Intraoperative Echocardiographic Study of Air Embolism During Cardiac Operations

Transcription

1 Intraoperative Echocardiographic Study of Air Embolism During Cardiac Operations Jens Tingleff, MD, Frederic S. Joyce, MD, and G6sta Pettersson, MD, PhD Department of Cardiothoracic Surgery RT, The National University Hospital, Rigshospitalet, Copenhagen, Denmark Background. Central nervous system damage remains a feared complication after heart operations. Air embolism (AE) is one of several possible causes of central nervous system damage. In previous studies, intraoperative transesophageal echocardiography (ITEE) has been used to detect AE, but identification of the periods of risk and the origin of AE is lacking. Methods. Two groups of patients undergoing elective heart operations were studied with ITEE. Group I consisted of 15 patients undergoing true "open heart" operations, either aortic or mitral valve. Group II consisted of 15 patients undergoing coronary artery bypass grafting. Results. In group I (valve operation), ITEE detected AE in all patients, particularly in the period between the release of the aortic cross-clamp and the termination of cardiopulmonary bypass. Furthermore, 12 of the 15 patients had new episodes of AE up to 28 minutes after termination of cardiopulmonary bypass. In the majority of cases, ITEE clearly demonstrated that the air originated in the lung veins and was not air retained in the heart. In group II (coronary artery bypass grafting) episodes of AE were only seen in the period between cross-clamp removal and the termination of cardiopulmonary bypass, and only in half of the patients. Conclusions. Careful standard cardiac deairing did not prevent AE caused by the delayed release of air trapped in the lung vessels. Routine use of ITEE is recommended to assess the thoroughness of deairing procedures. This will help eliminate AE or at least lead to an increased awareness of the problem of retained air. Minimizing AE during open heart operations should contribute to a reduction in central nervous system damage and improvement of intellectual function after heart operations. (Ann Thorac Surg 1995;60:673-7) entral nervous system (CNS) damage has always C been one of the most feared complications after heart operations [1, 2]. There are several possible causes of cerebral damage after heart operations, and embolization of air is one them. One would expect air embolism (AE) to be more common after an operation during which the heart has been opened, and in accordance with this CNS damage is found to be more frequent after valve operations than after coronary artery bypass grafting (CABG). Several procedures for deairing of the heart before discontinuation of cardiopulmonary bypass (CPB) have been established [3, 4]. The relative effectiveness of these procedures has, however, not been well studied and proved, and the problem of cerebral dysfunction after heart operations has not been eliminated [5, 6]. The use of intraoperative transesophageal echocardiography (ITEE) to detect intracardiac air has been reported by several authors [7-12]. The echocardiogram clearly visualizes air bubbles as white reflections because air bubbles are effective reflectors of sound waves due to vast differences in acoustic properties between air and blood [13]. Accepted for publication June 19, Presented at the Sixth International Symposium on Echocardiography in Cardiac Surgery, Washington, DC, Nov 9-11, Address reprint requests to Dr Tingleff, Department of Cardiothoracic Surgery RT-2152, The National University Hospital, Rigshospitalet, 9 Blegdamsvej, DK-2100 Copenhagen 0, Denmark. The aim of the present study was to use ITEE as a method to describe and quantify the problem of AE during routine cardiac operations. Two groups of patients were examined: one undergoing valve operations and the other undergoing CABG. The formal study was designed after insufficient deairing had been observed in too many patients despite several efforts to improve the standard deairing procedures. We hoped to identify periods of risk for AE during and after CPB and to localize the source of the air. Material and Methods Two groups of patients undergoing elective heart operations were studied (Table 1). Group I ("open heart" group) consisted of 15 patients; 3 underwent aortic valve and 12 mitral valve operations. Group II ("closed heart" group) consisted of 15 patients who underwent CABG. The CPB circuit included a hollow-fiber membrane oxygenator (Safe-II; Polystan, Copenhagen, Denmark). The system was primed with Ringer's lactate and carefully deaired. A prebypass filter was used but no arterial line filter. Standard cannulation included cannulation distally in the ascending aorta and a two-stage venous cannula except for mitral valve operations, in which bicaval cannulation was used. Antegrade crystalloid cardioplegia (St. Thomas' Hospital cardioplegic solution) was used in all patients. Deairing after valve operations (group I) included left 1995 by The Societv of Thoracic Surgeons ]95/$ (95)

