Providing bag-mask ventilation (BMV) during the apneic phase of RSI improved oxygenation compared to standard RSI and did not increase the rate of complications.
Why does this matter?
The point of rapid sequence intubation [or induction] (RSI) is to preoxygenate, quickly give the sedative and paralytic, and pass the tube without giving positive pressure ventilation during the brief apneic interval. The concern about BMV while apneic is that air may be insufflated into the stomach, increasing the risk of vomiting and aspiration. RSI was introduced in 1970. All aspects of RSI, including BMV, have since been continually debated. Will this RCT put the controversy to rest?
Bagging during RSI?
This was a multicenter RCT that randomized 401 critically ill patients who required intubation in the ICU to either standard RSI (no BMV) or BMV during the apneic phase. Median SpO2 was 96% in the BMV group vs. 93% for standard RSI. Although, the importance of SpO2 as the primary outcome has been debated. Severely low SpO2 (<80%) was also much less common in the BMV group than controls: 10.9% vs 22.8%, respectively (RR 0.48). Witnessed aspiration by the intubator or new infiltrate on CXR was no different between the groups. Patients thought to be at high risk of aspiration, with contraindication to ventilation (i.e. full stomach or pregnancy), were excluded. However, both groups still had about 60% of patients with at least one or more risk factors for aspiration, including about half in each group not known to be fasting in the 6 hours prior to intubation. Standard methods of preoxygenation were allowed in both groups. There was a trend toward the BMV group having a lower starting SpO2, which should have placed this group at a disadvantage for the primary outcome, though it didn’t seem to do so. Big differences between the groups were that BMV was more often used for preoxygenation in the BMV group (39.7% BMV vs 10.9% control, RR 3.65), and non-invasive ventilation or high-flow nasal cannula (HFNC) was more often used in the standard RSI group. This could have been because the group assignments obviously could not be blinded; i.e. patients assigned to the BMV group also used it more often for preoxygenation.
It’s also important to note that BMV was done correctly, which included, “oxygen flow rates of at least 15 liters per minute, a valve attached to the expiratory port of the bag-mask device to generate a positive end-expiratory pressure of 5 to 10 cm of water, an oropharyngeal airway, a two-handed mask seal performed by the intubating clinician with a head-tilt and chin-lift maneuver, and ventilation at 10 breaths per minute with the smallest volume required to generate a visible chest rise.” It’s quite another thing to engage in emotional bagging, squeeze the bag with savage force, and blow the patient’s stomach up like a balloon.
Another major difference between the groups is that patients in the BMV group all received supplemental oxygen from induction to laryngoscopy via the bag. On the contrary, just 78% of the control group received supplemental oxygen during this time (which would be considered apneic oxygenation – ApOx). ApOx may have been used less often in the control group, as this same research group previously found it didn’t work in the FELLOW trial. Both the FELLOW trial and ENDAO (seemingly named with the conclusion already in mind) found no benefit of ApOx. However, several studies have found benefit to ApOx, not to mention a host of anesthesia papers that clearly show it is effective in prolonging apnea times without desaturation. The authors report 78% of standard RSI patients received apneic oxygenation; however, the percentage who received effective ApOx was much lower. Oxygenation delivered during the apneic phase was done, “via non-rebreather (NRB) or nasal cannula (NC).” This is perplexing. A NRB mask doesn’t provide ApOx, as it requires active breathing to be effective. Standard NC was being used in 11.9% (24/202) for preoxygenation, and I can only assume was continued during the apneic phase, yet this would provide little to no impact on ApOx at standard NC flow rates. HFNC up to 70L/min was being used in 20.3% (41/202) for preoxygnation and, I assume, was continued during the apneic phase. Since both NRB and standard NC would contribute almost nothing to ApOx, in reality, only those continued on HFNC could be considered as receiving effective ApOx from induction to laryngoscopy. So, the percentage of patients who would have received any meaningful oxygen delivery during the apneic phase was only 20.3%. Could it be that because this same group conducted the FELLOW trial, in which they concluded ApOx didn’t work, that this technique was seldom used in the control group and contributed to the primary outcome of lower SpO2?
So, will this study change my practice? First of all, they did an excellent job. This paper provides high quality evidence that, in the right patient, BMV during the apneic phase can be done safely and reduces serious desaturation. Now I won’t hesitate to bag patients with poor preoxygenation and hypoxia during the apneic phase. It’s also an important reminder that if I’m going to bag, I need to do it correctly – emotional bagging kills. However, if I can adequately preoxygenate and provide ApOx with HFNC, I don’t plan to use BMV during the apneic phase in my patients. It’s also very important to remember that patients deemed high risk for aspiration were excluded from this study. That would make many of my patients in the ED ineligible for BMV. It would be interesting to see the trial repeated with 100% of the standard RSI patients getting NC at very high flow rates, 60-70L/min, to see how this impacted lowest SpO2 in the control group.
PulmCrit has an excellent post on this paper and goes into depth on the physiology of what is happening here. It is so worth your time to read.
Bag-Mask Ventilation during Tracheal Intubation of Critically Ill Adults. N Engl J Med. 2019 Feb 18. doi: 10.1056/NEJMoa1812405. [Epub ahead of print]
Open in Read by QxMD