Author + information
- Received June 6, 2016
- Revision received October 24, 2016
- Accepted November 3, 2016
- Published online May 15, 2017.
- John Silberbauer, MA, MD(Res)a,∗ (, )
- John Gomes, PhDa,
- Sean O’Nunain, MDa,
- Senthil Kirubakaran, MDa,b,
- David Hildick-Smith, MDa and
- James McCready, MBBSa
- aRoyal Sussex County Hospital, Brighton, United Kingdom
- bQueen Alexandra Hospital, Cosham, Portsmouth, United Kingdom
- ↵∗Address for correspondence:
Dr. John Silberbauer, Royal Sussex County Hospital, Eastern Road, Brighton BN2 5BE, United Kingdom.
Objectives This study assessed the feasibility of intentional coronary venous perforation and exit with subsequent pericardial carbon dioxide (CO2) insufflation as a novel method for assisting subxiphoid pericardial puncture in the setting of epicardial mapping and ablation for ventricular tachycardia. The technique required that coronary venous perforation would not lead to significant bleeding.
Background Widespread adoption of first-line endoepicardial ventricular tachycardia ablation has not been taken up because of the risk of lacerating coronary vessels and puncturing the right ventricle with direct subxiphoid puncture.
Methods A lateral branch of the coronary sinus was subselected using a diagnostic JR4 coronary catheter inside a steerable sheath, via femoral access, and a distal branch then perforated intentionally using a high tip load 0.014-inch angioplasty wire. Either a microcatheter or over-the-wire balloon was then passed over this into the pericardial space, allowing up to 150 ml of pericardial CO2 insufflation, which allowed direct visualization of subxiphoid anterior pericardial access using a microneedle technique.
Results Intentional coronary vein exit was achieved in all 12 patients. In 1 patient, this confirmed widespread pericardial adhesions and therefore only endocardial VT ablation was undertaken. The other patients underwent successful pericardial CO2 insufflation and subxiphoid access allowing epicardial ventricular mapping and ablation. The immediate pericardial aspirate was dry or contained serous fluid in all but 1 patient.
Conclusions We report the first human transcoronary vein exit procedure. Coronary vein exit and subsequent percutaneous subxiphoid anterior access using a microneedle puncture after CO2 pericardial insufflation can be achieved reliably and safely.
Access to the pericardial space may be required to undertake a successful ventricular tachycardia (VT) ablation. A percutaneous access method using the now-termed “large bore needle” technique was first described by Sosa et al. (1). Complications, occasionally requiring emergency sternotomy, associated with this approach have limited widespread first-line endoepicardial VT ablation (2). The overall complication rate ranges from 6% to 25%, with a right ventricular (RV) puncture rate of 5% to 17% in reported case series (3–6). Because of this, several articles have been written that aim to identify which patients might best benefit from a first-line endoepicardial VT ablation. Cardiomyopathy type, 12-lead VT morphology characteristics, pre-procedural imaging, endocardial unipolar mapping, and a prior failed endocardial VT ablation are some of the methods used to guide this decision (7–15). Subxiphoid access is inherently associated with RV puncture because the target space is small, not visualizable, and the right ventricle is a mobile structure. Using a direct anterior versus posterior subxiphoid approach may reduce the likelihood of damage to abdominal viscera but not RV perforation, transection, or coronary vessel damage (4). More recently, subxiphoid access using a micropuncture needle has been shown to reduce the incidence of large pericardial effusions, although the RV perforation rate remained the same (16). In the setting of suture ligation of the left atrial appendage, Rogers and Greenbaum have defined a new technique for accessing the epicardial space using an intentional right atrial appendage exit strategy (17,18). The back end of an angioplasty wire is used to exit the right atrial appendage and a microcatheter is passed over the wire into the pericardial space. The pericardial space is then insufflated with carbon dioxide (CO2) allowing fluoroscopic visualization of the separated visceral and parietal pericardial layers with subsequent subxiphoid access, affording minimal risk of RV perforation.
We hypothesized that a modification to this technique using an intentional coronary vein exit strategy would be relatively straightforward and safe. We hypothesized that we would be able to exit a lateral coronary vein reliably using a 0.014-inch angioplasty wire allowing for wire feedback and that upon withdrawal of the microcatheter, there would be no clinically significant bleeding from the coronary vein. We report on the first experience of intentional coronary vein exit with CO2 insufflation of the pericardial space and subxiphoid microneedle puncture for percutaneous epicardial access for VT ablation.