2 674 MINI-SYMPOSIUM TINGLEFF EF AL Ann Thorac Surg AIR EMBOLISM AND CARDIAC OPERATIONS 1995;60:673-7 Table 1. Patient Characteristics in Group I (Valve Operation) and Group II (Coronary Artery Bypass Grafting) Group I Group II Variable (n 15) (n = 15) Male 9 (60%) 12 (80%) Female 6 (40%) 3 (20%) Age (y) Median Range atrial, apical, and aortic puncture or venting during cardiac filling and pulmonary, compression until no fur~ ther air could be evacuated or observed. In all aortic valve patients either a left atrial or apical vent was used during the operation, and during deairing this vent was open to allow good transpulmonary flow (in mitral valve patients the left atrial suture line was open until deairing was finished). Deairing was followed by aortic root venting at the highest point. Aortic venting was continued after weaning from CPB until no new episodes of AE had been observed for 2 to 3 minutes. Protamine administration was delayed until the aortic vent had been removed. If a large amount (grade 2, see below) of air was observed later, further deairing by puncture of the aortic root and sometimes the left atrial appendix was done. In CABG patients aortic root air was evacuated through the combined cardioplegia and venting cannula. After completion of the proximal anastomoses aortic root air evacuation was allowed through an anastomotic site after release of the side-clamp. Each vein graft was punctured with a fine needle before removal of bulldog clamps from the grafts. Intraoperative transesophageal echocardiographic examination was performed with a commercially available Vingrned CMF 700 (Vingmed, Horten, Norway) connected to a 5-MHz mechanical annular phased-array transducer mounted in a sealed tip at the end of a conventional endoscope. Echocardiograms were continuously ob- tained and recorded on videotapes from the start of the deairing procedure to the end of the operation (closure of the sternum). All videotapes were later analyzed by a trained echocardiographer blinded to other patient data. The study period was divided into three periods: Period I was from the beginning of deairing to release of the aortic cross-clamp. This period of deairing occurred only in group I. Period 2 was from release of the aortic cross-clamp to termination of CPB. This period included reperfusion of the heart, rewarming, and performance of proximal anastomoses in CABG patients. Period 3 was from the end of CPB to the end of the operation. This period included hemodynamic stabilization, hemostasis, and closure of the sternotomy. Air bubbles in the heart on the echocardiogram were registered when characteristic white reflections were seen. The number of gas bubbles was arbitrarily graded on a scale from 0 to 2 (Fig 1): grade 0 = no air, grade I = air that did not dominate the echocardiogram, and grade 2 = air that dominated the echocardiogram. Every observed increase of at least one grade was judged as a new episode of AE. The total duration of the AE episodes was calculated for each period. For every new episode, the origin of the air was judged as follows: air bubbles seen only in the heart and not in the pulmonary veins were classified as coming from the heart cavity, air bubbles seen streaming in the pulmonary veins to the left atrium were classified as coming from the lungs, air bubbles streaming backward from the left ventricular outflow tract into the left ventricle were classified as coming from the aortic root, and episodes that could not be related to any of the above categories were classified as unknown. Results Group I (Open Heart Operations) The duration of period 1 varied from 2.5 to 12 minutes. Grade 2 air bubbles were detected in 4 of 15 patients and grade 1 bubbles in the rest of the patients. In 8 of the 15 A B Fig 1. Intraoperative transesophageal echocardiographic study during the period after termination of cardiopulmonary bypass in a patient undergoing aortic valve replacement. The white reflections in the left atrium (LA) and left ventricle (LV) represent detection of gas bubbles: (A) grade 1 and (B) grade 2 gas bubbles.

