Article Text
PDF
Abstract
Objective Percutaneous left atrial appendage (LAA) occlusion has been developed as a viable option for stroke and thromboembolism prevention in patients with non-valvular atrial fibrillation (NVAF) and at high risk for cerebral cardioembolic events. Data on device implantation and long-term follow-up from large cohorts are limited.
Methods 110 consecutive patients with NVAF and contraindications to oral anticoagulants (OACs) underwent LAA occlusion procedures and achieved a longer than 1 year follow-up. All patients were enrolled in a prospective registry. Procedures were performed using the Amplatzer Cardiac Plug or Amulet guided by fluoroscopy and intracardiac echocardiography.
Results Mean age of the population was 77±6 years old; 68 were men. Atrial fibrillation was paroxysmal in 20%, persistent in 15.5% and permanent in 64.5% of cases, respectively. Mean CHA2DS2-VASc and HAS-BLED scores were 4.3±1.3 and 3.4±1, respectively. Technical success (successful deployment and implantation of device) was achieved in 100% of procedures. Procedural success (technical success without major procedure-related complications) was achieved in 96.4%, with a 3.6% rate of major procedural complications (three cases of pericardial tamponade requiring drainage and one case of major bleeding). Mean follow-up was 30±12 months (264 patient-years). Annual rates for ischaemic stroke and for other thromboembolic events were respectively 2.2% and 0%, and annual rate for major bleeding was 1.1%.
Conclusions Our data suggest LAA occlusion in high-risk patients with NVAF not suitable for OACs is feasible and associated with low complication rates as well as low rates of stroke and major bleeding at long-term follow-up.
- Stroke
Statistics from Altmetric.com
Introduction
Non-valvular atrial fibrillation (NVAF) is the most common sustained arrhythmia, with an annual risk for stroke of 5% in non-anticoagulated patients. Oral anticoagulants (OACs)1–4 are the current gold standard for stroke and thromboembolism prevention. Nevertheless, a non-negligible number of patients have absolute/relative contraindications to OACs, and to warfarin in particular, as they have a high bleeding risk and often previous bleedings.5 On the other hand, up to 50% of patients are not well managed, because they do not take anticoagulants, have inadequate anticoagulation or discontinue warfarin, increasing their risk of thromboembolic complications.6
A percutaneous procedure has been developed in recent years aimed to provide circulatory exclusion of the left atrial appendage (LAA),7 the major source of cardiac thromboembolism.8 In the PROTECT-AF (WATCHMAN LAA System for Embolic Protection in Patients with Atrial Fibrillation) trial (WATCHMAN device, Boston Scientific), LAA occlusion demonstrated non-inferiority to OACs9 (annual rate for ischaemic stroke 1.4% vs 1.1% in the control group), though there was a 10% of major procedural complications and 14% of patients prolonged OAC during follow-up. Relatively few studies have evaluated LAA occlusion results in patients with absolute/relative contraindications to warfarin. The PLAATO (Percutaneous LAA Transcatheter Occlusion) study used the PLAATO device (ev3), a device that was withdrawn from the market.10–14 The first prospective multicentre non-randomised ASAP (ASA Plavix Feasibility Study With Watchman LAA Closure Technology) trial showed that LAA closure with the Watchman device can be safely performed without a warfarin transition, and is a reasonable alternative to consider for patients at high risk for stroke (Mean CHADS2 score 2.8±1.2), but with contraindications to systemic oral anticoagulation.15 Some recent studies, on patients with short-term follow-up, reported the feasibility of LAA occlusion with the Amplatzer Cardiac Plug (ACP) device (St. Jude Medical) in patients with absolute contraindications to OACs, showing a lower than expected rate of embolic and haemorrhagic events and acceptable procedural success.16 ,17 A Canadian multicentre study (52 patients), a single-centre study from Switzerland (152 patients, some with non-dedicated devices), as well as two registries, confirmed these results.18–21
In this paper, we analyse procedural and long-term follow-up results of LAA occlusion performed in a single centre, on a large prospective cohort of patients with contraindication to OACs.
Methods
Study population
A total of 110 consecutive patients with NVAF underwent LAA occlusion at Ospedale del Cuore Fondazione Toscana C.N.R., Gabriele Monasterio, Massa, between January 2009 and June 2014. All patients had (a) CHA2DS2-VASc score ≥2; (b) absolute/relative contraindication to OACs due to high bleeding risk or previous haemorrhage or stroke during treatment with OACs; (c) longer than 1 year follow-up. The CHA2DS2-VASc22 and HAS-BLED23 scores were calculated. The local ethic committee approved this prospective registry (local ethic committee approval no. 414, CARDIO-02-R).
Device and implantation procedure
Patients were implanted with the ACP or Amulet devices (St. Jude Medical, Plymouth, Minnesota, USA). Both devices consist of a lobe and a disc, made of a nitinol mesh and polyester patch, connected by a flexible waist. Percutaneous implantation is facilitated by a dedicated delivery system. The lobe is implanted within the LAA and has stabilising wires to retain its position. The disc seals the orifice of the LAA. The appropriate device size is selected from eight different lobe diameters available for the ACP, and eight for Amulet, based on the internal LAA diameter. Depending on the lobe diameter, the disc diameter is 4 or 6 mm larger that the lobe for ACP, while that for Amulet it is 6 or 7 mm larger. Amulet is the second-generation device with some new features to allow treatment of more anatomies and simplify device preparation (increased size dimensions, exclusion of the proximal end screw to have a flat disc and avoid thrombus formation, longer waist to provide more flexibility between the lobe and disc, preloaded device to facilitate and reduce device preparation and de-airing time).
Prior to the device implantation, transoesophageal echocardiography (TEE) and cardiac CT (CCT) were performed to exclude left atrial and LAA thrombus and determine the appropriate device size. The procedure was performed by right femoral vein approach and trans-septal atrial puncture. All procedures were guided by both fluoroscopy and intracardiac echocardiography (ICE), and performed with local anaesthesia. Implantation techniques and the use of procedural imaging have been described elsewhere.24 ,25 Heparin (100 U/kg) was administered after the trans-septal puncture, and the dose was adjusted to achieve an activated clotting time >250 s. No anticoagulation therapy was administered after the procedure. As a general rule, dual antiplatelet therapy with aspirin (250 mg intravenous load, followed by 100 mg daily) plus clopidogrel (300 or 600 mg load, followed by 75 mg daily) was given for 3 months after the procedure, with subsequent single antiplatelet therapy.
Technical success was defined as successful deployment and implantation of the device, while procedural success was defined as technical success without major procedure-related complications (major pericardial effusion/tamponade, stroke, systemic embolism, major bleeding and device embolisation). Pericardial effusion/tamponade and bleeding were considered significant if requiring respectively drainage or transfusion with ≥2 units of packed red blood cells or surgical intervention. The same operator (SB) performed all the procedures. All the patients remained in the hospital for at least 24 hours after the procedure.
Follow-up
Follow-up was performed by clinical visits or telephone follow-up at 1, 6 and 12 months, and then once a year. Patients were considered censored in case of death or after 5 years of follow-up. At 6 months postimplantation, CCT with iodinated contrast medium was performed to confirm stable device position, exclude device-related thrombus and assess residual leaks. A semiquantitative method was used to define leaks with CCT as opacification of more than one-third of LAA in the arterial phase. In patients that could not perform CCT, TEE was performed. A leak was classified as minor, moderate or major when the jet, observed by TOE, was <1 mm, between 1 and 3mm, and >3 mm, respectively. Stroke, transitory ischaemic attacks (TIA), peripheral thromboembolic events, major/minor haemorrhage and death from cardiac or any other causes were registered as events. Major bleedings were defined, according to the International Society on Thrombosis and Haemostasis criteria, as clinically overt bleedings accompanied by a decrease in the haemoglobin level of at least 2 units of packed red cells, occurring at a critical site or resulting in death. Minor bleedings were defined as clinically overt bleedings that did not satisfy the criteria for major bleedings and that led to hospital admission, physician-guided medical or surgical treatment or a change in antithrombotic therapy.26
Statistical analysis
Continuous data were expressed as mean±SD. Technical and procedural success rates were calculated as percentages of the total number of patients. Patient-years were calculated as the product of patient number and the mean years of follow-up. For proportions, numbers and percentages were used. Comparisons between observed and expected rates24 were assessed using binomial χ2 test, and relative risk reduction was calculated by dividing the absolute risk reduction by the event rate in the historical control group. Survival free from events curves were analysed using the Kaplan-Meier estimate. Statistical analysis was performed using SPSS (V.19, SPSS, Chicago, Illinois, USA); p<0.05 was considered significant.
Results
Baseline characteristics of the population are shown in table 1. Reasons for anticoagulation therapy contraindications are shown in table 2.
Patient baseline characteristics
Patients’ contraindications to anticoagulation therapy
Procedural results and in-hospital outcomes
ACP device was implanted in 91 patients, while Amulet was used in 19 patients. Technical success was achieved in all procedures. Procedural success was achieved in 96.4% of cases, with a 3.6% rate of complications (three cases of pericardial tamponade requiring drainage, one major bleeding). Six patients had minor bleeding. In two cases, the device that was chosen according to the preoperatory assessment was changed in size during the procedure due to instability at implant or at the pre-release stability test. No patient experienced device embolisation (table 3).
Procedural findings and outcomes
In 90.9% of cases (n=100), dual antiplatelet therapy with aspirin plus clopidogrel was given for 3 months after the procedure, with subsequent single antiplatelet therapy. In a few cases, single antiplatelet therapy was prescribed after the procedure (5.5%, n=6), or dual antiplatelet therapy was continued indefinitely after 3 months from the procedure (3.6%, n=4) (table 4).
Follow-up results
Follow-up
Mean follow-up was 30±12 months, comprising a total implant experience of 264 patient-years. Fourteen patients (12%) died during the follow-up period. One cardiac death occurred in a patient with hypertrophic cardiomyopathy. A patient on growth factors therapy for haematological disorders experienced an ischaemic stroke and died 1 month later. There were five other cases of ischaemic stroke at follow-up, while one patient experienced a TIA. Three cases of major bleeding occurred (one gastrointestinal, a haemorrhagic stroke and a subdural haematoma) (table 4). Among non-cardiovascular (CV) deaths, we report acute respiratory insufficiency, acute renal failure, lung cancer, liver cancer, urinary bladder cancer, three cases of pneumonia and two cases of accidental head trauma. The main clinical characteristics of patients with cardiovascular events during the follow-up period are shown in table 5.
Main features of patients with stroke/major bleeding at follow-up
The annual rates for ischaemic stroke and for other thromboembolic events were 2.2% and 0%, respectively. The annual rate for major bleeding attested at 1.1%. The Kaplan-Meier curves for freedom from death, thromboembolic and haemorrhagic events, respectively, are shown in figure 1.
Kaplan-Meier curves for adverse events (death, stroke/peripheral embolism, major bleeding) in 110 patients with non-valvular atrial fibrillation that underwent left atrial appendage occlusion, at 60 months’ follow-up.
In 89 out of 110 patients, a follow-up CCT was performed at 6 months. TEE was performed in patients that could not perform CCT. At CCT, the opacification of the appendage during the arterial phase was present in 44/89 patients, but only in 22 cases an incomplete apposition of the device was revealed. A possible explanation for LAA opacification in the other 22 cases could be incomplete endothelialisation of the device. At TEE, a leak was present in 4/20 patients. One patient died before the 6-month follow-up. In five out of six cases of ischaemic stroke in our follow-up, a leak was documented at 6 months CCT (four leaks) or TEE (one leak) (p=0.183 for comparison of stroke rate in patients with or without a residual leak). No case of device thrombosis or dislocation occurred. In a few cases, we repeated a second CCT to analyse peridevice leaks documented during first follow-up CCT, and no changes were observed in any cases (reduction or increase) in leak dimensions.
Discussion
The main findings of this single-centre report are (1) high technical and procedural success of LAA occlusion procedure; (2) long-term efficacy of LAA occlusion procedure in reducing thromboembolic events with low incidence of bleeding complications; (3) importance of accurate preprocedural, periprocedural and postprocedural imaging to avoid complications, optimise the procedure and potentially predict clinical events at follow-up; (4) all the procedures were done in local anaesthesia under ICE/fluoroscopy guidance. Remarkably, this report, with a 3-year follow-up outcome data, is one of the first studies assessing LAA occlusion results on a large single-centre sample.
Warfarin has for a long time been considered the gold standard therapy to reduce the occurrence of thromboembolic events. Nevertheless, warfarin exposes patients at risk of bleeding events, while many factors that promote thromboembolism also facilitate haemorrhagic events. In addition, it may be difficult to maintain an adequate anticoagulation. The net result is that a non-negligible number of patients that should be on warfarin have absolute/relative contraindications to it, reluctance to treatment or inadequate anticoagulation.22 Novel OAC drugs show to provide similar or better stroke prevention than warfarin at reduced, but still clinically relevant, bleeding rates.2–4 Percutaneous LAA occlusion procedure could represent a viable alternative to OAC in patients with unacceptable bleeding risk.8
In this single-centre series, we confirm and extend pre-existent data on LAA occlusion procedure safety and effectiveness.15–21 In our experience, the procedures guided by ICE and fluoroscopy warranted few cases of periprocedural major complications, with no cases of death related to the device, device embolisation, TIA, stroke or peripheral embolisation.
We confirm that a strategy of LAA occlusion followed by antiplatelet therapy is associated with a very low rate of embolic events, especially considering the expected rate on the basis of the score risk. Follow-up data were collected from 110 patients with high risk of bleeding, and with homogeneous clinical and procedural features, that achieved a follow-up longer than 1 year, yielding a cumulative follow-up experience of 264 patient-years, the longest to our knowledge with the ACP/Amulet device.17–20 ,27 Over a 30-month follow-up, as compared with patients with similar CHA2DS2-VASc score,28 without OAC therapy, our patients show a 60% and 71% relative risk reduction for ischaemic stroke and thromboembolic events, respectively. Interestingly, the event rate for thromboembolic events was also lower than historical populations with similar CHA2DS2-VASc score, on warfarin therapy,28 possibly due to inadequate anticoagulation range. At the same time, a 63% relative risk reduction for haemorrhagic events was shown as compared with the expected rate in patients with similar HAS-BLED score and on warfarin therapy.28 All haemorrhagic events in our cohort occurred in first 2 months of follow-up, when dual antiplatelet therapy was prescribed. Our findings are consistent with previous works;18–21 ,27 our very long follow-up time leads to a still higher appreciation of LAA occlusion procedure results.
Follow-up imaging was achieved in all patients. Since a gold standard imaging modality has not been defined so far, we usually prefer CCT instead of TEE due to less invasiveness. Any case of device thrombosis or dislocation was visualised. In this series, CCT shows high sensitivity for leak detection and a trend toward a higher rate of ischaemic events when a leak is present. These results merit further investigation on larger cohorts also in an attempt to define, according to individual risk, postprocedure antiplatelet therapy.
Study limitations
This is single arm prospective registry. Imaging follow-up was performed only once in the majority of patients. This may limit diagnostic accuracy.
Conclusions
In patients with NVAF at high risk of cardioembolic events and with absolute/relative contraindications to anticoagulation therapy, LAA occlusion followed by dual/single antiplatelet therapy was associated with a low rate of thromboembolic and bleeding complications at very long follow-up (mean 30 months). LAA occlusion suffered a low rate of procedural complications due to careful preprocedural and periprocedural imaging in this single-centre report. Even if a residual leak at follow-up CCT is quite common, a trend for an association between the presence of residual leak and ischaemic events may be hypothesised. Clinical trials with larger cohort of patients and a more complete imaging follow-up are expected to confirm these results, and possibly optimise postprocedural therapy. Randomised clinical trials, comparing LAA occlusion with other therapeutic strategies, are warranted.
Key messages
What is already known on this subject?
Percutaneous left atrial appendage (LAA) occlusion is a recognised option for stroke and thromboembolism prevention in patients with non-valvular atrial fibrillation and contraindications to oral anticoagulants.
What might this study add?
Among a cohort of 110 consecutive patients, LAA occlusion was associated with high technical success (100%) and a relatively low rate of periprocedural complications (3.6%). A low rate of thromboembolic and bleeding complications was observed at long-term follow-up (mean 30 months), with an annual rate for ischaemic stroke and for other thromboembolic events of 2.2% and 0%, respectively.
How might this impact on clinical practice?
Our data provide further support for consideration of LAA occlusion in adults with non-valvular atrial fibrillation who are not candidates for long-term oral anticoagulation.
Acknowledgments
Thanks to Ms Karin Joan Tyack for her linguistic and data collection support.
References
Footnotes
Contributors SB, GT, MR, ARDC, UP, AR and CP were personally involved in the procedures of LAA occlusion, contributed to data analysis and to the writing of the manuscript. LEP and MR acted as main data collectors and wrote the different versions of the manuscript. EC helped in data collection and analysis.
Competing interests SB is proctor for St. Jude Medical and Edwards Lifesciences. The other authors report that no potential conflicts of interest exist with any companies/organisations whose products or services may be discussed in this article.
Ethics approval Comitato Etico di Massa e Carrara.
Provenance and peer review Not commissioned; externally peer reviewed.