Latest posts by The Communiques (see all)

‘Sometimes you can perform a simple task a thousand times without a problem.Then just once, a rare and devastating event occurs that makes everyone involved question their practice of that previously viewed “simple” task. Gas embolism is one of those events – hard to diagnose, time critical, and difficult to treat. It does not discriminate between the old and the young, or the sick and the healthy’. The following case is published in  the  March edition of the  Clinical Communiqué

Case Number: 2012/2810 Qld Case Précis Author: Dr Gerard J Fennessy MBChB, FCICM CLINICAL SUMMARY

RC was an otherwise well 3½-year old girl who was referred to a regional hospital emergency department (ED), with flu-like symptoms and dehydration. She was admitted for observation, and rehydration with oral and intravenous (IV) fluids. A 1000ml bag of IV normal saline (NS) was being given via an IV giving set (“Set A”) and infusion pump into a peripheral cannula in the back of her hand. The infusion pump had a safety feature preventing infusion if air was detected in the line.

Soon after her presentation, a decision was made to transfer RC to a larger hospital for closer monitoring. This was via a 50-minute aeromedical helicopter flight. When the paramedic arrived, the fluid chart showed that she had received 900mls of NS over three hours. It was agreed that RC had received enough IV fluids so the infusion of NS was ceased and “Set A” was disconnected.

However, the IV fluid bag (containing 100ml of NS) remained with the patient. Due to limited space aboard the helicopter, there were no infusion pumps. Instead, IV fluids were given via a free running IV set (“Set B”) and opaque pressure bag. Prior to take-off, the paramedic removed “Set A” and “re-spiked” the original IV bag with “Set B”. He primed the new giving set, “Set B”, and discarded some NS onto the tarmac.

Following discussion with the ambulance clinician, and believing there was about 300mls of NS left in the bag, the paramedic decided to run IV NS at 250mls per hour, using the pressure bag to maintain IV flow. Midway through the flight, RC suddenly and rapidly deteriorated. She initially had a seizure, then went into cardiac arrest. Immediate resuscitation was attempted in the confined space, and the pilot landed seven minutes later. Resuscitation continued until arrival at the hospital ED. Resuscitation attempts for RC were ongoing in the ED, however these were unsuccessful.

A Chest X-Ray (CXR) showed air in both the heart and the right internal jugular vein. PATHOLOGY The hospital staff notified the coroners’ pathologist of the CXR findings. This allowed the pathologist to specifically examine for air within the body. Autopsy revealed air within multiple intravascular spaces including the right atrium and ventricle, superior vena cava and aorta, and superficial veins covering the brain. There was 70mls of air in the heart – a large amount for a small child. The pathologist concluded that RC had died of massive air embolism. Influenza A was also detected on laboratory tests.


The investigation centred around the mechanism by which air entered RC’s veins. It was established that this occurred during the flight. A paediatric intensive care specialist, called as an expert witness, found that there were a number of contributing circumstances.

These included: “re-spiking” the IV bag, which inadvertently enabled air to enter the IV bag; the use of an opaque pressure bag; lack of an infusion pump or “air” alarm; miscalculating the remaining IV fluid as 300mls instead of 100mls; and difficulties monitoring RC within the retrieval environment. The opaque pressure bag had the dual effect of obscuring the air from view, and forcing air into RC’s veins. The coroner heard that air is able to enter IV bags if they are “re-spiked”. This in itself would not have caused an air embolus, but for the pressure bag, as a gravity-fed drip would have stopped. Furthermore, a new bag of NS would not have contained air. The opaque pressure bag had the dual effect of obscuring the air from view, and forcing air into RC’s veins. Once again, this would not have been problematic had there been more fluid in the bag, or it had not been “re-spiked”. An infusion pump would cease to infuse once air was detected in the line. The “re-spiking” was not done for cost purposes – a bag of NS costs about $1.

Instead, paramedics preferentially used intravenous fluids from the hospital, so as to not deplete their own emergency stocks, enabling them to “maintain operational preparedness”. Once the air embolism occurred, it was unlikely that resuscitation would have been successful. The clinical deterioration would have occurred within seconds of air entering RC’s veins, and removal of the air would have been difficult, if not impossible. The situation was confounded by the helicopter environment and lack of a clear diagnosis.


The coroner made two recommendations at inquest:

  1. That IV saline bags be clearly labelled with ‘Single Spike Only’, in order to prevent air entry, concurrent with an educational program. (Note: although the educational program was implemented, the manufacturer of IV saline maintained that the current ‘Single Use Only’ label was sufficient.)
  2. That ambulance services implement new guidelines regarding priming giving sets and to stop “re-spiking”, and that opaque pressure bags are replaced with clear bags, and infusion pumps for IV fluid delivery are made available on aeromedical flights.


The administration of IV fluids is one of the most ubiquitous treatments in hospital settings, and is generally an uncomplicated and low-risk practice. The use of pressure bags is also widespread because they are effective, reliable and low cost. Air embolism is a rare but severe complication of intravenous access. It is rarely reported with peripheral IV access, more commonly associated with central IV access.

This case highlights a number of clinical practice changes, such as not re-spiking IV bags, and using fresh IV fluids, to prevent air embolus following IV access and fluid administration. Additional technology such as an infusion pump with “air” alarms would have prevented this death, and these should be universally used whenever IV fluids are required. These safety requirements are even more crucial when there are competing distractions for the staff involved in the care of the patient.

Did you know that you can sign up for free to receive the  Clinical Communiqué?  For more information and to register Click here to go to the site.

The thoughts of this blog are of the individual writer and not necessarily those of the Nurses for Nurses Network. To read our full disclaimer click here >>