Noradrenergic suppression to reduce electroencephalographic arousal after intubation: a randomised, placebo-controlled trial

Abstract

Background

Intraoperative awareness, without explicit recall, occurs after induction of anaesthesia in approximately 10% of persons under 40 yr of age. Most anaesthetic agents minimally suppress the noradrenergic system. We hypothesised that addition of dexmedetomidine, which suppresses noradrenergic activity, may reduce encephalographic (EEG) arousal in response to tracheal intubation; such an effect would lay the foundation for future studies of dexmedetomidine in reducing intraoperative awareness.

Methods

A single-site randomised, placebo-controlled trial with sex-based stratification was conducted. Participants, aged 18–40 yr old, undergoing intubation for general anaesthesia were eligible for recruitment and randomly allocated to receive dexmedetomidine or placebo. Dexmedetomidine (0.5 μg kg⁻¹) was given as a 5-min loading dose before induction. Bispectral index (BIS) values were collected during the induction phase of anaesthesia and the isolated forearm technique was used to assess patients' responsiveness before and after tracheal intubation. The primary outcome was the effect of dexmedetomidine on changes in BIS from pre-to postintubation.

Results

A total of 51 patients were recruited and included in the primary analysis. We did not observe an effect of dexmedetomidine on changes in BIS after tracheal intubation (mean difference –1.13, 95% confidence interval [CI] –4.87 to 2.62; p=0.556). Dexmedetomidine reduced the estimated plasma propofol concentration at loss of responsiveness (difference [dexmedetomidine – placebo]: –1.06 μg ml⁻¹, 95% CI –1.66 to –0.46; p<0.001) and before intubation (difference [dexmedetomidine – placebo]: –1.84 μg ml⁻¹, 95% CI –2.79 to –0.90; p<0.001). There was one patient in the placebo group who gave positive responses in the isolated forearm test before and after tracheal intubation.

Conclusions

Dexmedetomidine demonstrated an anaesthetic-sparing effect at induction of anaesthesia but did not prevent EEG arousal after tracheal intubation, as defined by an increase in the BIS value.

Clinical Trial Registration

Australia and New Zealand Clinical Trials Registry (Trial ID: ACTRN12622000754741).

Noradrenergic signalling is pivotal in modulating arousal via autonomic responses to noxious stimuli. Prima facie suppression of these functions seems core to the anaesthetic state, yet propofol and the volatile general anaesthetic agents offer limited suppression of the locus coeruleus.^1,2 Building upon this understanding, we have postulated that additional suppression of noradrenergic activity may assist in reducing the occurrence of intraoperative awareness.³ One potential approach to achieving suppression of central noradrenaline release is by the alpha-2 adrenergic agonist, dexmedetomidine. In the absence of other sedatives, dexmedetomidine is thought to induce an EEG pattern similar to stage 2 non-rapid eye movement sleep with an increase in delta power and spindling,^4,5 which is thought to reflect inhibition of locus coeruleus activity.⁶

Studies indicate that even a 0.5 μg kg⁻¹ loading dose of dexmedetomidine is effective in reducing propofol requirements with limited side-effects such as bradycardia and hypotension.⁷ Particularly noteworthy is the observation that combining dexmedetomidine with propofol at a reduced dose resulted in decreased propofol requirements for induction and notable reductions in blood pressure rises during tracheal intubation.^7,8 The utility of dexmedetomidine during tracheal intubation has been well-documented, consistently showing a decrease in the haemodynamic response compared with both placebo and alternative agents such as labetalol.9, 10, 11, 12, 13 Although other sympathetic modulators, such as esmolol, have been suggested to reduce the increase in bispectral index (BIS) with intubation,¹⁴ it remains uncertain whether dexmedetomidine can provide a comparable advantage.

Connected consciousness refers to intraoperative awareness with or without explicit recall. It manifests in 5–11% of patients after tracheal intubation during general anaesthesia.^3,15,16 This figure is substantially higher than the incidence of awareness with explicit recall,17, 18, 19, 20 thereby highlighting potential inadequacies in the prevailing emphasis on amnesia-centric ‘depth of anaesthesia’ monitors in meeting patient expectations regarding anaesthesia-induced unconsciousness.²¹ Implicit memory and intraoperative awareness have been associated with decreased postoperative satisfaction, the experience of dysphoria, and the development of post-traumatic stress disorder.^15,19,22 As such, connected consciousness, as assayed by the isolated forearm test, was added as a key secondary outcome in our study.

