Author + information
- Received August 5, 2015
- Revision received September 14, 2015
- Accepted September 24, 2015
- Published online January 25, 2016.
- Christian Frerker, MD∗∗ (, )
- Michael Schlüter, PhD†,
- Oscar D. Sanchez, MD‡,
- Sebastian Reith, MD§,
- Maria E. Romero, MD‡,
- Elena Ladich, MD‡,
- Jörg Schröder, MD§,
- Tobias Schmidt, MD∗,
- Felix Kreidel, MD∗,
- Michael Joner, MD‡,
- Renu Virmani, MD‡ and
- Karl-Heinz Kuck, MD∗
- ∗Department of Cardiology, Asklepios Klinik St. Georg, Hamburg, Germany
- †Asklepios Proresearch, Hamburg, Germany
- ‡CVPath Institute, Inc., Gaithersburg, Maryland
- §Department of Cardiology, University Hospital of the RWTH Aachen, Aachen, Germany
- ↵∗Reprint requests and correspondence:
Dr. Christian Frerker, Department of Cardiology, Asklepios Klinik St. Georg, Lohmühlenstr. 5, 20099 Hamburg, Germany.
Objectives This study sought to assess the feasibility and safety of using a filter-based cerebral protection system (CPS) during MitraClip implantation and to report on the histopathologic analysis of the captured debris.
Background Stroke is one of the serious adverse events associated with MitraClip therapy.
Methods Between July 2014 and March 2015, 14 surgical high-risk patients (age 75 ± 7 years; 7 men; median logistic EuroSCORE 21%) underwent MitraClip implantation employing cerebral protection with a dual embolic filter system. All patients had severe mitral regurgitation of predominantly functional origin.
Results All procedures were successfully completed for both CPS deployment/retrieval and MitraClip implantation. A total of 28 filters (2 from each patient) were analyzed. Microscopically, debris was identified in all 14 patients. The most common tissue types were acute thrombus and small fragments of foreign material, which were found in 12 patients (85.7%) each. Organizing thrombus was present in 4 patients (28.6%), valve tissue and/or superficial atrial wall tissue in 9 patients (64.3%), and fragments of myocardium in 2 patients (14.3%). No transient ischemic attacks, strokes, or deaths occurred peri-procedurally or during a median follow-up interval of 8.4 months.
Conclusions In this small study of patients undergoing MitraClip treatment with cerebral protection, embolic debris potentially conducive to cerebrovascular events was found in all patients. Debris was composed most often of acute thrombus, foreign material likely originating from the hydrophilic device coating, and valve/atrial wall tissue. Further studies are warranted to assess the impact of cerebral protection on the incidence of cerebrovascular events after MitraClip therapy.
Mitral regurgitation (MR) is the second most common manifestation of valvular heart disease in adults (1). Since the first MitraClip (Abbott Vascular, Santa Clara, California) implantation in 2008, several studies have attested to the safety and effectiveness of this new percutaneous treatment option in patients with moderate-to-severe or severe MR (2–4). Current guidelines consider the MitraClip system to be a treatment option for high-surgical-risk patients (evidence class IIb) (5,6). Occurrence of a stroke or a transient ischemic attack (TIA) is one of the potential complications during a MitraClip procedure. Incidences of 0.2% to 2.6% have been reported (2,3,7). Blazek et al. (8) showed that the MitraClip procedure resulted in new ischemic cerebral lesions on diffusion-weighted magnetic resonance imaging (MRI) in 23 (85.7%) of 27 patients studied.
Cerebral embolic protection devices for the prevention of cerebrovascular events have been introduced for transcatheter aortic valve replacement (TAVR) procedures (9). Van Mieghem et al. (10) reported on the histopathologic characteristics of debris captured and retrieved from a filter-based cerebral embolic protection device in 30 (75%) of 40 patients undergoing TAVR. The present study assessed the feasibility and safety of using a filter-based cerebral protection device during MitraClip implantation and reports on the histopathologic analysis of the debris captured.
