Venous Air Embolism: The Severity Depends on Many Factors
Introducing air into the venous system may not cause hemodynamic or neurologic consequence, because absent a right to left shunt, the lung is a superb filter for air bubbles.2 Small amounts of air can be removed from the pulmonary vascular bed by gas diffusion across the arteriolar wall and into the alveolar spaces. However, when the capacity of the lung to remove gas is exceeded (for the adult dog is about 20 mL/min),2 pulmonary outflow tract obstruction with or without concomitant arterial embolization can occur. The direct effect of a large venous air embolus is obstruction of the pulmonary vasculature resulting in circulatory collapse. The nonobvious damage is caused by bubbles in the microcirculation resulting in endothelial damage, cytokine release, microthrombosis, and the resulting tissue ischemia. The clinical consequences of a venous air embolism not only depends on the volume of air introduced, but also the speed, the geometry of the bubbles (large unified air bubbles versus microbubbles), the presence of a right to left shunt, and the patient’s baseline cardiac function.3
As early as 1809, Nysten estimated the lethal dose of air to be 40 to 50 mL in a small dog and 100 to 120 mL in a large dog.4 The lethal volume of air in an adult human is unknown, but is estimated to range from 200 to 300 mL from multiple case reports.3 We performed a simple calculation using data from Dr Varga’s article. If the speed of the IVF infusion is 300 mL/h, it will then take 200 minutes to infuse 1 L of room temperature IVF. That means during every minute of infusion, only about 0.03 mL of air will be released into the patient’s circulation, assuming all the air (4.7 mL per the study) will be released as it is warmed by body temperature. This volume of air introduction is significantly lower than the volume identified in animal studies or human case reports as potentially harmful. This air volume (4.7 mL/L IVF, that is, about 0.05 mL per 10 mL IVF) is also roughly 19 times less than the air volume used during a bubble study to detect intracardiac shunting, where 1 mL of air is mixed with 9 mL of normal saline.
If we rapidly transfuse blood to mimic a clinically emergent situation at a rate of 500 mL/min (the maximal infusion rate by a Belmont Rapid Infuser, Belmont instrument corporation, Billerica, MA), the theoretical outgassed air volume and speed should be about 5 mL and 5 mL/min, respectively. In this situation, without an external fluid warmer to trap part of the outgassed air, the air is released into the circulation at that rate may indeed overwhelm the natural lung filtering system. However, an air removal device is often in place to trap outgassed air.
In the setting of intracardiac or intrapulmonary shunting, the tolerance to venous air will be much lower, because the protective effect of lung filtering system is eliminated. It was suggested by Varga et al1 that the internally outgassed air may be released as small bubbles, but the exact size is unknown.