Stroke reduction utilizing left atrial appendage exclusion

Stroke remains a disabling and sometimes lethal disorder. It is estimated that 800,000 strokes occur each year in the United States; it is currently the fourth-leading cause of death in the country. Historically, strokes have been directly responsible for the deaths of nine U.S. presidents, and Franklin Roosevelt and Woodrow Wilson both had strokes while in the White House.

Stroke, dementia and orthopedic fractures due to falls remain the most frequent causes for institutionalization and prolonged nursing home care. The causes of stroke are varied and include hemorrhagic and ischemic mechanisms.

Particularly in the older population, nonrheumatic atrial fibrillation (AF) has become an increasingly important contributor to stroke incidence. More than 30 percent of strokes in individuals over the age of 75 years are due to AF. Multiple stroke risk factors in nonrheumatic AF patients have been identified; these factors are additive (similar to coronary risk factors) and have served as a basis by which practitioners decide to prescribe oral anticoagulants (OACs), chiefly warfarin, over the last several decades.

The CHADS2 scoring system (and more recently CHADS-VASc) has served as a guidepost for initiation of OAC therapy. Patients who have AF and a CHADS2 score of 2 to 6 are advised to initiate OAC therapy, as they have an annualized stroke risk of 4 to 18 percent. Risk reduction of 70 percent can be achieved with such a program, achieving annualized risks of 1.2 to 5.4 percent. The CHADS2 score is calculated as:

  • Congestive heart failure, 1 point
  • Hypertension, 1 point
  • Age > 75 years, 1 point
  • Diabetes, 1 point
  • Prior stroke or transient ischemic attack, 2 points

Persistent (requiring cardioversion) AF as compared with paroxysmal AF carries a similar risk of thromboembolic stroke. Thus, OAC recommendations are identical.

"Recent studies using sophisticated implantable cardiac rhythm devices have suggested that even as little as six minutes to six hours of AF can be a marker of enhanced stroke risk for patients," says Thomas M. Munger, M.D., electrophysiologist at Mayo Clinic's campus in Rochester, Minnesota.

It is anticipated that many more such monitoring devices will become available to patients in the near term, particularly patients with multiple risk factors or those who have already experienced cryptogenic stroke. As such, practitioners will soon be confronted by a larger group of patients seeking counsel about significant identifiable asymptomatic "AF burden" that poses a threat to them concerning thromboembolic stroke.

Warfarin and novel oral anticoagulants

The traditional approach to these patients has been warfarin (Coumadin) therapy, but this drug notoriously also poses a significant hemorrhagic hazard, particularly in older patients who are on multiple other medications. Warfarin has a narrow therapeutic index, interacts with several other drugs, is affected by dietary alterations, is difficult to regulate at times, and requires regular blood test monitoring with the international normalized ratio (INR).

The more recently released novel oral anticoagulants (NOACs) — dabigatran (Pradaxa), rivaroxaban (Xarelto), and apixaban (Eliquis) — have all displayed similar efficacy to warfarin for stroke prevention in nonrheumatic AF patients in large randomized controlled trials the last five years.

Additionally, apixaban was demonstrated to have had a lower bleeding risk when compared with warfarin in the ARISTOTLE trial. However, the full adoption of the NOACs as alternatives to warfarin has been tempered by:

  • Relative costs
  • Lack of reversal agents for bleeding complications (currently under development)
  • Renal clearance requiring dosing adjustments in the elderly
  • Rebound pro-coagulation phenomena upon drug cessation in some individuals

Nevertheless, the NOACs remain a promising class of pharmacological agents meriting further evaluation and adoption as an alternative to warfarin.

Many AF patients remain at high risk of bleeding complications from oral anticoagulation regardless of the agent, and thus remain untreated or minimally treated (antiplatelet agents such as aspirin or clopidogrel) prophylactically against thromboembolic stroke. The HAS-BLED bleeding risk scoring system can identify such patients.

"There remains a large group of patients who are not candidates for OAC therapy," says Dr. Munger. The first subgroup includes patients who have experienced life-threatening bleeding episodes — generally intracerebral, gastroenterological, or renal — despite therapeutic INR (2-3) and can be identified by HAS-BLED.

A second subgroup includes "OAC failures," which are patients with documented thromboembolic strokes despite being on therapeutic warfarin or NOAC treatment.

Finally, a third subgroup includes patients with other ancillary factors that the practitioner perceives would increase a patient's risk of bleeding on OAC, such as recurrent traumatic falls, occupation, polypharmacy, noncompliance, severe thrombocytopenia, and intestinal angiodysplasia.

Taken together, this group of patients with AF, CHADS2 ≥ 2, and who cannot receive OAC therapy due to contraindication or failure are exactly the individuals to consider for left atrial appendage (LAA) exclusion.

Left atrial appendage exclusion

LAA exclusion is not a new concept and has been performed since 1949 as an adjunct to cardiac surgery. In the 1980s, Cox utilized LAA amputation or exclusion along with the maze "cut-and-sew" atriotomy lesion sets for the prevention of AF and stroke. In 1999, he reported an extremely low 0.6 percent annualized rate of stroke in these surgical patients who were not treated with OAC.

Joseph L. Blackshear, M.D., cardiologist at Mayo Clinic's campus in Jacksonville, Florida, and co-workers in 1996 reported similar findings, and in 2003 also reported on the use of video-assisted thoracoscopic surgical isolation and resection of the LAA to prevent stroke in patients who were OAC failures.

At least 10 to 15 percent of new-onset AF patients not on OAC therapy will have demonstrable clot or debris within the LAA noted on transesophageal echocardiography; by removing or excluding the virtual space of the LAA, the propensity to develop clots within the arterial circulation is thought to diminish by > 90 percent.

Results from one surgical series did caution about the need for complete closure, as only 45 percent and 72 percent were completely closed using suture and stapling techniques, respectively. These patent left atrium-left atrial appendage (LA-LAA) channels could still act as clot repositories, susceptible to thromboembolization. Surgical devices such as the Tigerpaw and AtriClip can assist surgeons who choose to exclude rather than amputate the LAA in the operating room setting.

Available devices

Three percutaneous devices to facilitate LAA closure have been developed and are in use:

  • Watchman
  • Amplatzer
  • Lariat

The only randomized data so far generated using these devices is for the Watchman LAA plug.

The Watchman is a nitinol cage that is placed via a transseptal LA endocardial sheath into the orifice of the LAA. With release after deployment, the cage radially expands with 10 active fixation anchors to maintain positioning and closure of the LAA orifice.

The cage comes in five different sizes and will contour to most LAA anatomic variants. The device can be placed in patients with prior cardiac surgery. Device endothelization occurs within six weeks, and 45 days of OAC is generally prescribed until this healing process has ensued.

In 2009, David R. Holmes Jr., M.D., cardiologist at Mayo Clinic's campus in Rochester, Minnesota, and co-workers reported results from the randomized clinical trial PROTECT AF; noninferiority of the WATCHMAN device as compared with warfarin therapy for stroke prevention was shown. A beneficial effect on mortality appears to be emerging with multiyear follow-up of such patients. Food and Drug Administration approval of the device is tentatively expected this year.

The Amplatzer plug has been utilized for percutaneous patent foramen ovale (PFO) closure and also has been adopted as a method for LAA exclusion, though no randomized clinical trial of its use for this indication currently exists.

The Lariat device is placed using a combined percutaneous endocardial and epicardial approach. A transseptal left atrial endocardial sheath delivers an occlusive balloon catheter that allows angiographic imaging of the LAA and subsequent delivery to its tip of a magnetic wire element; simultaneously, the pericardial space is accessed from a percutaneous subxiphoid approach, and a second magnetic wire element is delivered to the outside of the LAA tip. The two magnets attract each other, and then act as a "rail" over which the LARIAT suture is conveyed to the base of the LAA, where the suture is tightened to facilitate LAA base closure.

The magnetic wires are then removed, and the suture cut. The LAA is thus isolated from the LA; it infarcts and gradually atrophies, leaving no virtual space. Closure rates of > 95 percent have been achieved; randomized clinical trials evaluating this technique are planned to begin in late 2014.

No OAC therapy is required after Lariat LAA closure. Not all LAA anatomies, as determined by cardiac contrast CT, are appropriate for this technique (LAA size must be < 40 mm and not superiorly directed). Additionally, prior cardiac surgery or pericarditis is a contraindication for this technique, as pericardial access will not be readily accomplished.

LAA exclusion will enhance stroke reduction in high-risk patients with nonrheumatic AF who are not candidates for anticoagulation due to OAC failure or high-risk bleeding potential. "In the next 20 years, the at-risk patient pool will balloon due to population demographics and increased detection of AF due to consumer-based monitoring devices and techniques," says Dr. Munger. "LAA exclusion therapy will become a more meaningful and important management tool for minimizing stroke potential in the future."