Between July 2014 and March 2015, a total of 14 surgical high-risk patients underwent a MitraClip procedure that employed cerebral protection with the dual-filter Sentinel system (Claret Medical, Santa Rosa, California) at 2 German sites (Hamburg [n = 10] and Aachen [n = 4]). The Hamburg patients were treated consecutively in 2 series of 6 and 4 patients; the Aachen patients were selected according to the patient’s propensity for thromboembolic events due to echocardiographically detected thrombus material on pacemaker or defibrillator leads. Four patients (29%) had an internal cardiac defibrillator (2 of them as a cardiac resynchronization therapy) and 1 patient (7%) had a two-chamber pacemaker. No patient had a porcelain aorta. Pertinent baseline patient characteristics are given in Table 1.
All patients had severe MR of predominantly functional origin. Left ventricular function was impaired, with a mean ejection fraction of 37%, and all patients presented with reduced physical capacity (New York Heart Association functional class III or IV). A history of atrial fibrillation and prior stroke was present in 8 (57%) and 2 (14%) patients, respectively. No patient was found to have a thrombus in the left atrial appendage or the left ventricle. Nine patients were on oral anticoagulation (phenprocoumon, n = 4; direct oral anticoagulants, n = 5) without interruption of the anticoagulation during the MitraClip intervention; 5 patients had dual antiplatelet therapy with aspirin and clopidogrel. Patients on dual antiplatelet therapy received a 300-mg loading dose of clopidogrel after the MitraClip procedure. Written informed consent was obtained from all patients.
Cerebral protection device
The Sentinel Cerebral Protection System (CPS) comprises 2 bag-like embolic filters attached back-to-back to the tip of a single 6-F compatible catheter, which is delivered from the right arm via radial or brachial artery access over a standard 0.014-inch coronary guidewire (Figure 1). The filter bags are made of polyurethane film with 140-μm laser-drilled holes, and each filter is mounted on a self-expanding nitinol wire loop. The system is advanced under fluoroscopic guidance such that the larger proximal filter (loop diameter: 9 to 15 mm) is deployed in the brachiocephalic trunk, and the smaller distal filter (loop diameter: 6.5 to 10 mm), which is in an articulating sheath inside the proximal filter, is placed in the left common carotid artery.
A standardized anticoagulation regimen with heparin was initiated with a loading dose of 70 IU/kg aiming at an activated clotting time (ACT) between 250 and 300 s. The bolus of heparin was given after the puncture of all peripheral vessels. Deployment of the cerebral protection device was subsequently performed in all patients via right radial access—that is, before the transseptal puncture required for MitraClip implantation. Left heart catheterization was not performed after placement of the CPS.
The procedural details of percutaneous mitral valve repair using the MitraClip have been described before (2). All procedures were performed under general anesthesia in a hybrid operating room. Procedural success was defined as discharge MR of maximally moderate severity. Procedural details are given in Table 2.
After MitraClip implantation, the embolic filters were withdrawn into the catheter and removed from the patient. Outside, the filters were cut and stored in 10% neutral buffered formalin solution and shipped to the CVPath Institute (Gaithersburg, Maryland) for analysis.
Histopathologic assessment of captured debris
A total of 28 filters (2 from each patient) were analyzed. The filters were photographed (Canon EOS Rebel XSi; Canon U.S.A., Melville, New York), examined grossly for visible debris, then cut open; all contents were filtered through a 40-μm nylon cell strainer (BD Falcon/Corning, Durham, North Carolina). The majority of particles detected were adherent to the filter, with only a small number of immersed particles retrieved from the fixative. The cell-strainer disc was photographed to document successful debris transfer, and measurements of the retrieved debris were performed. The specimens were dehydrated in a graded series of alcohols and embedded in paraffin, and sections of 5 to 6 μm thickness were cut, with 2 sections mounted on each glass slide. The sections were stained with hematoxylin and eosin and Movat’s pentachrome.
The sections were evaluated for the presence of thrombus, valve and atrial wall tissue, vascular structures with or without atherosclerotic changes, myocardial fragments, calcification, and foreign material. Thrombus was classified as acute if it showed platelets and fibrin with entrapped red blood cells and acute inflammatory cells, or as chronic if the thrombus showed the presence of spindle-shaped cells with or without macrophages that lined the thrombus, infiltrated it, or had any organization with matrix deposition interspersed between the fibrin/platelet thrombus. Two pathologists reviewed the slides independently, and the final diagnosis was based on unanimous agreement.
