Stroke prevention with left atrial appendage (LAA) occlusion is an important alternative
treatment for high-risk atrial fibrillation (AF) patients with contraindications to
oral anticoagulation (OAC).[1]
[2] The presence of LAA thrombus detected on preprocedural imaging usually precludes
implantation of the endocardial device because of a prohibitively high periprocedural
embolic risk.[3] Although successful resolution of LAA thrombus with OAC has been described, all
previous reports are in patients without contraindications to OAC.[4]
[5]
[6] Furthermore, long-term data in these patients are lacking. We hypothesized that
short-term OAC with apixaban could safely facilitate resolution of LAA thrombi even
in a high-risk population, thereby allowing safe LAA occluder device implantation,
with good outcomes over the long term.
This was a retrospective cohort study of 83 patients who underwent work-up for LAA
occlusion at our center between January 2015 and December 2017. LAA thrombi were documented
in 11 (13.3%) patients (mean age: 76.9 [ ± 6.9] years; 7 males) on preprocedural
imaging (9 on transesophageal echocardiography and 2 on cardiac computed tomographic
angiography) ([Table 1]). All patients with LAA thrombi had permanent AF, with median CHA2DS2-VASc and HAS-BLED scores of 4 (interquartile range [IQR]: 3–5) and 2 (IQR: 2–3),
respectively. Contraindications to lifelong OAC in this cohort were prior intracranial
hemorrhage in six patients (two on warfarin, one on dabigatran, one on heparin, and
two while not on anticoagulation), prior major gastrointestinal hemorrhage requiring
emergency hospital admission and blood transfusion in three patients (two on warfarin
and one while not on anticoagulation), severe unexplained anemia on dabigatran in
one patient, and failed OAC in one patient. At the time of imaging, none of the patients
were receiving OAC, and four (36.4%) patients were on a single antiplatelet agent.
There were no significant differences between patients who had documented LAA thrombus
compared with those without in terms of age (p = 0.80), sex category (p = 0.74), AF type (p = 0.11), comorbidities (hypertension [p = 0.50], diabetes mellitus [p = 0.31], and prior stroke or transient ischemic attack [p = 0.47]), and CHA2DS2-VASc score (p = 0.69).
Table 1
Characteristics of patients with left atrial appendage thrombus
|
Patients
|
Characteristics
|
|
Patient 1
|
Age: 82 y; eGFR: 42; CHA2DS2-VASc: 3; HAS-BLED: 2; recurrent GI bleed on warfarin; OGD revealed gastric angiodysplasia;
CKD; COPD; angina
|
|
Patient 2
|
Age: 76 y; eGFR: 74; CHA2DS2-VASc: 5; HAS-BLED: 2; recurrent GI bleed on warfarin; no source identified on OGD
and colonoscopy
|
|
Patient 3
|
Age: 75 y; eGFR: 90; CHA2DS2-VASc: 4; HAS-BLED: 3; massive GI bleed while off anticoagulation; OGD identified
variceal source, which was treated with banding; nonalcoholic liver cirrhosis; old
inferior MI; mild LV dysfunction; hypothyroidism
|
|
Patient 4
|
Age: 78 y; eGFR: 53; CHA2DS2-VASc: 5; HAS-BLED: 2; ICH on heparin for newly diagnosed atrial fibrillation; prior
ischemic stroke; CKD; HTN; hypothyroidism
|
|
Patient 5
|
Age: 81 y; eGFR: 64; CHA2DS2-VASc: 3; HAS-BLED: 3; ICH on warfarin (subtherapeutic international normalized ratio);
meningioma; epilepsy; HTN; prior pneumothorax
|
|
Patient 6
|
Age: 80 y; eGFR: 48; CHA2DS2-VASc: 4; HAS-BLED: 3; ICH on dabigatran; CKD; epilepsy; uncontrolled HTN; treated
breast cancer
|
|
Patient 7
|
Age: 68 y; eGFR: 47; CHA2DS2-VASc: 6; HAS-BLED: 2; ICH while off anticoagulation; cerebral amyloid angiopathy;
prior stroke; old MI; severe LV dysfunction; CKD; HTN
|
|
Patient 8
|
Age: 88 y; eGFR: 67; CHA2DS2-VASc: 3; HAS-BLED: 2; large ICH on warfarin; HTN
|
|
Patient 9
|
Age: 67 y; eGFR: 75; CHA2DS2-VASc: 5; HAS-BLED: 3; ICH while off anticoagulation; recurrent stroke and transient
ischemic attack; prior pulmonary embolism and deep venous thrombosis; seizures; left
ventricular hypertrophy
|
|
Patient 10
|
Age: 83 y; eGFR: 44; CHA2DS2-VASc: 4; HAS-BLED: 2; severe anemia of unknown origin while on dabigatran; CKD; left
bundle branch block; HTN
|
|
Patient 11
|
Age: 68 y; eGFR: 67; CHA2DS2-VASc: 5; HAS-BLED: 3; ischemic stroke on apixaban; cerebral amyloid angiopathy; prior
embolic stroke; polycythemia; COPD; alcohol excess; HTN
|
Abbreviations: CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease;
eGFR, estimated glomerular filtration rate; GI, gastrointestinal; HTN, hypertension;
ICH, intracranial hemorrhage; LV, left ventricular; MI, myocardial infarction; OGD,
esophagogastroduodenoscopy.
The potential risks of surgical thrombectomy and ligation were felt to outweigh the
benefits. Hence, following detection of LAA thrombi, off-label dose-adjusted apixaban
(2.5 mg twice daily for three patients and 5 mg twice daily for eight patients) was
prescribed for each patient. The choice of anticoagulation was based on evidence of
potential favorable characteristics of apixaban compared with other OAC in terms of
efficacy and safety profile.[7] Repeat imaging in the form of transesophageal echocardiography, or cardiac computed
tomographic angiography was scheduled at 6- to 8-week intervals. Complete resolution
of LAA thrombus was observed in 10 (90.9%) patients after apixaban treatment for a
median of 94 days (IQR: 44–126 days). During treatment with apixaban, one patient
(who received a dose of 5 mg twice daily) had a severe gastrointestinal bleed requiring
blood transfusion, and one patient suffered an ischemic stroke with subsequent full
recovery. There was no significant difference in the risk of bleeding according to
apixaban dosage (p = 0.73). One patient (patient 8, [Table 1]) had persistent LAA thrombus on repeated imaging, and a patient-centered decision
was ultimately taken for continued apixaban therapy despite prior major intracranial
bleed while on warfarin. In this patient, no bleeding complications were observed
over a follow-up period of 25 weeks.
The 10 patients with thrombus resolution underwent successful LAA occlusion with no
periprocedural complications. Apixaban was replaced by dual-antiplatelet therapy with
aspirin and clopidogrel for 6 weeks postprocedure followed by single (or no) antiplatelet
therapy. The decision to discontinue long-term antiplatelet therapy in three patients
was based on an individualized approach after accounting for their high bleeding risk.
All patients underwent cardiac imaging with transesophageal echocardiography or cardiac
computed tomographic angiography at 6 to 8 weeks postprocedure; no device-related
thrombus was observed in any patient. Across a median follow-up of 2.5 years (IQR:
0.6–3.3 years), one patient suffered a transient ischemic attack and another had an
episode of severe epistaxis despite not being on antiplatelet or OAC therapy at the
time. During this same period, four (40%) patients died.
Given the low number of patients, our study was underpowered and should therefore
be interpreted with caution. Furthermore, the incidence of LAA thrombus in our study
may not reflect those of the general AF population as many of our high-risk patients
were not on anticoagulation therapy at baseline. It is also worth noting that we adopted
a strategy of minimizing the exposure to OAC in this high-risk cohort by performing
LAA occlusion without delay in patients with a confirmed resolution of LAA thrombi.
Our experience suggests that short-term treatment with apixaban may be effective and
safe for aiding the resolution of documented LAA thrombus in high-risk patients who
are ineligible for lifelong OAC therapy. This strategy allows the LAA occlusion procedure
to be undertaken safely, thereby potentially reducing the long-term risk of thromboembolism.
