Correspondence Emory University Hospital, 1364 Clifton Road F606, Atlanta, GA 30322, USA

Email peter_block@emoryhealthcare.org

The case

A 73-year-old woman presented with persistent atrial fibrillation (AF), which had lasted more than 5 years. In an effort to control the ventricular response to AF, she had previously undergone catheter ablation of the atrioventricular junction and received a single chamber pacemaker. Anticoagulation therapy with warfarin was initiated at this time, but unfortunately the patient developed recurrent gastrointestinal bleeding complications requiring multiple blood transfusions. By using colonoscopy, the source of the bleeding was eventually localized to multiple arterial-venous malformations. To minimize gastrointestinal blood loss, warfarin anticoagulation was carefully controlled at slightly subtherapeutic levels of the international normalized ratio. Six months before admission to hospital, the patient had received a dual chamber pacemaker and dofetilide therapy for cardioversion. Although her sinus rhythm briefly returned to normal, the AF had returned. Cardioversion was attempted again using dofetilide and although the patient had a normal sinus rhythm for 2 days, she had persistent AF for the subsequent 4 months.

The patient had NYHA Class II symptoms of heart failure, mainly due to fatigue caused by AF and anemia. Her medical history was notable for a transient ischemic attack 10 years previously—possibly due to a thromboembolic event—that manifested as weakness and clumsiness in the right hand. No abnormalities were found on physical examination except for a grade 2/6 systolic murmur radiating from the apex to the axilla. Echocardiography confirmed that this murmur was caused by mild mitral regurgitation. No signs of atherosclerotic disease were found using carotid ultrasonography. Transesophageal echocardiography was performed to exclude other possible sources of thromboemboli. There was no evidence of patent foramen ovale, the left ventricle was normal in size and function, the transverse and descending aorta were free of significant atherosclerotic disease and no thrombus was detected in the left atrial appendage (LAA; Figure 1A). Furthermore, no spontaneous echo contrast was observed in the left atrium.

Figure 1 Transesophageal echocardiography of the left atrial appendage.

(A) A transesophageal echocardiogram of the left atrial appendage before placement of the PLAATO device (B) A transesophageal echocardiogram of the left atrial appendage after percutaneous occlusion. An arrow shows deployment of a PLAATO device in the left atrial appendage. Abbreviations: LA, left atrial body; LAA, left atrial appendage.

Full figure and legend (9K)Figures & Tables indexDownload Power Point slide (216K)

The patient was considered a poor candidate for long-term anticoagulation therapy with warfarin, because of recurrent gastrointestinal bleeding. Pulmonary vein ablation was considered, but this strategy was not advised because aggressive anticoagulation therapy is necessary after the procedure. Instead the patient was referred for percutaneous LAA occlusion as a participant in the Percutaneous Left Atrial Appendage Transcatheter Occlusion (PLAATO) trial. After informed consent, she was taken to the cardiac catheterization laboratory and sedated with fentanyl and midazolam. Baseline transesophageal echocardiography was repeated to rule out intercurrent development of LAA thrombus and was also used to monitor placement of an LAA occlusion device.

By using echocardiography, the neck of the LAA was estimated to have a diameter of 25 mm. The right femoral vein was cannulated using the Seldinger technique and a trans-septal puncture was performed with a Mullins sheath and Brockenbrough needle. Following intravenous administration of weight-based heparin, a Cook wire guide was advanced into the left atrium through the sheath; the Mullins sheath was then replaced with a preformed delivery sheath that accompanied the PLAATO device. A 32 mm occlusion device (ev3 Inc., Plymouth, MN, USA) was advanced through the delivery catheter and deployed under echocardiographic and fluoroscopic guidance (Figure 1B, Figure 2). Doppler echocardiography and contrast injections through the delivery catheter confirmed that the device was positioned so that the neck of the LAA was completely occluded, and the PLAATO device was then released. A final angiogram of the left atrium showed that the LAA had not filled with contrast agent (Figure 3). A 300 mg loading dose of clopidogrel and 325 mg of aspirin were administered. The patient tolerated the procedure well and was released from hospital the following day. Clopidogrel and aspirin therapy was continued for 6 months, and changed to aspirin only thereafter. Endocarditic prophylaxis was prescribed indefinitely, as judged by the treating physician.

Figure 2 A cineangiographic frame of the PLAATO occlusion device.

An arrow shows the PLAATO device immediately after deployment and before its release from the delivery catheter. Abbreviation: DC, delivery catheter.

Full figure and legend (6K)Figures & Tables indexDownload Power Point slide (212K)

Figure 3 A cineangiographic frame of the final left atrial angiogram.

No contrast agent leak can be seen around the device or in the left atrial appendage. An arrow indicates the PLAATO device after release. Abbreviation: LAA, left atrial appendage.

Full figure and legend (6K)Figures & Tables indexDownload Power Point slide (212K)

Clopidogrel therapy was stopped 1 month after the patient had undergone the PLAATO procedure, because of recurrent gastrointestinal bleeding. She continued to take 180 mg aspirin daily, but clopidogrel therapy was resumed 8 months later after an episode of blurred vision. A head CT scan carried out as part of the diagnostic work-up did not reveal any acute changes. Only small areas of encephalomalacia were noted, which were probably related to remote ischemia. Chest X-ray showed the device in the left atrial appendage. Carotid ultrasonography revealed only minimal atherosclerotic changes, without evidence of significant stenosis. One year later, the patient underwent endoscopic cauterization of colonic arterial-venous malformations, which successfully controlled the gastrointestinal bleeding. With iron supplementation alone, she now maintains a hematocrit of 35%. Two years after percutaneous LAA closure, the patient continues to do well clinically.

Discussion of diagnosis

Patients with AF have a five-fold increased risk of stroke compared with patients in normal rhythm.¹ When atrial thrombus is seen by transesophageal echocardiography in patients with non-rheumatic AF, 90% of the time it resides in the LAA. Anticoagulation with warfarin is an effective management strategy for patients with paroxysmal or fixed AF. This treatment reduces the risk of stroke by almost 70% and is more effective than aspirin or aspirin plus low-dose warfarin therapy.^{2, 3, 4} Patients with the highest risk of stroke gain the most benefit from anticoagulation therapy, yet these same patients are more likely to have contraindications, or might not receive such therapy.^{3, 4, 5} Of particular concern are patients with AF who have bleeding episodes while taking anticoagulant therapy, those with a high risk of injury that might be complicated by bleeding, (e.g. after a fall) and patients who are difficult to monitor adequately.

Treatment and management

Closure of the LAA seems to be a reasonable prophylactic treatment to help prevent thromboembolism. Although surgical closure of the LAA can be performed, it is invasive and associated with poor patient outcomes.^{6, 7, 8} Transvenous closure of the LAA is a new approach, and recent trials have tested its safety and efficacy.^{9, 10} A recent study by Ostermeyer et al. comprised two registries of 111 patients (aged 71 9 years).¹¹ Patients recruited to the trial had contraindications for anticoagulation with warfarin and one or more of these additional risk factors for stroke: presence of congestive heart failure; a left ventricular ejection fraction less than 40%; systolic hypertension greater than 160 mmHg; diabetes; age >65 years; previous myocardial infarction or known coronary stenosis of more than 50%; spontaneous echo density in the LAA or a blood flow velocity in the LAA of less than 20 cm/sec. Their data show that this investigational strategy is promising; one neurological death was reported within 30 days of the PLAATO procedure and three patients required in-hospital pericardiocentesis for hemopericardium during or following the procedure. After a follow-up of almost 10 months, only 2 out of 111 patients had experienced a stroke (2%). This was equivalent to a 60% reduction in stroke rate, as the anticipated incidence of stroke had been 6.3%, based on the patients' CHADS2 (i.e. congestive heart failure; hypertension; age >75 years; diabetes; stroke or transient ischemic attack) risk stratification scores.¹² Transesophageal echocardiography at follow-up showed that the device had not migrated and no mobile thrombi were seen in any patients. Five patients had major adverse events (new major or minor stroke, cardiac or neurological death, myocardial infarction or requirement for cardiovascular surgery) related to the PLAATO procedure. Although the risk of stroke was not abolished by LAA occlusion in this study, it does seem to have been reduced. A non-cardiac thromboembolism, a thrombus in the left atrial body, or other causes could account for continued stroke occurrence.

Only a randomized trial, however, can establish the actual reduction in stroke rate following LAA occlusion. A new trial called the WATCHMAN Left Atrial Appendage System for Embolic PROTECTion in Patients with AF (PROTECT AF) trial is underway to evaluate the stroke rate following LAA occlusion using the Watchman^® device (Atritech Inc. Plymouth, MN; Figure 4). The Watchman^® device is not immediately occlusive, but acts as a filter until endocardialization of its atrial surface occurs. Patients who enter the trial must be able to take warfarin for at least 45 days and aspirin thereafter. One in three patients receives warfarin therapy alone, while the remaining patients receive a Watchman^® device. This trial should give further insight into the efficacy of this strategy. If it yields positive results in terms of patient outcome, LAA percutaneous occlusion could be an attractive management option for many patients with AF. Such patients might not be ideal candidates for warfarin therapy or choose to take warfarin for only a short period of time after percutaneous LAA occlusion, before switching to aspirin therapy alone.

Figure 4 The Watchman^® occlusion device.

Barbs on the outer edge of the wires help to fix this device to the musculature of the left atrial appendage. An arrow shows which side faces the left atrium when the device is deployed. Abbreviation: B, barbs.

Full figure and legend (17K)Figures & Tables indexDownload Power Point slide (222K)

Conclusion

AF is associated with a significantly increased risk of stroke and necessitates treatment with warfarin anticoagulation. Few alternative options exist, however, for patients who cannot tolerate this treatment or who develop bleeding complications, as in the present case. Surgical left atrial ablation is one alternative and another option is antiplatelet therapy with aspirin, clopidogrel or a combination of these. Early results of a Phase I trial of percutaneous LAA occlusion are promising, and an ongoing prospective randomized trial will help to determine its usefulness as an alternative therapeutic strategy.

References

1. Wolf PA et al. (1991) Atrial fibrillation as an independent risk factor for stroke: The Framingham Study. Stroke 22: 983–988 | PubMed | ISI | ChemPort |
2. Atrial Fibrillation Investigators (1994) Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 154: 1449–1457 | PubMed | ISI |
3. van Walraven C et al. (2002) Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA 288: 2441–2448 | Article | PubMed | ChemPort |
4. Hart RG et al. (2003) Lessons from the stroke prevention in atrial fibrillation trials. Ann Intern Med 138: 831–838 | PubMed |
5. Bungard TJ et al. (2000) Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 160: 41–46 | Article | PubMed | ISI | ChemPort |
6. Blackshear JL and Odell JA (1996) Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation. Ann Thorac Surg 61: 755–759 | Article | PubMed | ISI | ChemPort |
7. Madden J (1948) Resection of the left auricular appendix. JAMA 140: 769–772
8. Johnson WD et al. (2000) The left atrial appendage: our most lethal human attachment! Surgical implications. Eur J Cardiothorac Surg 17: 718–722 | Article | PubMed | ChemPort |
9. Sievert H et al. (2002) Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation—early clinical experience. Circulation 105: 1887–1889 | Article | PubMed | ISI |
10. Hanna IR et al. (2004) Left atrial structure and function after percutaneous left atrial appendage transcatheter occlusion (PLAATO): six-month echocardiographic follow-up. J Am Coll Cardiol 43: 1868–1872 | Article | PubMed | ISI |
11. Ostermayer SH et al. (2005) Percutaneous left atrial appendage transcatheter occlusion (PLAATO System) to prevent stroke in high-risk patients with non-rheumatic atrial fibrillation. J Am Coll Cardiol 46: 9–14 | Article | PubMed | ISI |
12. Gage BF et al. (2004) Selecting patients with atrial fibrillation for anticoagulation: stroke risk stratification in patients taking aspirin. Circulation 110: 2287–2292 | Article | PubMed | ISI | ChemPort |

Contact the journal about this article