Histopathologic slides were scanned using a digital slide scanner (Axion Scan.Z1, Carl Zeiss, Thornwood, New York), and the morphometric analysis was performed using HALO digital image analysis software (Version 1, Indica Labs, Corrales, New Mexico). Quantitative measurements were performed on the 20 largest particles identified in each cross section. The cumulative area of debris was quantitated in each cross section using an internal algorithm for automated particle detection and was expressed as the cumulative area per case. Minimum and maximum diameters of individual particles were also determined.
Continuous variables are described as mean ± SD if normally distributed, or as median (interquartile range [IQR]) if not. The Mann-Whitney U test was used to assess between-group differences in maximum particle diameter, with p < 0.05 considered statistically significant. Categorical variables are described with absolute and relative frequencies.
The mean ACT in the 14 procedures was 289 ± 48 s. All procedures were successfully completed for both CPS deployment/retrieval and MitraClip implantation; 6 patients were discharged with no or trace residual MR, 5 had mild MR, and 3 patients were discharged with moderate MR. A single clip was implanted in 7 patients (Hamburg n = 5, Aachen n = 2); the other 7 patients received 2 clips each. The mean transmitral pressure gradient after clip implantation was 3.3 ± 1.4 mm Hg (range: 1 to 6 mm Hg).
The procedures lasted for a mean of 91 ± 43 min (range: 40 to 200 min), with a fluoroscopy time of 30 ± 16 min (range: 11 to 70 min). The total device time—the time from transseptal puncture to withdrawal of the clip delivery system from the left atrium—amounted to 58 ± 31 min (range: 15 to 130 min).
There were no peri-procedural TIAs or strokes. A single vascular complication occurred in a patient in whom the supra-aortic arteries were insufficiently visualized by angiography; the patient experienced bleeding from a small thyroid artery, which was treated conservatively by administration of protamine at the end of the procedure. During a median follow-up period of 8.4 months (IQR: 3.5 to 10.3 months), no patient died or experienced out-of-hospital TIA or stroke.
Microscopically, debris was identified in all 14 patients. The morphometric analysis of a total of 515 sections (242 from proximal, 273 from distal filters) revealed a median cumulative particle area per patient of 2.46 (IQR: 0.44 to 3.67) mm2 and a median maximum particle diameter of 295 (IQR: 104 to 509) μm; cumulative areas of debris captured in proximal and distal filters were 1.35 (0.30 to 2.09) mm2 and 1.07 (IQR: 0.13 to 2.21) mm2, respectively, with maximum diameters of 346 (IQR: 211 to 555) μm and 217 (IQR: 63 to 442) μm, respectively (p < 0.0001). The most common tissue types were acute thrombus (Figure 2A) and small fragments of nonpolarizable basophilic foreign material (Figure 3) that was morphologically consistent with hydrogel, which were found in 12 patients (85.7%) each. Organizing thrombus (Figure 2B) was present in 4 patients (28.6%), and 2 patients were devoid of any thrombus. Fibroelastic tissue with proteoglycan deposition, consistent with either valve tissue and/or superficial atrial wall tissue (Figure 4), was found in a total of 9 patients (64.3%). Two patients (14.3%) showed minute fragments of myocardium. Tissue particles exhibiting calcification were not observed in any of the filters.
Differentiations of histopathologic findings according to the number of clips implanted and the study site are given in Tables 3 and 4⇓⇓, respectively. No difference in the distribution of debris type was apparent in either analysis. However, the maximum particle diameters were statistically significantly larger in patients treated with 2 clips (median 402 vs. 134 μm [1 clip], p < 0.0001) and in patients treated in Aachen (median 411 vs. 262 μm [Hamburg], p < 0.0001).
The various combinations of types of debris as found in the individual patients are shown in Figure 5. The debris was composed of thrombus, valve/atrial wall tissue, and foreign material in 8 patients (57%), and of thrombus plus foreign material without any valve/atrial wall tissue in 3 patients (21%). In 1 patient each the constituents of debris were thrombus and myocardial fibers, foreign material and myocardial fibers, and valve/atrial wall tissue only.
The debris was contained in all 14 proximal and all 14 distal filters. The type of debris captured at the 2 filter locations is shown in Figure 6. Note the similarity of the distributions, with most filters at both locations containing acute thrombus and foreign material, followed by valve/atrial wall tissue. Two filters at either location contained organizing thrombus, and myocardial fibers were detected only in 2 distal filters.