In this RCT, we assessed whether the impact of adding dexmedetomidine to a propofol–remifentanil induction reduced EEG arousal after tracheal intubation, as indicated by an increase in the BIS value. Additionally, this trial sought to assess the impact of dexmedetomidine on isolated forearm technique (IFT) responsiveness.

Methods

Study design

Noradrenergic Suppression to Reduce EEG arousal after Intubation was a single-centre, randomised, placebo-controlled, parallel-group trial comparing the effects of dexmedetomidine vs placebo (saline) on EEG arousal. The protocol and supplementary documents were approved by the Human Research Ethics Committee (2021/ETH11073), Sydney Local Health District RPAH Zone, and informed consent was obtained from all participants before study activities. This project was prospectively registered on the Australia and New Zealand Clinical Trials Registry (Trial ID: ACTRN12622000754741).

Patient selection

Patients between the ages of 18–40 yr old, with an ASA of 1–3, and who required tracheal intubation for general anaesthesia were eligible for the study. Those with a history of allergy to dexmedetomidine, history of heart block, pregnant women, and/or those unable to provide informed consent were excluded from the trial. All participants were identified before their procedure and the study protocol was discussed with the treating anaesthetic and surgical team.

Randomisation

This study compared dexmedetomidine vs placebo (saline) as a loading dose of 0.5 μg kg⁻¹ i.v. over 5 min. Eligible patients were randomly allocated in a 1:1 ratio by a trained member of the research team using the REDCap randomisation module in the study database. The treating anaesthetic doctor preparing the syringe was informed of the outcome of randomisation. Both the patient and the research staff who collected the intraoperative data and followed the patient up after surgery remained blinded to study group allocation.

Trial procedures

Participants entering theatres were connected to standard monitoring (NIBP, Pulse oximeter, three-lead ECG) and BIS monitoring. BIS data were recorded onto a USB stick for the duration of the study activities. After randomisation and preparation of the anaesthetic drugs, the study drug (dexmedetomidine or placebo), drawn up in a 50-ml syringe, was administered as a 5-min loading dose (0.5 μg kg⁻¹). Using the Hannivoort model,²³ the estimated plasma concentration (CP) of dexmedetomidine at the end of the 5-min infusion was 2 ng ml⁻¹. This then decreased to a plateau concentration of 0.5 ng ml⁻¹ over the following 150 s. After the loading dose was completed, 200 μg of glycopyrrolate was administered to counteract potential bradycardia from the combination of dexmedetomidine and the combination of remifentanil and propofol.

After the 5-min study drug loading dose, remifentanil TCI Minto model at a 4 μg ml⁻¹ targeted infusion was commenced. When this concentration was achieved, propofol TCI Marsh model was commenced at propofol CP of 4 μg ml⁻¹. The Marsh model was chosen as it is ambivalent to participant sex which we subsequently wished to include as a variable in our statistical models. During these preintubation procedures, data were collected at specific time points including BIS and propofol CP. At the commencement of the propofol infusion, the participant was asked to hold up a syringe in one hand. The time of syringe drop (which we defined as the point of loss of responsiveness) and propofol CP were recorded. To prepare for the IFT, a sphygmomanometer was applied to one of the participant's forearms and inflated to 50 mm Hg above systolic blood pressure as measured by NIBP. Subsequently, the neuromuscular blocking agent, rocuronium (0.6 mg kg⁻¹) was administered by the anaesthetist.

Before tracheal intubation, the propofol infusion was increased by 1 μg ml⁻¹ every minute to achieve a target BIS of 40–50. Once the BIS target level was achieved, the blinded research staff conducted an IFT with an additional independent assessor acting as a second witness to visualise any activity. The patient's trachea was then intubated by the anaesthetic doctor at their discretion. After intubation, the IFT was repeated and outcomes documented. Additional data points were collected for 5 min after intubation, including BIS at 1, 2, and 5 min and propofol CP. Study procedures ended after this 5-min period and anaesthetic doctors could proceed with their case at their own discretion.

Participants were followed up in the PACU at 15 and 60 min after their procedure. Data were collected on pain, postoperative nausea and vomiting (PONV), anxiety score, and delirium assessment (using Confusion Assessment Method for the ICU [CAM-ICU] and Nursing Delirium Screening Scale [NuDESC] scores). At 24 h and 7 days after surgery, a modified Brice and satisfaction questionnaire were performed with patients, either in person or over the phone, to assess for awareness with postoperative recall and postoperative satisfaction.

