From the Departments of *Cardiac Thoracic & Vascular Anesthesia, and †Cardiology, Sri Chitra Tirunal Institute of Medical Science & Technology, Trivandrum, Kerala, India.
Accepted for publication October 2, 2013.
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site.
Address correspondence to Dinesh Kumar, US, DM, PDCC, Department of cardiac thoracic & vascular anesthesia, Sri Chitra Tirunal Institute of Medical science & Technology, Trivandrum, Kerala-695011, India. Address e-mail to firstname.lastname@example.org.
Paradoxical air embolism via an atrial septal defect (ASD) or a ventricular septal defect may occur if air bubbles are entrained into the right heart.1,2 The air in the aorta may embolize into the coronary arteries, most often the right coronary artery (RCA) because of its anterior origin. RCA air embolism manifests as acute deterioration of right ventricular (RV) function associated with ST-T changes in the right and inferior electrocardiograph (ECG) leads. We report a case of coronary air embolism (CAE) that was detected on coronary angiography, which occurred to the left anterior descending (LAD) artery after anesthetic induction in a patient scheduled for device closure of an ASD.
Our institutional ethical committee reviewed and approved this case report for publication.
A 50-year-old woman was scheduled for elective device closure of a secundum ASD in the cardiac catheterization laboratory. She presented with a history of dyspnea on exertion, New York Heart Association functional class–II, and palpitations for the past 10 years. Her heart rate was 68 beats per minute (bpm) and regular, arterial blood pressure (BP) was 110/80 mm Hg, and a systolic ejection murmur of grade 3/6 heard in the pulmonary area. The ECG revealed sinus rhythm without any ST-T changes. Her chest radiograph showed cardiomegaly and pulmonary congestion. Two-dimensional transthoracic echocardiography (TTE) performed before the procedure showed a 28-mm ostium secundum ASD surrounded by a 5-mm rim of atrial tissue, through which blood was shunting left to right. Other findings were volume overloading of right atrium (RA) and RV, left ventricular ejection fraction (LVEF) of 58%, and RV systolic pressure of 45 mm Hg derived from a grade II tricuspid regurgitation jet. After premedication with oral diazepam 10 mg routine monitoring including ECG, pulse oximeter and noninvasive BP was begun. Her baseline vital signs were heart rate 62 bpm, BP 110/80 mm Hg, and SpO2 98%. Anesthesia was induced with propofol 2 mg/kg, midazolam 0.1 mg/kg, fentanyl 4 μg/kg and vecuronium 0.2 mg/kg and maintained with oxygen, air, isoflurane, intermittent doses of fentanyl and vecuronium. IV fluid was administered through a standard transfusion set without an air filter. The syringes, extension tubing, and 3-way connectors had been de-aired before injecting drug boluses. After tracheal intubation, her heart rate suddenly decreased to 42 bpm and BP decreased to 60/40 mm Hg. She was treated with boluses of atropine 0.6 mg and phenylephrine 100 μg. This treatment resulted in sinus tachycardia, accompanied by ECG changes of ST ↑ in I, II, V2-5, and avL leads. Because her hypotension was worsening, an infusion of epinephrine was commenced at a rate of 0.05 mcg/kg/min, which increased her heart rate and BP to 120 to 140 bpm and 160/120 mm Hg, respectively. The TTE examination showed anterolateral wall hypokinesia with global dysfunction of the left ventricle (LV). Transesophageal echocardiography (Philips EnVisor, Philips ultrasound) revealed a LVEF of 35% and a cardiac index at the LV outflow tract (LVOT) of 1.6 L/min/m2. The inhaled gas mixture was changed to 100% oxygen. During this period, the cardiologist secured femoral vessel access, and coronary angiography showed slow flow phenomenon in the mid and distal LAD and its branches suggestive of a CAE (Fig. 1, Video 1, see Supplemental Digital Content 1, http://links.lww.com/AACR/A10). With inotropic support, her hemodynamic condition remained stable. Her heart rate decreased to 98 bpm and BP stabilized at about 120/80 mm Hg. The ST elevation decreased and LVEF improved, although regional wall motion abnormalities (RWMA) persisted. Repeat coronary angiogram revealed resolution of the LAD air embolism within 15 minutes, which was evident as spontaneous recanalization (Fig. 2, Video 2, see Supplemental Digital Content 2, http://links.lww.com/AACR/A11). The ASD was closed after deploying a 38-mm device (Cocoon ASD device, Vascular Innovation Company limited). After closure of the ASD, her LV end diastolic pressure (LVEDP) remained increased at 30 mm Hg, which did not decrease despite injection of a bolus of furosemide 60 mg and continued stabilization of vital signs for 30 minutes. Her Troponin-T evaluation performed after stabilizing vital signs were positive (Patient’s value, 0.28 ng/mL, and labarotory normal range is 0–0.2 ng/mL). In view of an acute coronary event and persistently increased LVEDP, we anticipated an increased risk of pulmonary edema had the device been left in situ. Hence, the device was retrieved, and the patient was transferred to cardiac care unit. Inotropic drugs were continued, and nitroglycerine was added to increase coronary perfusion. The TTE showed persistent RWMA of the mid, apical, anterior, and lateral walls along with a LVEF 40%. Her trachea was extubated after 12 hours of mechanical ventilation, and the inotropic drugs were withdrawn over the next 24 hours. The patient was discharged without any neurological deficits. TTE examination a week later revealed a LVEF of 55% and mild hypokinesia of the lateral wall. At the patient’s request, surgical closure of the ASD was performed uneventfully after 6 months.
