Scientific Advisory Committee on Medical Devices used in Cardiovascular Systems - Record of Proceedings – November 24, 2017
Committee Members Present: John Ducas (Chair), Renzo Cecere, Marino Labinaz, Alan Menkis, Brent Mitchell, Joaquim Miró, Raymond Yee
Regrets: Eric Cohen, Christopher Feindel, Barry Rubin, John Webb
Invited Guests: Mathieu Bernier, Yaariv Khaykin, Ratika Parkash, Marcel Ruzicka
Health Canada Representatives:
Office of Science: Hripsime Shahbazian
Medical Devices Bureau (MDB, TPD): Kevin Day, Ian Aldous, Marianne Ariganello, Ben Elliot, Ian Glasgow, Jianming Hao, Karen Kennedy, Mark Korchinski, James McGarrity, Roy Masters, Catherine Milley, Chris Schmidt, Maurice Sylvain, Lanyi Xu, Marc Lamoureux, Amanda Jones
Bureau of Cardiology, Allergy and Neurological Sciences (BCANS, TPD): Mick Gelsema, Timao Li
Regulatory Operations and Regions Branch (RORB): Barbara Harrison
Marketed Pharmaceuticals and Medical Devices Bureau (MHPD): Patrick Fandja, Hui Zheng, Shirley Chou, Elaine Wong
Abbreviations used in this record:
- six minute walk test
- ambulatory blood pressure monitoring
- acute coronary syndrome
- acute decompensated heart failure
- atrial fibrillation
- atrial flutter
- accessory pathway
- atrial septal defect
- biventricular assist devices
- B-type natriuretic peptide
- blood pressure
- below the knee
- bridge to transplant
- congestive heart failure
- cardiac implantable electrical device
- complex fractionated (atrial) electrogram
- contact force sensing
- circumferential pulmonary vein ablation
- cardiac resynchronization therapy
- computerized tomography
- cavo-tricuspid isthmus
- drug coated balloons
- Data and Safety Monitoring Board
- ectopic atrial rhythm
- emergency department
- ejection fraction
- emergency room
- glomerular filtration rate
- estimated glomerular filtration rate
- guideline-directed medical therapy
- heart failure
- heart failure with preserved ejection fraction
- heart failure with reduced ejection fraction
- interatrial shunt device
- implantable cardioverter defibrillator
- instructions for use
- Investigational Testing Authorization
- Interagency Registry for Mechanically Assisted Circulatory Support
- Kansas City cardiomyopathy questionnaire
- left atrium
- left atrial pressure
- left ventricular assist device
- left ventricular ejection fraction
- left ventricle
- major cardiovascular and cerebrovascular event
- major adverse cardiovascular or neurological event
- medical devices bureau
- Minnesota living with heart failure questionnaire
- magnetic resonance
- magnetic resonance imaging
- muscle sympathetic nervous system activity
- normal sinus rhythm
- National Heart, Lung, and Blood Institute
- New York Heart Association
- optimal medical management
- percutaneous coronary intervention
- pulmonary capillary wedge pressure
- pacing capture threshold
- percutaneous transluminal angioplasty
- pulmonary vein
- premature ventricular contraction
- pulmonary vein isolation
- quality of life
- pulmonary-to-systemic flow ratio
- right atrium
- renal artery stenosis
- renal blood flow
- remote catheter System
- randomized clinical trial
- renal denervation
- right-sided heart catheterization
- remote monitoring
- receiver operating characteristic (analysis)
- right ventricle
- right ventricular assist device
- right ventricular failure
- right ventricular myocardial infarction
- ventricular rate or response
- Scientific Advisory Committee on Medical Devices Used in the Cardiovascular System
- serious adverse events
- special access program
- systolic blood pressure
- Standard Deviation
- transcatheter aortic valve replacement
- total artificial heart
- transesophageal echocardiography
- transient ischemic attack
- transcatheter mitral valve replacement
- Therapeutic Products Directorate
- ventricular tachycardia
- ventricular assist devices
1. Opening Remarks & Welcome
Dr. John Ducas, Committee Chair
Dr. Ducas opened the meeting, welcomed the committee members and guest speakers. He invited Dr. Pat Stewart, the current Director General of the Therapeutic Products Directorate (TPD), to provide opening remarks. Dr. Stewart introduced himself, thanked the committee members and guest speakers for sharing their time and expertise and acknowledged the efforts all participants had made to attend the meeting. He noted that the Government is moving forward with number of initiatives to try and improve access and appropriate use of health products. In medical device area Health Canada is looking at innovative health technology, artificial intelligence, home monitoring. He provided updates regarding International Medical Device Regulators Forum (IMDRF) and discussions regarding relying more heavily on Real World Data to inform regulatory decisions. He noted the importance of the Scientific Advisory Committee on Medical Devices Used in the Cardiovascular System (SAC-MDUCS) to Health Canada's regulatory review and decision-making process and acknowledged the committee on its continued support. He concluded by wishing the committee members and participants a productive meeting.
2. Review of the Agenda, Affiliations and Interests Declarations, and Confidentiality Agreement
Dr. John Ducas, Committee Chair
The Chair reviewed the agenda items with the committee. The agenda was accepted with minor adjustments to accommodate last minute changes. Question 5: Treatment of Unprotected Left Main Coronary Artery Stenosis has been postponed to the next meeting. Instead a new topic was added to the agenda to address the use of Cardiac Implantable Electrical Devices (CIEDs) used in an MRI setting that was originally submitted for e-consultation. Members were reminded to disclose any conflicts prior discussion and as the meeting proceeds.
He invited Hripsime Shahbazian to address the next topic.
3. Review of Terms of Reference, Membership and members Affiliations and Interests
Hripsime Shahbazian, Office of Science, TPD
Hripsime Shahbazian informed the members that, as discussed at the last meeting in 2016, Health Canada conducted a membership renewal for this committee. The revised Membership List and Summary of Expertise, Experience and Affiliations and Interests for SAC-MDUCS are now posted on Health Canada Website.
She noted that further to discussion at the last meeting the Terms of Reference (ToR) for this committee was also revised to include the following statement:
"Before each committee meeting the chair will ask the members/invited experts to provide their consent to the recording of their views and opinions in the Record of Proceedings (RoP) and to the publication of the RoP on Health Canada's Website."
In addition, the revised ToR now includes a requirement to select a Vice-chair, who at the request of the Chair, or in the case of the Chair's absence, will chair the meeting(s) and take on the Chair's other tasks, as required.
The revised ToR was reviewed and officially accepted by the members on February 22, 2017 and is now posted on Health Canada Website.
In closing Hripsime Shahbazian congratulated the committee members for the Committee's 15th anniversary. She noted that some of the members have served on the committee since its inception, including Dr. Raymond Yee (past chair), Dr. John Ducas (current chair), Dr. Joaquim Miró and Dr. L. Brent Mitchell as well as two of the ad hoc members: Dr. Stephen Lownie and Dr. Chris Simpson.
She thanked all members and presenters for their active participation and contribution to the advisory process.
4. Summary of How Information from Previous Meeting Is Being Used
Kevin Day, Medical Devices Bureau, TPD
Mr. Kevin Day presented a brief summary of how previous advice and recommendations provided by the committee have been considered by Health Canada.
The following key discussion topics and their impact were noted:
CRT Response and Multi-Point Pacing have been very active:
- Working with recommendation to limit use to 'failed' CRT
- From previous SAC-MDUCS discussions, battery impact being included in IFU
- Allowing access to algorithms where there does not appear to be a safety concern
- Some on-going monitoring taking place
Drug Coated Balloons (DCBs) continue to be an active area:
- Good to assess superior effectiveness compared to PTA
- Good for reference that not all DCB are the same
- Additional attention given to Below the Knee (BTK) and Dialysis indications
- Challenge with Class II indications
Transcatheter Mitral Valve Replacement (TMVR) is an active area in terms of ITA and SAP:
- Better understanding of patient need and potential patient population to be treated
- Very good context for the emerging technologies
Transcatheter Aortic Valve Replacement (TAVR):
- Good context for upcoming trials in lower risk patient populations
- As access routes continue to evolve, this poses challenges in labelling and indications for use
- Transfemoral continues to be best
- Apical, direct Aortic, subclavian, axillary, vena cava, etc.?
- Note that most valves will be fine for 5-years
- It is both helpful and unsettling
- Support for issuing licences with longer term conditions
- Insights into MR Conditional Cardiac Implantable Electronic Devices and limits on PCT and MR impact
- What does MR conditional mean?
- He noted that the committee will have an opportunity to discuss this topic today.
- What does MR conditional mean?
Building on previous discussions the following areas were noted as well:
- Intra-atrial Shunting
- Renal Denervation
- Ventricular Assist Devices
- Cardiac Ablation and treatment of AF
Mr. Day thanked members for their ongoing dedication to the committee.
5. Update on Remote (Robotic) Navigation for Ablation and Other Aspects of the Treatment of AF (and Other Arrhythmias)
Invited Speaker: Dr. Yaariv Khaykin
Dr. Khaykin introduced himself, disclosed his affiliations and proceeded to address the questions posed by Health Canada.
OBJECTIVE: There continue to be numerous licence amendments to add new features to complex mapping and robotic navigation systems for ablation. At the same time, there is some clinical data coming out that suggests that outcomes associated with the use of these technologies, and their new features, is not providing a clear benefit to patients; certainly the data is not consistent. Health Canada would like an update to better understand the current challenges with these technologies, in the context of the current state of complex ablation (e.g. AF). This update is intended to provide us with the most current clinical viewpoint in order to best assess the safety and effectiveness of the technologies and the appropriate claims and indications for use.
1. Specific treatments associated with AF
Provide an update on the treatment of AF in Canada including the indications for use or patients that are treated with catheter ablation, the clinical outcomes achieved for these patients, and the optimization of the therapy given. Additionally, please address the following specific cases:
- There has been some data recently suggesting that identification of rotors for the treatment of AF (Focal Impulse and Rotor Modulation (FIRM) Ablation) is less effective than anticipated. It would be useful to discuss risks and challenges of these systems including:
- Discuss whether cardiac mapping system guided ablation is beneficial for treating patients with persistent or long-standing persistent atrial fibrillation. Contradictory results have come out in recent clinical studies, such as Substrate and Trigger Ablation for Reduction of Atrial Fibrillation (STAR-AFII) study (N Eng J Med 2015;372: 1812-1822), and Conventional Ablation for Atrial Fibrillation with or Without Focal Impulse and Rotor Modulation (CONFIRM) study (J Am Coll Cardiol 2012;60:628-636).
- Discuss how the current systems allow you to identify arrhythmia areas of interest and discuss the rates of clinical success associated with terminating AF when treating these areas.
- For patients with heart failure and Afib, are there special benefits for Afib ablation, such as observed in CASTLE-AF (Catheter Ablation vs. Standard Conventional Treatment in Patients With LV Dysfunction and AF (CASTLE-AF) ClinicalTrials.gov Identifier: NCT00643188)?
- What are the limitations of this study and what patients are most suited for this type of intervention?
2. Robotic Guidance and Mapping Systems
Provide an update on robotic guidance and electro-anatomic mapping systems for catheters for the diagnosis and treatment of arrhythmias. Please include any new clinical evidence related to safety and effectiveness of the technology, novel systems currently under development, and any future trends in this area. Discuss the successes and challenges associated with these systems. Please also address the following:
- Have features introduced in the past, such as semi-automated navigation, contact pressure feedback, image registration and integration with other systems, improved workflow, accuracy, and clinical outcomes for ablation procedures. Are systems now safer and/or more effective?
- What is the best current clinical evidence that has demonstrated the safety and effectiveness of this technology and how are outcomes impacted by the use of the robotic guidance and mapping systems.
Dr. Khaykin started by providing background information on remote (robotic) navigation for ablation and other aspects of the treatment of AF (and other arrhythmias). He explained that fluoroscopy was historically used to visualize simple ablation targets, but because of atrial anatomy complexity fluoroscopic images are inadequate to locate less obvious targets. He also highlighted the concern for radiation exposure for operators.