3 Ann Thorac Surg MINI-SYMPOSIUM T1NGLEFF ET AL ;60:673-7 AIR EMBOLISM AND CARDIAC OPERATIONS Table 2. Intraoperative Transesophageal Echocardiographic Detection of Air After Termination of Cardiopulmonary Bypass (Period 3) in 15 Patients After Heart Valve Operations (Group I) Last Episodes of Duration of Total Duration of Air Detected Patient Valve Period 3 Number of Air Episodes in After Termination No. Procedure (rain) New Episodes the Heart (min) of CPB (min) 1 Mitral Mitral Aortic Mitral Mitral Mitral Mitral Mitral Aortic Mitral Mitral Mitral Mitral Aortic Mitral CPB cardiopuhnonary bypass. patients, grade I air bubbles were still detected at the end of the period. Any air detected during this period was judged as representing retained intracardiac air. Period 2 lasted 15 to 46 minutes. In all patients, new episodes of air bubbles were detected. New episodes occurred up to 30 minutes after release of the aortic cross-clamp. Overall 34 new episodes of air bubbles were observed during this period. In 12 of the 15 patients ITEE clearly demonstrated bubbles coming from the pulmonary veins. In 3 patients the origin of the bubbles was unknown. Period 3 lasted from 18 to 63 minutes. The observations are summarized in Table 2. New" episodes of air bubbles were detected in 12 of 15 patients, with an onset within the first minute after termination of CPB in 9 patients. In 3 patients, the first bubbles were detected 4 to 23 minutes after termination of CPB. In 3 patients, new episodes of air bubbles were observed as late as 16 to 28 minutes after CPB had been terminated. During period 3 ITEE clearly demonstrated bubbles coming from the lung veins in 10 of the 12 patients. In 2 of the 3 patients who did not exhibit bubbles in this period, profuse collateral blood flow from the pulmonary, veins had been present throughout the operation. In 4 patients, new episodes of air bubbles resulted in air collections in the apex of the left ventricle or the left atrial appendage. This air was verified and removed by needle puncture in all 4 patients. In 3 patients, air bubbles appeared in the aortic cannula in immediate relation to the echocardiographic demonstration of intracardiac air. Group II (CABG) After CABG, period 1 (deairing after cross-clamp removal) did not exist. The duration of period 2 varied from 11 to 41 minutes. Episodes of AE were detected in 8 of the 15 patients. In 4 of the patients, grade 2 intracardiac air was detected in connection with the release of the aortic cross-clamp. In these cases, ITEE clearly demonstrated that the air came from the aortic root. In another patient, grade 1 AE was detected in connection with manipulation of the partial-occluding aortic clamp. Also in this patient, ITEE demonstrated that the air originated in the aortic root. In the remaining 2 patients, the origin of the AE could not be determined. In period 3, no episodes of air were detected in any patients. Normal clinical postoperative observation did not reveal any neurologic deficits in any of the patients in group I or group II. All patients were discharged alive and walked out the hospital. Comment The frequent observation of intracardiac air after termination of CPB in patients undergoing heart valve operations, in spite of several efforts to improve the deairing procedures, stimulated this systematic study to quantify the problem. Several studies have reported use of ITEE for detection of air bubbles during heart operations [7-12]. In the present study we have systematically studied the effectiveness of available deairing procedures and the periods of risk for AE, and we have examined the possible localization of entrapped air. Due to differences in acoustic impedance between blood and particles, particles will produce a cloud of white reflections on the echocardiogram. When the particles are air, they are called microbubbles [14]. The size of particles detectable with ultrasound has been variously estimated to be between 2 and 125 /~m. Several factors influence the detection sensitivity for microbubbles such as differences in acoustic properties, the