Outcomes

Primary outcome

The primary outcome of this study was the change in BIS from preintubation to postintubation. We present results averaged across all postintubation BIS values (10 s, 1, 3, and 5 min postintubation) and analysis of each time point individually, and of the peak postintubation BIS.

Secondary outcomes

Prespecified secondary outcomes included group-based differences in responsiveness to the IFT, haemodynamic changes (MAP and HR) after tracheal intubation, and propofol CP and BIS at loss of responsiveness and before intubation. We also assessed sex-based differences in propofol CP and BIS.

We analysed group-based differences in postoperative complications: PONV, with levels ‘none’, ‘nausea’, or ‘vomiting’; pain, using the numeric rating scale; anxiety, using a 0–10 scale; sedation or agitation, using the Richmond Agitation–Sedation Scale (RASS); and PACU delirium, using the CAM-ICU. Finally, we assessed patient-reported outcomes at 24 h and 7 days after surgery, all with levels ‘none’, ‘moderate’, or ‘severe’: drowsiness, surgical/anaesthesia site pain, thirst, hoarseness, throat soreness, nausea/vomiting, cold sensation, confusion, and shivering. All secondary outcomes should be considered hypothesis-generating.

Sample size justification

Prior work has suggested the standard deviation of postintubation BIS to be 11.¹⁴ Using this value, 50 subjects provide 90% power to show a difference of 10 points in BIS with alpha 0.05. Two extra patients were included to account for loss to follow-up, resulting in a sample size of 52 patients.

Statistical methods

We analysed our primary outcome using a mixed-effects Gaussian-family linear model. We used postintubation BIS as the outcome and controlled for preintubation BIS, as this approach is statistically more powerful for change scores and protects against regression to the mean.²⁴ The outcome was BIS at 10 s, 1, 3, and 5 min postintubation, and the independent variables were randomised group, time point of BIS measurement, a group∗time point interaction, preintubation BIS, and a random intercept for participant ID. Variance in BIS values across time points was analysed using Levene tests. Given preintubation BIS values differed between randomised groups, we also report analyses that are not adjusted for preintubation BIS. The parametric g-formula was used to compute the average difference in postintubation BIS between randomised groups at all time points.

Continuous secondary outcomes (BIS, propofol CP, MAP, HR) were analysed using fixed effect Gaussian-family models. Secondary outcomes that were ordinal (postoperative complications, and patient-reported outcomes at 24 h and 7 days) were analysed using proportional odds logistic regression. For ordinal outcomes, the method described by Brant²⁵ was used to assess the parallel regression assumption. No adjustments for multiple comparisons were made as these outcomes were hypothesis-generating.

Analyses were by intention to treat, with missing primary outcome data for one participant as described below. All analyses were performed in R (R Core Team [2024]; R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing, Vienna, Austria) via the RStudio integrated development environment. The ‘marginaleffects’ package was used for G-computation.²⁶

Results

In total, 52 participants were recruited to the study from May 2022 until April 2024. One participant's surgery was cancelled after recruitment and randomisation, leaving 51 participants (Fig 1). The characteristics of the cohort are shown in Table 1. There were no major differences between groups in the measured pretreatment characteristics.

Fig 1.

CONSORT diagram for the trial.

Table 1.

Characteristics of the study cohort. Data are presented as n (%) or median (interquartile range). IFT, isolated forearm technique.

Characteristic	Overall, N=51	Placebo, N=25	Dexmedetomidine, N=26
Age	32 (28–36)	30 (27–36)	34 (28–36)
Sex, n (%)
Female	37 (73)	17 (68)	20 (77)
Male	14 (27)	8 (32)	6 (23)
ASA
1	40 (78)	20 (80)	20 (77)
2	11 (22)	5 (20)	6 (23)
BMI	24.7 (22.7–29.6)	25.3 (23.1–30.8)	24.3 (21.6–28.5)
Surgery type, n (%)
ENT	3 (5.9)	3 (12)	0 (0)
General	5 (9.8)	1 (4.0)	4 (15)
Gynaecological	20 (39)	9 (36)	11 (42)
Neurological	14 (27)	8 (32)	6 (23)
Orthopaedic or trauma	2 (3.9)	1 (4.0)	1 (3.8)
Other	5 (9.8)	2 (8.0)	3 (12)
Plastic	1 (2.0)	0 (0)	1 (3.8)
Urological	1 (2.0)	1 (4.0)	0 (0)
History of anaesthetic awareness, n (%)	0 (0)	0 (0)	0 (0)
Preoperative anxiety scale	3 (1–4)	3 (1–4)	3 (2–4)
Responsive on preintubation IFT, n (%)
Responsive	1 (2.0)	1 (4.0)	0 (0)
Unresponsive	50 (98)	24 (96)	26 (100)
Responsive on postintubation IFT, n (%)
Responsive	0 (0)	0 (0)	0 (0)
Unresponsive	50 (100)	25 (100)	25 (100)
(Missing)	1	0	1

