• 
• 

Three Part Question

Does [the use of rocuronium] in [intubation in the emergency department] decrease [provision of post-intubation sedation compared to use of suxamethonium]

Clinical Scenario

You are working in a busy emergency department and intubate a patient using a rapid sequence induction with rocuronium as your neuro-muscular blocking agent (NMBA) of choice, there is a delay in setting up a sedative infusion and shortly after intubation, your patient becomes tachycardic and hypertensive.

You worry they are under-sedated and experiencing awareness so give a bolus dose of sedative as a temporising measure.
You wonder whether your choice of a longer acting NMBA compared to the more traditionally used suxamethonium could have delayed your recognition of awareness and led to under-sedation.

Search Strategy

Medline via Ovid interface 1946 to Week 2 June 2019
EMBASE via Ovid interface 1974 to June 13th 2019

(rocuronium AND (succinylcholine OR suxamethonium)) AND (ED OR emergency department) AND (sedati* OR awareness OR AAGA OR propofol OR midazolam OR ketamine OR etomidate OR hypnotic).mp

Limit to humans and English language

Search Outcome

A search of the Medline database found 12 results of which 3 were relevant to the 3 part question.
A search of the EMBASE database found 75 results of which 4 were relevant to the 3 part question.

In total 4 papers were found which were relevant to the three part question.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Lembersky et al 2019 USA	11,748 patients included in the NEAR database of ED intubations. Collected from 25 participating hospitals. Included all age groups and patients undergoing both RSI and non-RSI techniques.	Retrospective cohort study (prospectively collected database, retrospectively analysed)	Sedation (defined as propofol, ketamine, etomidate, midazolam, diazepam, fentanyl or morphine) given within 15 minutes of intubation, bolus or infusion.	Sedation given within 15 minutes of intubation in 9099/11748 (77.5%)	Applicability to question limited as included paediatric patients and non-RSI techniques. Outcome choice of sedation given at singular time point may not demonstrate sedation practice adequately. Sedation doses not recorded. Hypnotics and analgesics grouped together as sedation. 1526/11748 (13%) of patients intubated without drugs.
			Likelihood of post-intubation sedation within 15 min depending on NMBA choice (short acting vs long acting) (Whole cohort) Confidence intervals 95%.	Suxamethonium use associated with increased chance of post-intubation sedation compared with longer acting agents (rocuronium and vecuronium). Odds ratio 1.89 (CI 1.68-2.12)
			Likelihood of post-intubation sedation within 15 min depending on NMBA choice (short acting vs long acting) (Analysis with traumatic arrest and cardiac arrest cases (n=1787) excluded)	Suxamethonium use associated with increased chance of post-intubation sedation compared with longer acting agents (rocuronium and vecuronium). Odds ratio 1.95 (CI 1.72-2.22)
Johnson et al 2015 USA	106 patients ≥18 years of age intubated with either suxamethonium or rocuronium in a single centre tertiary, urban ED over a 3 year period. Exclusion criteria: sedation contraindicated, intubated prior to arrival, intubated without NMBA.	Retrospective cohort study	Number of patients intubated with suxamethonium vs rocuronium	Suxamethonium: 30/106 (28%). Rocuronium: 76/106 (72%)	Observational design risks confounding by indication; were perceived sicker patients given rocuronium and less sedation to avoid hypotension. No data collected on illness severity. No data on dose of sedative, intermittent sedative boluses not included. Wide confidence intervals.
			Mean time to initiation of continuous sedative infusion (intermittent boluses not included)	Time to sedation in rocuronium group longer at 34 +/- 36 min vs 16 +/- 21 min in suxamethonium. P <0.002
Korinek et al 2014 USA	254 consecutive patients intubated in a single centre academic tertiary ED ≥18 and ≤89 years of age. Excluded if not intubated with a NMBA or etomidate.	Retrospective cohort study	Number of patients intubated with suxamethonium vs rocuronium	Suxamethonium 127/254 (50%). Rocuronium 127/254 (50%)	39 had further doses of long acting NMBA, 31 of these had suxamethonium for RSI. Risks of confounding by indication, although data showed no significant difference in vital signs or GCS between groups. Using outcome of rate of infusion as time from initiation instead of time from intubation makes comparison of doses more challenging.
			Analgesic or sedative administered prior to RSI	Suxamethonium: additional analgesic 10%, additional sedative 21%. Rocuronium: additional analgesic 9%, additional sedative 19%.
			Post intubation sedation given during stay in ED (whole cohort)	218/254 (86%) Received post-intubation sedative infusion. 11/254 (4%) received intermittent sedative boluses only. Propofol (n=169) and midazolam (n=49) only agents used for sedative infusion.
			Time to initiation of sedation	Suxamethonium: Median time (IQR) 8min (5-15min) Rocuronium: Median time (IQR) 12min (12-91min) P=0.002.
			Rate of propofol infusion at 30 minutes post-initiation of infusion	Suxamethonium: Propofol - 42+/-24mcg/kg/min. Midazolam - 0.048+/-0.041mcg/kg/h. Rocuronium: Propofol 30+/-23mcg/kg/min P=0.002. Midazolam - 0.030+/-0.048mcg/kg/h P=0.057
			Of patients who received sedation (n=218), number who received at least one additional sedative bolus	Suxamethonium: 66.7% Rocuronium: 40.1% P<0.001
			Time to analgesic initiation post RSI (n=108 (42.5%) received analgesic infusion, n= 36 (14.2% received analgesic boluses but no infusion) All received fentanyl	Suxamethonium: Median time (IQR) 14.5min (4-41min) Rocuronium: Median time (IQR) 35.5min (12-91min) P=0.002
			Rate of fentanyl infusion at 30 minutes post-intubation	Suxamethonium: Mean 0.859 +/- 0.488mcg/kg/h Rocuronium: 0.645 +/- 0.545mcg/kg/h P=0.041
Watt et al 2013 USA	200 patients intubated in an urban, tertiary, academic ED. Aged ≥18 and ≤89 years. Excluded if not intubated with etomidate, not given NMBA or deemed not to require sedation.	Retrospective cohort study	Dose of induction agent (etomidate) and baseline characteristics	Suxamethonium: 0.3+/-0.1mg/kg etomidate. Rocuronium: 0.3+/-0.1mg/kg P=0.782 No statistically significant difference between groups in age, weight, rates of traumatic brain injury, vital signs or GCS	Limitations inherent to retrospective study. Reasons for choice of NMBA not documented and risks confounding by indication. No mention of whether analgesics given. Time to sedation a surrogate of awareness/under-sedation
			Proportion of patients receiving post-intubation sedation	155/200 (77.5%) initiated on sedative infusion (whole cohort) (propofol n=148, midazolam n=7). 45/200 (22.5%) received bolus doses only (Propofol n=20, midazolam n=24, lorazepam n=1)
			Mean time to post-intubation sedative use	Suxamethonium: 15 +/- 13minutes. Rocuronium: 27 +/- 29 minutes. P<0.001
			Proportion received sedation within 15 minutes post-induction	Suxamethonium: 69%. Rocuronium: 46%. P=0.001

