Ann Thorac Surg 2000;69:931-933
© 2000 The Society of Thoracic Surgeons
Case Reports
Successful treatment of massive arterial air embolism during open heart surgery
Stefan Huber, MDa,
Bruno Rigler, MDa,
Heinrich E. Mächler, MDa,
Helfried Metzler, MDb,
Freyja M. Smolle-Jüttner, MDc
a Division of Cardiac Surgery, University Clinic of Surgery Graz, Karl-Franzens-University, Graz, Austria
b Division of Cardiac Anesthesia, University Clinic of Surgery Graz, Karl-Franzens-University, Graz, Austria
c Division of Thoracic and Hyperbaric Surgery, University Clinic of Surgery Graz, Karl-Franzens-University, Graz, Austria
Address reprint requests to Dr Huber, Klinische Abteilung für Herzchirurgie, Univ. Klinik für Chirurgie Graz, Karl-Franzens-Universität Graz, Auenbruggerplatz 29, 8036 Graz, Austria
e-mail: stefan.huber{at}kfunigraz.ac.at
 |
Abstract
|
|---|
We report a case of a 5-year-old girl who suffered a massive arterial air embolism during surgical closure of an atrial septal defect. The risk of permanent neurologic deficits or even fatal outcome is significant (mortality rate, 31%). We successfully treated a proven arterial air embolism with intraoperative (retrograde cerebral perfusion) combined with postoperative procedures (deep barbiturate anesthesia and hyperbaric oxygenation). At discharge the girl had fully recovered from the initial neurologic defects.
 |
Introduction
|
|---|
Massive arterial air embolism accompanying cardiopulmonary bypass is a very rare complication. However, it can result in serious brain damage or even death. The key to controlling air embolism lies in prevention. Guidelines for managing gross air embolism should be established and each open heart team should be prepared if suddenly faced with this problem [1]. Different intraoperative and postoperative procedures have been described for managing air embolism. We report a case in which we successfully treated a proven arterial air embolism with a combination of retrograde cerebral perfusion, deep barbiturate anesthesia, and hyperbaric oxygenation.
A 5-year-old girl (body surface area, 0.77 m2) was admitted for surgical closure of an atrial septal defect. The circulatory assist device, a first-generation type from Stöckert (Munich, Germany; flow rate, 2,156 mL/min), and the oxygenator Shiley S-070 (Shiley, Irvine, CA) were used. The arterial line filter was a Pall LPE 1440 (Pall, St Germain, France). The blood level sensor was a first-generation detector from Stöckert (type 23-22-00). It had been placed in proper position (60 mL blood level) and no malfunction had been detected on a routine check before extracorporeal circulation. Just after cross-clamping of the aorta (esophageal temperature, 32°C), administration of crystalloid cardioplegia, and right atriotomy, the blood level inside the oxygenator suddenly dropped significantly due to frequent arterial pressure changes. However, the blood level sensor did not signal an alarm and a pump stop was not initiated. A large bolus of air over a length of 15 cm (calculated volume of 10.6 ccm [volume = radius2 x
x length; radius = diameter/2 = 0.95 cm/2 = 0.475; Length = 15 cm]) rushed through the arterial line (9.5 mm, inner diameter) and was detected visually. On instant alert, the surgeon clamped the arterial line and the perfusionist stopped the pump manually. Nevertheless, the surgeon did detect visually the entry of air into the childs aorta, so that cerebral air embolism had to be assumed. The venous line was clamped as well, and the child was placed in a steep head-down Trendelenburg position. The arterial cannula was removed and the arterial line was connected to the venous line. Retrograde cerebral perfusion was performed at 32°C with a flow of 300 mL/min directed up the superior vena cava. This was continued until the blood mixed with bubbles returning to the site of the aortotomy was free of air. After 5 minutes of retrograde perfusion, standard bypass was resumed and the atrial septal defect was closed quickly. At this stage the anesthetist reported the childs pupils dilated without any reaction to light. A second phase of retrograde cerebral perfusion was performed after the end of the aortic clamping period (15 minutes at 32°C). Subsequently, standard bypass was resumed until the end of extracorporeal perfusion (total extracorporeal perfusion time, 69 minutes). Simultaneously, steroids and diuretics were administered for cerebral protection. An incision of the tympanic membrane in the ear (paracentesis) was performed on both sides in preparation for hyperbaric oxygenation. The sternotomy was closed as quickly as possible. At the end of the surgical procedure the childs pupils were contracted. She was transferred straight away to the hyperbaric pressure unit and the first of five sets of hyperbaric oxygenation was started (duration, 67 to 74 minutes; depth, 15 to 20 m, at 2.56 ATA [ATA = atmosphere = 1.0/325 bar; 1 bar = 105 Pascal (Pa) 2.56 ATA = 2.59 x 105 Pa]). At the end of this procedure, an epidural pressure probe was placed for continuous control of cerebral pressure level changes. The child was held in deep anesthesia with barbiturates for 2 days. The intracranial pressure was kept at 11 to 15 mm Hg, the upper body was held in a 30-degree upright position, the blood sugar level was not allowed to exceed 150 mg/dL, and the childs surface was cooled to keep the body temperature at 37°C. During the first 2 postoperative days the childs pupils were dilated, however, a computed tomographic scan of the brain, performed on the 2nd postoperative day, showed normal ventricular sizes and normal differentiation of the cortex and the medulla. In addition, no signs of intracranial bleeding nor displacement of brain substance could be detected. On the 3rd postoperative day, electroencephalographic (EEG) measurements of evoked potentials showed sustained basic and cortical answers to stimulation, and the anesthesia was changed to dormicum. On the 5th day, the pupils were contracted, isocoric, and showed prompt reaction. The child moved all extremities asymmetrically, and showed intensive reflexes to external stimulation. The intracranial pressure probe was removed on the 6th day. On the 7th postoperative day extubation was done. The neurologic examination revealed pathologic signs, ie, spontaneous oral twitches and asymmetric movements of the extremities with weakness mainly on the left side. However, the childs pupils were contracted and the eye movements were symmetric. The child was conscious, though still in a reduced condition. Her general condition improved quickly, and on the 14th postoperative day she was discharged with no signs of neurologic defect.
Two years later the parents, both teachers, have not noticed any pathologic changes in the development of their child.
The checkup of the blood level sensor proved that the threshold of tolerance had been out of range.
 |
Comment
|
|---|
Massive arterial air embolism is a very rare complication during extracorporeal circulation. The most common reason for air to enter the arterial system during extracorporeal perfusion is that the blood level inside the oxygenator gets too low [1]. To prevent this, a blood level sensor must be placed in proper position and correct function must be guaranteed. In addition, a defoaming filter should prevent the occurrence of microbubbles. Upon detection of air bubbles, a system device should give either an audible or visual alarm to stop the perfusion. Constant attention to the oxygenator blood level is mandatory throughout the bypass procedure and instant access to the cutoff switch is imperative [1]. However, in most cases the speed of air passage is too fast, leaving not enough time to clamp the arterial line nor to pass the information to the perfusionist for a prompt pump stop. Automatic tubing occlusion in response to air bubble detection is not used routinely [2].
Additional reasons for massive arterial air embolism are mainly iatrogenic due to surgical procedures. Cardiotomy suction tubing wedged deep into the pulmonary artery resulting in air being drawn into the left atrium had been documented, as well as inadvertent reversal of left ventricular vent suction tubing [1]. Furthermore, one case has been reported in which a communication had been created accidently between the adherent lung parenchyma and the left atrium. This caused a massive influx of air from the lung into the systemic circulation causing fatal cerebral embolization [3].
As opposed to venous air embolism, in cases of arterial air embolism no clinical signs such as electrocardiographic changes, increase in pulmonary pressure, decrease in systemic blood pressure, decrease in end-tidal carbon dioxide concentration (CO2), increase in arterial CO2, or decrease in arterial oxygen tension occur intraoperatively. Arterial air bubble detection using transesophageal echocardiography or intracranial Doppler and EEG are not yet routine procedures during open heart surgery [4, 5].
Statistical data from Kol and colleagues [6] showed that out of 6 patients only 2 recovered completely; 2 patients died immediately [6]. Data from Mills and Ochsner [1] confirm the high risk of arterial air embolism (mortality rate, 31%; 4/13 patients). Neither deep barbiturate anesthesia nor use of hyperbaric oxygenation was employed in any of the 13 patients in that report [1].
It is obligatory for each open heart team to be prepared to deal with this situation should it arise. Considering our management we need to discuss if retrograde cerebral perfusion in a head-down Trendelenburg position will increase the risk of coronary artery air embolism. Further it is advisable to monitor the venous pressure of the innominate vein to adjust the retrograde perfusion rate. Additionally, it is beneficial to temporarily compress the carotids during retrograde perfusion so that the vertebral arterial system is purged of air.
However, we conclude that our case demands special attention, since a proven arterial air embolism was treated successfully (complete recovery with no permanent neurologic sequelae) with a combination of retrograde cerebral perfusion, deep barbiturate anesthesia, and hyperbaric oxygenation.
 |
References
|
|---|
-
Mills N.L., Ochsner J.L. Massive air embolism during cardiopulmonary bypass. Causes, prevention, and management. J Thorac Cardiovasc Surg 1980;80:708-717.[Abstract]
-
Paulsen A.W., Hargadine W.L., Lambert G.S., Long A.C. A technique for automatic tubing occlusion in response to air bubble detection when using a centrifugal pump. J Extra Corpor Technol 1990;22:98-100.[Medline]
-
Kole S.D., Sakensa D.S., Oswal D.H. Fatal cerebral air embolism during open heart surgery caused by lung parenchyma to left atrium communication. J Heart Valve Dis 1994;3:583-584.[Medline]
-
Hoka S., Okamoto H., Yamaura K., Takahashi S., Tominaga R., Yasui H. Removal of retained air during cardiac surgery with transesophageal echocardiography and capnography. J Clin Anesth 1997;9:457-461.[Medline]
-
Dalmas J.P., Eker A., Girard C., et al. Intracardiac air clearing in valvular surgery guided by transesophageal echocardiography. J Heart Valve Dis 1996;5:553-557.[Medline]
-
Kol S., Ammar R., Weisz G., Melamed Y. Hyperbaric oxygenation for arterial air embolism during cardiopulmonary bypass. Ann Thorac Surg 1993;55:401-403.[Abstract/Free Full Text]
Accepted for publication July 7, 1999.