have a propofol-sparing effect and potentially minimise the suppression of brain electrical activity. .................................................................................................................................................................

Correspondence to: M. G. Irwin Email: [email protected] Accepted: 15 November 2016 Keywords: bispectral index (BIS); pain; propofol; remifentanil; total intravenous anaesthesia (TIVA)

Introduction Remifentanil is a potent l-opioid receptor agonist and, therefore, a powerful analgesic. Like other opioids, it is a poor hypnotic in clinically relevant concentrations with minimal effect on cognitive function [1, 2]. The opioid effect on electroencephalography-derived variables is weak [3]. In contrast, propofol is a powerful hypnotic agent which rapidly and reliably causes loss of consciousness, probably through an action on GABA receptors [4]. Unlike commonly used volatile anaesthetic agents, propofol has long been considered to be non-analgesic [5]. However, if given in sufficiently large doses to suppress © 2017 The Association of Anaesthetists of Great Britain and Ireland

cerebral activity, propofol can ‘blunt’ the response to a painful stimulus [6]. Given these different properties, propofol and remifentanil are commonly administered contemporaneously for total intravenous anaesthesia (TIVA) [7], to individually titrate the two major components of the anaesthetic state, namely hypnosis and analgesia [8]. Given the very different pharmacodynamic effects of these drugs, it is inappropriate for a user to compensate for one drug with an excessive dose of the other, although this may occur if the end-point of titration, that is, hypnosis or loss to response to pain, is not well defined or understood. 479

Anaesthesia 2017, 72, 479–487

Our purpose in conducting this study was to determine whether remifentanil would significantly reduce the effect site concentration of propofol required to abolish the response to pain (but not to achieve loss of consciousness, in this case being defined as a loss of response to verbal and tactile stimuli). Conversely, we wished to assess whether using propofol alone to achieve loss of response to pain would be accompanied by a greater drop in bispectral index (BISTM) than when combined with remifentanil, indicative of greater suppression of brain electrical activity.

Methods Ethical approval was obtained from the Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster. After screening 113 patients for eligibility, written informed consent was obtained from 100 patients with ASA physical status 1 or 2, between the age of 18 and 66 years (Fig. 1). Patients did not receive premedication and were scheduled to undergo general anaesthesia for minor elective surgery, defined as procedures that were expected to last less than an hour, with no or minimal blood loss, and not requiring large incisions or entry into the cerebral or thoracic cavity. Exclusion criteria were: any known disease of the central nervous system; intake of any drug acting on the central nervous system; any known psychiatric illness; and any known adverse effects to the study drugs. Once the patient was in the operating room, standard monitoring was applied. Bispectral index was monitored (M-BIS Module BISXP software version 4.0; Datex-Ohmeda Division, Instrumentarium Corporation, Helsinki, Finland) via a four-electrode sensor (BISXP Quatro Aspect Medical Systems, Newton, MA, USA) which was placed on the forehead. A peripheral nerve stimulator (Innovator model NS 272; Fisher and Paykel, Auckland, New Zealand) was attached to the right wrist over the ulnar nerve with surface electrodes (Telectrode; Bio-Protech Inc., Wonju, Korea). Intravenous cannulation was obtained via the dorsum of the left hand or the antecubital fossa. Commercial infusion pumps (Injectomat TIVA Agilia; Fresenius Kabi, Cheshire, UK) were used for the propofol and remifentanil target-controlled 480