Comment(s)

The intubation of the sick and crashing patient in a busy emergency department is fraught with difficulty even at induction, with difficult decisions to be made about induction agent choice and dose to strike a balance between adequate anaesthesia and avoiding haemodynamic collapse. Subsequent management of the intubated patient requires close attention to adequate anaesthesia to avoid pain and distress. Although the zeitgeist in the critically ill patient is for light levels of sedation, aiming to ameliorate the negative effects of unnecessary deep sedation, for the patient with residual neuro-muscular blockade, deep sedation is required to avoid awareness and distress. Watt et al were perhaps the first to look at this in 2013 and found rocuronium was associated with both statistically and clinically significant longer times to initiation of sedation when compared with suxamethonium (27 +/- 29 min vs 15 +/- 13 min respectively). Rocuronium was also associated with a significantly lower proportion of patients receiving sedation within 15 minutes post-intubation, despite the induction sedative, etomidate, having a duration of action of only 3-5minutes. Both groups were well matched in terms of demographics and vital signs. Korinek et al built on this work, albeit in apparently the same ED (in a different time period), and found that even once a sedative infusion has been started, rocuronium is associated with significantly lower propofol infusion rate at 30 minutes than suxamethonium, the effect did not reach significance with midazolam. Amongst the subset initiated on a sedative infusion, a smaller proportion of patients given rocuronium received subsequent boluses of sedation when compared to the suxamethonium group. Korinek et al expanded on Watt et al by also looking at analgesic infusions and found significantly longer time to initiation of fentanyl infusion and decreased subsequent fentanyl infusion rate at 30 minutes amongst patients receiving rocuronium. Johnson et al found increased time to sedation amongst patients receiving rocuronium, with results that were consistent with those reported previously by Watt et al and Korinek et al. The largest study of the four, Lembersky et al, looked at proportion of patients receiving sedation within 15 minutes post-intubation. They found that suxamethonium was associated with an increased chance of receiving sedation within 15 minutes compared to longer acting NMBA with an odds ratio of 1.89 (CI 1.68-2.12), increasing to 1.95 (CI 1.72-2.22) when post-cardiac/traumatic arrests were excluded. There are limitations common to all of the aforementioned studies. Inherent to the retrospective cohort design, the use of rocuronium may be chosen in patients perceived to be sicker with higher chance of hyperkalaemia; such patients may also need more cautious use of sedation post-intubation, although it should be noted groups were generally well matched in the reviewed studies. Ultimately using sedative dose or time to sedation is only a surrogate of risk of awareness; as such it is impossible to draw the conclusion that these patients have experienced increased rates of awareness, something notoriously difficult to reliably demonstrate in any case. Regardless, there is consistent evidence in these studies that when rocuronium is used for intubation, compared to suxamethonium there is increased time to initiation of sedation, decreased provision of subsequent boluses and decreased rate of sedative infusion once initiated. There was also a delay in initiation of analgesic infusion and subsequent analgesic infusion rate amongst those given rocuronium. Although without prospective design it is difficult to state conclusively, there is concern that this effect is due to the masking of the signs of under-sedation by persistent neuromuscular blockade.

Editor Comment

CR

Clinical Bottom Line

The use of rocuronium for intubation in the emergency department is associated with delayed initiation of post-intubation sedation and analgesia, reduced provision of subsequent sedative boluses and reduced rate of infusions of both analgesia and sedation, when compared with suxamethonium. There is some cause for concern that such patients may be being under-sedated and exposed to risk of awareness whilst paralysed.

Level of Evidence

Level 3 - Small numbers of small studies or great heterogeneity or very different population.

References

1. O. Lembersky, D. Golz, C. Kramer, et al Factors Associated with Post-Intubation Sedation After Emergency Department Intubation Utilizing a National Airway Registry American Journal of Emergency Medicine May 2019; S0735-6757(19)30304-3.
2. Johnson et al Impact of Rocuronium and Succinylcholine on Sedation Initiation After Rapid Sequence Intubation Journal of Emergency Medicine July 2015. 49(1):43-9
3. Korinek et al Comparison of rocuronium and succinylcholine on postintubation sedative and analgesic dosing in the emergency department European Journal of Emergency Medicine 2014; 21:206-211
4. Watt et al Effect of paralytic type on time to post-intubation sedative use in the emergency department Emergency Medicine Journal 2013; 30; 893-895