Embolic stroke is an uncommon, but devastating perioperative complication of open heart surgery that is seen in up to 2.2% of patients undergoing valve surgery.1 This is partially attributed to the increase in perioperative atrial fibrillation (AF) seen in patients undergoing valve surgery. We present an unusual case where a potential embolic stroke was prevented due to the timely detection of a left atrial (LA) thrombus via transesophageal echocardiography (TEE) after removal of the aortic cross-clamp during a minimally invasive aortic valve replacement and LA appendage ligation via the AtriClip (Atricure, Inc, Westchester, OH) device. Written consent was received from the patient for publication of this case report.
An 84-year-old man with a history of hypertension, chronic AF, and aortic stenosis was scheduled for minimally invasive aortic valve replacement with LA appendage ligation using the AtriClip device. He was anticoagulated with daily apixaban for several years; his last dose was 2 days before surgery. The patient was admitted to the hospital the day before surgery and his anticoagulation was bridged with heparin therapy that was discontinued 6 hours before surgery.
The procedure was performed via a right anterior thoracotomy after the right lung was isolated with a bronchial blocker. Baseline cardiac rhythm was AF. A complete 2-dimensional (2D) intraoperative precardiopulmonary bypass (CPB) TEE examination was performed by an experienced echocardiographer. Findings included critical aortic stenosis (valve area of 0.6 cm2, peak gradient across the valve of 53 mm Hg) and normal biventricular function. Interrogation of the LA revealed enlargement (6.2 cm) and significant spontaneous echocardiographic contrast in the appendage (Figure A). 2D multiplanar evaluation of the entire LA appendage showed no evidence of thrombus. Flow velocities in the LA appendage were low (<40 cm/s). A bubble study revealed no evidence of a patent foramen ovale. The patient was heparinized before commencement of CPB, and an activated clotting time (ACT) >480 seconds was maintained throughout the bypass period. During CPB, surgical manipulation of the LA appendage to occlude and appropriately size the AtriClip device was visualized via TEE (Figure B). The surgical procedure was uneventful. The aortic valve was replaced with a bioprosthesis, and the AtriClip device was deployed across the base of the LA appendage, isolating it from the LA. After removal of the aortic cross-clamp, ventricular fibrillation was noted on the electrocardiogram. One defibrillatory shock converted the rhythm to asystole, V-pacing was started, and deairing maneuvers were commenced. TEE examination revealed no aortic insufficiency and no flow in the LA appendage. However, a large echodense object measuring 1.2 × 1.1 cm was visualized in the LA (Figure C) near the intra-atrial septum. The decision was made to rearrest the heart and retrieve the suspected thrombus via an incision in the LA. An untethered object with the consistency of a thrombus was removed from the LA. The atrial incision was closed and the patient was weaned from CPB. The surgery was completed without any further issues. The patient had an uneventful postoperative course and was discharged home without any complications.
Patients may develop thrombus in the LA appendage within a few days of being in AF,2 and therefore require lifelong anticoagulation to decrease the risk of stroke. Patients who undergo cardiac surgery for other indications can be freed from chronic anticoagulation via exclusion of the LA appendage from the circulation. The LA appendage can be isolated from the LA surgically when patients are undergoing cardiac surgery by internal or external suture ligation. These methods do not always offer permanent exclusion of the LA appendage as up to 36% of patients redevelop flow into the excluded appendage.3 The AtriClip is a self-closing titanium clip applied at the base of the LA appendage that excludes the appendage from the LA. It has successfully maintained separation of the LA appendage from the LA in the long term,4 and may potentially reduce the risk of thromboembolic events.5
In this case, we feel that a previously present thrombus was not detected by the pre-CPB TEE examination and that it broke free from the appendage into the LA during manipulation of the appendage for sizing, or during deployment of the AtriClip device. While we think this is the most likely explanation, we cannot completely exclude other etiologies of thrombus formation. Previous reports have described new intracardiac thrombi after CPB; however, most of them were after reversal of heparin with protamine.6,7 In our case, the thrombus was noted during CPB with full systemic heparinization. Prosthetic valve material can be thrombogenic, and may have played a role in the formation of the thrombus we detected. Because the ACT was always therapeutic and only the aortic valve was replaced, it is unlikely that blood contact with the bioprosthetic aortic valve led to a thrombus in the LA. Additionally, there was no atrial incision to contribute to thrombus formation because the LA appendage was closed externally with the AtriClip device. Kim et al8 reported a new LA thrombus after a mitral valve replacement in a patient with chronic AF and visible spontaneous echocardiographic contrast on the pre-CPB echocardiographic examination. Our patient certainly had similar risk factors for the development of a LA appendage thrombus; he had chronic AF, spontaneous echocardiographic contrast was visible on the pre-CPB TEE, and his anticoagulation had been stopped several hours before surgery. Considering risk factors for thrombus formation and the consistently therapeutic ACTs on CPB, it is likely that a thrombus was present pre-CPB and not formed during CPB. A similar finding in a patient with risk factors for stasis was described in a recent report by Liang et al9 where a ventricular thrombus not originally seen on TEE was dislodged during CPB on manipulation of the heart. The absence of an intra-atrial shunt in the pre-CPB examination in our case likely precludes the thrombus from being a paradoxical embolus from the venous circulation.
