Download this article's Factoid in PDF (& PPT for Gold Subscribers)

Among patients with atrial fibrillation (A-fib), the particular morphology of the left atrial appendage (LAA)—the source of most thromboemboli—appears to affect stroke risk, according to a study published in the August 7, 2012, issue of the Journal of the American College of Cardiology. If the finding is confirmed, determining LAA type could be factored into decisions regarding anticoagulation among patients at low to intermediate risk, the authors say.

Investigators led by Andrea Natale, MD, of the Texas Cardiac Arrhythmia Institute at St. David’s Medical Center (Austin, TX), looked at 932 A-fib patients who underwent imaging with cardiac CT (n = 499) or MRI (n = 433) in preparation for transcatheter ablation. Their LAAs were classified into 1 of 4 configurations and correlated with their history of stroke or TIA.

Chicken Wings Predominate

The prevalences of the descriptively named LAA morphologies were:

• Chicken wing (an obvious bend or folding back of the LAA on itself): 48%
• Cactus (dominant central lobe with secondary lobes): 30%
• Windsock (1 dominant lobe as the primary structure): 19%
• Cauliflower (limited overall length with more complex internal characteristics): 3%

Overall, 8% of the cohort had experienced a previous stroke or TIA, but the risk varied considerably across the LAA types:

• Chicken wing: 4%
• Windsock: 10%
• Cactus: 12%
• Cauliflower: 18%

Chicken Wing Morphology Less Risky

Interestingly, the most prevalent LAA shape, chicken wing, was least likely to be associated with stroke or TIA. When the LAA morphologies were broadly divided into chicken wing vs. non-chicken wing types, patients in the former category had one third as many strokes or TIAs as those in the latter (4% vs. 12%; P < 0.001).

After controlling for CHADS2 score (a measure of stroke risk), sex, left atrial size, and A-fib types, a multivariable model showed that patients with chicken wing morphology were 79% less likely to have a history of stroke or TIA (OR 0.21; 95% CI 0.05-0.91; P = 0.036). The results were unchanged when the CHADS2 scores was replaced with those of the novel CHA2DS2-VaSc index that includes additional stroke risk modifiers.

From the reverse perspective, when chicken wing LAA was taken as the reference, the likelihood of stroke or TIA was 4 times higher with the cactus shape, 5 times higher with the windsock shape, and 8 times higher with the cauliflower shape. Overall, having non-chicken wing morphology was an independent predictor of stroke or TIA (OR 2.95; 95% CI 1.75-4.99; P = 0.041).

Specifically among low-risk patients (CHADS2 score 0-1), those who also had chicken wing LAA showed a lower incidence of prior stroke or TIA (0.7% vs. 4.6% for non-chicken wing morphology patients; P = 0.001). After adjustment for sex, A-fib type, and left atrial size, non-chicken wing morphology remained a strong independent predictor of stroke (OR 10.1; 95% CI 1.25-79.7; P = 0.019).

On the other hand, in a high-risk cohort (CHADS2 score ≥ 2), although stroke and TIA were more common in non-chicken wing patients than in patients with the LAA morphology, the difference was not significant (46% vs. 29%; P = 0.061).

A Novel Idea

In a telephone interview with TCTMD, Robert J. Sommer, MD, of Columbia University Medical Center (New York, NY), called the paper “very interesting” in that it not only puts forward a novel idea but makes intuitive sense as well.

“If you look at the image of the chicken wing LAA, you can see that it has the smoothest contour and the fewest side branches or pockets where blood could pool,” he explained. “The other 3 configurations have little outcroppings off the main stalk where thrombus could potentially form.”

Moreover, Dr. Sommer said, the finding “has significant implications not only for management of patients with anticoagulation but for use of transcatheter closure procedures.”

Currently practitioners are “struggling with this whole link between the LAA and stroke risk in A-fib patients, and trying to assess the risk factors involved,” he commented. “If we prove that appendage closure is beneficial, and are able to add that [therapy] to our armamentarium for fighting stroke, we can start to risk-stratify A-fib patients in a more systematic way,” he said. The ultimate goal would be to “determine which patients might do best with antiplatelet therapy, which with anticoagulation, and which with closure, based on anatomy, among other factors,” he commented.

Meanwhile, “this [classification] is a good starting place for further investigation,” Dr. Sommer observed, although additional, preferably prospective information is needed to know “whether this anatomic variation really makes a big difference.” In that regard, “it would be interesting to analyze data from the PROTECT AF and PREVAIL trials to see if there are certain subgroups that are at even higher risk because of anatomy and thus would benefit most from closure,” he added.

Clinical Implications for Low-Risk Patients

In a telephone interview with TCTMD, lead author Luigi Di Biase, MD, PhD, also of St. David’s Medical Center, said the more immediate clinical value of the paper lies in its confirmation of the differential risk of the 2 broad LAA types in patients with low CHADS2 scores.

Current guidelines do not recommend anticoagulation for patients with scores of 0 or 1, Dr. Di Biase reported, “so basically [the decision regarding appropriate therapy] is left to the physician. Now when you don’t know whether to give just aspirin or a [more potent] drug like Coumadin or one of the newer anticoagulants, you can just do a CT or MRI and look at the LAA. If it is a non-chicken wing morphology, you should go for the stronger therapy, while if it is a chicken wing, you may choose aspirin.”

Although Dr. Di Biase said he would welcome a prospective study to confirm the morphological findings, he added that it would be challenging given the large patient sample required due to a low event rate. A more practical next research step, he suggested, would be to investigate the potential of using 3-D transesophageal echocardiography (TEE) to image the LAA instead of the more costly CT or MRI. “Since A-fib patients at some point have a TEE for cardioversion or other indications, 3-D TEE may be able to characterize the LAA morphology,” he said.

 

---

Source:
Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol. 2012;60:531-538.

 

 

Related Stories:

• Left Atrial Appendage Closure Strategies Compared
• New Oral Anticoagulants Top Warfarin in A-fib Stroke Prevention
• Apixaban Prevents Stroke in A-fib Patients Irrespective of Previous Stroke/TIA