LEFT ATRIAL APPENDAGE CLOSURE IN PATIENTS WITH MECHANICAL MITRAL VALVE PROSTHESIS: A MULTICENTER ITALIAN PILOT STUDY

STUDY POPULATION

This is a retrospective, multicenter study carried out at the Niguarda Hospital (Milan), San Raffaele Hospital (Milan) and Sant ’Eugenio Hospital (Rome), Italy. Patients were enrolled from January 2012 to January 2022. The following inclusion criteria were considered:

- AF patients on OAC therapy with VKA who underwent percutaneous LAAC;

- Recent history (<1 month prior to LAAC) of failure of appropriate VKA therapy. Events considered were:

- occurrence of transient ischemic attack (TIA), stroke or systemic embolism (SE);

- identification of LAA thrombus or moderate/severe sludge at transesophageal echocardiography (TOE);

As for the thromboembolic events, ischemic stroke was defined as a sudden onset of a focal or global neurological deficit, lasting >24h or <24h but with imaging-documented new or presumed new ischemic lesion; TIA was defined as a neurological dysfunction lasting <24h and without new alteration identified on imaging studies[16]. SE was defined as an abrupt vascular insufficiency associated with clinical or radiological evidence of arterial occlusion in the absence of another likely mechanism. LAA thrombus was considered as the presence of a organized hyperechogenic formation placed in the bottom of LAA. LAA sludge was graded according to universally accepted guidelines in case of an intracavitary echodensity consisting of a prethrombotic state with very pronounced spontaneous echocontrast but without being a thrombus formed[6].

The following exclusion criteria were considered:

- Identification of INR below the therapeutic range in conjunction with the event;

- Possibility of embolic origin different from LAA (PMV thrombosis, left ventricle [LV] apical akinesia, intracardiac shunts and severe carotid stenosis) identified at cardiac imaging.

Baseline clinical characteristics and therapy have been recorded for all patients. Written informed consent was obtained from each patient. The retrospective study was approved by the Ethics Committee of Niguarda Hospital and complies with the Declaration of Helsinki. Data were recorded in a dedicated database in compliance with the ethic committee of our center.

TECHNICAL ASPECTS OF PROCEDURES

To minimize the risk of complications, intra-procedural TOE monitoring was always conducted. LAAC procedure was performed directly by a senior operator for each center (P.M., M.M. and A.G.) according to universally accepted guidelines[17]. CPDs were used according to the operator’s discretion. Before the release of the device, its position, anchoring and sizing were evaluated. Procedural success was defined as technical success of device deployment without procedure-related major cardiac events: cardiac tamponade, device embolization, periprocedural stroke (within 24 hours from the procedure) or significant peridevice leak (PDL,

≥5 mm at a Nyquist limit of 20–30 cm/s).

POSTPROCEDURAL MANAGEMENT AND FOLLOW-UP

Information about periprocedural outcomes, including all procedure and device-related adverse events (<7 days) were extracted from patients charts and electronic records. Patients were discharged from the hospital with the aim of maintaining lifelong OAC therapy to prevent the recurrence of ischemic events with a hybrid strategy consisting of LAAC + OAC therapy. For 3 months after discharge, a single antiplatelet drug was added to OAC regimen according to the operator’s discretion. The first clinical follow-up (FU) was performed at 3 months from hospital discharge with a clinical visit + TOE to rule out major complications such as device-related thrombosis (DRT), device embolization, or significant PDL. Additional long-term FU was conducted by clinical visits in our outpatient clinic or, if not available, by telephone interviews with the patient every year. Data concerning medical therapy and adverse events during FU were recorded, in particular death (cardiovascular and all-cause), events (ischemic stroke, TIA, and SE) and major bleedings.

A call for therapeutic alternatives in patients with “valvular AF” and OAC failure

LAAC in patients with AF and failure of OAC therapy is a continually expanding indication due to proved safety and efficacy of the procedure and no need for antithrombotic therapy enhancement[18,19]. However, all LAAC devices that have acquired commercial approvement so far were tested only in the specific population affected by non valvular AF and anticoagulant contraindication. In our study, 167 (9.5%) patients referred to three major Italian centers underwent LAAC due to antithrombotic therapy failure and 12 of them were carriers of mechanical PMV. To be noted, 5 (42%) patients of PMV group had recent history of CE. Since VKAs are the only OAC available in this population, prevention of further adverse events remain a real-world critical concern[20]. LAAC effectiveness in this setting has been confirmed by a recent trial that compared the long-term benefits of LAAC concomitant to cardiac surgery[21]. Enhancement of medical therapy in such scenarios lacks of comparative trials and the decision to improve antithrombotic therapy should be weighted considering several factors such as the bleeding risk[22]. In our study, median HAS-BLED score of the enrolled population was 3, justifying an alternative therapeutic choice.

Left atrial appendage closure in carriers of mechanical mitral valve prosthesis

AF patients with mechanical PMV are characterized by some additional critical concerns compared to conventional AF patients such as the need for higher INR target and the proximity of the prosthesis to the appendage orifice. In this setting, preprocedural imaging by TOE or cardiac computer tomography are useful to evaluate the relationship between PMV and LAA. Indeed, in cases of little LAA orifice, protruding Coumadin’s ridge or too proximal deposition, the LAAC device may protruding out of the LAA for some millimeters. Despite protrusion of device shoulder by <40% to 50% of its total length is acceptable according to guidelines[23], hindering of valve prosthesis is very unlikely due to the anatomical orientation of LAA and mitral valve. In our study, the major part of PMV group underwent the procedure with an umbrella-shaped device (such as the Watchman), probably due to the higher medical staff experience while in two cases a deeper implant of the Amulet was chosen. Both procedure and fluoroscopy time were higher in PMV group due to the greater complexity and the need to repeated evaluation of the correct positioning of the device on various fluoroscopic and echocardiographic projections. Absence of complications and PDL have been reported in this group. In 35 cases judged at high risk of intraprocedural thromboembolism a TriGuard3 (Keystone Heart, Caesarea, Israel) or Sentinel (Boston Scientific, Marlborough, MA, US) CPD was used. They were constantly employed in cases of LAA thrombosis or severe LAA sludge while in presence of moderate to severe sludge (rating ≤3), a personal choice was performed by the specialist. The use of CPDs in LAAC with LAA thrombosis allowed to a more safe procedure and to reduce the use of specific techniques to minimize interventions within the LAA[24]. So far, a single case report have described a successful deployment of Watchman FLX in the context of LAAC performed with cerebral protection in a patient carrier of mechanical PMV affected by multiple CEs and LAA thrombosis despite therapeutical INR range[15].

Postprocedural antithrombotic therapy

Our study explored for the first time the effects of a possible hybrid strategy including LAAC plus OAC therapy with VKA in patients with mechanical PMV. All patients continued VKA after LAAC but nearly 50% added an antiplatelet drug for 3 months after LAAC according to the center clinical practice. No DRT were reported and no patients continued antiplatelet therapy beyond that period. Our decision to avoid long-term antiplatelet therapy was in accordance with the only case report present in literature describing a similar scenario[15]. In this case, antithrombotic therapy with VKA alone was continued after the procedure and no CEs during the FU were reported. Currently, the optimal medical therapy after LAAC is still an unresolved issue and varies by type of LAAC device. As this population is highly heterogeneous and characterized by variable thrombotic and bleeding risk, a standard approach is not consistently applied and patients are discharged with different therapeutic regimens according to clinician choice. According to a recent metanalysis, VKA was inferior to DOACs in both CEs and major bleedings endpoints[25]. However, no randomized studies have been performed in patients with indication to continue OAC as well as in patients with mechanical PMV. Since in our specific population the risk of bleeding was not negligible, the hybrid strategy of LAAC plus VKA without the increase of INR target or prolonging antiplatelet therapy was judged the most suitable. At long term FU, despite both higher ischemic and hemorrhagic risk of PMV group compared to control, no differences in terms of adverse events were reported.

Study limitations

This is an observational, multicenter, retrospective study; therefore, it has the inherent limits of the study design, and our results must be confirmed in a larger sample size. The choice to perform LAAC in patients with mechanical PMV and high-ischemic risk represents an off-label approach based on real-world practice. To compare procedural outcome of LAAC as well as its long term effects, a population with same indication but without mechanical PMV was selected. This choice brings the inevitable bias that PMV group needs higher INR levels than the control group and therefore intrinsically different ischemic and hemorrhagic risk profiles. Finally, so far there are no data available to compare our hybrid approach to an approach of anticoagulant intensification/switch treatment without LAAC: this comparison would indeed be of great importance to evaluate and, eventually, support the role of our strategy.

Conclusions

In this pilot study we reported the feasibility and safety and the long-term beneficial effects of LAAC in patients with mechanical PMV performed in high volume Italian centers. The present study is a proof of a new concept but further evaluation of this strategy in large prospective, controlled trials is warranted.