The trusted source for
healthcare information and
Left Atrial Septal Pouch and Stroke
Abstract & Commentary
By Michael H. Crawford, MD
Sources: Tugcu A, et al. Septal pouch in the left atrium and risk of ischemic stroke. J Am Coll Cardiol Img. 2010;3:1276-1283; Gurudevan SV, et al. Septal thrombus in the left atrium: Is the left atrial septal pouch the culprit? J Am Coll Cardiol Img. 2010;3:1284-1286; Chandrashekhar Y, Narula J. LA septal pouch as a source of thromboembolism: Innocent until proven guilty? J Am Coll Cardiol Img. 2010;3:1296-1298.
The source of embolic thrombotic material in patients with cryptogenic stroke is controversial. There are data supporting an association with atrial septal abnormalities, such as patent foramen ovale (PFO) and atrial septal aneurysm. Recently, attention has focused on the so-called left atrial (LA) septal pouch, which is created by incomplete fusion of the cranial part of the septum primum with the septum secundum without an intra-atrial communication. Thus, these investigators from Columbia University in New York sought to assess the relationship, if any, between a LA pouch and stroke in a study cohort of 255 stroke patients and 209 control subjects who underwent transesophageal echocardiography with contrast and the Valsalva maneuver within 3 days of stoke onset. Of the 464 individuals enrolled in the study, 89 had a PFO, 9 a closed pouch (incomplete fusion but no opening to either atria), 5 a right septal pouch (incomplete fusion of the caudal part of the septum with opening to the right atrium), and 4 an atrial septal defect. These patients were excluded. Technically, inadequate contrast echoes excluded 13 additional individuals. The remaining 344 consisted of 187 stroke patients and 157 controls. A LA pouch was found in 29%. Of the stroke patients, 37% were believed to have cryptogenic stroke after diagnostic evaluation. Stroke patients were older (71 vs. 67 years, p < 0.001), and more likely to have hypertension and atrial fibrillation. However, there was no difference in the prevalence of LA pouch (both groups 29%). Also, a LA pouch was not a multivariate predictor of cryptogenic stroke.
Cerebral vascular occlusion explains 90% of strokes, but the source of the thrombotic occlusion is often difficult to determine. It is believed that about 20% are cardioembolic, and the majority of these are LA appendage thromboembolism in patients with non valvular atrial fibrillation. Between 10% and 30% are deemed cryptogenic because the cause of the vascular occlusion is unclear. Cryptogenic strokes are more common in younger individuals, so much attention has been focused on determining their cause, so that preventive treatment can be applied. Initially there was considerable interest in PFO since paradoxical emboli had been caught transversing this defect by fortuitous imaging. Also, venous thrombus is much more common than other cardiac causes of thrombus formation in younger individuals. However, despite the frequency of PFO, the plausibility of the hypothesis and the implantation of many PFO closure devices, recent large population studies have failed to support an association and a recent PFO closure trial failed to show a benefit in stroke prevention. Now attention has focused on other sources of atrial thrombus, such as atrial septal aneurysms and pouches. Although there are case reports of thrombi found in these structures, their overall relationship to cryptogenic stroke is unclear. The LA pouch is intriguing because it was present in about a third of the subjects in this and other studies arguably more common than PFO. However, this study failed to support a role for the LA pouch in stroke.
The major limitations to this study were the small size, imperfect matching between cases and controls, and the relatively older age of the patients. LA pouches and cryptogenic stroke are more common in younger patients. Thus, another potential cause of embolic stroke seems to be eliminated. What is clear is that we do not understand the genesis of LA thrombi seen outside of the appendage and how important such clots are in the origin of cryptogenic stroke. I am guessing not very, but we need more and better data to make firm conclusions.