He identified and briefly discussed the challenges associated with ablation procedure in general:
- While reasonable outcomes are expected and tied to operational procedural volume and experience, the atria are complex anatomical structures, and there are few highly experienced centers in Canada; providing services to patients in remote areas remains difficult,
- There are conflicting theories regarding the etiology of the arrhythmia, as well as appropriate therapeutic targets, particularly in patients with persistent and longstanding persistent AF,
- There remains and inability to directly assess lesion sufficiency so that fibrillation does not return,
- There is a risk of damaging collateral structures,
- Patient and operator exposure to fluoroscopy, and
- Operator disability related to wearing lead aprons.
Dr. Khaykin identified the perceived benefits of complex mapping, such as the ability to:
- Reduce need for fluoroscopy,
- Log special sites and ablation lesions,
- Extrapolate physiological processes using activation maps, and
- Monitor catheter tissue interface to predict whether the lesion is sufficient and not damaging.
He also reviewed some benefits of robotic magnetic navigation:
- Reduction of fluoroscopy radiation exposure for the operator and patients,
- Reduction of MSK strain on the operator,
- Automation of workflow, and
- Possibility to provide access for remote patients (e.g. Stereotaxis system).
Complex mapping has evolved over time. Initially, the goal was catheter visualization in non-fluoroscopic 3D space to create 3D "primary" maps, signal propagation maps, and voltage maps (to identify scar tissue, which is important for ablation of ischemic VT). This approach was also intended to help visualize multiple electrodes with concurrent signal acquisition. Side-by-side image Sync and image reconstruction using pre-acquired CT and MRI images allowed ablation to be carried out on a previously acquired CT image. In 2007, real-time registration (alignment and reconstruction) of ultrasound images replaced the need to register pre-acquired images, and allowed for more precise localization of multiple catheters, as well as integration of tissue contact information to mitigate forces and pressures that could result in perforation or damage to collateral structures, but ensure sufficient force to create the lesion. Furthermore, merging images from multiple modalities enables the user to project ablation points into real life ultrasound images. Other types of derivative maps were created to help predict target ablation areas. These maps show fractionation, rotors, as well as other signal characteristics for morphology pace mapping (e.g. ventricular tachycardia and PVCs which are known to lead to cardiomyopathy in some patients). He acknowledged this type of mapping has been very challenging, but with the development of morphology and 3D maps, patient treatment may be helped.
He described early 3D mapping (early 2000) as slow and tedious as it required point by point system inputs for location interpolation in 3D space. Those early procedures were known to take four to six hours, and because patients were given a lot of fluid in the process, their anatomies could have changed and rendered the maps unusable. More recent is the integration of fluoroscopy, which allows the registration of those images for background mapping. This too has allowed operators to reduce patient exposure to fluoroscopy. Another very recent innovation is contact force imaging with the use of colour coded tags that represent force-time integrals. These tags are automatically generated based on predefined ablation parameters ("objectified lesions").
Next Dr. Khaykin discussed the significance of ablating tissues around the pulmonary veins (PVs), as they are known to generate extra beats. He indicated that electrically isolating PVs and preventing activity from entering the atrium can prevent atrial fibrillation in some patients. He noted that while PV isolation/ablation was once called a "curative" approach, he believes this approach modifies the statistical risk of a patient going out of rhythm.
In response to the question of whether mapping and robotic navigation systems provide clear outcome benefits to patients, Dr. Khaykin presented data demonstrating that procedure time has decreased and efficacy has improved over time.
Early data comparing non-fluoroscopic 3D imaging compared to conventional fluoroscopic guided procedures shows that acute success and complications are not significantly different (Sporton et al, 2004; Khaykin et al, 2012), however there was a significant reduction in fluoroscopy time and radiation dose/time across a number of different clinical conditions.
In Addition there has been an improvement in "case" (procedure) time, from 4 - 6 hours, to 1.5 - 2.5 hours in most centers, and attributed the improvement to ultrasound integration, the use of non-gated mapping, and catheter (tip) irrigation. In addition to shorter procedure time, delivery of targeted energy has also contributed to RF time reduction. Dr. Khaykin estimates an average reduction of 100 fold per case; therefore patient radiation exposure is now expected to be below that received during pacemaker implantation.
Dr. Khaykin noted that non-randomized data from a historical database (~400 PVIs) demonstrate that the introduction of new technologies (e.g. CARTO 3) makes a difference. Christoph et al. (2015) compared the CARTO 3 with the newer CARTO-UNIVU, and showed significant reductions in fluoroscopy time across multiple arrhythmia types with the newer system (AFL, AF, EAT, and VT; but not AP, and PVC).
Contact Force Sensing (CFS)
Dr. Khaykin introduced contact force sensing technology and described the TactiCath ablation catheter that uses light properties to measure pressure/force at the tip in contact with the tissue. The catheter has a 65 or 75 mm deflectable curve, and uses fiber optic technology for optimal stability during contact force measurement. There is risk of perforation if too much force is used.
Dr. Khaykin noted that there are considerable inter-operator differences in the amount of applied force during first time use. Some operators were using between 1 and 38 g, while others were using up to 123 g. These differences make it challenging to compare outcome data from different centers because each operator uses different levels of pressure, therefore randomized studies that compare multiple centers, are actually comparing different operator approaches that are not the same. It was also noted that insufficient force produces inadequate lesions; successful lesions were achieved using ≥ 20 g force (400 gs – force time integral).
He described comparative force data for TactiCath and SmartTouch catheters; however, he noted that it is questionable whether the forces expressed from different systems are measures of absolute pressure. Dr. Khaykin feels these forces are relative to the individual technologies and therefore cannot be used interchangeably between systems.
He presented discussed the TOUCH AF study (Conti et al, 2017) that involved the SmartTouch catheter, and compared CFS guided vs. CFS blinded ablation for persistent AF. The study showed that there were no differences in RF time, or single procedure freedom from arrhythmia at 12 months. However, an older technology was used, only very experienced operators were involved, and no CFS standard had been established. Using data from his center, his group set out to determine the impact of CFS on re-do procedures via ROC analysis to determine predictive value of recoverable lesions. They found a significant improvement in outcomes. He also presented data from a meta-analysis describing multi-center experiences (Zhou et al, 2017) and noted that collectively there is a significant trend towards positive outcomes with CFS ablation. A similar trend was observed for major, and to a lesser extent, minor complaints. In addition, fluoroscopy time and radiation dose were overall statistically lower in CFS ablation procedures.
There was a concern that if CFS was introduced with navigation systems that moves the user away from feeling the contact force of the catheter that may result in increase of perforation risk. There was a question whether there are any factors of successful contact force measurement. Dr. Khaykin indicated that at his center, ablation catheters are used manually (feeling the contact force), as they have no other way. He noted that there are a few centers in Canada that have the robotic and magnetic navigation technology. With respect to perforation risk and in the case of the Stereotaxis system, which can apply a maximum of 10 g per spot, this amount of force, is at the minimum range to create stable lesions. It was noted that 10 g is barely enough to create a stable lesion and this may be one of the reasons why there has not been broad adoption of this technology.
Health Canada asked what level of evidence would be required to support specific targeted levels of force. Dr. Khaykin recommended that manufacturers should suggest contact forces that are within the specifications of their own technology and avoid extrapolating from or referencing other systems. He noted that while it would be nice to establish force levels based on randomized studies, two levels of data collection would be required: retrospective data collection and evaluation to forecast parameters, and then conduct a randomized study. He felt that currently there is enough data to start randomized study to determine whether these systems truly offer an advantage.
There was a question whether there are any animal and human tissue specific data that correlate to the amount of (real or virtual) force used (g) with the completeness of a tissue lesion. Dr. Khaykin indicated that the objective is to link force with lesion completeness, and highlighted the problem of most data coming from bench studies of a perfused thigh muscle, which are very controlled, and not indicative of in vivo conditions.
There was a comment that some element of safety is needed, however the data can't come from humans because of AE fistula risk; if too much pressure is applied to the back wall with too much time it may damage structures. It is important for operators to have that information.
There was a concern that if contact force varies for different areas of the heart (e.g. at the cavotricuspid isthmus), it may not be useful to have recommendations that are not dependent on the anatomical region. Dr. Khaykin agreed that operators first need to collect bench type data. It was recommendation that animal and human data are needed to develop recommendations before conducting a randomized study.
There was a question regarding how large the pressure safety factor is between optimal ablation and adverse events. Dr. Khaykin noted that based on animal data, the manufacturer of the SmartTouch indicates that perforation risk is increased when the user reaches an average input of 120 g, however, it does not get close to that level. The biggest concern with AF ablation is esophageal injury, because the esophagus happens to run right in the area that we are trying to ablate: between the pulmonary veins in the back of the left atrium. Also, because the anatomy of the esophagus and thickness of the atrium vary between people, there's absolutely no way to ablate the pulmonary veins without at some point doing ablation over the esophagus. However, risk of esophageal injury is very low; it's estimated at 1 in 2000. His center hasn't seen many injuries to surrounding structures, particularly with evolving technologies, and the adverse event rates are much lower than they used to be. Nevertheless, we are still concerned about esophageal injuries. He noted that at this time we do not have the technology to predict, prevent or study AEs, as they remain a low risk complication.
It was agreed that ablation force recommendations will have to be location specific since there is a large variation of tissue thickness.
It was noted that while there is a regulatory need to better define and understand the safety of force contact technologies, there is little motivation for manufacturers to run randomized studies at this time.
Magnetic and Robotic Navigation
Dr. Khaykin described the Stereotaxis navigation system that uses a magnetic vector to direct the ablation catheter tip. He noted that this technology has existed since 2000.
Several non-randomized comparisons involving older generation systems show non-significant decrease in procedural, fluoroscopy, or ablation times for magnetic and robotic systems compared to conventional ablation, across various types of arrhythmia.
Dr. Khaykin noted that long term outcomes of navigation for atrial fibrillation in 112 patients were published this year (Yuan et al, 2017). These non-randomized data show some improvement in symptoms, without significant differences in procedure times, or complication rates. A meta-analysis (Proietti et al, 2013) of non-randomized data shows no significant freedom from arrhythmia with remote magnetic navigation, but perhaps a small reduction in complications. Dr. Khaykin highlighted that the greatest amount of force used with remote navigation is 10 g, and that the risk of perforation and esophageal damage is expected to be low.
Dr. Khaykin presented data from a comparison of robotic navigation vs. conventional ablation (Steven, 2010) that show significant reductions in fluoroscopy time, operator radiation exposure and exposure time with robotic systems; however, procedure duration trended towards a significant increase. He also described a worldwide survey of robotic AF ablation (Bai et al, 2012) involving 1728 procedures across 12 centers, which demonstrated an overall complication rate of 4.7% and 67.1% success rate after 18 (±4) months follow-up. In addition, another meta-analysis of non-randomized data by Zhang et al (2014) showed no significant improvement, a trend towards more complications, but reduced fluoroscopy time with robotic systems.
In a randomized comparison between the Hanson and manual SmartTouch (Russo et al. 2016), the use of touch feedback improved success at 1 year and reduced fluoroscopy time, however he did not believe the one year value of 90%. He noted his use of the Hansen system with animals, but found there was no contact force or manual feedback.
Next he described the findings from The Man and Machine Trial (Rillig et al. 2017). In this study a comparison between manual and robotic ablation showed no difference in procedural clinical outcomes at 15 months, but the use of robotic systems was associated with longer procedure times. He highlighted that a fistula was found in one patient treated with the robotic system. This individual had a reddened area in the esophagus without symptoms, clinicians intervened and attempted repair. During continued monitoring in hospital the lesion fistulised and the patient died. Dr. Khaykin noted that lesion is a very challenging complication, one that operators worry about, and that is extremely difficult to predict and treat.
There was a question if there is any new data that indicate an increased risk of pericardial effusion with the Hanson system. Dr. Khaykin indicated he wasn't aware of any new data, but cited The Man and Machine study in asserting there was no difference in pericardial effusions or other complications with that system, apart from the AE fistula, that was an isolated event.