4 676 MINI-SYMPOSIUM TINGLEFF E~I AL Ann Thorac Surg AIR EMBOLISM AND CARDIAC OPERATIONS 1995;60:673-7 operating ultrasound device's frequency, and interactions between the microbubbles. In the present study we discriminated between microbubbles and spontaneous echo contrast ("smoke") seen in heart cavities with low flow velocities. The sensitivity of microbubble detection has been shown to be very high in an experimental study [15l. Today, no ideal method for quantification of microbubbles is yet available. In our study, we arbitrarily graded the amount of air bubbles on a scale from 0 to 2. This arbitrary scale made it possible to register new episodes of microbubbles during periods when microbubbles were still present (a change from grade 1 to 2). Although other reflectors could produce a similar echocardiogram, it is probable that the most significant reflector in this study was air. During the initial deairing procedure it was obvious. Presence of air was confirmed at every echocardiographically guided needle puncture of presumed intracardiac air. In addition, bubbles appeared in the aortic cannula immediately after an episode of microbubbles in the heart in a few patients. In CABG patients, no episodes of air bubbles were seen after weaning from CPB, nor did we see air in the aortic cannula. In the present study, ITEE demonstrated that intracardiac air bubbles in patients undergoing ()pen heart operations are likely to show up even after a careful standard deairing procedure has been performed, both during the period of partial CPB (period 2) and after termination of CPB (period 3). In fact, new episodes of air bubbles appeared in the heart more than 20 minutes after termination of CPB. The two most critical periods for air bubble release were otherwise immediately after removal of the aortic cross-clamp and during the first minutes after weaning from CPB. These observations are in agreement with the results of a previous study using transcranial Doppler echography to identify cerebral AlE during heart operations I161. Because ITEE revealed a greater prevalence of microbubbles during true open heart operations (mitral and aortic valve operations) than after CABG, it seems most likely that the bubbles seen represent air admitted to the left ventricle and left atrium through the cardiotomy. Intraoperative transesophageal echocardiography demonstrated that late new episodes of microbubbles came to the heart from the pulmonary veins, suggesting that air had entered and been trapped in the lung vessels while the heart was open. It seems that entrapped air is not mobilized until normal blood flow through the lungs has been reestablished after weaning from CPB. In 2 of 3 patients in whom no microbubbles were seen after CPB, profuse blood flow through the pulmonary veins had been observed during the operation. It is probable that this continuous transpulmonary blood flow prevented air from being trapped in the lungs. The importance of air trapping in the pulmonad' veins during heart operations was reported many years ago [171. It seems to us, however, that its importance as well as the difficulties associated with its removal by standard deairing procedures have been underestimated. At least this was true for ourselves. Microbubbles occurred also after CABG even though the heart had not been opened. Air appeared immedi- ately after release of the aortic cross-clamp in most patients. In 6 of 8 patients it could be determined that the intracardiac air came from the aorta. This air could have entered the aorta at a cannulation or anastomotic site. The other episodes are of uncertain origin and more difficult to explain. Air from venous infusion lines is normally prevented from reaching the systemic circulation by filtration in the pulmonary vasculature. The filtration capacity of the lungs can be exceeded, however, when the rate of venous air infusion reaches a critical level [18]. Moreover, factors such as anesthetic agents and puhnonary vasodilators may alter the pulmonary filtration capacity 119, 201. An in-vivo study has also shown that small, sonicated microbubbles may pass unhindered through the pulmonary circulation [21]. These mechanisms may explain the presence of microbubbles in the heart in all categories of patients undergoing cardiac or other operations. Overall, however, there was no doubt that the total amount of air observed in the heart in CABG patients was very small compared with what was observed in valve patients. No CABG patient demonstrated air after weaning from CPB. It should be emphasized that this study followed a period during which we regularly observed that our deairing procedures did not result in complete evacuation of intracardiac air and that episodes of AE occurred still late after CPB termination. During this period, every recommended method was tried to eliminate the problem. This means that already when we started the formal study our deairing was performed more carefully than before we first observed the problem. Today a standard, careful deairing is performed and includes a short period when a high transpulmonary flow is maintained. Deairing is followed by aortic root venting at the highest point (250 ml/min for at least 5 minutes). Aortic root venting will not totally prevent air bubbles from appearing in the descending aorta as observed by ITEE, but will hopefully reduce the amount of air that reaches the periphery including CNS. Protamine administration is delayed until termination of venting. Intraoperative transesophageal echocardiography is used whenever possible to confirm and guide the deairing procedures. Even if the incidence has decreased, neurologic complications after open heart operations are still at a significant level, ranging from 1% to 5% [5, 22]. When psychometric tests or morphologic studies (computed tomographic scans) are used, evidence of CNS damage has been found to be even more prevalent [23, 24]. That our patients did clinically well does not prove they had no cerebral injuries. Central nervous system complications occurring after cardiac operations could be related to the operation, CPB, intrinsic patient factors resulting in local or general hypoperfusion of the brain, or a combination of these. The most important causes of CNS damage during CPB are altered cerebral perfusion and embolism [22, 25, 26]. Emboli may consist of air or particles such as platelet aggregates, fibrin, fat, tissue, calcium, or other particles related to the operation or the perfusion [27]. The clinical manifestation of cerebral injury may indicate a focal or diffuse injury. Air embolism in the form of large numbers of microbubbles is expected to cause a

5 Ann Thorac Surg MINI-SYMPOSIUM TINGLEFF ET AL ;60:673-7 AIR EMBOLISM AND CARDIAC OPERATIONS diffuse injury and global cerebral dysfunction. Microembolism during heart operations has been demonstrated in patients by retinal fluorescein angiography technique [28, 29]. A neuropathologic study revealed ubiquitous alterations in the brain consisting of localized areas with dilatation of capillaries and small arterioles in patients who had died after a heart operation [30]. The sources of the microembolism observed in those studies were not clear, but air was considered. The number of microemboli observed by the retinal fluorescein angiography technique shows a correlation with the results of psychometric tests done before and after operation, indicating that the severity of the damage in case of microembolism is a question of quantity [29, 31]. All evidence suggests that diffuse microembolism may impair neurologic function without producing visible cerebral infarctions. A few of our patients (3 heart valve patients and 4 CABG patients) were simultaneously studied by ITEE and transcranial Doppler echography, and in all of these patients, the transcranial Doppler echography indicated cerebral embolization simultaneous with ITEE demonstration of intracardiac air bubbles. The following conclusions may be drawn from our study. First, even after careful deairing of the cardiac chambers has been performed, there still is a high risk of AE. The main cause seems to air trapped in the pulmonary veins. This air is not mobilized until full blood flow through the lungs has been reestablished, and mobilization of such air may continue for a long period of time. In patients undergoing CABG, AE is much less prevalent compared with true open heart operations. Improved deairing strategies and better methods to prevent air from reaching the central nervous system should be developed. Routine use of ITEE during heart operations is recommended and will lead to increased awareness of this problem and will contribute to increased efforts to prevent or minimize CNS damage after a heart operation. The fact that observed AE is not always followed by a detectable neurologic functional deficit should not be used as an excuse for not taking the problem seriouslv. Air is still harmful to the brain and other organs, large amounts more than small amounts. We do not know how many air bubbles are too many. References 1. Gilman S. Cerebral disorders after open-heart operation. N