Consecutive patients undergoing VT ablation between January and May 2016 at the Sussex Cardiac Centre for scar-related, anti-arrhythmic–resistant sustained VT or implantable cardioverter defibrillator shocks were invited to participate in the study protocol. All patients provided written informed consent and had a conventional indication for epicardial mapping and ablation. The protocol was approved by the institutional review board. Procedures were undertaken under general anesthesia with 600 mg of intra-venous teicoplanin antibiotic prophylaxis at induction. When possible, anti-arrhythmic medication was withdrawn at least 5 half-lives before the procedure.
Intentional coronary vein exit method
Double right femoral venous and single right femoral arterial access was initially obtained using 8-F sheaths. The arterial sheath was used for invasive arterial pressure monitoring. A medium or large curl 71 cm Agilis NxT steerable introducer (St. Jude Medical, St. Paul, Minnesota) was then placed in the region of the coronary sinus (CS) ostium (Figure 1). An ablation catheter (Thermocool Smarttouch Surround Flow D-F curve, Biosense Webster, South Diamond Bar, California, or Tacticath Quartz F curve, St. Jude Medical) was then used within the Agilis introducer to advance the Agilis introducer into the CS. After withdrawal of the ablation catheter, fluoroscopic CS venography was undertaken in left and right anterior oblique views using iodinated contrast injection. A lateral CS branch or, if absent, an anterolateral branch, was then identified for intentional perforation and exit. A 5- or 6-F JR4 diagnostic coronary catheter was used to subselect the target venous branch. A high tip load 0.014-inch angioplasty wire (Asahi Confianza Pro 12 or MiracleBros 12, Abbott Vascular, Santa Clara, California) was then used within the JR4 catheter to perforate and exit a small distal lateral vein (Online Video 1). Once the wire was seen to loop around the cardiac silhouette, either an over-the-wire balloon, without inflation (Ryujin Plus 1.25/1.5 mm, Terumo Interventional Systems, Somerset, New Jersey [2.5-F distal shaft diameter]) or a microcatheter (Asahi Corsair [2.6-F distal shaft diameter]) was then passed over the angioplasty wire into the pericardial space. The angioplasty wire was then withdrawn and a 2- to 5-ml contrast injection performed to confirm pericardial positioning in this series, although this is not an essential step for this technique to be undertaken safely if the wire is seen to loop around the cardiac silhouette.
A CO2 tank was connected using sterile tubing to a 50-ml Luer-Lok syringe and filled at a flow rate of 1 l/min. Up to 150 ml of pure filtered CO2 (1.5 ml/kg) was then injected via the microcatheter or over-the-wire balloon, using the hand-held syringe method, into the pericardial space observing the insufflation with left lateral fluoroscopy (19). CO2 was used as it has the unique properties of high solubility, low viscosity, and buoyancy (19). Invasive arterial pressure monitoring was used to prevent capno-tamponade. Once sufficient space was visible, the pericardium was then punctured anteriorly using a xiphisternal approach with a Quincke type 12-cm 22-G spinal needle (Vygon, Écouen, France). The initial 2 cases were undertaken using a conventional 18-G Touhy needle (marketed by multiple manufacturers); however, we changed to the aforementioned microneedle when we observed rapid CO2 loss and pericardial deflation upon initial subxiphoid puncture. The microneedle was pre-loaded with a soft-tipped hydrophilic extra support 0.014-inch angioplasty wire (High Torque Whisper ES, Abbott Vascular), allowing both rapid access and prevention of CO2 loss. After tenting, the needle can be visualized entering the pericardial space and the angioplasty wire is advanced and looped in the pericardial space (Online Video 2). A 5-F radial artery sheath (Glidesheath Slender, Terumo Interventional Systems, Somerset, New Jersey) is then passed over the angioplasty wire into the pericardial space. Once the radial sheath dilator is withdrawn, this is “double-wired” with the 0.032-in. Agilis Epi steerable introducer wire (St. Jude Medical). After the 0.032-inch wire is looped in the pericardial space the angioplasty wire is withdrawn and the Agilis Epi steerable introducer is advanced into the pericardial space. Aspiration is then undertaken to remove CO2 and any blood in the pericardial space and connected to a pre-vacuumed closed wound drainage system once irrigated mapping and ablation is undertaken. The microcatheter or over-the-wire balloon is withdrawn after successful epicardial access and the Agilis NxT steerable introducer is either used for RV endocardial mapping with right-sided cardiomyopathy or for subsequent transseptal left ventricular (LV) access with left-sided cardiomyopathy. In the case of transseptal access, this is undertaken under transesophageal guidance with subsequent heparinization targeting an activated clotting time of 350 s.