The committee members were asked to comment if it would be sufficient for manufacturers of robotic systems to demonstrate non-inferiority to manual procedures to support regulatory approval. Dr. Khaykin indicated that demonstrating non-inferiority of complication rate was probably more important than non-inferiority of success rate. He added that thesetechnologies have been around for 10-12 years, and recruitment of 200 to 300 patients is achievable to answer these questions.
There was a comment that all atria are different, and these differences dictate catheter "loiter" times required to create lesions; perhaps this would be better achieved by the use of robotics. Dr. Khaykin indicated that AE fistulas are seen with every technology used in the atrium, including cryoablation, laser ablation, andRF balloons. Therefore, every type of energy has resulted in unpredictable fistula formation. He agreed that it would be great to come up with a set-it-and-forget-it-formula, but we're not quite there yet.
Next Dr. Khaykin described AMIGO Remote Catheter System (RCS), which uses modern technology and consists of two components, the Amigo RCS and the Amigo remote controller.
Other Remote Navigation Technologies were mentioned:
- MRI driven catheter manipulation
- Stereotactic arrhythmia radio ablation (STAR)-using much higher radiation dose
He showed a screenshot of the CyberKnife software in which a pre-acquired MRI image is used to target ablation. He indicated it delivers a much higher radiation dose as compared to fluoroscopy, but the radiation is used to create contiguous lesions.
Benefits of cardiac mapping system guided ablation for patients with persistent or long-standing persistent AF
Dr. Khaykin discussed mechanisms of atrial fibrillation and reiterated that ablation began with PVI, because of the firing that occurs there, and it is known that some occurs outside of the PVs. Also, chronic fibrillation is described in terms of multiple wavelets where there are rotors for entry, and parasympathetic ganglia heavily innervate all around the back of the atrium, and right around the pulmonary veins. The typical approach with PV isolation presumably works to prevent and insulate those extra beats coming from the pulmonary veins. Other techniques involve: ablating across the roof or floor of the atria to isolate the back wall, which is histologically part of the pulmonary vein antrum; and ablating the mitral isthmus line (a number of these patients develop a flutter around the mitral annulus so that the line from the left anterior pulmonary vein to the mitral annulus is considered the 'classical' mitral isthmus line).
Dr. Khaykin presented studies for ablation of long-standing persistent AF. A number of studies have reviewed combinations of lines. Earley et al. (2006) reported that 42 patients with permanent AF were treated with CPVA with roof line and CTI line, and 52% of patients required a second procedure, but up to 74% success was reported with cumulative procedures. In another study (Oral et al. 2006) 146 patients randomized to amiodarone + cardioversion with or without CPVA. The proportion of patients treated with CPVA that achieved NSR improved at each successive time point as compared to patients that did not receive CPVA.
Other ablation strategies address areas outside of the PVs. In particular, areas of "fractionation" or unusual electrical activity have been targeted, as these areas are thought to be important in maintaining atrial fibrillation. Nademanee et al. (2004) treated 64 patients with chronic AF by targeting CFAEs; patients were ablated until 95% converted to a sinus rhythm (20% had received ibutilide which likely assisted with conversion). The major complication rate was 5%, and redo rate was 30%, which he considered acceptable for this population.
Oral et al. (2009) treated 119 patients using incremental left atrial complex fractionated atrial electrograms (CFAE) ablation, and reported 79% success with AF termination during APVI, and 36% with cardioversion. In a comparison of biatrial vs. left atrial ablation, comparable to a maze procedure, patients that underwent biatrial ablation did much better, although procedural times were longer, number of complications was greater and so was fluoroscopy exposure.
Dr. Khaykin noted that this is not feasible in all patients. He indicated that incremental ablation was better than incremental biatrial ablation. He explained the hierarchical approach begins with PV isolation, and from there a roof line may be added, and if a patients is still in AF, they ablate a line to the left atrial appendage, and may add a floor line and then isthmus line. After that other sites are explored. The data demonstrate that as operators move incrementally through these targets, the cumulative incidence of AF termination increases (Haissaguerre et al. 2005). The study only involved 60 patients and 87% had their AF terminate during ablation, and overall success at 11 months was 95%. However, Dr. Khaykin noted that his early experience with AF ablation in 2004 and 2009 (Khaykin et al. 2010) involved PVI followed by CFAE ablation in patients who were still in AF, and the results were less than expected; he reported a 44% success rate at one year following the first procedure.
The STAR AF trial (substrate versus trigger ablation for reducing atrial fibrillation: a randomized multicenter trial) randomized 100 high-burden paroxysmal/persistent patients to PVI, CFE, or hybrid PVI+CFE ablation (using proprietary St. Jude algorithm) (Verma et al. 2010). The hybrid combined PVI+CFE ablation approach resulted in statistically superior freedom from AF. Similarly, several studies have demonstrated the superiority of stepwise and combination PVI and CFAE approaches (Brooks et al. 2010).
As part of the larger STAR AF II trial, 589 patients were randomized to PVI, PVI+CFE, and PVI + linear roof and mitral valve isthmus line ablation. Results were surprising: patients treated with PVI alone did marginally better than the other approaches at 18 months, and data show no apparent benefit of adding CFAE and linear ablations. However, Dr. Khaykin cautioned that these findings should be interpreted in the context of the navigation system(s) used: the addition of CFE with the proprietary algorithm was not helpful, and no contact force information was available to operators.
Arrhythmia areas of interest
Dr. Khaykin indicated that when considering specific abnormal electrograms, it is unclear whether they identify areas important for maintaining atrial fibrillation, or whether they represent wave front collisions. He noted that there are a number of mapping systems that are focussed on time and frequency domains, as well as proprietary measures that seek to identify those abnormal points such as: CARTO, EnSite, Rhythmia, Topera and Acutis and there are a number of algorithms available as part of these systems that attempt to interpret these electrograms in the time domain. Dr. Khaykin does not believe these algorithms are able to pinpoint the most important areas responsible for the arrhythmia maintenance, and cited the inability to reproduce results of identification of abnormal rhythms. He briefly described the CARTO, Rhythmia and EnSite Precision systems.
He discussed the non-randomized CONFIRM study (Narayan et al. 2012/2014) that included 107 patients and showed that patients treated with FIRM+PVI ablation vs PVI alone did better over time with 56% AF termination, and the study extension showed patient successes were durable. He indicated there are no randomized studies at present, but several registries, the largest group of which showed 59% of patients exhibited AF termination; however, patients were mixed with respect to previous procedures (PVI). Interestingly, the authors reported there was no association between AF termination and eventual actual freedom from fibrillation. He suggested that we need to wait for proper randomized studies (REAFFIRM and REDO-FIRM) to determine whether this algorithm can be relied upon.
He described the use of 'basket' technology (US MEDICAL) that incorporates ultrasound crystals and 48 electrodes, and is intended to reconstruct the 3D anatomy of the chamber from beat to beat in real time, and projects activation. Patients with persistent or long-standing persistent AF at 6 sites were treated with an algorithm intended to identify local firing and rotors. They found multiple underlying mechanisms; on average the algorithm identified 6.2 mechanisms per patient.
Atrial fibrillation and heart failure
Dr. Khaykin described the CASTLE-AF trial that enrolled 397 patients with AF, CHF and EF < 35% that were randomized to ablation or medical therapy (Marrouche et al. 2017). The primary outcome (mortality and HF related hospitalization) and each individual endpoint were significantly reduced in patients treated with ablation. Similarly, the Ablation vs. Amiodarone for Treatment of Atrial Fibrillation in Patients With Congestive Heart Failure and Implanted ICD/CRTD – (AATAC) Trial (Di Biase et al. 2016) showed that significantly more patients with AF and CRT-D achieved freedom from AF following ablation, than those that did not. Also, CHF related hospitalization was 31% vs. 58% (P>0.001), and mortality was 8% vs 18% (P=0.037) in favour of ablation. Also, as part of an older study- Pulmonary Vein Antrum Isolation versus AV Node Ablation with Bi-Ventricular Pacing for Treatment of Atrial Fibrillation in Patients with Congestive Heart Failure (PABA-CHF) (Khan et al. 2008) 81 patients with AF, NYHA II to III, and EF <40% were randomized to PVI vs. AV node ablation and biventricular pacing. Patients treated with PVI showed significant improvement in QoL questionnaire, 6MWT, and ejection fraction at 6 months. He also presented the findings of a meta-analysis (Zhu et al, 2016) that excluded the CASTLE-AF or AATAC data, but also confirmed significant improvement in EF and QoL in ablated CHF patients. He added that his experiences ablating CHF patients with defibrillators were positive, despite lower success rates, and the need for multiple ablations.
There was a question whether the presence or absence of anticoagulation was critical. Dr. Khaykin indicated he does not know whether ablation lowers the risk of stroke; however, anticoagulation increases the risk of post-procedural bleeding, and that patients do better without anticoagulation. He acknowledged that the current guidelines suggest ablation should be ignored and anticoagulated as if the patient has never had an ablation. He also stated there were no temporal associations between AF and stroke, and that lots of work needs to be done in this area.
In conclusion he noted that:
- There are multiple technologies on the market that allow for the precise reconstruction of the anatomy with secondary maps (activation, voltage, fractionation), but their usefulness hast not been validated; he believes further research is necessary.
- Contact sensing technology has already shown improved outcomes, with the caveat that individual technologies are not comparable.
- There is lack of consensus on the best approach to treat persistent and long standing persistent AF. Ablation is not curative, but simply reduces the statistical probability of an event. Clinical reductions in symptoms and event frequency should be considered wins.
- There is evidence to consider AF ablation is superior to standard treatment in patients with CHF and AF.
- Current magnetic navigation technologies are expected to produce similar results to manual catheter manipulation, but offer reduced fluoroscopy exposure.
- Current robotic navigation technologies produce similar clinical outcomes, with reduced patient and operator fluoroscopy exposure, but at the cost of longer procedures.
Dr. Khaykin concluded his presentation.
Dr. Cecere arrived to the meeting.
6. Update for the Renal Denervation Systems for the Treatment of Hypertension
Speaker: Dr. Marcel Ruzicka
Dr. Ruzicka introduced himself, disclosed his affiliations and proceeded to address the questions posed by Health Canada.
OBJECTIVES: In March 2012, based on data from HTN-1 and HTN-2, Health Canada licenced the Symplicity Renal Denervation System in Canada. Prior to licensing, applications had also been received by Health Canada under the Special Access Program. Following the results of HTN-3 and the lack of effectiveness that was observed, Renal Denervation was not actively performed and the originally licensed Symplicity system licence was cancelled. At our meeting on 2014-11-28, Dr. Raj Padwal discussed the outcomes and the limitations that were seen in the testing of the renal denervation systems, including data from HTN-3. As there is now renewed clinical activity associated with Renal Denervation, Health Canada is requesting input and recommendations on how the assessment of new systems should be considered in the context of applications for Clinical Trials and Licences.
More than three years have now past since publication of the SYMPLICITY HTN-3 trial (Renal Denervation in Patients with Resistant/Uncontrolled Hypertension) results in 2014. A number of weaknesses in the trial data, as well as in the technology and its use have been discussed. Based on the lessons learned from the clinical testing of the Symplicity Renal Denervation System, please discuss the following:
- Provide a brief overview of the current state of development and testing or use of renal denervation systems, including the Symplicity Spyral and EnligHTN systems.
- Based on the lesson learned from the SYMPLICITY HTN series of trials including HTN-3, if a new renal denervation system is proposed for investigational testing, what critical information should be assessed and required, such as inclusion/exclusion criteria, measures for truly resistant hypertension, trial design with a sham control group, medication adherence, specific endpoints such as ambulatory blood pressure, physicians' experience, duration of follow-up, etc. Are there key pre-clinical data requirements?
- Discuss any new clinical evidence supporting or refuting the beneficial effects of Renal Denervation on resistant hypertension;
- Has there been any new progress in the methodology for monitoring the effectiveness of a renal denervation procedure.
- Are there any techniques other than Radio Frequency, such as using ultrasound or chemical denervation, used for renal denervation, and if so, what are the advantages and disadvantages of these techniques? What kind of critical information should be assessed if a clinical trial using such a technique is proposed?
