introduction
Bilevel Sequence Intubation (BSI) refers to initiation of noninvasive bilevel positive pressure ventilation with a backup rate prior to intubation (either using a BiPAP machine or a full-featured mechanical ventilator). BSI is distinct from traditional rapid sequence intubation (RSI), since BSI involves the delivery of machine-initiated, pressure-controlled breaths following administration of sedation and paralytics.
I've been using BSI for about a decade for optimization of the sickest patients, prior to intubation. It's safe and highly effective. The PREOXI trial recently demonstrated that BSI is beneficial among all subsets of critically ill patients (even patients on room air). PREOXI is a high-quality, pragmatic RCT that will change practice. The trial has already been reviewed (e.g., EMCrit, First10EM). When you read the study, please bear in mind that, due to the Hawthorne effect, the control group did an outstanding job of pre-oxygenation – so implementing protocolized BSI in real-world ICUs would cause greater benefit than seen in the PREOXI trial.
BSI will soon the be the standard practice for intubation of critically ill patients. Currently, pre-oxygenation is achieved using a hodgepodge of different techniques that are often ineffective and operator-dependent. Replacing this provider-dependent potpourri with a single, uniform, safe, and effective practice will improve patient safety.
I wrote about BSI in 2015 here. That post is reasonably good, and maybe worth looking at for folks who aren't familiar with BSI. However, I've learned more about BSI over the past decade, so it's time for a fresh post on the nuances of this technique.
getting started: benefits of BSI
The primary benefits of BSI are:
- Administration of 100% FiO2 for several minutes with a pressure-reinforced mask seal (if there is a mask leak, pressurized leak compensation from the ventilator will cause oxygen to leak out of the mask – rather than allowing the patient to suck air into the mask).
- Recruitment of lung tissue using positive pressure.
- Prevention of de-recruitment during sedation/paralysis.
- Early identification of hemodynamic instability induced by positive pressure ventilation (allowing for vasopressor initiation prior to intubation).
- Early identification of patients who are difficult to bag-mask ventilate.
key concept: CO2 clearance from mechanical breaths is probably irrelevant
You might be wondering why I didn't list CO2 clearance above. It's because it probably doesn't matter.
Apnea causes the pCO2 to rise about 2-3 mm per minute. Anyone who has done an apnea test to confirm brain death will recognize that pCO2 rises slowly (in order to achieve a pCO2 rise of >20 mm consistently, a 10-minute apnea test is required).
In BSI, the duration of time when the machine is providing breaths is about one minute, which might equate to a pCO2 rise of ~2-3 mm. Providing gentle breaths during this minute might provide about 25-50% of a normal minute ventilation, so mechanical breaths might blunt the rise of pCO2 by 25-50% (leading to an absolute pCO2 improvement of ~1-2 mm).
Hypercapnia is generally very well tolerated, so the degree of pCO2 rise required to cause clinical harm is usually high. This renders the tiny pCO2 gains provided by mechanical breaths (~1-2 mm) wholly irrelevant.
OK… so why provide any machine breaths at all? A few reasons:
- Machine breaths may improve recruitment and oxygenation (e.g., by increasing the mean airway pressure, and providing spikes in airway pressure that help pop open atelectatic alveoli).
- This is what has been demonstrated to work in the PREOXI trial.
- Providing machine breaths confirms that the patient can be successfully bag-mask ventilated (more on this below).
early evaluation of whether the patient can be bag-mask ventilated
Most patients will maintain a patent airway during BSI (with ongoing tidal volumes recorded on the machine) – which is reassuring. This indicates that if laryngoscopy fails, the patient can be re-oxygenated with bag-mask ventilation.
Rarely during BSI, as paralysis and sedation take effect the airway will occlude (in the final ~20-30 seconds prior to laryngoscopy). Practitioners should anticipate potential airway occlusion during this period and attempt to keep the airway open using a jaw thrust. Nonetheless, airway occlusion will sometimes occur. This should be managed as follows:
- Don't panic. The airway occlusion has trapped positive-pressure oxygen in the patient's lungs – so they won't desaturate rapidly. Just wait until the patient is fully relaxed and proceed with intubation.
- If the first intubation attempt fails, then re-oxygenation should be performed with the use of an airway adjunct (e.g., bag mask ventilation with an oral airway, or laryngeal mask airway).
early identification of hemodynamic instability
Bilevel positive pressure ventilation is often initiated ~5-15 minutes before intubation. This provides some time to evaluate how the patient responds hemodynamically to increased intrathoracic pressure. Some patients will experience hypotension during this period of time, allowing for immediate initiation of a vasopressor infusion.
Early identification of tenuous hemodynamics prior to intubation is useful, because this allows for hemodynamic stabilization before administration of sedation and full conversion to positive-pressure ventilation.
nuts & bolts: how to optimize BSI performance
Based on the above concepts, we can set the ventilator in a rational fashion:
try to maximize the expiratory pressure (aka ePAP, aka PEEP)
- A key principle here is that we are fundamentally trying to maximize the ePAP.
- Expiratory pressure is probably providing the bulk of the benefit of BSI:
- ePAP is the primary driver of mean airway pressure and recruitment.
- ePAP is the primary factor that affects preload, allowing for early assessment of hemodynamic deterioration.
- In order to maximize the ePAP while providing some degree of ventilation, it might be reasonable to set the ePAP at 4-5 cm below the iPAP.
try to keep the inspiratory pressure (iPAP) at a safe level
- Excessive iPAP may increase the risk of gastric insufflation and aspiration. Most literature suggests that keeping iPAP below ~20 cm is safe, but this precise cutoff remains poorly defined. (34992795)
- Overall, a range of 10-18 cm iPAP seems reasonable for most patients (selected based on assessments of the risk of aspiration, as compared to benefits from improved recruitment).
- For patients with refractory hypoxemia, higher settings may occasionally be reasonable (e.g., iPAP of 25 cm). The use of higher iPAP may be especially reasonable for patients who have a nasogastric tube placed to suction (which may help to reduce the risk of gastric insufflation).
examples of how this would work:
- [1] Conservative setting (e.g., increased concern for aspiration): 10 cm iPAP / 5 cm ePAP.
- The PREOXI trial required at least 10 cm of iPAP and 5 cm of EPAP. Thus, pressures below this shouldn't generally be utilized.
- [2] Generic setting: 15 cm iPAP / 10 cm ePAP.
- [3] Severe hypoxemia setting: 18 cm iPAP / 14 cm ePAP.
- [4] Maximally aggressive for refractory hypoxemia: 25 cm iPAP / 21 cm ePAP.
machine respiratory rate:
- Patients are often tachypneic prior to intubation (e.g., respiratory rate 25-35). It's often reasonable to set the machine's respiratory rate slightly below the patient's native respiratory rate (so the ventilator doesn't trigger while the patient is still awake). A rate of ~20 breaths/minute is often reasonable.
- The PREOXI trial required a respiratory rate of ≧10 breaths/minute, so the respiratory rate should not be lowered below 10 breaths/minute.
- As the patient's breathing slows following sedation and paralysis, the machine will seamlessly start providing breaths.
- BSI (bilevel sequence intubation) has been shown to avoid hypoxemia and peri-intubation cardiac arrest. Adoption of this as standard care will improve patient safety and cognitively off-load practitioners (so that they can focus on other aspects of airway management).
- Carbon dioxide clearance by machine-triggered breaths is miniscule and unlikely to contribute to the benefits of BSI.
- The primary benefits of BSI are likely the administration of 100% FiO2 and positive mean airway pressure that maintains recruitment. Consequently, maximization of ePAP (i.e., PEEP) makes sense.
- Inspiratory positive airway pressure (iPAP) is limited to below roughly ~20 cm, to reduce the risk of aspiration.
- Ventilator settings may be determined based on a balance of the risk of hypoxemia versus the risk of aspiration:
Acknowledgement: Thanks to Dr. Scott Weingart (@emcrit) for thoughtful comments on this post.
more information on PREOXI:
- Manuscript at NEJM.
- Scott Weingart (EMCrit) on PREOXI: Initial cast and follow-up wee.
- Critical Care Time podcast.
- Justin Morgenstern (First 10 in EM).
- PulmCrit: ADAPT and SCREEN trials are full of sound and fury, signifying little - December 13, 2024
- PulmCrit: How to quickly create a useful professional account in BlueSky - November 28, 2024
- PulmCrit Wee: Why MedTwitter should move to Bluesky - November 15, 2024
Excellent. Thank you
Great review Josh.
In the UK we typically use a Mapleson C which would provide the ePaP and then ventilate through apnoea using the same device.
Would NIV machinery offer much benefit over a Mapleson C as seemingly doing the same thing ?
Excellent physiologic why and how.
Im anesthesiologist from Poland and what really bugs me is this. During my anesthesiology training this protocol (more or less) was standard before every general anesthesia with ETT placement (and i think it is pretty much anywhere else). First CPAP 5cm for preoxi then after sed/paralitic switch to biLevel. In healty lungs you have up to 5 min before desat even occurs.
I’d argue that one other benefit is that, with waveform capnography, the presence of a good trace during mandatory apnoeic ventilation demostrates both that the airway is open and that there is a reasonable/partial seal. The loss of trace indicates either loss of seal or loss of airway patency, both of which can be then rapidly rectified. Without mandatory ventilation, this isn’t immediately obvious. Great post.
Such a great article! I would be interested in your thoughts on performing BSI in patients with TBI – especially i’d like to know if PRVC ventilation could come in handy (train of thought = volume guarantee for co2-consistency + pressure regulation for controlling ICP)
Ive recently studied this in theory but have very little practical experience in this field as of today so any comments / insights / experiences would be much appreciated! 🙂
PRVC could of course be introduced including ASB for preoxygenation (which is why in my understanding this could be suitable for BSI)
In patients with a severe metabolic acidosis would the ventilation during the apneic period make a relevant difference to Co2 clearance? I believe the Co2 would be rising more quickly with apnea in this group because of the increased metabolic activity and Co2 production. Prior to using BSI, in the past we used to do a modified RSI with some breaths given during the apneic period for patients in this subgroup. Great post!