At the end of the VT ablation procedure, 2 mg/kg of triamcinolone acetate was administered into the pericardial space to reduce the risk of pericarditis and subsequent pericardial adhesion formation; the steerable sheath was then immediately removed without placement of a pigtail catheter (20). Protamine was administered to reverse heparinization and all the other sheaths removed. A total of 1.5 mg/kg enoxaparin, if rewarfarinization was required, or 20 mg rivaroxaban was given 4 h later for left-sided endocardial ablation procedures.
All patients underwent high-density automated epicardial mapping (Confidense, Carto 3, Biosense Webster; or Automap, Ensite Precision, St Jude Medical) using a multipolar mapping catheter (Pentaray Nav or Decanav, Biosense Webster, or Livewire duo-decapolar, St. Jude Medical). If indicated, epicardial ablation was undertaken at 30 to 40 W, LV endocardial at 40 to 50 W, and RV endocardial/aortic cusp at 30 to 40 W. Ablation was directed at achieving complete substrate abolition and noninducibility from at least 2 sites with up to 4 extrastimuli down to a 200 ms coupling interval or ventricular refractoriness (21).
Data were collected and analyzed using Excel, version 2010 (Microsoft, Redmond, Washington). Data are presented as mean ± SD for normally distributed data, otherwise as median (range).
Twelve patients had attempted pericardial access using this technique, of which 11 were successful between January and May 2016 (Table 1). The median age was 65 (range: 37 to 81) years, with a male preponderance (9 males, 3 females). Diagnoses were ischemic cardiomyopathy in 5 patients, suspected or confirmed arrhythmogenic RV cardiomyopathy in 4 patients, and dilated cardiomyopathy in 3 patients. One patient was on aspirin, 1 patient on aspirin and rivaroxaban, 1 patient on clopidogrel and warfarin, 2 patients on rivaroxaban alone, and 2 patients on warfarin alone. Four patients were not anticoagulated. All patients on pre-procedural anticoagulation had this withdrawn pre-procedurally such that they were not anticoagulated at the moment of epicardial puncture. LV ejection fraction ranged from 15% to 61%.
Intentional coronary vein exit was successful in all 12 patients. We were unable to obtain subxiphoid pericardial access in 1 patient because of pericardial adhesions. In this patient, multiple different venous exit points were assessed, all giving a characteristic “disc-shaped” dye hold up indicative of widespread LV pericardial adhesions. This was confirmed post-procedurally with abnormal pericardial thickening and nodularity seen around the left ventricle on computed tomography scanning. Because it was apparent that epicardial access would not be feasible without surgical adhesiolysis, an endocardial-only ablation was undertaken in this patient (Figure 2).
In the other 11 patients, the mean volume of CO2 insufflated was 143.3 ± 16.1 ml. There was no significant change in heart rate (74 ± 15 beats/min vs. 76 ± 17 beats/min; p = 0.45) or systolic blood pressure (108 ± 13 mm Hg vs. 108 ± 11 mm Hg; p = 0.76) before and after CO2 insufflation. During our early experience, we encountered intramyocardial wire tracking (Figure 3). Later, it became apparent that the angioplasty wire is felt to “pop” out the vein and should immediately and easily track around the cardiac silhouette. If the wire tip is seen to curl, then this is indicative of an intramyocardial course and it should be withdrawn and advanced in a different direction, without the need for dye injection. The initial 2 patients underwent subxiphoid access using a conventional 18-G Tuohy needle. Rapid pericardial deflation was observed with loss of subxiphoid pericardial access in 1 patient requiring repuncture (Figure 4). During deflation, the right ventricle was seen to recoil toward the Tuohy needle, making this a potentially dangerous phenomenon, although, in these 2 cases, the 0.032-inch wire was advanced before RV contact with the Tuohy needle. Subxiphoid access was rapid and straightforward in the 9 cases using a pre-loaded microneedle. In all cases, tenting and puncture of the parietal pericardium was visualized fluoroscopically without the need for contrast indicating the correct moment to advance the angioplasty wire. The mean time taken from CS access with the deflectable sheath to subxiphoid pericardial access was 28 ± 12 min within a median procedure time was 310 min (range 190 to 380 min). Initial subxiphoid access to the pericardium revealed either straw-colored fluid or only CO2 gas in all patients. Upon withdrawal of the coronary vein microcatheter or over-the-wire balloon, a small amount of short-lived bleeding (<30 ml) was seen in all patients. In no patient (n = 8) did this prevent transseptal access and subsequent heparinization to a target ACT of 350 s. No rebleeding was seen after heparinization. In these patients, a heparin bolus was administered about 10 min after pericardial access and coronary vein microcatheter withdrawal upon obtaining left atrial access.
In 2 patients, an anterolateral coronary venous exit was used. In 1 patient, because of no lateral branches, and in another as the only lateral branch had a very tortuous ostium that could not be subselected (Figures 5A and 5B).
Seven patients had pre-existing biventricular defibrillators with coronary venous LV leads. There were no lead displacements or changes to LV lead parameters in any patient. In 1 patient, no suitable lateral or anterolateral branches were seen on CS venography and CS exit was therefore undertaken alongside the LV lead, which was used as a road map to guide coronary vein exit (Figure 5C).
Hemopericardium without hemodynamic compromise was observed on one occasion. A total of 150 ml of blood was aspirated after pericardial Agilis Epi introducer insertion. There was no active bleeding, and it was not clear whether there had been a prior bleed from a recent defibrillator extraction procedure or whether there had been bleeding from catheter manipulation within the body of the CS that had already stopped bleeding. In this case, transseptal access was still undertaken with subsequent heparinization and no active bleeding was seen. One patient had clinical pericarditis that was successfully treated with nonsteroidal anti-inflammatory drugs and colchicine. There were no other epicardial access or procedural complications.
Epicardial substrate ablation was undertaken in 9 of the 11 patients. Noninducibility was achieved in 11 of the 12 patients.
To our knowledge, this is the first cardiac procedure using intentional coronary vein exit. This was feasible in all 12 patients without any clinically significant complications. Therefore, it is possible to make a deliberate small coronary vein perforation without ongoing bleeding despite anticoagulation. This result was expected and is in keeping with the work by Greenbaum and Rogers, who undertook the first study on intentional cardiac exit, via the right atrial appendage, in the setting of left atrial appendage ligation (18). There are also anecdotes of wire exit during cardiac resynchronization implantation without clinical sequelae.
It seems clear from this work and that of Greenbaum and Rogers that pericardial CO2 insufflation virtually eliminates the risk of inadvertent coronary vessel and RV puncture as the target for puncture is both large and directly visualizable (18). This, in combination with an anterior microneedle puncture, also reduces the risk of infradiaphragmatic trauma. The downside of this technique is that it adds about 15 min to the procedure and requires an extra technical step: coronary vein exit. Once the coronary vein exit has been performed, however, the subxiphoid puncture is then straightforward. That this is our first experience, and that we managed to exit a lateral/anterolateral coronary vein successfully in all 12 patients, suggests that the technique is not especially challenging. This will be particularly true for electrophysiologists who are familiar with both VT ablation and cardiac resynchronization device implantation. Furthermore, we believe that with increased experience and further refinement, our access times will be reduced. The main delays have been attributed to persisting with lateral branch subselection because we were keen to avoid approaching the LV summit or anterior interventricular groove because the potential for damage to the left main stem or left anterior descending coronary arteries with a stiff angioplasty wire after coronary vein exit. Whether this risk is real or only perceived is as yet unknown. In our later experience, we moved more quickly to an anterolateral branch if a suitable lateral branch could not be found. If exiting at any point within the coronary venous system is safe, then the technique will clearly be more efficient.
Apart from exiting the coronary vein rather than the right atrial appendage, another difference between these 2 techniques is that we have used the front end rather than the back end of an angioplasty wire. To achieve this, we chose a high tip load wire capable of vessel perforation. Very little curve, if any, is required at the end of the wire because if it is important that the wire does not loop, which will keep it intraluminal. The most important feedback is given by the tip curling when being directed intramyocardially, at which point the wire should be withdrawn and exited elsewhere. Once the wire is in the epicardial space, it should track freely around the cardiac silhouette confirming correct placement.
Another potential advantage of coronary vein exit over right atrial appendage exit is that it allows an invasive assessment of LV pericardial adhesions before subxiphoid access. This aspect is inherently more relevant to VT ablation than left atrial appendage ligation. Although we only had 1 patient in this small series with pericardial adhesions, we were able to reliably perforate out the coronary venous system in various locations and inject contrast to assess for local adhesions. Pericardial adhesions at the point of subxiphoid puncture are also known to increase the risk of direct RV perforation because the potential space is lost, perhaps conveying benefit over the microneedle technique as well.