The main findings of this small study of using a dual-filter CPS during MitraClip implantation are as follows:
• Using the CPS during a MitraClip procedure is feasible and safe.
• Debris was found in all patients, with larger particles identified in proximal filters, in patients treated with 2 clips, and in patients at an increased risk of thromboembolic events.
• The most prevalent types of debris were acute thrombus and foreign material consistent with hydrogel coating.
It has been shown that percutaneous interventional procedures bear the risk of new peri-procedural cerebral ischemic events (11). In a recent consensus document, the Mitral Valve Academic Research Consortium (MVARC) stated that stroke and TIA are “clinical endpoints to be collected in all trials of mitral valve therapies” (12). The incidence of stroke or TIA after a MitraClip procedure of up to 2.6% (7) justifies any attempt to prevent such complications.
In this study, the Claret Sentinel device was used for cerebral protection during MitraClip procedures. Except for 1 minor vascular complication, no complications occurred while using this CPS. The overall procedure duration of 91 min compares favorably with procedure times reported in the Pilot European Sentinel MitraClip Registry (138 min), the MitraClip post-approval ACCESS-EU registry (100 min), and the German transcatheter mitral valve interventions (TRAMI) registry (103 min) (3,4,13).
In our series of patients without any peri-procedural cerebrovascular events, the filters captured debris in all patients. A recent diffusion-weighted MRI study of patients after MitraClip implantation without cerebral protection revealed an 86% incidence of newly acquired microembolic cerebral lesions (8), suggesting a strong relationship between embolic debris captured in filters during MitraClip implantation with cerebral protection and ischemic cerebral lesions detected by MRI after MitraClip implantation without cerebral protection. Although most of the MRI-detected embolic lesions remained clinically silent, other studies have shown that clinically silent cerebral lesions were associated with neurocognitive impairment and the development of dementia (14,15).
Potential origins of emboli during transcatheter mitral valve repair
Feld et al. (16) have shown that the Brockenbrough transseptal needle generates particles when advanced through the dilator and transseptal sheath. The particles in that study were not examined further for the type of material generated. In our study, foreign particles found in the filters consisted of nonpolarizable basophilic material that was morphologically consistent with hydrogel. To date, no evidence has been found of foreign material arising from the Sentinel filter itself and/or its coating. Even after we manually scraped the surface of filters, no foreign material was observed under high-magnification microscopy.
Hydrophilic coating is frequently used on medical devices for interventional procedures to decrease friction between sheaths or catheters and vessel walls (17). Mehta et al. (18) identified hydrophilic polymer emboli in 9 patients who underwent various vascular interventions using different coated devices. Furthermore, multiple findings of hydrophilic-coating material in different organs have been described (19–21). For the MitraClip procedure several catheters coated with hydrophilic materials are used, such as the transseptal sheath for transseptal puncture and the guide catheter for the MitraClip system. However, it is unknown from which catheter the majority of foreign material originated.
The MVARC stated that transcatheter mitral valve therapies may predispose patients to the formation of thrombus (12). Acute thrombus could develop at the transseptal sheath as well as at the guiding catheter and the clip delivery system due to their thrombogenic nature. Maleki et al. (22) reported up to 9% thrombus formation on regular transseptal sheaths despite adequate anticoagulation. In our study, anticoagulation was adequate (as reflected by a mean ACT of 289 s throughout the procedures). In a cohort of 81 patients who underwent TAVR, thrombotic material was found in 74% of patients (23). However, in that trial the ACT was suboptimal at a mean of 230 s (23).
The origin of the valve and atrial wall tissue captured in 9 patients (64%) could be generated from the transseptal puncture and the advancement of the MitraClip into the left ventricle across the mitral valve. Furthermore, during grasping of the mitral leaflets some valve tissue might be embolized.
Calcific debris was not found in any of the filters analyzed in our study. This finding appears to be consistent with the MRI study by Blazek et al. (8) who reported no impact of mitral valve calcification on the occurrence of embolic lesions.
