Controlled rapid sequence induction & intubation


Hey folks

This interesting paper came up in my journal reading this month.

Controlled rapid sequence induction and intubation : an analysis of 1001 children

Its a retrospective audit of an anaesthesia protocol introduced in 2007 into a Swiss hospital for intubation of children who were at risk of aspiration ( unfasted or condition predisposing to gastric regurgitation. This included emergency and elective cases.

They called it the ” controlled rapid sequence induction and intubation” protocol. Essentially a complete modification of the traditional RSI technique. They mostly chose to not use cricoid pressure, there was standard mask ventilation using pressure limit whilst paralysis onset awaited. Strangely they chose in the main to use a slow acting paralytic , atracurium, rather than suxamethonium or rocuronium, justifying that the mask ventilation meant time was not crucial!

In many respects this is not RSI at all, not even a modified version..its simple induction of anaesthesia like any elective patient

Now the interesting part is their complications profile. They report only one episode of regurgitation but no evidence of pulmonary aspiration. This is in over a 1000 cases of their protocol. Seems reassuring! Also the incidence of serious hypoxia was very low (0.3%)

I asked some FOAMEd colleagues to comment on the paper too so here are their thoughts!

I encourage you to read the full paper if you can access it!

Reuben Strayer of EMUpdates

I can only speak for emergency department patients – this is an OR study, which is a different population, using a different set of tools (e.g. sevoflurane), by providers that have a different skillset.

peri-induction bag mask ventilation is appropriate when the likelihood of the patient developing clinically important hypoxia during the intubation attempt, if peri-induction ventilation is omitted, exceeds the likelihood of inducing vomiting and aspiration by doing bag mask ventilation. since kids desaturate faster than adults, the younger the patient, the more the balance tips to PIV. The other half of the equation is quite important, however, and it is relevant, if one were to consider bringing a controlled RSII paradigm to the emergency department, that anesthesiologists probably spend 1000 minutes performing bag mask ventilation for every minute an emergency physician spends performing bag-mask ventilation.

that said, this paper reminds us in patients with an oxygenation deficit, peri-induction ventilation is a strategy worth considering. ideally this would be done using an NIV mask and a ventilator, which isn’t susceptible to operator catecholamine-induced overbagging, but a fair substitute for the sublime indifference of a machine is the phrase bag slowly and gently, slowly and gently.

regarding cricoid pressure, what’s that? sounds dangerous.

regarding atracurium, a fair alternative to succinylcholine, if faster-onset rocuronium isn’t available.

Scott Weingart

Key point is the PIP limitation on the breaths. This is why Jim’s oxylator stuff is so cool as well. PIP limited breaths are fine in the peri-intubation. I accomplish the same with Vent-As-Bag. This ignores ApOx which would suffice in these elective op kids and avoid bagging altogether. Levitan bags all of his RSIs.

Scott Weingart, MD
ED Critical Care Blog & Podcast

Nicholas Chrimes of ClinicalCred

Hi Minh,

I’m assuming you won’t be surprised to see that my response to this is characteristically longwinded…

RSI is more about the principle behind it than the precise technique used to achieve it. That principle is to minimise the risk of pulm aspiration by:
Adopting strategies to reduce the risk of regurgitation occuring
Adopting strategies that minimise the time between removing the patients ability to protect their own airway and the placement of a cuffed ETT in their trachea (the “rapid” bit) – to decrease the window of opportunity for aspiration to take place if regurgitation did occur.

When clinicians say “I’m going to do an RSI” I think they mean that they are going to use a technique which adheres to the above principles, rather than necessarily rigidly committing to any specific set of strategies.

As stated in the article the typical standard elements are:
Cricoid pressure
Sux (or Roc these days)
No ventilation

There are probably some other reasonably constant standard elements they’re not explicit about.
Suction under the pillow: I do this for all inductions whether RSI or not anyway – while the likelihood of regurgitation may be lower in an apparently fasted patient, the speed at which I’ll need to respond to it if it occurs is unchanged.
Pre-determined dose of induction agent: rather than the slow titration to effect that characterises elective intubation of a fasted patient.
Immediate “chase” of induction agent with relaxant: rather than waiting to confirm the dose of induction agent has been adequate & patient is asleep as would occur in the elective intubation of a fasted patient prior to administering NMBDs.