- What evidence would you require in order to introduce Renal Denervation into regular clinical practice in Canada?
Dr. Ruzicka noted that he will discuss the renal sympathetic denervation for treatment of hypertension post Symplicity HTN-3 Trial. He will look at:
- Efficacy and Safety Endpoints of Symplicity HTN-3 Trial
- Relevance of Renal Sympathetic Hyperactivity in Hypertension
- Limitations in Internal Validity of pre Symplicity HTN-3 Trial
He noted that the Symplicity HTN-3 Trial was a prospective, single blind, randomized (2:1), and sham-controlled trial, with 6 months follow-up.
Primary efficacy endpoint was:
- The mean change in office SBP from baseline to 6 months in denervation group compared to mean change in office SBP in sham control group (superiority margin 5 mmHg)
Secondary efficacy endpoint (which study was powered for) was:
- The change in mean 24-hr ambulatory SBP at 6 months
Primary safety endpoint was:
- A composite of major adverse events defined as death from any cause, ESRD, and embolic event resulting in end-organ damage, renal artery or other vascular complications, hypertensive crisis within 30 days, new renal artery stenosis ≥ 70% within 6 months
Next he discussed inclusion and exclusion criteria.
Inclusion criteria were:
- Before referral: SBP ≥ 160 mmHg (average of 3 measurements at an office visit) on ≥ 3 BP lowering drugs at maximally tolerated doses including diuretic, with no changes in BP medications within 2 weeks before referral
- Post referral: 2 weeks of home BP monitoring (AM and PM) and diary of adherence to BP medications
- Confirmatory screening visit: SBP ≥ 160 mmHg, review of adherence to medication, and average SBP from 24-hr ABPM ≥ 135 mmHg
Exclusion criteria included:
- known secondary causes of hypertension
- multiple renal arteries
- renal artery diameter < 4 mm and/or length < 20 mm
Study treatment (2:1 randomization and blinding):
Treatment group: renal artery denervation (Symplicity renal denervation catheter, Medtronics)
Sham control: renal angiography
Blood pressure medication: No changes to BP lowering drugs allowed during the 6-month follow-up.
Intention to treat principle:
- 316 patients required on the basis of the 9.8% safety performance criterion to provide 80% power, with the use of a one-sided significance level of 0.05.
- Considering expected patient attrition, authors calculated that 530 patients will need to be enrolled.
- A superiority margin of 5 mmHg for the primary efficacy end point was considered a clinically meaningful BP reduction based on the observed decreases in cardiovascular morbidity with small reductions in SBP (2-5 mmHg) with pharmacotherapy.
The trial concluded that:
"This blinded trial did not show a significant reduction of systolic blood pressure in patients with resistant hypertension 6 months after renal-artery denervation as compared with a sham control."
This trial did show that RDN is a relatively well tolerated procedure, with a low major adverse event rate (1.4%).
Why renal sympathetic denervation is a reasonable hypothesis for treatment of hypertension?
Dr. Ruzicka listed milestones in treatment of hypertension:
- Walter Kempner – Salt Free Diet
- Lumbo-Dorsal Sypathectomy
- Diuretics, Sypathetic Ganglion Blockers
- Central Sympatholytcs
- α – Blockers, ß-Blockers
- Calcium Channel Blockers
- Ace-Inhibitors, ANG II Receptor blockers,
- Re-discovery of Aldosterone Blockers
He described the Renal Sympathetic Nervous System:
Renal Sympathetic Afferents
- Mechanoreceptors/Chemoreceptors within kidney
- Central Projections to Hypothalamus
- Modulation of Central Sympathetic Outflow
Renal Sympathetic Efferents
- Central Sympathetic Outflow to kidneys
- β1 - juxtaglomerular granular cells- renin release rate- stimulates increase release rate of renin
- α1a - intrarenal vasculature- RVR and RBF- stimulates decrease in GFR and RBF
- α1b - renal tubular epithelium - tubular sodium absorption- stimulates sodium and water retention
Role of RSNA in Experimental HTN:
- Studies have been inconsistent in determining whether renal denervation prevents/delays the development of HTN or not.
He noted limitations to internal validity of the pre Symplicity HTN -3 studies:
- Patient Selection Process
- Does this patient have sympathetically driven HTN?
Dr. Ruzicka explained that we do not know, and there is no easy way to administer a test to directly assess central sympathetic outflow and/or renal sympathetic activity. If a decrease in central and renal sympathetic activity in response to RDN is the major mechanism for BP lowering, one could consider "acute BP response to clonidine" to predict response (or a lack thereof) to RDN. MSNA might also be a predictor for elevated sympathetic activation.
Regression to the mean – in pretty much all of the studies reported decreases in BP were much larger compared to those based on day-time average from 24-hr ABPM (and this could be related to significant number of patients with white coat effect and/or true to regression to the mean).
Dr. Ruzicka explained that this issue was addressed in Symplicity HTN-3 as patients underwent 2-week period of daily ambulatory BP measurements and had then 24-hr ABPM.
He noted that the placebo effect – that is well documented in HTN trials could have played a role in earlier trials as they were either not controlled or the control arm was not blinded. He indicated that this issue was addressed in Symplicity HTN-3 with the sham control as patients were blinded to whether they received renal angiogram only or renal angiogram with RDN.
Adherence to BP lowering drugs
Dr. Ruzicka noted that one could also speculate that patients post RDN who are not blinded may increase their adherence to "lower" number of BP lowering drugs and/or in believe that they "(semi)cured". Symplicity HTN-3 trial did not include rigorous methods of assessments of compliance with BP lowering drugs.
Mechanistic explanations for lack of BP response to Renal Sympathetic Denervation
Dr. Ruzicka described the technical aspects of catheter-based renal sympathetic denervation:
- Distribution of renal sympathetic nerves
- Highly operator dependent technique
- New radio ablation catheters
To address the question whether the treatment does what it is supposed to do Dr. Ruzicka provided examples of old and new catheters for renal sympathetic denervation.
Regarding new evidence for safety and efficacy of renal sympathetic denervation for treatment of resistant hypertension he noted that at this time all we have is the Global Symplicity Registry, which is a prospective, multi-center, non-randomized international registry for RDN that provides safety and efficacy data up to 36 months post procedure. It provides excellent safety record, and similarly excellent efficacy record, however many patients are lost to follow up.
He indicated that there are three major clinical trials underway:
- SPYRAL HTN OFF/ON MED - Global Clinical Study of Renal Denervation With the Symplicity Spyral™ Multi-electrode Renal Denervation System in Patients With Uncontrolled Hypertension in the Absence of Antihypertensive Medications (SPYRAL HTN-OFF MED)
- Design: prospective, randomized, single blind trial,
- Participants: patients with resistant HTN
- Procedure: sham procedure vs RDN
- Device: Symplicity SPYRAL multi-electrode RDN system, sponsored by Medtronic Vascular Corp.
- Primary outcomes: acute and chronic safety, change in BP by 24-hr ABPM from baseline to 36 months
- Number of participants: 100
- Study completion: 9/2020
- Primary completion date: 1/2018 (final data collection)
- The RADIANCE HTN – This is a study of the ReCor Medical Paradise System in Clinical Hypertension (IRB no. 42715)
- Design: prospective, randomized, triple blind
- Participants: patients with resistant HTN
- Procedure: sham procedure vs RDN
- Device: The Paradise RDN Ultrasound System, Sponsored by ReCor Medical, Inc.
- Number of participants: 292
- Primary Outcome: mean reduction in daytime average ambulatory SBP from baseline to 2 months post procedure
- Completion date: 8/2021
- Primary Completion Date: 8/2018 (final data collection for primary outcome)
- REDUCE HTN REINFORCE - REDUCE-HTN: This study is designed to demonstrate the effect of Boston Scientific Vessix Renal Denervation System
- Design: prospective, randomized, single blind
- Participants: patients with resistant HTN
- Procedure: sham procedure vs RDN
- Device: Vessix RDN – sponsored by Boston Scientific Corp.
- Number of participants: 51
- Primary outcome: mean reduction in average 24-hr ambulatory systolic blood pressure (baseline to 8 weeks post procedure)
- Study Completion: 3/2021
- Primary Completion Date: 2/2018 (final data collection date for primary outcome)
Catheter-Based Renal Sympathetic Denervation for Resistant Hypertension: A Multicenter Safety and Proof of Principle Cohort Study was discussed next. It is a prospective observational study with:
- Resistant HTN (SBP>160 mmHg while on 3 BP lowering drugs including diuretics
- GFR (MDRD) > 45 ml/min/1.73m2
- Renovascular abnormality on angiogram
- Central Sympatholytic (Clonidine/Moxonidine)
- One renal artery dissection prior to radioablation (Treating with stent)
- No RAS (IN 1-6 mths F/U)
- Visceral interabdominal pain during radioblastation
The study concluded that:
"Catheter-based renal denervation causes substantial and sustained BP reduction without serious events in patients with resistant hypertension. Prospective randomised controlled trials are needed to investigate the usefulness of this procedure in the management of this condition."
Dr. Ruzicka described the Symplicity HTN-2 Trial:
- Prospective, open-label, randomized (1:1), controlled (renal sympathetic denervation + pharmacotherapy vs pharmacotherapy only) trial
- Follow-up: 6 months
Primary efficacy end point:
- The mean change in office SBP from baseline to 6 months in denervation plus pharmacotherapy group compared to mean change in office SBP in pharmacotherapy only group
Secondary efficacy end points:
- the change in mean 24-hr ambulatory SBP at 6 months
- the change in home blood pressure at 6 months
- number of patients with decrease in SBP ≥ 10 mmHg
- number of patients with achieved office SBP < 140 mmHg
Secondary safety endpoint:
- acute procedural safety
- chronic procedural safety (reduction in eGFR > 25% or new RAS > 60%)
- a composite cardiovascular end point (acute myocardial infarction, sudden cardiac death, new onset CHF, stroke, aortic or lower limb revascularization procedure, lower limb amputation, death from aortic or peripheral arterial disease, hemodialysis, death from renal failure, hospital admission for atrial fibrillation or hypertensive emergency)
- SBP ≥ 160 mmHg on ≥ 3 BP lowering drugs at maximally tolerated doses including diuretic, with no changes in BP medications within 2 weeks before referral
- 2 weeks of home BP monitoring and diary of adherence to BP medications
Exclusion criteria included:
- eGFR (MDRD formula) < 45 ml/min/1.73 m2
- renal artery diameter < 4 mm and/or length < 20 mm
Statistical analysis – primary end point:
- 50 patients per group required for 80% power to show benefit of renal denervation over control intervention assuming at least 12 mmHg difference between the groups with a 21 mmHg SD of the change in SBP from baseline to 6 months (a two-sided significance level of 0.05).
Acute procedural safety
- Femoral artery pseudoaneurysm (n=1), back pain (n=1), intraprocedural bradycardia (n=7) and hypotension (n=1)
Chronic procedural safety (reduction in eGFR > 25% or new RAS > 60%)
- Decrease in eGFR > 25% (RDN n=2, controls n=3)
- One progression of RAS (43 out of 49 patients had imaging at 6 months)
A composite cardiovascular end point or serious adverse events
- Hypertensive emergency (RDN n=3, controls n=2)
- TIA (RDN n=1, controls n=2)
- ACS-PCI (RDN n=1, controls n=1)
Health Canada noted that after receiving two randomised trials, a decision was made to licence the device. However when the HTN-3 results came out, they were not in agreement with the early findings. Given some of the history of these devices and the fact that we anticipate receiving future data on this technology providing additional trial data with complex patient population with large confounding factors and discordant observations that we have seen so far how do we approach new data that would be received for assessment of this technology?
One of the members commented that making the technology available to users based on available evidence at that time was a positive decision not a failure. The evidence provided at that time indicated the safety and benefits of this technology. The efficacy would be established as the technology is used.
It was also noted that true resistant hypertension is rare and likely only present in around 5% of hypertension patients.
There was a short discussion regarding the regulatory requirements around the safety and effectiveness. It was suggested that perhaps this could be a topic for discussion at a future meeting.
This concluded Dr. Ruziicka's presentation.
Dr. Ruzicka and Yee left the meeting.
7. Home Monitoring in Canada
Invited Speaker: Dr. Ratika Parkash
Dr. Parkash introduced herself, disclosed her affiliations and proceeded to address the questions posed by Health Canada.
OBJECTIVES: As medical devices become more autonomous, automated home monitoring systems are becoming increasingly important as a means of updating physicians as to patient and device status. Recently, Health Canada has received license amendment applications to add important device safety improvements to currently licensed implantable devices which depend on the patient having an automated home monitoring system. There is some concern that in certain cases, this may not be an effective a risk mitigation mechanism. Health Canada would like to better understand the prevalence of, ease of access to, and effectiveness of automated home monitoring systems in Canada in order to better understand the risks and benefits of using these systems as a vehicle for mitigating risks associated with device hazards and failures.
Provide an update on the current state of automated home monitoring of medical devices in Canada with specific reference to life-sustaining implantable devices such as pacemakers/defibrillators.
Please address the following:
- How prevalent is home monitoring of device-implanted patients in Canada and how significant a risk is there that a pacemaker-dependent patient does not have access to home monitoring (and thereby safety improvements)?
- How effective are home monitoring systems at communicating device or patient issues to a physician who is able to respond? Under what circumstances an important report might be missed or overlooked?
- What is the current process for receiving, reviewing, and responding to home monitoring reports? How easy is it to identify a serious or urgent case? How quickly can health care professionals respond in the case of an urgent report (e.g. imminent device or battery failure, or sudden change in patient condition)?
- Does increasing the amount of transmitted data increase the risk that important reports are not identified quickly (e.g. is the amount of data transmitted appropriate or useful)?
- What, if any, considerations should Health Canada be requiring from manufacturers to ensure that home monitoring solutions both cover the affected population and provide useful alerts to clinicians in a timely and appropriate manner?
- What kind of information should be being transmitted and to whom
- should prioritization mechanisms be incorporated into the reporting
- should companies be responsible for information triage
Dr. Parkash began her presentation by stating that in 2011 there were 25,000 pacemaker (PM) implants and 7000 implantable defibrillator (ICD) implants yearly in Canada, with approximately 120 000 patients living with these devices.
She explained that the recommended follow-up for these devices is based on consensus and position statements, with the majority of patients requiring at least yearly visits to a specialized device clinic during the lifetime of their device. She noted that there are many nuances for the need for follow-up. During the life of these patients, many issues may arise, such as atrial or ventricular arrhythmias that may result in syncope, stroke or sudden death, need for increased monitoring resulting from device advisories, or minor programming adjustments to improve device performance, or simply the need for enhanced surveillance as the device battery depletes and replacement is anticipated.
Dr. Parkash described the remote monitoring technology components:
- CIED is equipped with a micro-antenna that communicates with a small external device (at bedside or wearable) commonly known as the transmitter.
- The information transfer to the communicator can occur at preset time intervals with the participation of the patient (waving a wand over the device) or it can be sent automatically (wirelessly) without their participation.
- The encrypted data are then uploaded to an Internet-based database on a secure central server.
- The data processing facilities at the central database, depending on the clinical urgency, can trigger an alert for the physician(s) that can be sent via email, fax, text message, or phone. It can take several hours between the event occurring and the clinicians being notified.
- The details are also posted on the secure website for viewing by the physician (or their delegate) at their convenience.
She noted that remote monitoring technology has changed over time:
- Original design/technology did not permit real-time communication
- Data transfer can take several hours, depending on mode of data transfer ie cell-based, internet-based
- Newer technology is now using a smart phone app and data transfer is performed using cell towers
- Disadvantage is requirement of cell towers for this to occur
- In Canada, this is an issue in rural communities – we need to continue having support for land-line solutions.
How prevalent is home monitoring of device-implanted patients in Canada and how significant a risk is there that a pacemaker-dependent patient does not have access to home monitoring (and thereby safety improvements)?
Dr. Parkash noted that:
- Device clinics integrate RM capability into their routine functions and include this service as part of the standard of care for ICD/CRT patients (Strong Recommendation, Moderate quality evidence)
- For one manufacturer (Medtronic) at one Canadian facility, approximately 90% of ICD patients have RM, but only 7% of pacemaker patients have RM.
- Other manufacturers (Abbott) has less support provided
- RM is generally not supported for pacemaker patients in Canada as these are excluded from the standard contract with the device manufacturers.
- A recent position statement by Yee et al recommended that for CIED patients in whom no device issues are identified, routine follow-up assessment during the maintenance phase should blend remote monitoring (RM) with in-clinic assessments beginning after the 3-month post-implant assessment, alternating assessments between in-clinic and RM transmissions in a 1:1 ratio (Conditional Recommendation, Low-Quality Evidence).
How prevalent is home monitoring of device-implanted patients in Canada and how significant a risk is there that a pacemaker-dependent patient does not have access to home monitoring (and thereby safety improvements)?
She noted that:
- Use of remote monitoring for pacemakers is extremely low:
- Only 37% of patients (34,259) used RM (in the United States)
- Pacemaker-dependent patients are highly unlikely to be on remote monitoring unless they have an implantable defibrillator in place
- The largest trial on remote monitoring excluded pacemaker dependent patients for safety reasons
She explained that one of the barriers to increased use of remote monitoring is cost:
- Device contracts don't include device remote monitoring, so it is an added cost.
- There is no billing for remote monitoring in Canada (so no budget to cover those costs).
Other barriers include:
- Clinic flow
- Ability to deal with all the remote devices.
- Privacy issues (because the information goes through the US, and information is available for use by the American government due to the patriot act.)
- Some patients decline it because they would prefer to see the doctor in the clinic.
How effective are home monitoring systems at communicating device or patient issues to a physician who is able to respond? Under what circumstances an important report might be missed or overlooked?
Dr. Parkash noted that:
- Effectiveness as outlined in the clinical trials demonstrates:
- Reduction in time to actionable events
- Significant benefit for advisory-related issues
- Patient-related factors are reflected in:
- Success rates of 89-100% in the clinical trials in the literature
- Patient satisfaction – excellent, Quality of life improved over standard of care
- Compliance – 79% in real-world data (Rosenfeld et al PACE 2014;37:820-7), in the studies >85%
- There are some issues with patients not being around their home monitor for an alert to be transmitted, these are uncommon, and less likely to occur in the future with smart phone-based technology
- Potential for delayed in clinic response – generally next business day, however there could be delays if the transmission did not work.
- Failure of the technology to transmit is a very low, the chances of a report being missed or overlooked is generally very low (although we don't have a lot of data about this).
She provided information on current knowledge for RM – ICD (Implantable cardioverter defibrillator) and discussed results of systematic review studies.
Longest follow-up is 2 years; most are 1 – 1.5 yr. follow-up. All adverse events (syncope, death, etc.) are lower in the active, monitored group, and these studies have been powered for non-inferiority.
- Significant reduction in ICD shocks and battery longevity is longer due to lower number of capacitor charges required for those shocks
- Only one study showed reduction in mortality, most systematic reviews border on significance for mortality
- Reduction in time to clinical event (mean of 25 days)
Next Dr. Parkash presented registry data of 69 556 patients that showed 50% reduction in mortality in RM patients. She also provided results of Comparative Follow-up Schedule with Home Monitoring (COMPAS) study looking at pacemaker data for 2012 patients (the only randomized RM trial), for 18 months which confirmed non-inferiority of remote monitoring and a mean 117-day gain in the event detection.
Next Dr. Parkash discussed advantages and disadvantages for remote monitoring:
- A reduction in travel time and in-clinic wait-time, although this benefit would be especially great for patients living in rural and remote communities.
- The potential earlier detection of events could result in quicker physician intervention when needed and improved patient safety.
- The immediate follow-up after these events could provide reassurance or a perception of greater security for patients.
- In the longer term, earlier detection and treatment may result in improved disease management, better health, and potentially less or shorter hospital admissions.
- Some patients may be technically challenged or live in areas with communication quality issues related to fixed telephone lines or poor coverage for mobile networks.
- Patients may also have a concern for the privacy and security of their medical and personal information on the Internet.
- They may also have a false sense of security when using RM and avoid following up with a physician when the need arises.
- Constant surveillance with RM may have uncertain effects on patients' quality of life and may result in adjustment issues involving increased anxiety or the fear of constant observation. (This is a potential disadvantage, has not been proven.)
What is the current process for receiving, reviewing, and responding to home monitoring reports? How easy is it to identify a serious or urgent case? How quickly can health care professionals respond in the case of an urgent report (e.g. imminent device or battery failure, or sudden change in patient condition)?
Dr. Parkash provided information on after hours care. During office hours, if there was a remote transmission they would generally respond to it immediately. Anecdotally (though there is no data for this), the number of calls that need to be responded to within a 24/7 timeline is low. Calgary and Ottawa may be the only two clinics that have a 24/7 response system for remotes.
Dr. Mitchell noted that one other issue is that in patients that live in other cities, the only way that the ER doctor knows what is going on with the device is by receiving the transmission report from the patient's doctor, as the report is not sent to the ER.
She noted that eight of the clinics (36%) routinely call patients after a remote transmission is received. Most only call patient when there is an issue. Currently no direct line from the patient to their device clinic if they present to the ED, it is up to the discretion of the ED physician whether the device clinic learns of the patient's visit to the ED. This is the current state of operation in Canada.
Next she provided results from the patient perspective – patient survey n=512:
- 16.7% sometimes/always feel nervous when attending clinic
- 33% felt secure not going to hospital to have device checked, 20% were unsure, 47% preferred coming to hospital
- 75.8% of patients know what to do if they experience a shock (191/252)
- 90.8% of patients feel that a phone call after the remote transmission was very important
- 52% are nervous or scared when they receive an alert
Does increasing the amount of transmitted data increase the risk that important reports are not identified quickly (e.g. is the amount of data transmitted appropriate or useful)?
Dr. Parkash explained that:
- This does not seem to be an issue
- Workload is currently increased with the current system of blended RM and in-clinic visits
- Overall, as clinical studies demonstrate, effectiveness of RM is proven, irrespective of increased amount of data
- The use of alerts is critical to ensuring that important reports are not missed and to manage information overload
- a sensitivity of 100% and a specificity of 97% (positive predictive value 96%, negative predictive value 100% (Osca et al Rev Esp Cardiol 2009;62:886-95)
It is a huge burden, we haven't streamlined: we need to sort out which alerts need to be reviewed and which ones do not.
What, if any, considerations should Health Canada be requiring from manufacturers to ensure that home monitoring solutions both cover the affected population and provide useful alerts to clinicians in a timely and appropriate manner?
Dr. Parkash provided the following recommendations:
- Ensure that remote monitoring is available in all areas of Canada
- The changes in technology using cellular-based transmission is an advancement but not having technology that permits patients who live in areas where cell towers don't provide coverage is an issue. Manufacturers are removing the option of non-cellular based transmission; Dr. Parkash believes this is an issue.
- Technology should advance to permit more rapid monitoring of patients i.e. in the event of an appropriate ICD shock, remote monitoring technology should permit more rapid response, rather than having delays that don't permit near real-time response. We don't need absolute real-time response but near real-time would be desired.
- Ideally information should be transmitted to patient, family physician and specialty provider in forms that are relevant and understood by each of these users, particularly so that patients understand what is going on with their device when they have an alert or a remote transmission is sent.
- Mechanisms to handle the amount of data transmitted through RM would be of significant benefit, such as machine learning to assist in interpretation of transmissions where a triage system permits more urgent issues to be dealt with
- The responsibility of where this lies is unclear – ideally information triage should be device agnostic, or universal, rather than device company specific
- Patient care should be performed by trained personnel, rather than device companies.
She indicated that in Canada we have not streamlined the process yet. Establishing servers in Canada has major cost implications. We need to address many aspects of this process, such as:
- Remote monitoring
- Remote programming
- Privacy issues
- Some patients are not comfortable with use of this technology.