Scott et al. | The effect of remifentanil on propofol requirements

infusion (TCI). The infusion pumps were attached to the intravenous cannula using a non-reflux valve infusion system (TIVA Drop Air Matic P; Fresenius Kabi, Graz, Austria) that was primed with 1% propofol (Propofol-Lipuroâ; B Braun, Melsungen, Germany). The propofol infusion pump was programmed using the modified Marsh effect-site concentration model with a ke0 of 1.21.min 1 [9] and the remifentanil pump was programmed using the Minto model [10]. Remifentanil (GlaxoSmithKline Manufacturing, Torrile, Parma, Italy) was diluted saline to produce a concentration of 50 lg.ml 1. The responsible anaesthetist, who was not involved in the study, prepared this infusion as well as programmed and operated the pumps throughout the study phase while the investigator assessed the patient’s response. Patients were randomly allocated using a computer-generated sequence with the allocation concealed in an envelope to receive remifentanil at an estimated effect-site concentration of either 0 ng.ml 1 for the propofol alone group or 3 ng.ml 1 for the remifentanil plus propofol group. The remifentanil pump and the drug infusion lines were then covered and faced away from the investigator so as to blind them to the remifentanil infusion. Patients were pre-oxygenated and once end-tidal oxygen reached 80%, the study period began. The propofol pump was initially programmed with an effect-site concentration target of 1 lg.ml 1. When the effect-site concentration of propofol reached the target, the investigator shook the subject’s shoulder and said ‘say your name’ and ‘open your eyes’ using a constant force and tone until the subject responded appropriately. If the patient responded appropriately, the propofol effect-site target concentration was then increased by 0.2 lg.ml 1 and 2 min was allowed for the effect site to reach the new target. Two minutes was chosen, as the equilibration of the effect site with the new target concentration typically requires around 90 s. At the 2-min mark, the patient’s response to the above commands was re-assessed. If the patient was still responsive, this process was repeated until the patient reached our first end-point of loss of response to verbal and tactile stimuli and the propofol effect-site concentration was recorded. At this point and while © 2017 The Association of Anaesthetists of Great Britain and Ireland


Anaesthesia 2017, 72, 479–487

Assessed for eligibility (n = 113)

Enrollment

-Scheduled for general anaesthesia for minor elecve surgery -Between 1st March and 31st July 2015 -Invesgator available -ASA 1 or 2 -Age between 18 and 66 years

Excluded (n = 13) -Declined participation (n = 8) -CNS / psychiatric disease (n = 1) -CNS medicaon (n = 2) -Given premedicaon (n = 2)

Randomised (n = 100)

Allocation Allocated to propofol plus remifentanil group (n = 50)

Allocated to propofol alone group (n = 50)

Withdrawn BIS monitoring failure (n = 6) Age exceeding 66 years (n = 1)

BIS monitoring failure (n = 2) Error in programming TCI pump (n = 1) Leakage of propofol at cannula site (n = 1) Needed other medication and airway intervention (n = 1)

Analysis Analysed (n = 42)

Analysed (n = 46)

Figure 1 CONSORT diagram. CNS, central nervous system; BIS, bispectral index; TCI, target-controlled infusion. maintaining this concentration, a pain stimulus was administered via a peripheral nerve stimulator and the patient’s response was assessed. This was a tetanic stimulus (50 Hz, 80 mA, 0.25-ms pulses for 4 s) to the ulnar nerve on the right wrist. If the patient responded, the propofol effect-site target concentration was increased by 0.2 lg.ml 1 and the response reassessed after 2 min for equilibration. Loss of response to pain was defined as the point at which the subject © 2017 The Association of Anaesthetists of Great Britain and Ireland

stopped withdrawing, extending or grimacing to the painful stimulus, at which point the study was complete. As we were assessing motor response, no neuromuscular blockers or other drugs were given until this completion point when all relevant data were collected. Following this, the drug doses could be adjusted and further drugs administered, as required, to secure the airway and maintain anaesthesia for surgery. Ventilation and oxygenation was maintained throughout the 481