There are also some other features that are less consistently included:
No prior sedative agents: anaesthetists almost invariably violate this using fentanyl/alfentanil +/- midazolam, but my experience of our emergency physicians is that RSI tends to be pretty much just induction agent & sux/roc
Some practitioners would add to this strategies to minimise the impact of aspiration if it did occur such as use of non-particulate antacids such as Na citrate (administration of which, for some inexplicable reason, is almost routine practice in Australia for GA caesarian section but rarely undertaken for any other situation in which RSI is performed).
Some would also add strategies to decrease gastric volume (prokinetics, NG suction).

Uncertainty about the degree to which any of the above elements actually achieve the desired goal allows the specific technique of “RSI” to be varied whilst still potentially upholding the fundamental principle behind it. Additionally in various patient groups the balance of risks may change leading to a justifiable variation in technique. I’m going to look at each of the elements in “controlled RSI” to see whether I think the “RSI” principle remains sound in the face of them:

Elements of “Controlled RSI”:

Gentle mask ventilation (<12cmH2O): this is part of my usual practice and I think widely accepted by many anaesthetists to be safe (ie. not cause gastric insufflation with increased regurg risk) in RSI. Some points:
In OT we have soft bags, APL valves and airway manometry. On top of this we intubate & mask ventilate much more frequently than most ED/ICU physicians. This certainly doesn’t mean that in a panic anaesthetists in OT are beyond applying excessive ventilation pressure and inflating the stomach, but it provides us with some safeguards that may not be present in ED. The more stressful context, lack of “feel” from a self inflating bag and absence of manometry put the ED physician at greater risk of causing gastric insufflation, so I could understand why, despite gentle ventilation being accepted for RSI in OT patients, it may be considered unsafe by ED physicians. Conversely as Scott & Reuben suggest there is equipment to address this even in ED – and no reason Mapleson Circuits and airway manometry couldn’t become part of routine ED equipment (unlikely to be practical for prehospital context though).
Apart from the benefits of GMV in the rapidly desaturating patient (which could be at least partly addressed by NODESAT principles without needing to mask ventilate) my main rationale for using GMV is that it puts you “one step ahead” from a Vortex perspective. Once you induce the patient and enter the “funnel” of the Vortex the “clock starts ticking” on your time to critical desaturation (whether this is likely to be short or long). GMV during the time you are waiting for your relaxant to take effect means that you are using it to initiate implementation of optimisation strategies for FM ventilation. This prevents this first minute of oxygen consumption being wasted and provides greater situational awareness if intubation is subsequently difficult. If good FM ventilation is achieved prior to intubation, the knowledge that the “green zone” can be achieved provides reassurance to the airway operator that alveolar O2 delivery is achievable regardless of the outcome of the intubation attempt and potentially allows them to take a more measured, less panicked, and thereby hopefully less traumatic & more successful approach to laryngoscopy. Conversely if FM ventilation is unsuccessful and laryngoscopy subsequently fails the airway operator (& team) has a greater situational awareness that they might be in real trouble and are already at least part way through optimising FM ventilation (having already ascertained, for example, that OPA is required and thus immediately requesting it) – or possibly finished with an optimal attempt at FM ventilation completely (if it has been attempted towards the time of full muscle relaxation taking effect) allowing them to recognise they are already 1 strike down on the Vortex. This allows them to direct the time remaining more efficiently – by only having to have optimal attempts at the 2 remaining NSA techinques as well as escalating their “CriCon” status.
Overall I’m all for GMV during RSI.

No cricoid: despite having a reputation for being “pro-cricoid”, I’m not really particularly invested provided practitioners demonstrate a consistent rationale for their practice. I perform cricoid because it has a reasonable rationale, is expected practice of the Australian & New Zealand College of Anaesthetists and the Victorian Consultative Council of Anaesthetic Morbidity & Mortality and does no harm (although it can occasionally obscure view, this is easily addressed by easing/removing it – and in all honesty the need to do this is rare, probably less than 1 in 20 RSIs in my experience). They’re the same reasons I perform MILNS.
I accept though that there’s no evidence for cricoid pressure and reasonable argument can be made that it can be omitted from RSI without violating the overall philosophy behind RSI – but only if you believe that it is of NO VALUE in preventing regurgitation. It is this final point that I typically take issue with, as in my experience most people who “don’t believe cricoid helps” will still use it once there is visible evidence of regurgitation occurring (which is some ways is the least logical time to use it because the “horse has already bolted”). If you think it will help prevent “more horses bolting” once you have visible evidence of regurgitant material in the airway, you should be using it prophylactically to prevent the “first horse bolting” in patients you have assessed to be “at risk”. Conversely if you don’t think cricoid helps at all, I can’t see why you would suddenly invoke it because you see fluid regurgitating into the pharynx. It’s a bit like turning to God on your death bed.
Overall I have no problem with omitting cricoid in RSI providing the above rationale is consistently applied. This study fails on this criterion though. They identify a number of situations in which they think cricoid SHOULD be applied (bleeding tonsil, achalasia, Zenker diverticulum, colon interposition for oesophageal replacement). If they believe that cricoid needs to be applied for these conditions and is therefore of value in decreasing the risk of regurgitation, then omitting it for their “controlled RSI” violates the underlying philosophy of RSI.

Atracurium: I have real issues with this.
Despite the large dose they’re giving they report times of around 2-3 mins to intubating conditions, thus significantly prolonging the time between loss of spontaneous airway protective reflexes and protection of the airway with a cuffed tube. This delays provides a longer time during which there is an opportunity for aspiration of any regurgitated material and violates the underlying philosophy of RSI. It removes the entire “rapid” component.
Curiously I can’t work out how switching to atracurium contributes to their goal of decreasing hypoxic episodes, CV deterioration or difficult intubation anyway. It seems a pointless variation which theoretically increases aspiration risk for no foreseeable gain. If they wanted to avoid the complications of sux (which they make no mention of as a goal) they could have used rocuronium.
Their only justification for using atracurium in the article reveals a failure of the investigtors to understand the reason for using a rapidly acting NMBD in the first place. They state that “as hypoxemia & hypercapnia are prevented by gentle ventilation…the time to ideal intubating conditions is no longer critical… (and the) debate on whether to prefer succinylcholine or non-depolarising NMB agents becomes pointless”. Sux is used in RSI to decrease the time till intubating conditions to minimise the time till the airway is protected with a cuffed ETT, not just to because you won’t be ventilating in the interim.
Also what’s the plan if they CAN’T achieve FM ventilation with inspiratory pressures 12cmH2O and risk gastric insufflation/aspiration or allow hypoxaemia to occur. The delay to intubating conditions caused by using atracurium rather than sux/roc is problematic on many fronts.
I must say I’m also surprised by their inclusion of mivacurium in their protocol for short procedures. I didn’t think anyone used this drug any more. Apart from still having the same problem of a slow onset, my experience is that it has then neuromuscular blocking properties of normal saline (although it is an excellent mast cell degranulator – causing almost universal histamine release and rash)!

Interpretation of Results: the investigators report a decreased incidence of adverse events cf. previous studies with “standard RSI”. Whilst that may be beneficial I don’t know if that equates to their technique being safe from the point of view of preventing aspiration – so we don’t know what the trade off is to achieve this decreased incidence of adverse resp events. Aspiration is a low risk event, even in a patient with a full stomach – but its consequences can be serious/fatal, particularly if particulate matter is aspirated. Does their study of ~1000 children have adequate power to demonstrate that their technique is safe if none of their participants aspirate? I don’t know enough about statistics or the risk of aspiration to answer this.

