Endotracheal intubation is hazardous

Which is no surprise, I hope, to any of us.

Neil Finer has been a leader in the field of recognizing and quantifying the adverse physiologic effects of endotracheal intubation, and of finding ways to reduce those effects using premedication.

It was largely as a result of his work that professional societies now recommend using medications prior to intubation; medications that reduce pain, help to stabilize the babies physiology and facilitate the intubation.

We have recently, in our NICU, restricted the performance of endotracheal intubation in our most fragile patients to only those professionals who have already demonstrated their competence with larger, more stable babies. Infants under 29 weeks gestation are now only intubated by physicians, NNPs, or respiratory therapists who have shown that they can intubate larger infants. Babies with diaphragmatic hernia are also only intubated by a restricted list of people. I think that the sickest babies are not the place where a first-timer (or a second-timer) should be learning what is a difficult skill. Although junior residents who may in the future become the physician covering a delivery service in a peripheral hospital, and may well become the NRP team leader, do need to learn how to perform this skill well, our first priority has to be the babies, and making sure the most at-risk, tiniest, babies have the most skillful person performing procedures. It would be better, I think, to also have restrictions for the sickest larger babies with very stiff lungs who often desaturate severely and very quickly during intubation, but we have to find ways to ensure that residents leaving the training program have enough exposure to become competent.

We have submitted an abstract with our data to the next PAS meeting, and I will reveal our results at a later date, except to say that more experienced intubators are much more likely to intubate on the first attempt.

Does this matter? Well a new observational study from Hatch and co-workers in Vanderbilt studied 273 intubations in 162 patients. Adverse events occurred in 107 (39%) intubations with nonsevere and severe events in 96 (35%) and 24 (8.8%) intubations, respectively.

Nonsevere events included : Esophageal intubation with immediate recognition, Mainstem bronchial intubation (confirmed by chest radiograph), Oral/airway bleeding, Difficult bag-mask ventilation, Emesis, Chest wall rigidity

Severe complications included: Hypotension receiving treatment, An urgent or elective intubation becoming an emergency, Chest compressions, Code medications (presumably this means needing an epinephrine bolus), or Pneumothorax.

Adverse events were much more frequent for emergency, rather than elective or urgent intubations, and the odds of having an adverse event were doubled if there was more than one attempt at intubation. Infants who needed 3 or 4 attempts had about a 75% chance of an adverse event. Novice intubators had 22% success on the first try while for experienced intubators that was up to 57%.

Hypoxia and bradycardia were not counted as adverse events, as the authors wanted to be able to compare their data to studies looking at intubation for older children. But those two “secondary” outcomes were very common. 44% of the babies desaturated to below 60% saturation, and 24% had a bradycardia to less than 60 per minute.

We continue, in our NICU, to allow less experienced intubators to be the first to attempt intubation on larger babies as we think that the consequences, in the  long term, for the larger baby are probably less important, but are there long term consequences of all these adverse events in very immature babies?

A study from Stanford by Wallenstein and colleagues suggests that there are serious consequences. They compared the outcomes of babies under 1000g who needed intubation in the delivery room between those who were successfully intubated on their first attempt to those who required more than 1 attempt.

There were 88 babies over a 6 year period who were in their cohort.  40% were intubated on the first attempt and 60% required multiple attempts. Babies who were not intubated the first time were more likely to need chest compressions after the first attempt, were more likely to develop a grade 3 or 4 IVH, were more likely to develop NEC, a pneumothorax or PVL. Differences which remained after adjusting for other risk factors, and each of which individually was not “statistically significant”.

Death or neurodevelopmental impairment occurred in 29% of infants intubated on the first attempt, compared with 53% of infants that required multiple attempts, adjusted odds ratio 0.4 (95% confidence interval 0.1 to 1.0), P<0.05. Which was due to a more than doubling in the odds of dying (from 11 to 25%), and nearly a doubling in the odds of “neurodevelopmental impairment”.

What should we do about this?

Firstly, I think that our approach, of restricting intubation attempts by inexperienced personnel to only the more stable babies, is the way to go. More experienced personnel have a much higher rate of success on the first attempt, and fewer attempts means much fewer complications.

Secondly, we should all have strict protocols for premedication for elective intubations. The evidence of benefit, and the reduction of adverse events during premedicated intubations is very clear. One criticism of the data from Vanderbilt is that almost none of the intubations were preceded by muscle relaxation, and those that were pre-medicated used an opiate and a benzodiazepine. This despite the good evidence that muscle relaxation facilitates and shortens intubation, improves intubation conditions, and improves success on the first attempt; and the complete lack of evidence for a benefit of benzodiazepines.

Thirdly we need to find ways of training junior staff in this skill that do not include exposing critically ill babies to inexperienced intubators with a low success rate, including much more extensive use of simulation, simulations which are much more realistic, video-laryngoscopy, improved video-laryngoscopes, and so on.

Fourthly, we need to continue to investigate ways to make intubation less traumatic, less painful, less frequently unsuccessful, and faster.

Finally whenever we are about to intubate a baby, we should ask ourselves if we are in the best place to do it (can we wait till we get the baby to the NICU, place an IV and pre-medicate?) with the best personnel (do we have rules that there is always an experienced person present for every birth of an extremely preterm infant?) the best environment, the best equipment, and the best monitoring.

About keithbarrington

I am a neonatologist and clinical researcher at Sainte Justine University Health Center in Montréal
This entry was posted in Neonatal Research and tagged , , . Bookmark the permalink.

One Response to Endotracheal intubation is hazardous

  1. Pingback: Videolaryngoscopy to teach intubation | Neonatal Research

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s