For financial reasons alone, we have switched from using a microcatheter to an over-the-wire balloon for CO2 insufflation. It is probably also important to observe the feedback given by these devices when exiting the cardiac vein over the angioplasty wire. The device should pass relatively easily into the pericardial space and if not it may be that the exit is not “clean” and it is then sensible to exit elsewhere with the angioplasty wire to reduce the risk of transecting other structures.
This feasibility study allows for a more rigorous multicenter registry to be undertaken in which this technique can be compared with other epicardial access methods. It is hoped that with safer pericardial access more first-line epicardial VT ablation procedures along with other epicardial interventions may be undertaken.
This is a single-center case series; a larger multicentre registry will be required to confirm both the safety and efficacy of the presented technique. The theoretical risk of raised defibrillation thresholds during CO2 insufflation has been raised by Greenbaum and Rogers (18). In keeping with their study, we also found that the time from CO2 insufflation to pericardial access is <1 min, thus making prolonged cardiac arrest in the event of failed internal and external shock unlikely. We did not find the procedure proarrhythmic in any patient despite their adverse arrhythmia history. The technique is more complex and time-consuming than the standard subxiphoid approach. Nonetheless, when undertaken by an electrophysiologist also trained in cardiac resynchronization implantation, it remains relatively straightforward and takes approximately the same time as vascular and transseptal access combined.
We report the first human transcoronary vein exit procedure. Coronary vein exit and subsequent percutaneous subxiphoid anterior access using a microneedle puncture after CO2 pericardial insufflation can be achieved reliably and safely. The use of this novel technique to safely access the pericardial space safely has the potential to expand the indications for first-line epicardial access for VT ablation, affording a more comprehensive mapping and ablation procedure.
COMPETENCY IN MEDICAL KNOWLEDGE: Intentional coronary vein exit and subsequent subxiphoid anterior pericardial microneedle puncture after pericardial CO2 insufflation is technically feasible and appears relatively safe in the setting of ventricular mapping and ablation. Multicenter studies will be necessary to compare this approach to the intentional right atrial exit and microneedle approaches.
TRANSLATIONAL OUTLOOK: This initial experience is aimed at devising safer approaches to epicardial access than the conventional large bore and microneedle techniques that risk inadvertent RV puncture. Possible uses include cardiac ablation for other arrhythmias, annuloplasty, left atrial appendage ligation, and percutaneous drainage of traditionally inaccessible posterior effusions.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- coronary sinus
- left ventricle
- right ventricle
- ventricular tachycardia
- Received June 6, 2016.
- Revision received October 24, 2016.
- Accepted November 3, 2016.
- 2017 American College of Cardiology Foundation
- Della-Bella P.,
- Brugada J.,
- Zeppenfeld K.,
- et al.
- Koruth J.S.,
- Aryana A.,
- Dukkipati S.R.,
- et al.
- Sacher F.,
- Roberts-Thomson K.,
- Maury P.,
- et al.
- Acosta J.,
- Fernández-Armenta J.,
- Penela D.,
- et al.
- Cano O.,
- Hutchinson M.,
- Lin D.,
- et al.
- Soejima K.,
- Stevenson W.G.,
- Sapp J.L.,
- Selwyn A.P.,
- Couper G.,
- Epstein L.M.
- Berruezo A.,
- Mont L.,
- Nava S.,
- Chueca E.,
- Bartholomay E.,
- Brugada J.
- Vallès E.,
- Bazan V.,
- Marchlinski F.E.
- Wijnmaalen A.P.,
- van der Geest R.J.,
- van Huls van Taxis C.F.B.,
- et al.
- Codreanu A.,
- Odille F.,
- Aliot E.,
- et al.
- Andreu D.,
- Berruezo A.,
- Ortiz-Pérez J.T.,
- et al.
- Gunda S.,
- Reddy M.,
- Pillarisetti J.,
- et al.
- Greenbaum A.B.,
- Rogers T.,
- Paone G.,
- et al.
- Cho K.J.
- Dyrda K.,
- Piers S.R.D.,
- van Huls van Taxis C.F.,
- Schalij M.J.,
- Zeppenfeld K.
- Silberbauer J.,
- Oloriz T.,
- Maccabelli G.,
- et al.