TEE guidance is typically used to verify that a thrombus is not present before deployment of the AtriClip and that the LA appendage is occluded after deployment of the device.4 In a prospective cohort study of 231 patients undergoing TEE before elective mitral valve repair, mitral valve replacement, or LA tumor excision, TEE was highly accurate (100% sensitivity, 99% specificity) in identifying LA thrombi.10 In our case, however, an experienced echocardiographer was only able to identify spontaneous echocardiographic contrast and no clear thrombus during the initial TEE despite using multiplanar 2D echocardiography with adequate windows. Newer 3D techniques are increasingly routinely being used to identify LA appendage thrombi and provide greater sizing information before LA appendage occluder or excluder device placement.11 2D TEE provides a higher frame rate, and therefore higher resolution than 3D TEE but inadequate imaging planes can negate this advantage. Recent experience suggests that 3D TEE can improve differentiation of thrombi from pectinate muscle and other surrounding structures.12,13 In our case, 3D TEE may have identified the thrombus earlier. Examining the LA appendage with 3D TEE in addition to standard 2D imaging may be preferred in procedures involving LA excluder devices.
At the end of CPB, a focused TEE examination is performed to ascertain whether there is a need to reestablish CPB for additional repairs. In our case, we examined the aortic valve to look for any correctible issues, and evaluated the LA appendage to ensure it was isolated. We were not specifically scanning the LA or LV for dislodged thrombus and could have potentially missed it; the thrombus could have embolized systemically at any moment.
This case specifically highlights that even though TEE is excellent at detecting LA thrombus, further scrutiny during the post-CPB examination is necessary to identify any thrombus dislodged during the manipulation of the LA appendage. Detection of this thrombus in our patient before separating from CPB prevented a potentially deadly embolic event. Additionally, newer 3D techniques to evaluate the LA appendage for thrombi in addition to standard 2D techniques may be preferred before the deployment of an excluder device.
Name: Kinjal M. Patel, MD.
Contribution: This author helped prepare the manuscript and figure, and review the literature.
Name: Michael Rosenbloom, MD.
Contribution: This author helped prepare and revise the manuscript.
Name: Muhammad Raza, BA.
Contribution: This author helped prepare and design the manuscript, and review the literature.
Name: Sean Stevens, MS.
Contribution: This author helped prepare and design the manuscript, and review the literature.
Name: Jonathan Rost, MD.
Contribution: This author helped prepare and design the manuscript, review the literature, and prepare the figure.
Name: Ahmed Awad, MD.
Contribution: This author helped prepare and review the manuscript.
Name: Ronak G. Desai, DO.
Contribution: This author helped prepare the manuscript and figure and edit the manuscript.
This manuscript was handled by: Hans-Joachim Priebe, MD, FRCA, FCAI.
1. Murdock DK, Rengel LR, Schlund A, et alStroke and atrial fibrillation following cardiac surgery. WMJ. 2003;102:26–30.
2. Stoddard MF, Dawkins PR, Prince CR, Ammash NMLeft atrial appendage thrombus is not uncommon in patients with acute atrial fibrillation and a recent embolic event: a transesophageal echocardiographic study. J Am Coll Cardiol. 1995;25:452–459.
3. Katz ES, Tsiamtsiouris T, Applebaum RM, Schwartzbard A, Tunick PA, Kronzon ISurgical left atrial appendage ligation is frequently incomplete: a transesophageal echocardiograhic study. J Am Coll Cardiol. 2000;36:468–471.
4. Emmert MY, Puippe G, Baumuller S, et alSafe, effective and durable epicardial left atrial appendage clip occlusion in patients with atrial fibrillation undergoing cardiac surgery: first long-term results from a prospective device trial. Eur J Cardiothorac Surg. 2014;45:126–131.
5. Sakellaridis T, Argiriou M, Charitos C, et alLeft atrial appendage exclusion-Where do we stand? J Thorac Dis. 2014;6(suppl 1):S70–S77.
6. Stone ME, Silverman SH, Nomoto KIntracardiac thrombosis and acute right ventricular failure following complex reoperative cardiac surgery with aprotinin and deep hypothermic circulatory arrest. Semin Cardiothorac Vasc Anesth. 2007;11:177–184.
7. Patanè F, Sansone F, Campanella A, Asteggiano F, Rinaldi MAcute bioprosthetic thrombosis immediately after aortic valve replacement. J Cardiovasc Med (Hagerstown). 2009;10:167–169.
8. Kim SY, Song JW, Jang YS, Kwak YLFormation of intracardiac thrombus during cardiopulmonary bypass despite full heparinization and adequate activated clotting time -A case report-. Korean J Anesthesiol. 2012;62:571–574.
9. Liang Y, Chaichana K, Pretorius M, Eagle S, Jiang YIncidental detection of an intracardial floating thrombus by echocardiography during coronary artery bypass surgery. Anesth Analg. 2017;124:1783–1785.
10. Manning WJ, Weintraub RM, Waksmonski CA, et alAccuracy of transesophageal echocardiography for identifying left atrial thrombi. A prospective, intraoperative study. Ann Intern Med. 1995;123:817–822.
11. Nucifora G, Faletra FF, Regoli F, et alEvaluation of the left atrial appendage with real-time 3-dimensional transesophageal echocardiography: implications for catheter-based left atrial appendage closure. Circ Cardiovasc Imaging. 2011;4:514–523.
12. Marek D, Vindis D, Kocianova EReal time 3-dimensional transesophageal echocardiography is more specific than 2-dimensional TEE in the assessment of left atrial appendage thrombosis. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2013;157:22–26.
13. Anwar AM, Nosir YF, Ajam A, Chamsi-Pasha HCentral role of real-time three-dimensional echocardiography in the assessment of intracardiac thrombi. Int J Cardiovasc Imaging. 2010;26:519–526.