Anaesthesia 2017, 72, 479–487

study using a facemask, 100% oxygen and manual ventilation if required. The following measurements were recorded every 2 min effect-site concentration; heart rate; systolic blood pressure; diastolic blood pressure; and BIS value. The primary outcome was the propofol effect-site concentration at loss of response to verbal and tactile stimuli. In a previous study involving 20 volunteers designed to measure the effect-site concentration of propofol at which patients stopped responding to verbal stimuli, the mean (SD) value was 2.0 (0.9) lg.ml 1 [11]. We therefore calculated that 43 subjects would be required in each group to have 95% power to detect a difference of 0.5 lg.ml 1 in effect-site concentration at loss of response to verbal and tactile stimuli with a p < 0.05 significance. It was, therefore, decided that 50 subjects would be randomly allocated to each group, using a computer-generated sequence, in order to account for any subject withdrawal. Differences between the two groups for propofol effect-site concentration at loss of response to verbal and tactile stimuli and loss of response to pain were determined using the Mann–Whitney U-test. Differences between the two groups in the BIS value were determined using the t-test.

Results Eighty-eight subjects completed the study. The demographic and baseline haemodynamic data for the two groups are shown in Table 1. There were no significant differences between the two groups. Of the 12 subjects who did not complete the study, eight were not studied due to an inability to get a clean and stable BIS value (six in the propofol group and two in the remifentanil and propofol group), one due to an error in programming the TCI pump correctly (remifentanil and propofol), one due to a leakage of propofol from the cannula, one subject was not studied due to the need for further drugs for airway intervention (remifentanil and propofol) and one that was recruited into the propofol group whose age just exceeded 66 years old. The proportion of patients remaining responsive to vocal and tactile stimuli and to pain at different propofol effect site concentrations is shown in Figs. 2 and 3, respectively. As can be seen, the propofol effect-site concentration at which 50% of patients 482


Table 1 Patient characteristics and baseline haemodynamic data. Values are presented as mean (standard deviation) or number.

Age; years Male/female ASA 1 ASA 2 Comorbidities 0 1 2 3 4 Medications 0 1 2 3 4 5 SBP; mmHg DBP; mmHg HR; beats.min

Propofol alone n = 42

Remifentanil + propofol n = 46

45.6 (13.7) 18/25 22 20

42.7 (15.0) 16/30 30 16

21 13 5 2 1

1

29 10 1 1 0 1 135 (18) 79 (16) 75 (12)

27 15 4 0 0 41 3 1 1 0 0 129 (19) 74.4 (10.0) 77 (12)

DBP, diastolic blood pressure; HR, heart rate; SBP, systolic blood pressure. SBP, DBP and HR were taken when effect site concentration = 1 lg.ml 1.

demonstrated loss of response to verbal and tactile stimuli was 2.9 lg.ml 1 in the propofol group and 2.4 lg.ml 1 in the remifentanil and propofol group, representing a difference of 0.5 lg.ml 1. In contrast, the difference for loss of response to pain was 1.7 lg.ml 1, with it being 4.4 lg.ml 1 in the propofol group and 2.7 lg.ml 1 in the remifentanil and propofol group. The observations made for BIS values are shown in Table 2 and the BIS at different propofol effect-site concentrations can be seen in Fig. 4. BIS values at loss of response to verbal and tactile stimuli were significantly lower in the propofol group compared with the propofol and remifentanil (p = 0.0009). The BIS values at loss of response to pain were significantly lower in the propofol group, compared with the remifentanil and propofol group (p < 0.0001). There was a negative correlation between propofol effect-site concentrations and BIS values at loss of response to verbal and tactile stimuli (R = 0.426, p < 0.001), and a negative correlation between © 2017 The Association of Anaesthetists of Great Britain and Ireland


Anaesthesia 2017, 72, 479–487

Figure 2 Proportion of patients still responsive vs. concentration of propofol at loss of verbal response for propofol alone and propofol + remifentanil and loss of response to pain for propofol alone and propofol + remifentanil . Note the shift in response with added remifentanil is small for verbal response, but very large for response to pain. propofol effect-site concentrations and BIS values at loss of response to pain (R = 0.714, p < 0.001). None of the patients had any recollection of the tetanic stimulation when questioned postoperatively.

Discussion The intent of this study was not to merely replicate the apparent propofol ‘sparing’ effect of remifentanil in reducing the response to pain, a concept that is already well established [12]. Rather, the purpose was to provide insight into appropriate interpretations of this phenomenon. Our main finding was that achieving loss of response to command was not strikingly affected by remifentanil, since its addition to propofol only modestly reduced the levels of propofol required to achieve this end-point (Fig. 2). In contrast, remifentanil markedly reduced the amount of propofol required to obtund response to pain (Fig. 2). © 2017 The Association of Anaesthetists of Great Britain and Ireland

Correspondingly, the BIS value at loss of response to pain was significantly lower for the propofol-only group compared with remifentanil and propofol (Fig. 3). Notably, both the propofol effect-site concentration and BIS values seen in the propofol-only group, when loss of response to command occurred, were remarkably similar to the values seen in the remifentanil and propofol group when loss of response to pain occurred. This implies that the remifentanil dose can be increased in the presence of a painful stimulus to limit any change in brain electrical activity, without increasing propofol levels. Using propofol alone in the presence of a painful stimulus achieves the same effect but with greater suppression of cerebral electrical activity. This last notion is very different from the conventional approach to dosing (in volatile anaesthesia) by the minimum alveolar concentration (MAC), itself 483

Anaesthesia 2017, 72, 479–487


100

% patients

75

50

25

0

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Bispectral index

Figure 3 Proportion of patients still responsive at a given BIS value (data from Fig. 2) at loss of verbal response for propofol alone and propofol + remifentanil and loss of response to pain for propofol alone and propofol + remifentanil . Note the shift in BIS response with added remifentanil is small for verbal response, but very large for response to pain. The dotted lines show the BIS values at loss of response in 50% patients. Table 2 Bispectral index (BIS) at loss of response to commands (LRVTS) and loss of response to pain (LRP). Values are mean (SD). BIS value at LRVTS

BIS value at LRP


Remifentanil plus propofol n = 46


Remifentanil plus propofol n = 46

66 (12)

74 (9)

47 (16)

68 (12)

defined as the dose required to prevent movement in response to a standardised painful stimulus [13]. However, volatile agents produce powerful analgesia even at subanaesthetic doses, so loss of response to pain does not guarantee loss of conscious awareness [14]. This is, at least in part, why accidental awareness during general anaesthesia may occur [15] and also, perhaps, why patients can move to command but not to painful surgical stimulus in the isolated forearm construct. In contrast, propofol is predominately a 484

hypnotic drug with only mild analgesia through mechanisms linked to both central hypnotic effects and direct peripheral analgesic action [16]. High doses of propofol may provide analgesia, but only in a non-specific way through intense depression of the central nervous system. This may delay recovery from anaesthesia and have other adverse effects such as postoperative neurocognitive deficit and delirium [17–19]. One problem in maintaining a steady dose of propofol while increasing the dose of remifentanil to © 2017 The Association of Anaesthetists of Great Britain and Ireland


Anaesthesia 2017, 72, 479–487

110 100 90

BIS value

80 70 60 50 40 30 20 10

0.8

1.2

1.6

2.0

2.4

2.8

3.2

3.6

4.0

4.4

4.8

5.2

5.6

6.0

6.4

Propofol concentration (μg.ml–1)

Figure 4 Mean (SD) Bispectral index (BIS) at different propofol concentrations. painful stimuli is that there is a trade-off of one shortcoming of deep anaesthesia with another related to high-dose opioids. In our study, the dose of remifentanil was not particularly high and yet the attenuation of pain response was quite marked. Although both tolerance and hyperalgesia can occur with remifentanil, it is not clinically manifested with average doses around 3–6 ng.ml 1 (0.15–0.35 lg.kg.min 1) and it would be unusual for patients to require higher doses than this during surgery [20–22]. Another insight from our study relates to modelling the interaction between propofol and remifentanil. Response surface modelling can be used to guide optimal drug concentration. This assumes that the two (or more) drugs being investigated have additive, synergistic or infra-additive effects, and isobolograms and concentration effect curves can be constructed for a given endpoint, for example, loss of response to laryngoscopy. A number of commercial anaesthesia systems have been developed which claim to assist the anaesthetist in this regard, for example, the Dr€ager ubeck, SmartPilotâ (Dr€agerwerk AG & Co. KGaA, L€ Germany) and the GE-Navigator (GE Medical Systems Ltd, Pollards Wood, UK). In relation to the interaction between volatile agents such as sevoflurane and remifentanil, there may well be an additive or synergistic effect on analgesia because sevoflurane is well © 2017 The Association of Anaesthetists of Great Britain and Ireland

Propofol and remi.

Propofol only.

known to have analgesic properties and response surface modelling may have a useful role in that regard [3, 23, 24]. However, in relation to the interaction between propofol and remifentanil, they assume that a higher dose of one drug will compensate for a lower dose of the other and vice-versa. From our study and the work of others [25, 26], this would appear to be incorrect since propofol does not directly affect the end-point of response to pain. There are several methodological considerations and potential limitations that warrant comment. In the design of this study, it was important to assess the response to a painful stimulus more typical of surgery in addition to assessing loss of response to verbal and tactile stimulus. A tetanic electrical stimulation to a level of 50 mA is suitable for this purpose, as demonstrated in previous trials, and is non-traumatic, noninvasive, objective and quickly repeatable [27, 28]. We used target-controlled infusion in the effect-site mode rather than in the plasma mode as used in some previous studies. To the best of our knowledge, the combination of tetanic stimulus and effect-site targeting from TCI in the fashion employed in this study has not been used before in dose–response studies. Using the effect-site concentration as the target will result in the pump delivering a larger bolus compared with the plasma-targeting mode for a 485

Anaesthesia 2017, 72, 479–487

concentration increase. This causes a temporary ‘overshoot’ of the plasma propofol concentration over the effect-site target value, creating a large concentration gradient between plasma and effect-site, which enables a more rapid attainment of the target effect-site concentration. The converse is true for a decrease in effect-site concentration, where the plasma concentration is allowed to drop below the target to create a larger reverse gradient for egress of propofol from the brain. This compares with the plasma-targeting mode, where the desired value is set for the plasma concentration only, with drugs delivered at a rate that does not allow for the plasma value to exceed the target. The effect concentration will, therefore, rise and equilibration occurs at a slower rate than that achievable in the effect site mode. In previous studies, especially ones using the DiprifusorTM (AstraZeneca Limited, Macclesfield, UK), a device that can only be employed in the plasma-targeting mode), values for the effect site concentration were recorded as they were rising or falling towards the plasma value. Thus, it may appear that the 2-min interval used is too short for equilibration when in fact this is well within the calculations of the model. The main limitation is the reliance on the reliability of the TCI model operating in the effect-site mode, which may exacerbate equilibration inaccuracies from assumptions made by the effect site elimination rate constant (keo) used in the calculations [29]. We used the ‘modified’ Marsh model used with a keo of 1.21 min 1 that gives a t½ keo of 0.573 min as calculated by Struys et al. using BIS values as their endpoint [30]. As effectsite concentration cannot be measured, without concurrent measurements of plasma drug levels, accuracy will always be a concern. However, our chief aim in this study was to demonstrate a dose separation rather obtaining accurate absolute numbers at the different end-points and the same inherent inaccuracies apply to both groups equally. A similar argument can be made for the use of a keo derived from the use of a continuous end-point (i.e. BIS values) to evaluate a dichotomous end-point in this study. The use of patients scheduled for surgery meant we could only conduct measurements on one occasion during induction. It could be argued that we did not take into account the intra-individual variability, 486