The number of patients enrolled in this study was low. Most patients (11 of 14; 79%) had functional MR. Degenerative MR might be associated with more valve-related debris. We made no differentiation with respect to the type of MR. No brain imaging or assessment of neurocognitive function before and after the intervention was performed. Both methodological additions to the study protocol as well as a comparison with a matched cohort of patients undergoing MitraClip implantation without cerebral protection would have been desirable.
In this study of 14 patients undergoing MitraClip treatment using a CPS, embolic debris potentially conducive to cerebrovascular events was found in all patients. The debris was composed most often of acute thrombus, foreign material likely originating from the hydrophilic device coating, and valve/atrial wall tissue. No peri- or post-procedural strokes or TIAs occurred. Further studies are warranted to assess the impact of cerebral protection on the incidence of cerebrovascular events after MitraClip therapy.
WHAT IS KNOWN? MitraClip implantation is a valid therapeutic option to relieve symptoms in elderly patients with significant MR who are deemed inoperable or at high surgical risk. Stroke has been observed, if rarely, in association with MitraClip therapy, and diffusion-weighted MRI studies have revealed a high incidence of new cerebral ischemic lesions after the intervention.
WHAT IS NEW? The present study of a dual-filter cerebral protection system employed during MitraClip therapy has shown that embolic debris, composed predominantly of acute thrombus and foreign material likely originating from the hydrophilic device coating, was released in all patients.
WHAT IS NEXT? Our findings warrant further studies to assess the impact of cerebral protection on the incidence of cerebrovascular events and possibly identify patients in whom cerebral protection should be mandatory during MitraClip implantation.
The authors thank Torie Samuda, PA, for her assistance in the histopathologic analyses.
Dr. Frerker has received lecture honoraria from Abbott Vascular, Inc., and Claret Medical, Inc. Dr. Joner has received honoraria from OrbusNeich, Abbott Vascular, Boston Scientific, and Biotronik; and research grants from OrbusNeich, Abbott Vascular, BioSensors International, SinoMedical, Terumo Corporation, CeloNova, W.L. Gore, Medtronic, Microport, Boston Scientific, and Biotronik. Dr. Virmani has served as a consultant to 480 Biomedical, Abbott Vascular, Medtronic, and W.L. Gore; has received honoraria from 480 Biomedical, Abbott Vascular, Boston Scientific, Cordis J&J, Lutonix Bard, Medtronic, Merck, Terumo Corporation, W.L. Gore, Cardiovascular Systems Inc., Meril Life Sciences, and Spectranetics; and has received institutional grants/research support from 480 Biomedical, Abbott Vascular, Atrium, BioSensors International, Biotronik, Boston Scientific, Cordis Johnson&Johnson, GlaxoSmithKline, Kona, Medtronic, MicroPort Medical, CeloNova, OrbusNeich Medical, ReCore, SINO Medical Technology, Terumo Corporation, W.L. Gore, Spectranetics, Cardiovascular Systems Inc., Lutonix Bard, Surmodics, and Meril Life Sciences. Dr. Kuck has received consultant fees and research grants from St. Jude Medical, Medtronic, and Edwards Lifesciences; and has received research grants from Abbott Vascular, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- activated clotting time
- cerebral protection system
- interquartile range
- mitral regurgitation
- magnetic resonance imaging
- Mitral Valve Academic Research Consortium
- transcatheter aortic valve replacement
- transient ischemic attack
- Received August 5, 2015.
- Revision received September 14, 2015.
- Accepted September 24, 2015.
- American College of Cardiology Foundation
- Iung B.,
- Baron G.,
- Butchart E.G.,
- et al.
- Nickenig G.,
- Estevez-Loureiro R.,
- Franzen O.,
- et al.
- Maisano F.,
- Franzen O.,
- Baldus S.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Vahanian A.,
- Alfieri O.,
- Andreotti F.,
- et al.
- Glower D.D.,
- Kar S.,
- Trento A.,
- et al.
- Van Mieghem N.M.,
- Schipper M.E.,
- Ladich E.,
- et al.
- Stone G.W.,
- Adams D.H.,
- Abraham W.T.,
- et al.
- Ghanem A.,
- Müller A.,
- Nähle C.P.,
- et al.
- Allan R.W.,
- Alnuaimat H.,
- Edwards W.D.,
- Tazelaar H.D.
- Van Mieghem N.M.,
- El Faquir N.,
- Rahhab Z.,
- et al.