Primary outcome

On average, the BIS increased from the preintubation value to the peak postintubation value (Beta 3.86, 95% confidence interval [CI] 1.54–6.18; p=0.002). The median (interquartile range [IQR]) for BIS before intubation was 35.0 (14.2) in the dexmedetomidine group and 43.0 (8.0) in the placebo group. The median (IQR) for postintubation peak BIS was 38.0 (12.5) in the dexmedetomidine group and 46.0 (7.0) in the placebo group. After adjusting for BIS before intubation, we did not observe an effect of dexmedetomidine on postintubation BIS when averaged across all time points (difference [dexmedetomidine – placebo] –1.13, 95% CI –4.87 to 2.62; p=0.556) or when focusing on peak postintubation BIS (difference –1.55, 95% CI –5.86 to 2.76; p=0.482) (Table 2). We did not observe evidence of a difference in the relationship between dexmedetomidine and postintubation BIS between the different time points of BIS measurement (Supplementary Table S1). Analyses that were not adjusted for baseline BIS are shown in Supplementary Table S2. BIS values were lower before intubation in the dexmedetomidine group (Table 3) and remained relatively lower postintubation (Fig 2). Importantly, only 26 of 51 participants were in the target BIS range (40–50) before intubation, indicating difficulty in titrating anaesthesia to a stable BIS level before intubation. Fewer participants were in the target range in the dexmedetomidine group compared with the placebo group (8/26 vs 18/25, odds ratio [OR] 5.79, 95% CI 1.80–20.5; p=0.004). Changes in BIS for individual time points are plotted in Supplementary Figure S1. We did not observe evidence that the variance in BIS values differed between groups at each time point or between each time point across all participants (Supplementary Table S3).

Table 2.

Results for the primary outcome - change in BIS postintubation. Results are shown for linear mixed effect models with BIS at 10 s, 1, 3, 5 min, and peak postintubation as the dependent variable, and group allocation, time point of BIS measurement, a group∗time interaction, and preintubation BIS as the independent variables. Analyses include N=51 participants for peak BIS, with N=199 datapoints for the multilevel model used to calculate the effect size at other time points. ∗‘Estimate’ refers to the difference in BIS changes postintubation in the dexmedetomidine group minus the placebo group. Negative values suggest the change in BIS was more negative in the dexmedetomidine group. Estimates are calculated using the parametric g-formula (averaging across all predicted contrasts for each row of the dataset). The 95% confidence interval (CI) is calculated using the delta method. BIS, bispectral index.

	Estimate (95% CI)∗	P-value
Averaged over all time points	–1.13 (–4.87 to 2.62)	0.556
Immediately postintubation	–1.36 (–5.90 to 3.17)	0.555
1 min postintubation	–1.61 (–6.17 to 2.95)	0.490
3 min postintubation	0.05 (–4.53 to 4.64)	0.983
5 min postintubation	–1.57 (–6.15 to 3.02)	0.503
Peak postintubation BIS	–1.55 (–5.86 to 2.76)	0.482

Table 3.

Differences in randomised groups for the outcomes of propofol effect site concentration (μg ml⁻¹) and BIS, both at the time of loss of responsiveness and before intubation. A Gaussian-family linear model with a single predictor (group allocation) is used in all cases. AIC, Akaike information criterion; BIC, Bayesian information criterion; BIS, bispectral index; CI, confidence interval; CP, plasma concentration; df, degrees of freedom; No. Obs., number of observations. ∗Model diagnostics: No. Obs. =50; Log-likelihood =–72.6; AIC=151; BIC=157; Residual df=48. ^†Model diagnostics: No. Obs. =50; Log-likelihood =–202; AIC=411; BIC=417; Residual df=48. ^‡Model diagnostics: No. Obs. =51; Log-likelihood =–97.9; AIC=202; BIC=208; Residual df=49. ^¶Model diagnostics: No. Obs. =51; Log-likelihood =–179; AIC=364; BIC=369; Residual df=49. ^§∗p<0.05; ∗∗p<0.01; ∗∗∗p<0.001.

Characteristic	Propofol CP at time of loss of responsiveness (μg ml−1)∗			BIS at time of loss of responsiveness†			Propofol CP before intubation (μg ml−1)‡			BIS before intubation¶
	Beta	95% CI	P-value§	Beta	95% CI	P-value§	Beta	95% CI	P-value§	Beta	95% CI	P-value§
Randomised group
Dexmedetomidine – placebo	–1.06	–1.66 to –0.46	<0.001∗∗∗	–2.88	–10.9 to 5.18	0.476	–1.84	–2.79 to –0.90	<0.001∗∗∗	–4.62	–9.25 to 0.02	0.051

Fig 2.

Plot of the bispectral index (BIS) in both randomised groups across four postintubation time points. Lines connect individual patients, coloured by group allocation. The shaded ‘raincloud’ plots show the density distribution of the values in each group.

Secondary outcomes

Only one participant (in the placebo group) was responsive to the IFT before intubation (BIS at the time was 44), and one participant in the placebo group was responsive postintubation (BIS at the time of responsiveness was 32). This observation occurred outside of the initial window immediately after intubation, occurring immediately before cuff removal at 137 s postintubation. The participant was asked to respond at that point in line with the usual practice of the consultant administering the anaesthesia. Regardless, regression analysis was not performed owing to the low incidence of responsiveness. There were no cases of awareness with postoperative recall.

The propofol CP at loss of responsiveness and before intubation was lower with dexmedetomidine than with placebo. There did not appear to be a difference in BIS between groups at the time of loss of responsiveness (Table 3; Supplementary Fig. S2). Our results suggested that dexmedetomidine reduced BIS before intubation; however, it did not reach the threshold for statistical significance (beta –4.62, 95% CI –9.25 to 0.02; p=0.051). We did not observe evidence of a difference in haemodynamic response to intubation (as measured by changes in MAP: beta 6.70 mm Hg, 95% CI –2.29 to 15.7; p=0.140; and HR: beta –6.35 beats min–1, 95% CI –15.8 to 3.06; p=0.181) between the groups (Supplementary Table S4).

Adjusting for randomised group allocation, we did not observe sex-based differences in propofol CP at the time of loss of responsiveness (Beta 0.52 μg ml⁻¹, 95% CI –0.14 to 1.18; p=0.124) or before intubation (Beta –0.13 μg ml⁻¹, 95% CI –1.18 to 0.93; p=0.814), nor sex-based differences in BIS at the time of loss of responsiveness (Beta –2.62, 95% CI –11.73 to 6.48; p=0.572) or before intubation (Beta 1.36, 95% CI –3.76 to 6.48; p=0.602) (Supplementary Table S5). We also did not observe sex-based differences in the haemodynamic response to intubation (as measured by changes in MAP: Beta –4.90 mm Hg, 95% CI –15.0 to 5.24; p=0.334; and HR: Beta –4.46 beats min⁻¹, 95% CI –13.2 to 4.27; p=0.308) (Supplementary Table S6).

Brant tests did not suggest violation of the parallel regression assumption in any proportional odds logistic regression models assessing postoperative outcomes at 15 min and 1 h (Supplementary Table S7) or 24 h and 7 days (Supplementary Table S8). We did not observe evidence of group-level differences in PONV (Supplementary Table S9), postoperative pain (Supplementary Table S10), anxiety (Supplementary Table S11), RASS (Supplementary Table S12), or CAM-ICU scores (Supplementary Table S13) at 15 or 60 min after surgery. Group-level medians are plotted in Supplementary Figure S4. Both drowsiness (OR 0.33, 95% CI 0.10–0.98; p=0.045) and nausea/vomiting (OR 0.15, 95% CI 0.02–0.66; p=0.011) at 7 days were reduced in the dexmedetomidine group compared with the placebo group (at the threshold of an uncorrected p<0.05). No other group-level differences were observed for other patient-reported outcomes at 24 h or 7 days after surgery (Fig 3). The number of patients in each group with each outcome at 24 h and 7 days are plotted in Supplementary Figure S3.

Fig 3.

Ordinal regression models for outcomes at 24 h and 7 days after surgery. Brant's test suggested that the parallel regression assumption was not violated for any model. An odds ratio (OR) <1 suggests a lower odds of a more severe outcome (e.g. reduced odds of ‘severe’ thirst) with dexmedetomidine (DEX).

