December 30, 2015

I had a situation at work recently that got me thinking on the topic of intraosseous cannulation, and whether it is as widely taught and used as it should be. I was called in at 0200hrs to the hospital Emergency Department of the regional facility that I work at to help out with the resuscitation and transfer of a patient who had presented in an advanced state of septic shock. By the time I got there the patient’s temperature was sitting at 40.5 C, her blood pressure was 70/40mmhg and her pulse was 120. Her conscious state was drifting off, and despite multiple attempts at intravenous cannulation, the highly experienced medical officer on duty had been unable to establish access due to the patient’s shutdown vascular state. The patient needed intravenous antibiotics and fluids, and she needed them right then and there. The answer was obvious to me, and came in the form of two intraosseous cannulas, in that case a couple of Bone Injection Guns (BIG), that were swiftly fired into the patient’s proximal tibias, and antibiotics and fluids were being pressure-infused within minutes.

 I don’t recall ever being taught to do intraosseous cannulation as a medical student, and then as a junior doctor I put it in the same category as surgical airways and chest tubes, as one of those procedures to be feared, and done strictly as a last resort. It has only been through my deployed experience with the army that the process of intraosseous cannulation has been demystified for me, and its amazing simplicity and utility been revealed. So much so, there were many instances in Afghanistan where I didn’t even bother trying to establish an IV line and simply drilled in an EZI-IO instead. To validate that statement, these situations were often when there was a tactical situation to consider, low light, on a helicopter, or in a shocked casualty where time was of the essence and finding a decent vein to cannulate was considered unlikely.

 As an instructor of military medics I would encourage them to have a very low threshold to use IO also. The main reason for this was the simplicity and speed of IO insertion, and its reproducibility under duress. It has been an observation of mine that in high-stress situations when the adrenaline is flowing freely that my fine motor skills were not always as reliable as I would have liked them to be. For that reason, the relatively gross motor skills required to drill an IO into a tibial plateau were far more reliable than the finer motor skills required for IV cannulation. I figured if I was finding this in myself, as someone relatively experienced in IV cannulation, than a medic with less IV experience is likely to be even less reliable in establishing access under duress. Indeed, literature suggests that first attempt venous cannulation by experienced medical practitioners can be as low as 76-82% (Benson 2015; Nadler 2015) with the equivalent success rates for first attempt EZI-IO insertion being 90-100% (Santos 2013; Torres 2013), usually achieved in less than 30 seconds (Torres 2013). Another interesting recent pre-hospital study suggests that once an initial intravenous cannulation attempt has failed, the chance of success drops off dramatically with subsequent attempts, with success rates of 50% and 20% for fourth and fifth attempts respectively (Nadler 2015). Whilst the low success rates likely reflects the severity of the casualty rather than the skill of the operator, it suggests that after one or two failed attempts it might be best to put the IV cannula away and get out the IO instead if access is a necessity.

 The following video illustrates an EZI-IO being drilled into the tibia of a shocked Improvised Explosive Device (IED) victim in Afghanistan who had intra-abdominal bleeding from shrapnel wounds. By the time we got to the casualties in the video, the surviving ones were shocked to the point where they no longer had palpable radial pulses. The casualty in the video had an IV line established prior to our arrival, however it had tissued. I apologise in advance for my failure to prepare the casualty’s skin prior to insertion of the device. Whilst he was peppered with shrapnel wounds already and one of the first things to go through the IO was 2 grams of Ceftriaxone, there’s no excuse for not prepping the skin prior to inserting any form of access, especially when it is being drilled into bone.

Figure 1. Correct insertion angle demonstrated above, with incorrect, superior angulation shown below.

So there it is, my twenty cents worth on NPAs, with just a little bit of evidence-based support from the literature. Like so many things in pre-hospital medicine I feel that the best approach isn’t to make our protocols idiot proof, but rather to train our medics to not be idiots! NPAs are an extremely useful, cheap, quick and effective airway adjunct, and with appropriate training can be safely used in the pre-hospital and Emergency Department environment alike to rapidly establish a patent airway. The key is in understanding the potential risks associated with their use and to know when to back out and try something different.

Thanks for your interest, comments and questions are welcome. Cheers, dan pronk


Ellis, D, Lambert, C, Shirley, P 2006, ‘Intracranial placement of nasopharyngeal airways: is it all that rare?’, Emergency Medicine Journal, vol. 23, no. 8, p. 661.

 Kundra, P, Parida, S 2014, ‘Awake airway control in patients with anticipated difficult mask ventilation’, Indian Journal of Anaesthetics, vol. 58, no. 2, pp. 206-208.

 Muzzi, D, Losasso, TJ, Cucchiara, RF 1991, ‘Complication from a nasopharyngeal airway in a patient with a basilar skull fracture’, Anaesthesiology, vol. 74, no. 1, pp. 366-368.

 Steinbrunger, D, Mazur, R, Mahoney, PF 2007, ‘Intracranial placement of a nasopharyngeal airway in a gun shot victim’, Emergency Medicine Journal, vol. 24, no. 4, p. 311.

 Stoneham, M 1993, ‘The nasopharyngeal airway. Assessment of position by fibreoptic laryngoscopy’, Anaesthesia, vol. 48, pp. 575-580.