but taking the average of a population of subjects having the same technique applied to them for both groups should minimise the effect this may have on the different end-points. Second, we could have blinded the investigator by using a placebo solution of saline in place of remifentanil. However, an anaesthetist not involved with the study was solely responsible for the programming and operation of the pumps, and the pump panels faced away from the investigator. We examined only one dose of remifentanil. We found in our pilot study that higher remifentanil doses rendered patients apnoeic and, occasionally, induced a degree of truncal rigidity. This required airway and ventilatory support that in turn altered the level of stimulus to the patients. As such, we used a more modest dose of remifentanil of 3 ng.ml 1 in order to maintain some spontaneous ventilation during the experimental intervention. However, even at this dose, there was a marked attenuation in the response to pain. A more typical target during surgical procedures would be an effect site concentration of remifentanil of between 3 ng.ml 1 to 6 ng.ml 1. We cannot, therefore, determine from these data the interaction between propofol and higher doses of remifentanil. A further, but unavoidable, limitation is that the BIS readings were taken on non-paralysed patients and, consequently, there may be a variable but significant contamination of the BIS value from electromyographic activity [31]. In conclusion, we showed that when delivering TIVA, loss of response to pain is achieved at a much lower concentration of propofol when an appropriate dose of remifentanil is co-administered, than with propofol alone.

Acknowledgements Registered at ClinicalTrials.gov (identifier: NCT02287181). MI is an editor of Anaesthesia and SWC is a statistical advisor for the journal, so this article has undergone additional external review. No external funding or other competing interests.

References 1. Iselin-Chaves IA, Flaishon R, Sebel PS, et al. The effect of the interaction of propofol and alfentanil on recall, loss of consciousness, and the Bispectral Index. Anesthesia and Analgesia 1998; 87: 949–55. © 2017 The Association of Anaesthetists of Great Britain and Ireland

Scott et al. | The effect of remifentanil on propofol requirements 2. Shiokawa H, Yamaura K, Karashima Y, Hoka S, Megumu Y, University KHS. Evaluation of analgesic and hypnotic effects of remifentanil by in vivo patch-clamp recordings: 14AP8-7. European Journal of Anaesthesiology 2013; 30: 225. 3. Heyse B, Proost JH, Hannivoort LN, et al. A response surface model approach for continuous measures of hypnotic and analgesic effect during sevoflurane-remifentanil interaction: quantifying the pharmacodynamic shift evoked by stimulation. Anesthesiology 2014; 120: 1390–9. 4. Trapani G, Altomare C, Liso G, Sanna E, Biggio G. Propofol in anesthesia. Mechanism of action, structure-activity relationships, and drug delivery. Current Medical Chemistry 2000; 7: 249–71. 5. Grounds RM, Lalor JM, Lumley J, Royston D. Morgan M Propofol infusion for sedation in the intensive care unit: preliminary report. British Medical Journal 1987; 294: 397–400. 6. Black E, Campbell SG, Magee K, Zed PJ. Propofol for procedural sedation in the emergency department: a qualitative systematic review. Annals of Pharmacotherapy 2013; 47: 856–68. 7. Vuyk J, Mertens MJ, Olofsen E, Burm AG, Bovill JG. Propofol anesthesia and rational opioid selection: determination of optimal EC50-EC95 propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology 1997; 87: 1549–62. 8. Absalom AR, Mani V, De Smet T, Struys MM. Pharmacokinetic models for propofol-defining and illuminating the devil in the detail. British Journal of Anaesthesia 2009; 103: 26–37. 9. Marsh B, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol in children. British Journal of Anaesthesia 1991; 67: 41–8. 10. Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 1997; 86: 10–23. 11. Iwakiri H, Nishihara N, Nagata O, Matsukawa T, Ozaki M, Sessler DI. Individual effect-site concentrations of propofol are similar at loss of consciousness and at awakening. Anesthesia and Analgesia 2005; 100: 107–10. 12. Kern SE, Xie G, White JL, Egan TD. A response surface analysis of propofol-remifentanil pharmacodynamic interaction in volunteers. Anesthesiology 2004; 100: 1373–81. 13. Aranake A, Mashour GA, Avidan MS. Minimum alveolar concentration: ongoing relevance and clinical utility. Anaesthesia 2013; 68: 512–22. 14. Klomp T, van Poppel M, Jones L, Lazet J, Di Nisio M, LagroJanssen AL. Inhaled analgesia for pain management in labour. Cochrane Database Systematic Reviews 2012; 9: CD009351. 15. Pandit JJ, Andrade J, Bogod DG, et al. 5th National Audit Project (NAP5) on accidental awareness during general anaesthesia: summary of main findings and risk factors. Anaesthesia 2014; 69: 1089–101. 16. Bandschapp O, Filitz J, Ihmsen H, et al. Analgesic and antihyperalgesic properties of propofol in a human pain model. Anesthesiology 2010; 113: 421–8. 17. Andresen JM, Girard TD, Pandharipande PP, Davidson MA, Ely EW, Watson PL. Burst suppression on processed