Discussion

These data suggest that addition of dexmedetomidine to a remifentanil–propofol anaesthetic reduced propofol requirements at predefined endpoints such as loss of responsiveness but did not alter signs of postintubation arousal in the primary outcome (BIS) or haemodynamics. Indeed, we did not find evidence for differences in the outcomes between the two groups, suggesting that the addition of dexmedetomidine adds little to the remifentanil–propofol combination. These data contrast with accumulating literature on the utility of dexmedetomidine in this context, although prior studies (to our knowledge) have not tested dexmedetomidine with a remifentanil–propofol anaesthetic.

Our study did not mirror a similar study with esmolol. The reasons for this difference are unclear, but distinct pharmacological differences including mechanisms of action and blood–brain barrier permeability differences may be important. Other notable differences include the use of remifentanil in this study that may have also blunted the sympathetic response. More broadly, in this small study the incidence of IFT responsiveness was low with only one responder in the placebo group (1/25, 4% incidence rate). Using similar inclusion criteria, the incidence of IFT responsiveness was 11% in the ConsCIOUS2 study.¹⁶ Although our study herein is very small, a 4% responsiveness rate is consistent with the concept that the continuous provision of anaesthesia before intubation is appropriate for reducing IFT responsiveness. Because of the low numbers, we can make no comment on the utility of dexmedetomidine to further reduce this rate. Similarly, although two secondary outcomes (drowsiness and nausea/vomiting at 7 days) suggested potential differences between the groups, the combination of small sample size and testing multiple outcomes, without correction for multiple tests, mandate these outcomes be treated as hypothesis-generating. Future studies should consider capturing similar data.

Despite some preclinical evidence for sex differences with dexmedetomidine, we did not find that sex modulated any of the outcomes by group.²⁷ This is critical as sex is an important determinant of awareness with explicit recall and connected consciousness.²⁸ It is important to note that hormonal fluctuations at different stages of the menstrual cycle may further influence this susceptibility^29,30 and therefore future studies that are large enough to look at varying hormonal cycles will be required to better understand the influence of sex on anaesthetic requirements and outcomes.

This study had some limitations. Our power analysis was based on a similar study of esmolol that found a large effect size. This effect size was considered appropriate as we expected a strong signal would be required to change clinical practice. It is important to note that we did observe a smaller increase in BIS after intubation than anticipated from the prior study (4 units in our study vs 10 in theirs¹⁴). Nonetheless, although this was a small study, it showed little signal that addition of dexmedetomidine to a propofol–remifentanil anaesthetic is beneficial. Our data indicate that pharmacological trials of IFT responsiveness would be possible (with the proviso that the 4% incidence rate observed here would require a very large trial to establish reduced occurrence of responsiveness). We also struggled to titrate the anaesthetic to a stable BIS level before tracheal intubation (only 26/51 were in the target range), particularly in the dexmedetomidine group, and BIS levels were lower before intubation in participants given dexmedetomidine. This indicates difficulties in titrating to BIS in general, particularly over short time periods. This also required we conduct models adjusting for preintubation BIS as a covariate. Moreover, the BIS is an imperfect (and proprietary) index of EEG arousal, and it is possible that clinically significant regional changes in neuronal activity would have been observed with a 64-electrode research EEG. Our data do not suggest that addition of dexmedetomidine (albeit in the presence of a lower propofol CP) reduces the risk of arousal from intubation. Although dexmedetomidine exposure was associated with a lower propofol CP and lower BIS, we cannot exclude that higher doses of dexmedetomidine may have been more effective. However, the concomitant reduction in propofol dosing may mitigate any benefit.

Conclusions

These data do not suggest that the addition of dexmedetomidine to an induction with propofol and remifentanil attenuates the increase in BIS after intubation. Further studies of the approach to induction are required to limit the incidence of connected consciousness detected on IFT.

Authors’ contributions

Study design: RDS

Conducted the study in consultation with RDS: KK, KL, CT, AW, HB, NM, SM, JB, MB

Data analysis: TP

Manuscript drafting in consultation with RDS and KK: TP

Critical feedback on the manuscript: all authors

Funding

National Institutes of Health (R01 AG063849-01 and NHMRC 2024134 to RDS); funding for this project was from departmental resources.

Declaration of interests

RDS is an Editor at the British Journal of Anaesthesia.

Handling editor: Phil Hopkins

Data availability statement

The original data is available from the authors upon reasonable request.

Appendix A. Supplementary data

The following is the Supplementary data to this article: