Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation.

BACKGROUND Left atrial appendage obliteration was historically ineffective for the prevention of postoperative stroke in patients with rheumatic atrial fibrillation who underwent operative mitral valvotomy. It is, however, a routine part of modern "curative" operations for nonrheumatic atrial fibrillation, such as the maze and corridor procedures. METHODS To assess the potential of left atrial appendage obliteration to prevent stroke in nonrheumatic atrial fibrillation patients, we reviewed previous reports that identified the etiology of atrial fibrillation and evaluated the presence and location of left atrial thrombus by transesophageal echocardiography, autopsy, or operation. RESULTS Twenty-three separate studies were reviewed, and 446 of 3,504 (13%) rheumatic atrial fibrillation patients, and 222 of 1,288 (17%) nonrheumatic atrial fibrillation patients had a documented left atrial thrombus. Anticoagulation status was variable and not controlled for. Thrombi were localized to, or were present in the left atrial appendage and extended into the left atrial cavity in 254 of 446 (57%) of patients with rheumatic atrial fibrillation. In contrast, 201 of 222 (91%) of nonrheumatic atrial fibrillation-related left atrial thrombi were isolated to, or originated in the left atrial appendage (p < 0.0001). CONCLUSIONS These data suggest that left atrial appendage obliteration is a strategy of potential value for stroke prophylaxis in nonrheumatic atrial fibrillation.