There was a comment that now that this data has been received by the physician, there is an expectation and obligation to assess the transmissions, despite the sheer volume of information, because it has been shown that there is a benefit to the patient by analyzing the data. It was noted that physicians are getting "snowed in" by all the information.
While not shown with a superiority study, it seems likely that transmissions will provide increased safety. It is going to become very important to determine how the algorithms will work and how we will assess their safety.
There was a question regarding a change in the criteria of the action event. It was explained that there have been changes with how we deal with atrial arrhythmias, but now they have become actionable events, where previously it was not clear what to do. All other events have remained the same.
Depletion of the battery as a result of remote monitoring is now minimal, so this is not a concern.
All studies have demonstrated a reduction in "in-clinic" visits, however they don't account for the increase in workflow.
It was explained that since each province has their health information protection act, this causes major delays (years of delays) to get remote monitoring approved from a privacy standpoint.
In Canada, there is currently a study (~ 100 patients) to look at Remote view: reprogramming from afar (in another clinic, not at home). But device companies have backed off of this significantly due to the length and cost of privacy approval by institutions in Canada.
This concluded Dr. Parkash's presentation.
Dr. Parkash and Dr. Khaykin left the meeting.
Ventricular Assist Devices
Speaker: Dr. Renzo Cecere
Dr. Cecere introduced himself, disclosed his affiliations and proceeded to address the questions posed by Health Canada.
OBJECTIVES: Health Canada has evaluated a number of Left Ventricular Assist devices, with more limited exposure to Right Ventricular Assist Devices. At our meeting on 2015-06-26, Dr. Vivek Rao provided an overview of Mechanical Circulatory Support with a focus on the clinical impacts of left ventricular support using continuous flow or pulsing flow, and how these may promote or reduce pump thrombosis. Health Canada is requesting additional clinical insights on the therapy provided for right heart failure in the context of end stage heart failure patients. Health Canada is also requesting a general update on new developments within the area of circulatory support in the past two to three years of left, right, and bi-ventricular heart failure, especially as improved outcomes may lead to an increase in the use of this type of technology in Canada.
The left ventricular assist devices (LVADs) have become a standard of care among patients with end stage heart failure. However, right ventricular failure (RVF) after LVAD implantation or cardiac shock is frequent, with reported rates as high as 50% of LVAD recipients, and with very poor prognosis. RVF represents a major cause for postoperative multi-organ failure and death.
- What are the risk factors for RVF? What are the hemodynamic characteristics during RVF in comparison with LVF and bi-ventricular failure?
- What treatment options for RVF are available (with focus on treatment using medical devices)? What is appropriate timing for RVAD implantation?
- Provide an update on current evidence and clinical trials on both left and right ventricular assist devices (e.g., MOMENTUM 3 trial) as well as experience in bi-ventricular support.
- Provide an update on how and when devices are currently being used in Canada.
- Discuss the risks and benefits of ventricular assist devices. Particularly, discuss the risk of pump stop and pump thrombosis using ventricular assist devices and risk mitigation strategies.
Dr. Cecere described the evolution of Ventricular Assist Devices:
- Generation 1 devices: HeartMate, Novacor (1994-2000) –pulsatile, positive displacement, large size, placement in abdominal cavity
- Generation 2 devices: Jarvik, HeartMate II, INCOR (2000-2006) –continuous, axial flow with bearings, medium size, pericardial placement
- Generation 3 devices: HeartMate III, HVAD, MVAD, HeartAssist5 (2006-2015) –continuous, Levitating rotor- centrifugal or axial, miniature, placement in pericardium
He discussed some of the current Ventricular Assist Devices (VAD) with pulsatile flow pumps versus continuous flow pumps, centrifugal pumps versus axial pumps.
He presented the results of ENDURANCE Designation Therapy Trial- HeartWare HVAD for the treatment patients with advanced heart failure ineligible for cardiac transplantation.
He described the fully magnetically levitated circulatory pump for advanced heart failure.
Next he described the Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3 (MOMENTUM 3) trial which was:
- Non-blinded randomized trial that compared the centrifugal-flow pump (HeartMate 3) with the axial-flow pump (HeartMate II) in patients with advanced heart failure
- The HeartMate 3 (HM3) was associated with a higher rate of survival free of disabling stroke or survival free of reoperation to replace or remove the device at 6 months after implantation than was implantation of the HeartMate II (HM2) among patients with advanced heart failure, irrespective of their eligibility for transplantation.
- The incremental benefits associated with the HM3 observed in this 6-month analysis were due to the absence of suspected or confirmed pump thrombosis leading to surgical pump exchange or urgent transplantation.
- There were no significant differences between the two pumps in the associated rates of other major complications
- There was a trend toward a higher rate of disabling strokes with HM3 than with HM2
- Higher rate of right heart failure events managed with a right ventricular assist system with the axial-flow pump than with the centrifugal flow pump.
Dr. Cecere noted a number of limitations with this study:
- There was no blinding
- The surgeons at most, if not all, of the participating centers had long-term experience in the implantation of HM2, and thus the surgical and medical outcomes were potentially biased against the HM3
- The decision to remove or replace a pump for suspected or confirmed pump thrombosis was informed by the lactate dehydrogenase level or evidence of pump dysfunction but was at the discretion of the local site investigators
- The trial findings should not be extrapolated to a broader population with less severe heart failure
Trends in LVAD Use: The INTERMACS Registry Data was discussed next.
Dr. Cecere noted that experience with VADs is different from Province to Province thus it is difficult to provide Canada wide information. He provided information on the evolution in the use of long-term LVAD in Quebec: 2010-12 versus 2013-15 and comparison with the INTERMACS Registry:
He noted that use of LVADs in patients in cardiogenic shock has decreased over the last 6 years in Quebec. During the same period, LVAD has been used more frequently as destination therapy (DT) compared to bridge to transplant (BTT), with approximately 46% of patients now being DT candidates. In BC, funding is only provided for BTT patients. Changes in patient selection have been associated with improvements in clinical outcome. However, LVADs are still inferior to heart transplants. Funding is a significant issue as the devices cost approximately $120,000 and a full set of battery replacements costs $40,000.
Continued monitoring, timely feedback and participation in a large international registry likely contribute to optimizing quality of care. Quebec has recently developed a consensus document regarding standards of LVAD programs, in collaboration with clinical experts from each active institution. A companion summary of quality indicators, in line with INTERMACS, will be used to evaluate performance in the coming years.
Dr. Cecere noted that in Canada we are very fragmented in data collection and availability.
He discussed the Ontario Health Technology Assessment Series which has published on Left Ventricular Assist Devices for Destination Therapy in 2016. Ontario Health Technology Advisory Committee recommends that:
- Continuous-flow left ventricular assist devices (LVAD) are publicly funded as permanent therapy (also known as destination therapy) in patients with end-stage heart failure who are ineligible for heart transplantation.
- The Cardiac Care Network and Trillium Gift of Life Network provide guidance regarding which hospitals should offer this procedure and which patients should be eligible.
- The Cardiac Care Network and Trillium Gift of Life Network ensures data is collected on survival and quality of life for individuals receiving continuous flow LVAD as permanent therapy, and that this data be reviewed by Ontario Health Technology Advisory Committee in 2 years.
The full report is available at: Health Quality Ontario: Left Ventricular Assist Devices for Destination Therapy: OHTAC Recommendation (www.hqontario.ca)
Dr. Cecere discussed the NHLBI Randomized Evaluation of VAD InterVEntion before Inotropic Therapy (REVIVE-IT) pilot study.
- In 2008 it was believed that field of advanced HF treatment was ready for a trial to assess the use of long-term LVAD therapy in patients less ill than patients currently eligible for destination therapy VAD
- The specific objective was to compare LVAD therapy with optimal medical management in patients with less advanced HF than current LVAD indications, and
- To determine if wider application of permanent LVAD use for less ill patients would be associated with improved survival, quality of life, or functional capacity
- VAD therapy will improve functional status at 12 months and all-cause mortality will be no worse than that in the OMM arm of the study
- Estimated mortality rate 20%– 30% at 1 year
- Patient population inclusion criteria:
- hospitalization in past 6 months
- NYHA class IV or advanced class III
- maximal evidence-based therapies for 3 months
- duration of HF of at least 1 year
- no inotropic support for 30 days before enrollment
- Randomized between VAD and OMM arms.
- Primary end-point: a composite of survival and functional outcome
- Secondary end-points: safety, secondary functional and physiologic markers, health-related quality of life indicators, neurocognition, and cost-effectiveness.
- Patients should be followed for at least 2 years
First patient enrolment started in December 2014, in January 2015 the FDA notified that REVIVE-IT trial was placed on clinical hold after series of consecutive pump thrombus events was observed in patients receiving the device (HM II) for current FDA-approved labeling indications outside of the context of REVIVE-IT study at a REVIVE-IT clinical site. In March 2015 the DSMB met to discuss issues leading to the clinical hold, and in April 2015 REVIVE-IT study discontinued after NHLBI acceptance of DSMB recommendations
Dr. Cecere noted that for a trial similar to REVIVE-IT to move forward in the future, several important issues need to be addressed, including:
- Identification of a study device with a safety profile that satisfies clinical equipoise for trial investigators and regulatory agencies
- An improved understanding of patient characteristics that will permit more accurate assessment of HF prognosis
- New technology will be required to advance LVAD therapy into a population with less advanced HF.
VAD for Right Heart Dysfunction
Decision to implant dual HVADs (13 pts off-label use on the right side) (reference: J Heart Lung Transplant 2016; 35:466-473) was based on a combination of:
- Hemodynamic parameters:
- RV stroke work index,
- Mean pulmonary arterial pressure
- Trans pulmonary gradient
- Echocardiographic markers:
- Degree of RV dilation
- RV systolic impairment
- Estimated pulmonary arterial systolic pressure
- Severity of tricuspid regurgitation
RVF was defined in the HeartMate II Clinical Trial as either:
- The need for an RVAD in addition to the LVAD (group 1),
- Continuous inotropic support for at least 14 days after implantation (group 2), or
- Late inotropic support starting 14 days after implantation (group 3).
- Data from groups 1 and 2 were combined to form an early RVF group, whereas group 3 patients were examined separately (late RVF group)
The post-operative outcomes showed that 6 out of 13 (46%) patients died, and 4 out of 13 (31%) patients had RVAD thrombosis. It was stated that a combination of hemodynamic and clinical variables can be used to identify patients who might be at high risk for RVF.
Device options include direct and indirect RV bypass, as well as intracorporeal (Impella RP) and extracorporeal (Tandem RV assist device (RVAD), Protek Duo, venoarterial extracorporeal membrane oxygenation (VA-ECMO)) options.
Although there is some use of LVADs, including the HVAD, in Bi-ventricular support, the designs are not optimized for this application (e.g. cannula can be too long), and outcomes are poor. It isn't clear that Bi-VAD support is better than TAH.
TandemHeart in Right Ventricular Failure: nine (9) patients with MR-RVF between 2008-2010 received a TandemHeart RV support device. The etiologies included sepsis, post-cardiotomy, and acute RVMI. Results suggested that earlier support using right ventricular assist device could result in better clinical outcomes.
In general, RVF is a major problem with LVAD use. There are RV architecture changes due to septal shift with LV unloading and LVAD support. Recent efforts are to minimize septal shift that compromizes RV filling. There is less septal shift with new continuous flow devices, but this also means there is less ventricular recovery (e.g. for bridge to recovery). Variability in flow rates over the cardiac cycle is important.
Dr. Cecere discussed the RECOVER RIGHT Trial that is FDA approved prospective, multicenter, single arm study that evaluated the safety and probable benefit of a novel percutaneous VAD ( Impella RP) in patients with RVF refractory to medical treatment and deemed to require hemodynamic support.
In the RECOVERY RIGHT trial, the Impella RP, which is a size 22 Fr pump on an 11 Fr catheter and flow rate > 4 L/min was used. The results showed that hemodynamic improvement was reproducible in most patients as shown by increased Cardiac Index and decreased Central Venous Pressure.
