This retrospective, single-center study included consecutive patients with HCM and non-valvular AF who underwent LAAC between April 2017 and December 2019. Inclusion criteria: patients were over 18 years, presented with HCM and non-valvular AF, and were complicated with at least one of the following situations: (a) a high bleeding risk with HAS-BLED score ≥ 3; (b) an unwillingness to receive regular oral anticoagulation (OAC) therapy; (c) intolerance to chronic OAC; (d) stroke or thromboembolism even under OAC treatment. The diagnosis of HCM was based on a two-echocardiographic demonstration of a hypertrophied and non-dilated left ventricle (wall thickness ≥ 15 mm), in the absence of another cardiac or systemic disease capable of producing a similar magnitude of hypertrophy (9, 10). Documentation of AF was based on an electrocardiogram or Holter obtained either after the acute onset of symptoms or during a routine examination. The exclusive criteria included severe left ventricular outflow tract obstruction needing surgery or alcohol septal ablation, intraventricular obstruction, ventricular aneurysm, previous alcohol septal ablation or surgical myectomy due to outflow tract obstruction, and requirement for long-term anticoagulation therapy for reasons other than AF.

The retrospective study was approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine and complies with the Declaration of Helsinki. Written informed consent was obtained from each patient.

Transesophageal echocardiography (TEE) was performed prior to the procedure to rule out LAA thrombi. A cardiac computed tomography (CT) scan and 3-dimensional reconstruction of the left atrium were available in 90.6% (29/32) of the patients for preprocedure planning.

The LAAC procedure was performed as described previously (11). In brief, the procedure was performed under local anesthesia and fluoroscopy guidance, and TEE was introduced under deep sedation after device deployment to reconfirm the position of the device before release. A mean left atrial pressure above 10 mmHg was obtained after transseptal puncture. A Watchman device (Boston Scientific, Marlborough, MA, USA) with appropriate size (21, 24, 27, 30, and 33 mm) was chosen, generally 10–30% oversizing based on the ostial width of the LAA measured by angiography or cardiac CT. The device was then advanced into the delivery sheath and deployed by sheath retraction guided by fluoroscopy. Preliminary assessment was performed by angiography and tug test under fluoroscopy to check the device position and stability. TEE was then performed to reconfirm the position with minimal (<5 mm) to no residual peri-device leaks and a proper compression ratio under deep sedation. The device was released if it was verified by the assessment of “PASS” criteria.

Post discharge, office or transtelephonic visits were scheduled at the 3, 6, and 12 months following the procedure and once half a year thereafter. Adverse events reported during follow-up visit, based on the Percutaneous LAA occlusion Munich Consensus Document (12), including mortality, thromboembolic events (stroke and systemic embolism) and bleeding. TEE was performed to assess device occlusion safety and efficiency, including peri-device flow, device-related thrombus (DRT), device embolism, and pericardial effusion at 45 days and 6 months of follow-up time points. At discharge, OAC with Vitamin K antagonist or novel oral anticoagulant (NOAC) was recommended for all patients for at least 45 days unless there were contraindications. Dual antiplatelet therapy (aspirin and clopidogrel) was recommended for another 4.5 months and then life-long aspirin was prescribed if follow-up TEE confirmed satisfactory seal (jet <5 mm in width). If unsatisfactory seal or DRT was detected, OAC was continued until satisfactory seal was achieved or DRT was resolved by the evaluation of TEE repeated every 3–6 months.

Continuous variables were described as mean ± standard deviation (median [interquartile range] for non-normal data) and compared using Student's t-test (Mann–Whitney U test if normality not satisfied). Categorical variables were presented as percentages and analyzed using chi-square test or Fisher exact test where appropriate. All analyses were performed using SPSS version 22.0 (IBM Software Inc., Armonk, NY, USA). Two-sided P-values of <0.05 were considered statistically significant.

Data from 837 patients who underwent LAAC between April 2017 and December 2019 was reviewed. Thirty-nine patients who were diagnosed with HCM and AF were identified. Two patients with severe left ventricular outflow tract obstruction and one patient with intraventricular obstruction and apical aneurysm were excluded. In total, 36 cases with HCM and AF who underwent LAAC were included in this study, of whom 22 were for primary stroke prevention and 14 were for secondary prevention. Indications for LAAC included high bleeding risk (33.3%), unwilling to take OAC (47.2%), intolerance to chronic OAC (13.9%) and stroke or thromboembolism even under OAC treatment (5.6%).

Baseline clinical and demographic characteristics are presented in Table 1. The mean age was 68.8 ± 8.3 in the Primary Prevention Group and 69.5 ± 7.1 in the Secondary Prevention Group. Women composed 36.4% and 35.7% of the two groups, respectively. Seven patients (31.8%) in the Primary Prevention Group and six patients (42.9%) in the Secondary Prevention Group had a prior failed AF ablation. There were no significant differences in comorbidities, except prior stroke. Patients in the Secondary Prevention Group had higher CHA2DS2-VAS scores (5.1 ± 1.4 vs. 2.6 ± 1.6, P < 0.001) and higher HAS-BLED scores (2.8 ± 1.0 vs. 1.5 ± 0.9, P < 0.001) compared with those in the Primary Prevention Group.

The parameters of ultrasound cardiography showed no differences between the two groups (left atrial diameter, 47.1 ± 5.5 mm vs. 50.0 ± 5.8 mm; ejection fraction, 60.1 ± 6.5% vs. 63.2 ± 4.0%; max wall thickness, 19.5 ± 3.1 mm vs. 20.5 ± 4.1 mm; rest left ventricular outflow tract gradient ≥30 mmHg, 9.1% vs. 14.3%). The majority of patients had non-obstructive HCM (86.4 and 85.7% in the two groups, respectively).

The procedural characteristics are demonstrated in Table 2. Successful closure with satisfactory seals (residual leak ≤ 5 mm) was achieved in all patients, with mean procedure time of 53.4 ± 6.7 min and 52.1 ± 7.4 min for the Primary and Secondary Prevention Groups, respectively. Complete occlusion was achieved in 86.4% of the Primary Prevention Group and 92.9% of the Secondary Prevention Group. The median number of devices used was 1 (1.0–1.0), with mean device sizes of 28.7 ± 4.1 mm in the Primary Prevention Group and 29.6 ± 3.3 mm in the Secondary Prevention Group.

Two patients had procedure-related complications. One patient in the Primary Prevention Group had pericardial effusion, which required percutaneous drainage, and one patient in the Secondary Prevention Group had mild groin hematoma and recovered without surgery.

All patients underwent scheduled 45-day TEE. At 45 days, TEE showed satisfactory seals in all patients, with complete occlusion in 77.3% of the Primary Prevention Group and 78.6% of the Secondary Prevention Group (Table 3). No patients had residual flow of > 5 mm or DRT. At 6 months, 21 (58.3%) patients underwent a second follow-up TEE and 15 (41.7%) patients underwent a follow-up cardiac CT. Complete seals were identified in 86.4 and 85.7% of the two groups, respectively. One patient in the Secondary Prevention Group revealed asymptomatic DRT by TEE at the 6-month follow-up. The patient was a 75-year-old female with a CHA2DS2-VAS score of 6 and a HAS-BLED score of 3. OAC was discontinued in this patient after the 45-day TEE demonstrated complete occlusion of the LAA. The thrombus was resolved 3 months after reinitiated OAC at the TEE reassessment. Lifelong OAC was recommended for this patient and no thromboembolism or bleeding was recorded afterwards.

All patients completed at least 12 months of follow-up and the average follow-up was 29.4 ± 10.3 months in the Primary Prevention Group and 26.7 ± 11.1 months in the Secondary Prevention Group (Table 3). There were no thromboembolic events or deaths during follow-up. One patient in the Primary Prevention Group had a bleeding event (gastrointestinal bleeding while on dual antiplatelet therapy of aspirin and clopidogrel) 4 months after the procedure. No bleeding events were reported in the Secondary Prevention Group.

The pre- and post-implant antithrombotic treatment is shown in Figure 1. Prior to the procedure, 13 patients (59.1%) in the Primary Prevention Group and 10 (71.4%) in the Secondary Prevention Group received OAC. At 3 months, all patients switched to either dual or single antiplatelet therapy. During the latest follow-up, only one patient (1/36, 2.8%) in the Secondary Prevention Group remained on OAC because of previous DRT. Most of the patients were on single antiplatelet therapy (72.7% in the Primary Prevention Group and 78.6% in the Secondary Prevention Group).

This study has several limitations. This is a single-center retrospective study with a small sample size due to the low morbidity of HCM, which is one of the main limitations. Nevertheless, considering the lack of studies investigating LAAC in stroke prevention in the HCM population, this study offers valuable evidence on the use of LAAC for primary and secondary prevention of stroke. Patients with HCM are quite heterogeneous. Sudden cardiac death is more common in younger patients aged 10–35 years, heart failure death is more common in middle-aged patients, and stroke due to HCM-related AF is more common in older patients. Patients with HCM in this study were highly selective, mainly with old age, non- or mild obstruction and without prior surgery or ventricular aneurysm. Therefore, these selected patients had low risk of sudden cardiac death but an increased risk of stroke, and so did not represent the entire population of patients with HCM. Besides, the indications for most patients in the present study were unwilling to receive OAC therapy or having a high bleeding risk. How to select patients with HCM and AF who can benefit from LAAC therapy remains an open question. Further studies are needed to confirm the results of this study.