Engl J Med 1965;10: Taber RE, Maraan BM, Tomatis L. Prevention of air embolism during open-heart surged,: a study of the role of trapped air in the left ventricle. Surge D, 1970;68: Gallo JI, Martinez R, Ruiz B, Duran CM. A simple technique for removal of residual air after cardiopulmonary bypass (CPB). Thorac Cardiovasc Surg 1982;30: Tsuji HK, Redington JV, Mendez A, Kay JH. The prevention of air embolism during intracardiac surgery. J Thorac Cardiovasc Surg 1970;59: Fessatidis I, Prapas S, Hevas A, et al. Prevention of perioperative neurological dysfunction. A six year perspective of cardiac surgery. J Cardiovasc Surg Torino 1991;32: Furlan AJ, Sila CA, Chirnowitz MI, Jones SC. Neurologic complications related to cardiac surged,. Neurol Clin 1992;10: Krebber HJ, Hanrath P, Janzen R, Ritoff M, Rodewald G. Gas emboli during open heart surgery'. Thorac Cardiovasc Surg 1982;30: Rodigas PC, Meyer FJ, Haasler GB, Dubroff JM, Spotnitz HM. lntraoperative 2-dimensional echocardiography: ejection of microbubbles from the left ventricle after cardiac surgery. Am J Cardiol 1982;50: Topol EJ, Humphrey LS, Borkon AM, et al. Value of intraoperative left ventricular microbubbles detected by transesophageal two-dimensional echocardiography in predicting neurologic outcome after cardiac operations. Am J Cardiol 1985;56: Oka Y, Inoue T, Hong Y, Sisto DA, Strom JA, Frater RW. Retained intracardiac air. Transesophageal echocardiography for definition of incidence and monitoring removal by improved techniques. J Thorac Cardiovasc Surg 1986;91: Diehl JT, Ramos D, Dougherty F, Pandian NG, Payne DD, Cleveland RJ. Intraoperative, two-dimensional echocardiography-guided removal of retained intracardiac air. Ann Thorac Surg 1987;43: Meloni L, Abbruzzese PA, Cardu G, et al. Detection of microbubbles released by oxygenators during cardiopulmonary bypass by intraoperative transesophageal echocardiography. Am J Cardiol 1990;66: Richardson PD. Qualitative and quantitative methods for investigating gas emboli in blood. Med lnstrum 1985;19: Butler BD, Hills BA. Transpulmonary, passage of venous air emboli. J Appl Physiol 1985;59: Duff HJ, Buda AJ, Kramer R, Strauss HD, David TE, Berman ND. Detection of entrapped intracardiac air with intraoperative echocardiography. Am J Cardiol 1980;46: Van der Linden J, Casimir-Ahn H. When do cerebral emboli appear during open heart operations? A transcranial Doppler study. Ann Thorac Surg 1991;51: Fishman NH, Carlsson E, Roe BB. The importance of the pulmonary veins in systemic air embolism following openheart surgery. Surge D, 1969;66: Butler BD. Biophysical aspects of gas bubbles in blood. Med Instrum 1985;19: Butler BD, Hills BA. The lung as a filter for microbubbles. J Appl Physiol 1979;47: Katz J, Leiman BC, Butler BD. Effects of inhalation anaesthetics on filtration of venous gas emboli by the pulmonary vasculature. Br J Anaesth 1988;61: Feinstein SB, Shah PM, Bing RJ, et al. Microbubble dynamics visualized in the intact capillary circulation. J Am Coll Cardiol 1984;4: Furlan AJ, Breuer AC. Central nervous system complications of open heart surgery. Stroke 1984;15: Muraoka R, Yokota M, Aoshima M, et al. Subclinical changes in brain morphology following cardiac operations as reflected by computed tomographic scans of the brain. J Thorac Cardiovasc Surg 1981;81: Aberg T. Effect of open heart surgery on intellectual function. Scand J Thorac Cardiovasc Surg 1974;Supph Henriksen L. Evidence suggestive of diffuse brain damage following cardiac operations. Lancet 1984;1: Gilman S. Neurological complications of open heart surgery. Ann Neurol 1990;28: Taylor KM. Brain damage during open-heart surgery. Thorax 1982;37: Blauth CI, Arnold JV, Schulenberg WE, McCartney AC, Taylor KM. Cerebral microembolism during cardiopulmonary bypass. Retinal microvascular studies in vivo with fluorescein angiography. J Thorac Cardiovasc Surg 1988;95: Blauth C, Arnold J, Kohner EM, Taylor KM. Retinal microembolism during cardiopulmonary bypass demonstrated by fluorescein angiography. Lancet 1986;2: Moody DM, Bell MA, Challa VR, Johnston WE, Prough DS. Brain microemboli during cardiac surgery or aortography. Ann Neurol 1990;28: Blauth C, Smith P, Newman S, et al. Retinal microembolism and neuropsychological deficit following clinical cardiopulmonary bypass: comparison of a membrane and a bubble oxygenator. A preliminary communication. Eur J Cardiothorac Surg 1989;3:135-8.