Discussion on the risk factors, detection, and management of pump thrombosis followed.
There was a question whether the right ventricular failure is due to pulmonary hemodynamic changes or truly due to right ventricular itself. Dr. Cecere noted that it could be both: hemodynamic changes and right ventricle itself (e.g. right ventricular myocardial infarction) could contribute to development of right ventricular failure.
A question was asked whether the flow rates are clinically different (between the HM2 and HM3). It was explained that there is no difference.
Regarding the rate of driveline infection it was noted that it is relatively frequent and could be up to 20%, but of minor clinical impact.
There was a question regarding the accuracy of flow rate measurement of the right ventricular assist device considering the pressure difference between the left and right ventricles. It was explained that the flow rates are measured by algorithm (indirect measurement). Dr. Cecere noted that there are lots of ways for the flow rate measurement; however, we need to look at the patients' conditions and look at the clinical picture, not just the measurement.
This concluded Dr. Cecere's presentation.
Dr. Cecere and Dr. Mitchell left the meeting.
Dr. Yee returned to the meeting.
9. Use of Cardiac Implantable Electrical Devices (CIEDs) used in an MRI setting.
Discussion lead: Dr. Yee with contributions from Dr. Parkash and Dr. Mitchell.
This topic was originally circulated for e-consultation to Dr. Yee, Dr. Parkash and Dr. Mitchell.
Mr. Kevin Day introduced the topic to the committee member.
Acceptance Criteria for MR Conditional labeling
Until recently, applications to label Cardiac Implantable Electronic Device (CIED) system as "MR Conditional" have been supported by evidence that there is a very small likelihood of a chronic rise in pacing capture threshold (PCT) > 0.5V and an extremely remote likelihood of a rise in PCT > 1.0V due to a Magnetic Resonance (MR) scan induced RF heating of the lead electrode at the prescribed conditions. Available evidence suggests that no permanent damage is done to the underlying cardiac tissue when the rise in PCT is < 0.5 V. It is also noted that a change in PCT of < 0.5V is generally accepted as falling within expected background variation of PCT that may be observed clinically even without undergoing an MR scan.
Recently, Health Canada has received applications to amend licensed cardiac leads, including LV leads, which are currently contraindicated to MR scans and have them re-labelled as "MR conditional". For some of these leads, modelling estimates predict a significant possibility of a chronic rise in PCT > 1.0V and increases in PCT of 2.5V in worst case scenarios allowed by the labelled conditions.
Health Canada wishes to better appreciate how labelling can be used to effectively provide clinicians with the information needed to make informed decisions about the use of CIED systems in an MR environment, and to better understand what outcomes might be considered as unacceptable. For example, a device labelled as 'MR Conditional' has generally been interpreted to mean that under the conditions provided, the device will be safe to use and there is an expectation that no harm will be done to the patient. In cases where the chronic PCT increases by more than 1.0 V, this may be indicative of permanent scarring of the endocardial tissue. Although CIEDs can pace at higher voltage levels and overcome an increase in PCT, some permanent damage has been done, both in terms of endocardial damage as well as a significant reduction in battery life for the CIED.
Health Canada acknowledges that in some specific individual cases, MR scans can be a highly desirable imaging modality, without which adequate information about an underlying serious condition may not be obtained.
Health Canada requested an e-mail consultation to gather information from the clinical community so that Health Canada can gauge the level of concern associated with the use of CIED systems in an MR environment, and the potential impact to the endocardial tissue and PCT.
Specifically, Health Canada requested feedback regarding the following questions:
- Do you believe that for a device to be labelled as MR Conditional, that the predicted rise in PCT should not be greater than 0.5 V, for at least a large percentage of the 'worst-case' scanning conditions allowed by the MR Conditions provided in the labelling? If so, what percentage of patients should this be?
- Do you believe that devices labelled as MR Conditional should not cause permanent damage to endocardial tissue?
- Would you be comfortable with term 'MR Conditional' meaning that after the MR Scan, it will continue to be possible to pace the patient with a PCT less than the maximum possible output of the CIED, even if this means some damage to the underlying endocardium may occur?
- Do you believe it is appropriate to include additional data in the labelling outlining the probability of damage and rise in PCT that may occur if the proposed MR parameters are exceeded (such as a larger Specific Absorption Rate (SAR))? This information might be used in cases where moving forward with an MR scan was considered medically necessary.
- Are there other considerations from your clinical perspective that you feel Health Canada should be aware of as we try and define appropriate limits for MR conditional parameters for active implantable devices?
Dr. Mitchell and Dr. Parkash had to leave the meeting early. Dr. Yee agreed to lead the discussion. He noted that he conferred with Dr. Mitchell earlier at the meeting and has Dr. Prakash's input and will try to respond to these questions on behalf of the group.
Dr. Yee noted that that there are various approaches from center to center between cardiologists and radiologists and these approaches may differ significantly. He proceeded to address the questions in order posed by Health Canada.
Do you believe that for a device to be labelled as MR Conditional, that the predicted rise in PCT should not be greater than 0.5 V, for at least a large percentage of the 'worst-case' scanning conditions allowed by the MR Conditions provided in the labelling? If so, what percentage of patients should this be?
Dr. Yee provided the following recommendations:
- Yes, predicted rise in PCT should not be greater than 0.5 V. Clinicians are comfortable with 0.5 V limit, however, they feel that we are setting the bar very high. In clinical setting we are not seeing a dramatic increase.
- In medicine the confidence interval surrounding any given statement is usually 95%. Accordingly, it would be reasonable that the percentage of patients not harmed by an interventional purported not to be harmful should be at least 95%. Dr. Parkash suggested 99% in her response.
Do you believe that devices labelled as MR Conditional should not cause permanent damage to endocardial tissue?
The following recommendation was provided:
- Yes and no. It may be too high a standard to meet if we phrase it as "no permanent damage". It is preferable to qualify the statement by saying no tissue damage of sufficient severity as to lead to clinical deterioration of the patient or clinically significant deterioration in device function, of which PCT is the most important parameter.
- Dr Parkash indicated that they should not cause permanent damage and that creation of scar in endocardial tissue can lead to life-threatening arrhythmia if it occurs in the ventricle. It may also compromise the ability to pace safely. However, Dr Yee also pointed out that scarring already occurs where the lead is implanted, and it is unclear whether MRI-induced damage would be different.
- This would have to be looked at in context and the clinical need for the patient to undergo the MRI.
Would you be comfortable with term 'MR Conditional' meaning that after the MR Scan, it will continue to be possible to pace the patient with a PCT less than the maximum possible output of the CIED, even if this means some damage to the underlying endocardium may occur?
Dr. Yee provided the following suggestions:
- No. It is unacceptable to label any device as "MRI Conditional" if the CIED is susceptible to experiencing sufficient damage after exposure to MRI environment that PCT rises to a degree anywhere approaching the maximum pacing output of the CIED. That degree of tissue damage effectively shortens the life of the device and results in premature generator and lead replacement which exposes the patient to added risk of harm. In other words, just being able to say that the device will continue to pacing is at maximum output is not truly safe and is antithetical to the whole concept of MRI safety.
Do you believe it is appropriate to include additional data in the labelling outlining the probability of damage and rise in PCT that may occur if the proposed MR parameters are exceeded (such as a larger Specific Absorption Rate (SAR))? This information might be used in cases where moving forward with an MR scan was considered medically necessary.
Dr. Yee recommended that:
- If we already have a requirement in the labelling that the "MRI Conditional" designation is only applicable for the same MRI conditions under which the device was tested and shown to be "safe" then inclusion of such a statement should not be included.
- There may not be a benefit in additional labelling.
Are there other considerations from your clinical perspective that you feel Health Canada should be aware of as we try and define appropriate limits for MR conditional parameters for active implantable devices?
Dr. Yee felt that:
- Use of leads and pulse generators from different manufacturers, and whether the labelling would apply in these cases, if both had MR conditional labelling. However, all systems licensed so far are specifically for leads and cans from the same manufacturer, as this is the combination that is used in the evaluation of safety.
Dr. Yee noted that while he supports loosening the MRI Conditional label to allow for unrestricted rises in PCT up to the maximum output of a device, he also thinks that applying an arbitrary limit of ≤ 0.5V rise in PCT may also be overly restrictive. It really depends upon the baseline PCT prior to MRI exposure. For example, if the baseline PCT pre-MRI is 0.5V and the PCT rises by 0.6V post-MRI, that degree of change is over 100% but is of minimal clinical consequence since that rise has minimal impact upon battery longevity. Even a rise of 1-2 V, while undesirable, is not catastrophic and might well be worth it for the patient if the need for the MRI is strong. On the other hand, a rise of 1V or less in someone with a high baseline PCT (say 3.5V) may have a much greater impact. It may not be the absolute amount of PCT rise that is critical but, rather, the percentage rise. At a low baseline PCT (say, less than 1.5V) a 30% rise in PCT with MRI might be acceptable but the same percentage rise would be unacceptable at baseline PCT above 3-4V, let's say. Having said that, he recognizes that the PCT is not the focus solely because of the clinical consequences of such a rise but that a PCT rise is also representative of permanent tissue injury and the desire is to ensure that MRI exposure does as little harm as possible. However, he considers that some degree of tissue damage is acceptable the clinical setting and the debate is really about how much and how best to quantitate it.
Dr. Yee concluded the discussion.
10. Atrial Shunt/Flow Devices for the Treatment of Heart Failure
Speaker: Dr. Mathieu Bernier
Dr. Bernier introduced himself, disclosed his affiliations and proceeded to address the questions posed by Health Canada.
OBJECTIVES: Atrial shunt devices for the treatment of heart failure are new and novel devices that Health Canada would like to have a better understanding of the technology and patient populations that may benefit from implantation with such a device. This topic was previously presented at our 2015 meeting by Dr. Sebastien Bergeron (Laval), however the technology is still considered new and is evolving. Currently, a few technologies have been authorized through SAP and ITA, but since there is little clinical experience with these devices, it is a challenge to review these applications. A better understanding would help to determine requirements for clinical trial designs and possible future licensing decisions.
Provide an update on the devices used for atrial shunting. This should focus on interatrial left to right shunting in heart failure patients with preserved left ventricular ejection fraction (HFpEF). Other applications, such as left atrial pressure decompression into the coronary sinus or right to left atrial shunting, may also be discussed. Please include an overview of clinical evidence related to safety and effectiveness of the technology, novel systems currently under development, and any future trends in this area. In your discussion, please address the following:
- Provide a brief overview of the principle concept of interatrial shunting for the treatment of heart failure as background information. Also, discuss any safety risks associated to interatrial shunting.
- Provide a summary of atrial shunt/flow devices, in particular those intended to treat heart failure with preserved LVEF. For example, such devices include interatrial shunting from left atrium to right atrium either directly (trans-septal) or through the coronary sinus (a novel concept for this approach: a device system recently being requested via SAP). Devices could also include applications for pediatric patients.
- Discuss the history of the different device designs and clinical uses.
- Discuss the rationale of device designs, including uni-directional flow versus bi-directional flow devices.
- Provide an update of clinical evidence of safety and effectiveness. Discuss any lessons learned in regards to patient selection (including hemodynamic parameters), the population that will benefit the most, and best measures of safety and effectiveness.
- Provide recommendations on clinical trial designs including which endpoints are critical to measure the safety and effectiveness of these devices. Are randomized studies possible and what is the best control group?
- Provide recommendations for clinical data requirements that would demonstrate the safety and effectiveness of these devices for licensing purposes. Discuss which measures of effectiveness are useful and how much clinical improvement should be demonstrated.
- Provide your opinion on the future trends of novel devices for heart failure.
Dr. Bernier started by indicating that there is not much experience with atrial shunting in heart failure. There is very limited data. He will provide an update based on their clinic's experience in this area.
He explained that heart failure can be associated with either low EF or normal EF, however both have the same symptoms: shortness of breath and low exercise capacity. Days or weeks before a patient presents with decompensated HF, the left atrial pressure (LAP) is ≥ 25 mmHg, as measured by intracardiac monitors. Increased LAP is the cause of lung congestion in acute decompensated heart failure (ADHF). The main treatment for elevated LAP is diuretics to relieve symptoms.