Venous air embolism usually follows infusion of bubbles adhering to the IV infusion set, extension tubing, 3-way injection ports, and residual air in drug-filled syringes. Preventative measures include de-airing the syringes and IV fluid administration sets and using bubble filters. Although we did not witness air entering the venous circulation, we could not exclude the possibility of small bubbles, the visibility of which could have been masked by the propofol emulsion.
Paradoxical air embolism has been reported to occur after venous air emboli traverse small defects such as patent foramen ovale.3 In adults with an ASD and pulmonary hypertension, RA pressure remains elevated above the left atrial pressures during ventricular systole, which increases the possibility of RA air bubbles migrating paradoxically into the left heart.4
Once into the ascending aorta, the RCA, which originates more anteriorly than the left coronary artery (LCA), is more prone to air embolism in a supine patient. We are not certain why in our patient, the air entrainment occurred selectively into the LCA. Her coronary angiogram revealed that there were no congenital anomalies of the LCA that might have made it susceptible to receive air bubbles. However, a possible reason for entry of air into the left coronary circulation may be explained by an altered direction of flow in the aortic root. Paulsen et al.,5 during analysis of velocity in the ascending aorta in humans, showed that at the peak of systole, the highest blood flow is directed from the LVOT to the posterior wall of the ascending aorta. With ongoing systole, this area of major flow rotates 90° counterclockwise along the noncoronary leaflet. The interventricular septum usually moves toward the LV in systole and rightward during diastole. This movement of the interventricular septum becomes paradoxical, that is, leftward during diastole in the presence of a large ASD due to RV volume overload. The entire LV moves anteriorly during systole.6 This altered ventricular movement and the LV geometry probably changes the influence of shear forces on the LVOT laminar blood flow and alters the direction of the flow current in the ascending aorta. Another possible mechanism to explain the altered ascending aortic flow currents could be its transient extrinsic compression induced by the systolic expansion of an enlarged pulmonary artery and right pulmonary artery.