Overall Impression: I’m not really clear what “technique” it is they’re describing. They’ve removed a number of elements of RSI (some justifiably, some not) and ended up with an induction that just looks like every elective intubation I would do on a fasted patient. It’s not RSI, its not controlled RSI (if they’re not using sux/roc I can’t see any justification for using the term “rapid”), it’s not DSI, it’s just plain old “induction of anaesthesia & intubation”. There is no element in their technique that really addresses potential aspiration risk beyond what we would usually do in a fasted patient (even when FM ventilating elective, fasted patients, anaesthetists would aim to avoid inspiratory pressures greater than 15cmH2O and insufflating the stomach). Apart from suctioning a NGT if it is already present, I can’t see what separates this intubation from that of a fasted patient. Even if what they’ve done is can be proven to be safe from the point of view of aspiration risk, what they’ve described is not a new technique, its simply getting rid of the RSI concept entirely. Perhaps that’s justifiable and the RSI concept is unnecessary, with “gentle intubation” posing no additional aspiration risk (to be honest it wouldn’t surprise me if this were true – I suspect deep anaesthesia & ensuring adequate paralysis before instrumenting the airway are probably the most significant factors in preventing regurgitation) – but I don’t think this study demonstrates that, nor was it designed to.

Still awake?- Nick

Seth Trueger of MDAware

was typing my response when I got Nick’s, which I agree with almost entirely.

Key points:

-atracurium’s time to onset is too long; 2-3min both in textbooks (per google) and reported here. nothing about that is helpful

-this isn’t DSI, as the paralytic is given. I’m not an anesthetist but this strikes me as nearly identical to standard induction. DSI is sedative, oxygen, paralytic, with the general idea being that if the patient is too delirious to preox for RSI, it’s probably safer to calm them down and maintain their airway reflexes and hopefully even their spontaneous ventilation for preox before paralyzing them


Seth Trueger, MD
Health Policy Fellow
Emergency Medicine

Jim DuCanto

Very interesting and pertinent to our collective further development of clinical practice–the “State of the Art” needs an update.

I was in the process of pulling reference number 5 when the case I was providing anesthesia for abruptly ended–it is the reference to a 2010 study of classic RSII in the pediatric population. I will get it today and share with you guys.

Ever since I got the Oxylator in 2008, I’ve been using it for what they call cRSII. It’s almost tailor made for that application.

I’m coming up with a prototype cRSII “checklist” as it were. It underscores the need for two person BVM, but furthermore underscores that using automated ventilation devices will maintain the safety of the technique (limit pressures, flows, use two handed ventilation technique for better mask fit (pulling the face UP to the mask in edentulous patients is one example). More to come.


2 thoughts on “Controlled rapid sequence induction & intubation

  1. Great comments above.

    I would add that I think that the fact that they are not performing true RSI actually makes the findings MORE significant. Despite NOT optimising all other RSI related factors, providing “gentilation” and not using cricoid resulted in no adverse outcome. i.e. even when creating great conditions for aspiration to occur we still see none due to cricoid or gentilation.

    I disagree re cricoid: I don’t think one needs to believe that it has NO VALUE at all to justify not using it. One just needs to believe on the balance of the evidence that it should not be used and the evidence shows definite harms and no benefit. Further if one believed that it might have some benefit but this is generally outweighed by the harms in most situations then one could argue for using it only in those situations. So I don’t think it is necessarily as black & white as Chrimes points out but certainly I can see his point regarding the issue with using it in some situations only and largely agree with the horse has bolted concept. Having said all that I don’t believe we should use cricoid at all.

  2. I do think we need to develop a consensus on accepted modifications to RSI – reason being, there have been practitioners ‘hung out to dry’ by performing RSI that differs from the technique described by Stept and Safar

    …despite the fact, as Nick Chrimes says, many of us mean RSI as “giving a pre-determined dose of induction agent as bolus immediately followed by a rapid onset paralysing agent, typically roc or sux ….whilst performing manoeuvres to reduce risk of aspiration”

    I’ve made a faltering start to this in Crit Care Horizons last year – Airway Management of the Critically Ill Patient: Modifications of Traditional Rapid Sequence Induction and Intubation

    It is my hope that a consensus can be determined to help guide those performing RSI outside of OT, acknowledging accepted changes in traditional thio/sux/tube RSI and offering a degree of defence when medicolegal opinions still refer to TST as the Gold standard…

    Shouldnt be that hard to get an agreement, eh?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: