Abstract
An 25-year-old man was buried by an avalanche during off-slope skiing. He was rescued by his companions and resuscitated by mouth-to-mouth ventilation. The emergency physician from a helicopter based emergency medical service placed two venous lines in both external jugular veins and secured the airway with a tracheal tube. When transferred to the emergency department an additional central venous catheter was inserted via his right femoral vein. The subsequent computed tomography scan revealed several small air bubbles adjacent to the endothelium of the brachiocephalic vein. In an experimental setting, it was shown that air could enter the circulation via a central venous catheter within a few seconds, but measured values of embolising air were smaller than the calculated values when applying the law of Hagen-Poiseuille. Nevertheless, it is important to keep the lumens of a central venous catheter filled with saline before any manipulation in order to prevent or attenuate venous air embolism.
BACKGROUND
Air embolism abruptly entering the central venous circulation may cause pulmonary artery embolism, acute right heart failure or cerebral air embolism via a patent foramen ovale and can lead to death.1,2 In addition, fatal outcome has been reported following cerebral air embolism ascending retrogradely to the brain via the venous circulation.3,4 As venous air embolism (VAE) may occur during insertion,5 maintenance6 and removal7 of central venous catheters (CVCs), physicians need to be aware of this potential hazard.8
CASE PRESENTATION
An 25-year-old man was buried by an avalanche during off-slope skiing. His companions managed to rescue him within 15 min. As the victim did not breath spontaneously but had a palpable peripheral pulse, mouth-to-mouth ventilation was initiated. The emergency physician from a helicopter based emergency medical service placed an 18 gauge venule in the left external jugular and a 14 gauge venule in the right external jugular. He secured the airway with an tracheal tube and transferred the patient to the emergency department. On admission the patient presented in a haemodynamic stable state, artificially ventilated with first degree frostbite of the fingertips on both hands and severe hypothermia of approximately 30°C. Rewarming with a convective blanket system was started. The physician on duty decided to place an additional CVC via the right femoral vein. The procedure was carried out in the emergency room with facilities of cardiocirculatory and respiratory monitoring, but neither x ray nor ultrasound guidance were needed during CVC insertion.
The modified version of the Seldinger technique was applied under sterile barrier protection. The syringe and catheterisation needle were filled with saline before puncture and the catheter was filled with saline before insertion, as recommended in the institutional guidelines. After disinfection of the skin a 16 gauge catheterisation needle was inserted and gentle suction was applied to the syringe during insertion and withdrawal of the needle. Rapid flush of venous blood into the barrel of the syringe indicated that the vein was successfully entered at first attempt. As the syringe was designed to allow direct insertion of the guidewire the physician advanced the wire smoothly through the syringe into the vein without detectable resistance.
With the guidewire in place, the syringe and needle were withdrawn from the insertion site and a dilator was used to dilate the skin opening and subcutaneous tissue. The catheter was then advanced while the guidewire was removed. The insertion depth of the catheter was kept at 24 cm. Blood was easily aspirated from each of the catheter lumens and confirmed the intravascular position. The catheter was secured to the skin with suture and a sterile dressing was applied. Computed tomography (CT) scan was performed and showed signs of lung contusion and a small pneumothorax on the right side. Furthermore, the CT scan revealed several small air bubbles adjacent to the endothelium of the brachiocephalic vein (fig 1).
Figure 1.
Transverse computed tomography image of the chest at the level of the upper mediastinum, showing a small intravascular air bubble in the brachiocephalic vein (arrowhead).
INVESTIGATIONS
In a preliminary experimental setting we investigated the amount of air that can pass through a 16 gauge lumen of a CVC (CS-12123-E, ARROW, Reading, Pennsylvania, USA) under varying pressure gradients. We used a vertical silicone tube filled with citrated pig blood (haematocrit 35%) and a standard cardio-technical pump to simulate blood circulation.4 The CVC was inserted through the tube wall and directed caudate. After the CVC was filled with air an attached T bore stopcock was opened to allow air bubbles to enter via the 16 gauge distal lumen (single orifice at the catheter tip). Experiments with water circulation were done for control. Measured values were compared to predicted values by the law of Hagen-Poiseuille: Q=ΔPπr4(8 η L)−1. At subatmospheric pressure gradients between −3 and −7 cm H2O, air influx through the 16 gauge lumen into circulating water and blood was between 43–180 ml/min (R2=0.97) and between 24–156 ml/min (R2=0.99), respectively. Our results show that a potentially fatal amount of air embolism exceeding 50 ml can enter central venous circulation within a few seconds. However, the actual amount of air was much less than anticipated when applying the law of Hagen-Poiseuille. Measured values of embolising air significantly followed a linear regression and were smaller than the calculated values (fig 2).
Figure 2.
Airflow rates through a 16 gauge tube, 1.19 mm in diameter and 32 cm in length, as calculated according to the law of Hagen-Poiseuille as well as measured through a real 16 gauge lumen of a central venous catheter (CVC) in simulated blood circulation.
OUTCOME AND FOLLOW-UP
The patient was successfully weaned from the ventilator the next day and discharged in good health 2 days after admission to the hospital. During his stay he did not develop any neurologic, respiratory or circulatory symptoms. Beside some numbness in his left hand, his follow-up was uneventful.
DISCUSSION
VAE following venous access, especially insertion, maintenance or removal of CVCs is not a rare event. In our case we cannot exactly attribute the VAE to the CVC alone as both venules in the external jugular veins could have offered access of air to the central circulation. Neither the emergency physician nor the physician on charge in the emergency department had observed any events during puncture and insertion of the catheters. Furthermore, following puncture of the femoral vein there was no need to remove the syringe as the wire was directly inserted through the syringe. Finally, VAE following lung contusion or as a consequence of the pneumothorax is unlikely in the brachiocephalic vein.
Radiologists describing asymptomatic VAE could show that even after injection of intravenous fluids at peripheral venous sites, numerous venous air emboli in the head and neck veins can be detected in various locations of the venous system such as the internal jugular vein, cavernous sinus, frontal or temporal scalp, infratemporal fossa, superior ophthalmic vein, strait sinus, and the superior sagittal sinus.9,10 In particular, air embolism entering the cerebral vessels may be associated with severe residual neurological deficit in up to 38% of cases and mortality may be as high as 23%.2
The amount of VAE is influenced by many factors, including, intrathoracic pressure gradients (depending on volume status, sitting position of the patient, spontaneous breathing), lumen shape and diameter of the CVC, among others. When applying the law of Hagen-Poiseuille it is difficult to determine the true metric internal diameter of the CVC lumen when the lumens are non-circular or kinking (for example, subclavian access). Furthermore, fluid residues such as blood or saline inside the lumen may affect laminar flow at the boundary surfaces. In addition, the calculation disregards entrance and exit effects that were expressed by different air influx values into water and blood. Theoretically, the small Borda-Carnot pressure loss at the catheter tip due to the sudden expansion from the tube into the bubble has to be considered as well. The increased pressure difference due to surface tension at the 16 gauge catheter tip (approximately 2.2 cm H20 for blood and 2.5 cm H2O for water) serves to explain why no measured air influx was detected under pressure gradients of −1 and −2 cm H2O. Presumably, the initial negative driving pressure difference was diminished by this capillary effect in a small circular CVC. This underlines the importance of keeping the CVC filled with saline before insertion in order to prevent or attenuate VAE.
LEARNING POINTS
Air embolism can arise from venous lines in particular during central venous catheter manipulations. Spontaneous breathing in sitting patients is particularly risky.
Flushing the catheter lumen with sterile saline/intravenous fluid before and after use may help to avoid venous air embolism.
The actual amount of venous air embolism through a venous access might be less than the estimated amount of embolising air when applying the law of Hagen-Poiseuille.
Footnotes
Competing interests: none.
Patient consent: Patient/guardian consent was obtained for publication
REFERENCES
- 1.Ordway CB. Air embolus via CVP catheter without positive pressure: presentation of case and review. Ann Surg 1974; 179: 479–81 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Heckmann JG, Lang CJ, Kindler K, et al. Neurologic manifestations of cerebral air embolism as a complication of central venous catheterization. Crit Care Med 2000; 28: 1621–5 [DOI] [PubMed] [Google Scholar]
- 3.Ploner F, Saltuari L, Marosi MJ, et al. Cerebral air emboli with use of central venous catheter in mobile patient. Lancet 1991; 338: 1331. [DOI] [PubMed] [Google Scholar]
- 4.Schlimp CJ, Loimer T, Rieger M, et al. The potential of venous air embolism ascending retrograde to the brain. J Forensic Sci 2005; 50: 906–9 [PubMed] [Google Scholar]
- 5.Inamasu J, Nakamura Y, Saito R, et al. Cerebral air embolism after central venous catheterization. Am J Emerg Med 2001; 19: 520–1 [DOI] [PubMed] [Google Scholar]
- 6.Brouns R, De Surgeloose D, Neetens I, et al. Fatal venous cerebral air embolism secondary to a disconnected central venous catheter. Cerebrovasc Dis 2006; 21: 212–4 [DOI] [PubMed] [Google Scholar]
- 7.Moorthy SS, Tisinai KA, Speiser BS, et al. Cerebral air embolism during removal of a pulmonary artery catheter. Crit Care Med 1991; 19: 981–3 [DOI] [PubMed] [Google Scholar]
- 8.Ely EW, Hite RD, Baker AM, et al. Venous air embolism from central venous catheterization: a need for increased physician awareness. Crit Care Med 1999; 27: 2113–7 [DOI] [PubMed] [Google Scholar]
- 9.Woodring JH, Fried AM. Nonfatal venous air embolism after contrast-enhanced CT. Radiology 1988; 167: 405–7 [DOI] [PubMed] [Google Scholar]
- 10.Rubinstein D, Dangleis K, Damiano TR. Venous air emboli identified on head and neck CT scans. J Comput Assist Tomogr 1996; 20: 559–62 [DOI] [PubMed] [Google Scholar]