© 2017 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2017, 72, 479–487

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

electroencephalography as a predictor of postcoma delirium in mechanically ventilated ICU patients. Critical Care Medicine 2014; 42: 2244–51. Deiner S, Luo X, Silverstein JH, Sano M. Can intraoperative processed EEG predict postoperative cognitive dysfunction in the elderly? Clinical Therapeutics 2015; 37: 2700–5. Radtke FM, Franck M, Lendner J, Kruger S, Wernecke KD, Spies CD. Monitoring depth of anaesthesia in a randomized trial decreases the rate of postoperative delirium but not postoperative cognitive dysfunction. British Journal of Anaesthesia 2013; 110(Suppl. 1): i98–105. Martini CH, Boon M, Broens SJ, et al. Ability of the nociception level, a multiparameter composite of autonomic signals, to detect noxious stimuli during propofol-remifentanil anesthesia. Anesthesiology 2015; 123: 524–34. Maurtua MA, Deogaonkar A, Bakri MH, et al. Dosing of remifentanil to prevent movement during craniotomy in the absence of neuromuscular blockade. Journal of Neurosurgical Anesthesiology 2008; 20: 221–5. Yu EH, Tran DH, Lam SW, Irwin MG. Remifentanil tolerance and hyperalgesia: short-term gain, long-term pain? Anaesthesia 2016; 71: 1347–62. Heyse B, Proost JH, Schumacher PM, et al. Sevoflurane remifentanil interaction: comparison of different response surface models. Anesthesiology 2012; 116: 311–23. Manyam SC, Gupta DK, Johnson KB, et al. Opioid-volatile anesthetic synergy: a response surface model with remifentanil and sevoflurane as prototypes. Anesthesiology 2006; 105: 267–78. Wang LP, McLoughlin P, Paech MJ, Kurowski I, Brandon EL. Low and moderate remifentanil infusion rates do not alter targetcontrolled infusion propofol concentrations necessary to maintain anesthesia as assessed by bispectral index monitoring. Anesthesia and Analgesia 2007; 104: 325–31. Smith C, McEwan AI, Jhaveri R, et al. The interaction of fentanyl on the Cp50 of propofol for loss of consciousness and skin incision. Anesthesiology 1994; 81: 820–8; discussion 26A. Rantanen M, Ypparila-Wolters H, van Gils M, et al. Tetanic stimulus of ulnar nerve as a predictor of heart rate response to skin incision in propofol remifentanil anaesthesia. British Journal of Anaesthesia 2007; 99: 509–13. Irwin MG, Hui TW, Milne SE, Kenny GN. Propofol effective concentration 50 and its relationship to bispectral index. Anaesthesia 2002; 57: 242–8. Cortinez LI. What is the ke0 and what does it tell me about propofol? Anaesthesia 2014; 69: 399–402. Struys MM, De Smet T, Depoorter B, et al. Comparison of plasma compartment versus two methods for effect compartment–controlled target-controlled infusion for propofol. Anesthesiology 2000; 92: 399–406. Schuller PJ, Newell S, Strickland PA, Barry JJ. Response of bispectral index to neuromuscular block in awake volunteers. British Journal of Anaesthesia 2015; 115(Suppl. 1): i95–103.

487