Clinical presentation of a CAE depends on the side of coronary circulation involved, size of the air bubbles, and the intracoronary air dose.7 Porcine studies have demonstrated that injection of air bubbles with diameters 75, 150, or 300 μM in a volume of 2 μl/kg resulted in transient depression of regional myocardial function despite maintenance of heart rate, LV pressure, and mean BP.8 Common features include hypotension, ECG changes of myocardial ischemia, and arrhythmias. With involvement of the RCA, the predominant features are ST-T changes in ECG leads II, III, and aVF, which are accompanied with severe RV hypokinesia on echocardiography. The ECG changes in anterior and lateral wall territories, RWMA, and global LV systolic dysfunction are associated with a LCA air embolism. Ventricular arrhythmias and cardiac arrest may ensue when a large volume of air obstructs the coronary perfusion. The diagnostic modalities for detecting CAE include fluoroscopy and coronary angiogram. The air emboli usually divide into smaller bubbles as the air progresses distally in the arterial tree, resulting in a slow flow phenomenon in the epicardial vessels. The typical angiographic appearance of air in the coronary artery during contrast injection is the abrupt termination of the leading edge of the contrast media with a hazy appearance that pulses back and forth on a meniscus of air. There is a temporary cessation of flow in the main artery and its branches.9
The deteriorating hemodynamic condition after CAE should be stabilized with inotropic support, coronary vasodilators, and an intra-aortic balloon pump. Cardiopulmonary resuscitation may be necessary in the presence of ventricular arrhythmias or severe LV dysfunction. The coronary air may be directly aspirated using a catheter under fluoroscopic guidance or dislodged by a guidewire or forceful injection of saline, which causes fragmentation of the large occlusive air bubbles.10 Although the mechanism by which the air emboli dissipates is not clearly proven, it is believed that the bubble shrinking occurs due to nitrogen diffusion into blood and surrounding tissues. This diffusion is significantly accelerated by increasing the mean BP and administration of 100% oxygen.11 Ventilation with 100% oxygen decreases the tissue nitrogen concentration by 90% after approximately 30 minutes, resulting in diffusion of nitrogen from air embolus to tissue, which reduces the bubble size.12
Our patient’s increased LVEDP and impaired LV function prompted us not to proceed with deployment of an ASD device. Elective surgery may be deferred until the myocardium recovers from ischemic damage, and ventricular function returns to normal.
In summary, prevention is far superior to the treatment of CAE; however, once recognized, rapid and aggressive treatment should be initiated to achieve the best outcome. Although rare, air embolism to the LAD may occur in patients with a large ASD and RV volume overload. Clinicians should be alert to the possibility of this complication, which should be treated with hemodynamic resuscitation, adequate maintenance of mean BP, and administration of 100% oxygen. Coronary vasodilators such as nikorandil or nitroglycerine should be used to increase the coronary perfusion. Planned intervention or surgery may be deferred until after the myocardium recovers from the ischemic insult.
1. Sodhi KS, Das PJ, Malhotra P, Khandelwal N. Venous air embolism after intravenous contrast administration for computed tomography. J Emerg Med. 2012;42:450–1
2. Lee HY, Yoo SM. A case of paradoxical air embolism in the coronary artery through a patent foramen ovale demonstrated by coronary CT angiography. Clin Imaging. 2013;37:167–9
3. Yeddula K, Ahmad I, Mohammed SH, Hedgire S, Venkatesh V, Abbara S, Kalva SP. Paradoxical air embolism following contrast material injection through power injectors in patients with a patent foramen ovale. Int J Cardiovasc Imaging. 2012;28:2085–90
4. Oki T, Iuchi A, Fukuda N, Tabata T, Hayashi M, Tanimoto M, Manabe K, Kageji Y, Sasaki M, Hama M. Assessment of right-to-left shunt flow in atrial septal defect by transesophageal color and pulsed Doppler echocardiography. J Am Soc Echocardiogr. 1994;7:506–15
5. Paulsen PK, Nygaard H, Hasenkam JM, Gormsen J, Stødkilde-Jørgensen H, Albrechtsen O. Analysis of velocity in the ascending aorta in humans. A comparative study among normal aortic valves, St. Jude Medical and Starr-Edwards Silastic Ball valves. Int J Artif Organs. 1988;11:293–302
6. Ferlinz J. Left ventricular function in atrial septal defect: are interventricular interactions still too complex to permit definitive analysis? J Am Coll Cardiol. 1988;12:1237–40
7. Dib J, Boyle AJ, Chan M, Resar JR. Coronary air embolism: a case report and review of the literature. Catheter Cardiovasc Interv. 2006;68:897–900
8. Van Blankenstein JH, Slager CJ, Schuurbiers JC, Strikwerda S, Verdouw PD. Heart function after injection of small air bubbles in coronary artery of pigs. J Appl Physiol (1985). 1993;75:1201–7
9. Heupler FA Jr, Ferrario CM, Averill DB, Bott-Silverman C. Initial coronary air embolus in the differential diagnosis of coronary artery spasm. Am J Cardiol. 1985;55:657–61
10. Patterson MS, Kiemeneij F. Coronary air embolism treated with aspiration catheter. Heart. 2005;91:e36
11. Crank J The Mathematics of Diffusion. 19562nd ed Oxford, UK Oxford University Press:2–5
12. Groom AC, Farhi LE. Cutaneous diffusion of atmospheric N2 during N2 washout in the dog. J Appl Physiol. 1967;22:740–5