Shunting is considered an alternative to diuretics for lowering LAP, particularly in patients already receiving optimal medical therapy, not eligible for advanced HF treatments, and in whom symptom relief and quality of life improvement is needed.
He described the V-Wave Interatrial Shunt that is an hourglass shape device made of polytetrafluoroethylene (PTFE) and a self-expending Nitinol frame. He noted that the first generation device included a porcine pericardium one way valve that only allowed shunting from the left to the right atrium, through a 5 mm middle orifice, while right to left flow was prevented. The shunt is placed at the thinnest part of the intra-atrial septum, in the center of the fossa ovalis, which is the easiest location to access.
First-in-man Clinical Data:
Dr. Bernier explained that they have a fairly large HF clinic and follow approximately 1000 patients, out of which 60% are low EF patients. While all patients receive optimal medical therapy, some remain symptomatic without any other treatment options. These patients were identified for potential shunting. The V-Wave was implanted in a cohort of 22 patients on compassionate grounds using Health Canada's Special Access Program (SAP).
Data was collected from these patients to evaluate the feasibility and safety of the V-Wave system for HF treatment. Collected data included:
- Procedural success (successful device implantation with no peri-procedural death); and
- Safety (device/procedure-related MACNE - device related death, stroke, device embolization, pericardial effusion requiring intervention and device related re-intervention or surgery), at 3 and 12 months.
Additional data included:
- MACNE and SAEs at 12 months; and, exploratory efficacy - clinical and hemodynamic parameters (swan-ganz and TEE, 6 Min.HWT) after 12-months of follow-up.
Main Inclusion Criteria
- Chronic HF of ischemic, or non-ischemic etiology, irrespective of LVEF (80% low EF)
- NYHA Class III, or ambulatory Class IVa
- On guideline driven maximally tolerated medical, and device therapy
- Elevated BNP/NT-proBNP
Main Exclusion Criteria
- Isolated right-sided HF
- Moderate-severe RV dysfunction
- Severe pulmonary hypertension
The gradient between LA and RA was between 5 to 15 mmHg.
- Procedures performed by trans femoral venous approach, under general anaesthesia and TEE guidance
- Patients received anticoagulation therapy (Coumadin) for at least 3 months following the procedure
- Follow-up (1, 3, 6, 12 months and yearly thereafter to 5 years)
- NYHA class
- Six-minute walk test
- Quality of life questionnaires (KCCQ; MHLF)
- Right heart catheterization (baseline, 3, and 12 months)
- Transthoracic Echo
- Transesophageal Echo (1, 3, and 12 months)
- Through Canadian Special Access Program
- 22 patients enrolled (16 HFrEF, 6 HFpEF)
- First-In-Man Multicenter Feasibility Study
- Israel and Spain (5 centers)
- 16 patients enrolled (15 HFrEF, 1 HFpEF)
- Total study population: 38
- All subjects completed 12 month follow-up (none lost to follow-up)
The study provided the following results:
- Procedural Success = 38/38 (100%)
- No device malpositioning, dislodgement/embolization or device replacement
- Procedure time = 72 ± 24 min
(Includes: TEE, RHC, transseptal, shunt placement, all study measurements)
- Median LOS: 2 days (IQR: 1-3)
- In-Hospital Adverse Events:
- 1 Cardiac tamponade (treated by pericardiocentesis)
- 3 Vascular access complications (all treated non-invasively), none at the Canadian center.
- Primary safety: 97.4% freedom from Device/Procedure-Related MACNE at 3 months
- Death: 0, Stroke: 0, MI: 0, Device Embolization: 0, Cardiac Tamponade: 1
- Secondary safety: 92.1% freedom from any MACNE at 12 months
- Death: 2 (CV – not device or procedure related related to arrhythmia), Stroke: 0, MI: 0
Other Serious Adverse Events included acute decompensated Heart Failure (HF) (9), bleeding (4), renal impairment (3) and Ventricular Tachycardia (VT) (1).
Although small patient numbers introduces considerable variability, there was a 'clear' improvement from baseline at which all patients were Class III to IV. Between 3 and 12 months a significant proportion moved to Class I to II.
Quality of Life (QoL) Change: 70%, 59%, and 70% improvements from baseline at 3, 6 and 12 months respectively.
6MWT Change (m): Compared to baseline at three months, patients could walk on average 40 m more. At the 12 month follow-up, patients could walk on average 28 m more.
Echo follow-up: key observation is the decline in shunt patency rate after three months; colour Doppler shows flow from left to right atrium. By 12 months, almost 20% were not patent.
There was no change in hemodynamic parameters, eGFR and NT-proBNP from baseline to 12 months follow-up.
Members asked how patient improvements are explained without changes in wedge pressure (PCWP). Dr. Bernier explained that wedge pressure is dynamic and changes throughout the day in both directions, and average changes are small. Approximately 80% of patients actually exhibited at least 5 mmHg decreases from baseline. He noted that this is an imperfect measure to use as a comparator.
It was noted that this procedure is vulnerable to placebo effects. Would there be a sham control for the next study? Dr. Bernier explained that the chance of a placebo effect is high, and the next RCT will include a sham control.
There was a comment regarding the finding that NT-proBNP levels did not change from baseline. It was suggested that this finding might just indicate the device is not effective. Dr. Bernier indicated that the future RCT will shed additional light on this issue.
It was noted that the hole that is created by the device may not be large enough. It was also noted that if wedge pressure is decreased significantly you would expect the cardiac output to decrease as well. Dr. Bernier explained that this isn't always clear, based on where you are on the Starling Curve, but as stated earlier, BNP and wedge pressure are imperfect measurements.
There was a concern regarding taking the patient off anticoagulants after three months. Dr. Bernier noted that patients that are already taking those medications will continue to do so, and for the others, they are not sure what the best approach is.
Dr. Bernier presented comparison of baseline characteristics between the V-Wave patients and a historical control cohort (CHAMPION). He noted that V-Wave population was older, sicker and better treated, or had higher risk baseline characteristics than CHAMPION. It was also noted that the "unofficial" comparison of hospitalization rates of the current cohort to the CHAMPION control shows HF hospitalization rates were low relative to historical well-matched advanced Class III HF populations. Similarly, this difference was shown in HF patients with reduced EF.
He indicated that 1st, generation device contained a valve, however in the first 21 patients, device narrowing over time was observed.
Legacy V-Wave Shunt: Valve Function at 1 Year:
- Absence of L-R shunt flow was observed in 5/36 (14%) surviving patients
- Color doppler vena contracta jet in the valve region was narrowed or skewed off-axis in 13/36 (36%) pts
- The absence of shunt function was associated with worsening symptoms
- Qp:Qs 1.17 ± 0.12 at implant fell to 1.10 ± 0.13 at 1 year (P=0.04)
- Bioprosthetic leaflets developed neointimal proliferation (pannus) with thickening, as observed on 2.5 year explant specimens: commissural fusion, fixation and stenosis - not associated with acute deterioration or thromboembolic events
- The higher the wedge pressure at baseline, the less likely the shunt was to be closed at follow-up, and less likely the patients would present with symptoms. Patent shunts were associated with improved hemodynamics.
- Patent shunts were associated with reduced mortality, HF-Hospitalization and all-cause hospitalization (mean 26-month FU).
There was a question if patients receive the valved device. Dr. Bernier explained that all 38 patients received the valved device.
Generation 2: V-Wave Interatrial Shunt:
- The next device will be a shunt without a valve. The device has been re-designed to be similar, but without a valve, not unlike an ASD.
- As shown in animal testing, unlike valved shunts, Generation 2 Shunts were superior in Normal Physiology (2-3 mmHg L-R gradient) as evidenced by no late lumen loss at 6 months.
RELIEVE-HF Pivotal RCT is designed to establish the efficacy of the non-valved conduit.
- Sham-controlled, blinded (patient and HF team), adaptive design
- ~400 randomized + ~120 roll-in patients at 60 sites (40 US + 20 OUS)
- NYHA class III or ambulatory class IVa on GDMT (eligibility committee review)
- HFrEF or HFpEF – no specific LVEF criteria
- Hospitalization for worsening HF within 1 year or elevated BNP/NT-proBNP
- Outcome Measures:
- Primary Effectiveness: hierarchical comparison of mortality, transplant/LVAD, HF hospitalization, and 6MWD using Finkelstein-Schoenfeld/Win Ratio
- Primary Safety: Device-related MACNE at 30 days (performance criteria)
- Health economic metrics
- The main issue will be blinding and internal validity.
There was a question regarding how can HF team be blinded? Dr. Bernier explained that the device is almost impossible to see on X-ray. They'll have to figure out how to blind from echo readings. It will be a challenge.
In conclusion Dr. Bernier noted that the V-Wave Interatrial Shunt:
- Is safe to implant and can be implanted with high procedural success and low morbidity
- Improves symptoms, QoL, and left-sided hemodynamics without worsening right-sided cardiac function
- Has been associated with reduced heart failure hospitalization and mortality compared to well-matched historical controls on GDMT
- Appears well-suited for class III/ambulatory class IV, HFrEF and HFpEF patients, with high-risk characteristics
Dr. Bernier noted that a pivotal RCT is needed to for the Generation 2 V-Wave support the previous findings, and will begin soon to evaluate the safety and efficacy in patients with HFrEF and HFpEF.
There was a question related to diastolic dysfunction. Dr. Bernier noted that a paper in the Lancet reported that REDUCe Elevated Left Atrial Pressure in Patients with Heart Failure (REDUCE LAP-HF) trial on a small number of patients with HFpEF implanted with the interatrial shunt device (IASD, Covia Medical) showed improvement. An improvement in wedge pressure was demonstrated with exercise. An RCT for diastolic dysfunction involving that device is expected to begin soon.
To address the question of a device implanted in the coronary sinus Dr. Bernier indicated it is unknown how a CRT device might interact with the coronary sinus shunting device and further study is required.
There was a question regarding the ideal pressure gradient for the pulmonary artery, and the ideal flow rate. Dr. Bernier explained that the shunt is a fixed size at 5 mm, and the gradient between the atria is generally very small. Patient inclusion criteria were arbitrarily chosen: > 5 mmHg and < 15 mm Hg. They don't know what the ideal gradient or flow across the device should be. There was a concern about the potential for shunting to contribute to right heart failure. Dr. Bernier indicated that patients with right heart failure were excluded.
It was noted that in the history of HF, there have been no studies in Class III patients as they have been so well managed with medical treatment (as well as ambulatory Class IV), therefore this device may only offer improvement for those that would improve anyway. Dr. Bernier noted that while not representing a definitive therapy, difference would be worthwhile.
There was a suggestion that the static difference may not be observable, but present at exercise (where wedge pressures are higher), which is typically not well studied. The patients with preserved EF are difficult to treat, as no medical therapy (diuretics) works well.
There was a concern that the V-Wave design has largely stayed the same and will protrude into the atria. Would this result into a tissue ingrowth that would close the shunt? Also, could the device be closed if need be? Dr. Bernier explained that the device could be plugged, or during LVAD implantation, could be closed with a patch. In addition there was a concern for thromboembolic events on the left side. Dr. Bernier indicated that potential for thromboembolic events is unknown. He noted that he has not closed a device, but that it could be done. There was a comment that nitinol mesh devices, like other atrial flow devices (e.g. Occlutech), are generally not associated with thrombus, but others may be.
There was a question whether the loss of the valve will cause any complications. It was noted that the presence of a valve is associated with more turbulent flow, and risks are likely fewer without a valve.
It was suggested that if this procedure works it would be a bridge to another therapy; it would buy more time for the patient.
Dr. Bernier concluded his presentation.
11. Next Steps, Closing Remarks and Adjournment of Meeting
Dr. John Ducas, Committee Chair
The Chair thanked committee members and invited speakers for their participation and valuable input. Members will be canvassed to select a date for the next meeting.
Report a problem or mistake on this page
- Date modified: