Khaldoun G. Tarakji and Bruce L. Wilkoff
CARDIAC PACING
Cardiac pacing is the only definitive therapy for symptomatic bradycardia. Whether iatrogenic, ischemic, or intrinsic conduction system disease is present, cardiac pacing can be a temporary bridge to recovery, a backup safety therapy, or a permanent therapy, depending on the clinical scenario. What follows is a review of major topics in cardiac pacing.
Indications
Indications for cardiac pacing vary with the clinical scenario. The major determinant of need for permanent pacing is the anticipated duration of the pacing indication. For example, symptomatic bradycardia associated with a toxic ingestion of a nodal blocking drug (e.g., digitalis) can be anticipated to resolve as the drug is cleared. Temporary pacing may be indicated in the short term, but a permanent device should not be needed. Alternatively, a transient neurocardiogenic (cardioinhibitory) bradycardic episode may resolve spontaneously, and temporary pacing should not be needed. However, if episodes recur on medical therapy to the point of causing recurrent syncope, a permanent pacemaker is indicated to protect the patient from subsequent syncopal episodes.
Temporary Pacing
In the emergency department setting, transcutaneous pacing can be used as a bridge to a more definitive transvenous temporary pacing system in the setting of symptomatic bradycardia of any etiology with hemodynamic compromise.
In the critical care setting, temporary pacing can be a life saving bridge to recovery or, further, to a definitive therapy for the underlying cause of bradycardia. The indications can roughly be divided into those related to ischemia, and all other categories.
Acute myocardial infarction can be associated with bradycardia due to either sinus bradycardia, which does not require therapy unless it is causing hemodynamic compromise, or due to AV block or intraventricular block. AV block can be (a) intranodal, which is usually associated with inferoposterior infarcts (right coronary artery [RCA] 90%, left circumflex artery [LCX] 10%), manifests as first degree or Mobitz I pattern, is usually transient with benign prognosis, and rarely requires temporary pacing and almost never requires permanent pacing, or (b) infranodal, which is usually associated with anteroseptal infarcts (left anterior descending artery [LAD]), manifests as Mobitz II or third- degree block, is usually transient but may persist, and carries a poor prognosis as it signifies extensive infarction; it often requires temporary pacing and if it persists permanent pacing.
In general, “high-degree” heart block such as Mobitz type II second-degree heart block and third-degree heart block warrant temporary pacing during the acute phase of anterior (LAD territory) infarcts or inferior (RCA territory) infarcts. Further, new bifascicular block or alternating bundle branch block reflects ischemia within the interventricular septum and warrants temporary pacing as a backup in case of progression to complete heart block. Refractory bradycardia in the setting of an infarct in any territory necessitates temporary pacing.
In the absence of an acute myocardial infarction, symptomatic bradycardia with or without AV dissociation and third-degree AV block with ventricular escape warrant temporary pacing. Backup pacing indications include temporary ventricular pacing during right heart catheterization in the setting of preexisting left bundle branch block (LBBB), new bundle branch block or AV block in the setting of endocarditis, and essential pharmacologic therapies that may induce or exacerbate bradycardia.
Temporary pacing systems with temporary epicardial atrial and ventricular wires are routinely used in the setting of open heart surgery. These systems are used to optimize cardiac output coming off cardiopulmonary bypass, and subsequently as a backup system in case AV nodal conduction block occurs postoperatively, especially in the setting of valvular heart surgery.
Permanent Pacing
The indications for permanent pacing are listed in detail in the ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices). This document is summarized in Table 31.1.
TABLE
31.1 Indications for Permanent Pacemakers
AV atrioventricular; MI, myocardial infarction; SND, sinus node dysfunction; VT, ventricular tachycardia; LV left ventricular; EPS, electrophysiology study; CRT, cardiac resynchronization therapy; NYHA, New York Heart Association; HV Histo ventricular conduction time; DCM, dilated cardiomyopathy; LVEF, left ventricular ejection fraction; HOCM, hypertrophic obstructive cardiomyopathy.
From Writing Committee Members Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices). Circulation. 2008;117:2820-2840, with permission.
Device Features
Single-chamber devices that only pace the ventricle or atrium have fallen by the wayside in favor of more sophisticated atrioventricular pacing devices that have the ability to track the sinus node rate when appropriate and pace the ventricle after a set delay. They also switch modes to ventricular backup pacing when the atrial signal falls outside of set parameters, as in paroxysmal atrial fibrillation and sick sinus syndrome. Further, cardiac resynchronization therapy (CRT) with biventricular pacing has a significant beneficial role in patients with symptomatic heart failure (New York Heart Association [NYHA] Class III and ambulatory Class IV) and evidence of left ventricular dysynchrony (EF ≤ 35% and wide QRS duration ≥120 milliseconds). Despite the predominance of dual-chamber pacemakers, there are increasing data suggesting that right ventricular stimulation increases the incidence of heart failure, hospitalization, and death in various patient subsets. However, if the ventricle needs to be stimulated, the vast majority of patients tolerate right ventricular stimulation, as dual-chamber stimulation is preferred over sole ventricular stimulation. Figure 31.1 shows a schematic of pacemaker timing cycles.
FIGURE 31.1 Schematic of important device timing cycles and impulses.
Rate-Adaptive Pacing
A variety of methods have been employed to allow for implantable pacemakers to increase their pacing rate in the setting of metabolic demand for increased cardiac output. The most commonly employed methods include activity sensors (vibration, acceleration), or minute ventilation sensors. Other sensors include peak endocardial oxygen sensors and right ventricle (RV) impedance-based sensors, which have the advantages of responding to nonexertional stimuli (emotions). These techniques utilize vibration, acceleration, minute ventilation, or other measurements as a surrogate for increased metabolic demand for oxygen delivery. In patients with chronotropic incompetence, or the inability to increase cardiac output in response to exercise, rate-adaptive devices can utilize these surrogates to increase the pacing rate and therefore increase cardiac output.
There are advantages and disadvantages to each type of sensor system. Vibration sensors and accelerometers provide an almost immediate rise in rate and therefore cardiac output when they detect activity. However, they can be “fooled” by stimuli external to the patient that mimics patient activity (i.e., turbulence during flight, etc.). The accelerometer tends to respond more specifically to patient activity than does the motion sensor. The advantage of the minute ventilation sensor is that it responds specifically to the patient’s respiratory rate—a parameter that is controlled by the brainstem. Although this parameter perhaps more reliably reflects the degree of patient exertion, it tends to lag behind the initiation of strenuous activity. Dual-sensor systems that utilize data collected from more than one sensor modality may actually be best suited for effective rate-adaptive pacing.
Mode Switching
Another feature of dual-chamber cardiac pacemakers that allows the devices to respond to changes in the physiology of the patient is mode switching. Mode switching is the ability of the device to revert to a separate, backup pacing mode in the event that the primary pacing mode no longer best serves the patient’s pacing need. For example, in a patient with AV nodal block, a dual-chamber device may be programmed to sense or track the patient’s intrinsic sinoatrial rate and to pace the ventricle after a set delay within the range of 60 to 120 beats/min (bpm). If the atria begin to fibrillate, the sensed atrial rate would exceed the rate parameter and the device would switch modes to a backup ventricular-only mode with a set rate sufficient to prevent hemodynamic compromise. If the patient reverted to sinus rhythm subsequently, the device would recognize the atrial rate back within the set parameter range and switch back to the primary mode, tracking the atrium and pacing the ventricle. Mode switching allows for the maximum responsiveness to the patient’s intrinsic rhythm. These devices are most commonly programmed DDDR and revert to VVIR during periods of high atrial rates.
Other Programmable Features
Modern pacemakers now include a myriad of programmable features to better match the patient’s physiologic status. They can be programmed to pace at a lower “sleep” rate during typical sleeping hours, with absence of strenuous activity confirmed by the devices metabolic sensing system. Programmable pulse width and output allow the programmer to optimize the impulse specifications to ensure capture while preserving battery power. Diagnostic information and event data including mode-switching data can be stored and retrieved later to assess for the presence and prevalence of atrial arrhythmias and other events. Atrial and ventricular electrograms can be obtained and stored. Device status data can also be retrieved, including battery usage and projected battery life given current settings.
Leads
The pacing leads conduct the electrical pacing impulse to the myocardium, and conduct the intrinsic electrical activity of the myocardium to the sense amplifiers within the device. Unipolar leads have a single electrode at their tip, and therefore they direct current from their tip to the can of the device through the patient’s tissues, or vice versa. For this reason, problems such as pectoral, intercostal, or diaphragmatic stimulation are more likely to occur, particularly in implants requiring higher outputs to capture the ventricle. Bipolar leads have two electrodes with close proximity at their tip and direct current proximal to distal or distal to proximal over much smaller distances. These leads can achieve capture of the myocardium with lower output energies and thus are more efficient. They are capable of unipolar function as well, but with the same limitations as standard unipolar leads. Of note, bipolar leads are by necessity larger and stiffer than unipolar leads, and have been historically more prone to mechanical failures than unipolar leads.
Coronary sinus leads are small, highly flexible unipolar or bipolar leads. They can be directed from the right atrium via the coronary sinus into a branch cardiac vein for the purpose of pacing the left ventricle in synchrony with the RV in patients with ventricular dysfunction and delayed intraventricular conduction, usually manifest as a LBBB.
Epicardial leads can be placed surgically using minimally invasive techniques or during open heart surgery for another indication and subsequently utilized instead of transvenous leads for standard pacing or more commonly for CRT (biventricular pacing). Often two leads are placed at the time of surgery and one of the two is subsequently utilized for biventricular pacing, depending on the thresholds and pacing characteristics of each at the time of device implant.
A variety of fixation techniques are utilized to maintain the contact of the lead tip with the myocardium. Active fixation leads employ a fixed extended or retractable screw to engage the myocardium. These systems allow for better localization of the lead tip at the desired site of implantation during deployment of the fixation helix. Passive fixation systems utilize plastic projections near the distal electrode that entrap in the trabeculations of the right ventricle or the right atrial appendage to maintain the position of the lead. As lead implants “mature” over time, pacing thresholds tend first to rise due to inflammation and then improve as healing continues and the inflammation resolves. Most leads have a small amount of steroid impregnated at the lead tip that reduces the size of the fibrotic tissue capsule and reduces the chronic thresholds.
Basic Concepts of Impulses and Timing
Following is a review of some basic concepts in pacemaker theory that are central to an understanding of the clinical application of pacing technology.
Stimulation threshold: The minimum amount of electrical energy that consistently produces a cardiac depolarization. The energy is a combination of voltage and pulse duration. It can be expressed in terms of amplitude (milliamperes or volts), pulse duration (milliseconds), charge (microcoulombs), or energy (microjoules).
Voltage output: The amount of voltage being delivered to the heart every time the pacemaker emits a stimulus. It is expressed in volts.
Pulse width (or pulse duration): The length in milliseconds the voltage is delivered to the heart.
Strength-duration curve: The hyperbolic relationship between the voltage output and the pulse width that defines the stimulation threshold (Fig. 31.2).
FIGURE 31.2 The strength–duration curve.
Sensing: Sensing occurs when the electrical wave front through the myocardium passes directly underneath the electrode
Atrial sensitivity: A programmed parameter that defines the largest signal that will be ignored by the device and thus determines which signals are detected by the pacemaker or implantable cardioverter/defibrillator (ICD) in the atrial channel. Atrial sensing in the dual-chamber pacing mode, DDD, will inhibit the atrial stimulus which would occur at the end of the atrial escape interval (V-to-A interval), initiate the AV interval, and trigger the ventricular output at the end of the AV interval.
Ventricular sensitivity: A programmed parameter that defines the largest signal that will be ignored by the device and thus determines which signals are detected by the pacemaker or ICD in the ventricular channel. Ventricular sensing in the dual-chamber pacing mode, DDD, will inhibit both atrial and ventricular stimuli that were scheduled to be output at the end of the atrial escape interval (atrial) or AV interval (ventricle) and initiate a new atrial escape interval (V-to-A interval).
Atrial oversensing: Sensing on the atrial channel that occurs due to signals on the atrial lead either related to signals originating outside the atrium, such as far-field ventricular signals, myopotentials from the pectoralis major muscle or diaphragm, or from noise originating from a dysfunctional lead (insulation or conductor fractures or a loose set screw). Depending on the mode of pacing, atrial oversensing will either inhibit or trigger atrial and/or ventricular stimuli.
Ventricular oversensing: Sensing on the ventricular channel that occurs due to signals on the ventricular lead either relating to signals originating outside the ventricle, such as myopotentials from the pectoralis major muscle in a unipolar lead system or from lead dysfunction secondary to insulation or conductor fracture or loose set screws. Sometimes the ventricular channel will oversense the atrial paced output and inhibit the ventricular output. This is called crosstalk inhibition and is usually prevented by a blanking of the ventricular sensing amplifier during the atrial paced outputs.
Chronotropic competence: The ability to match cardiac output to the metabolic needs of the body by appropriate modification of the heart rate.
Minimum rate: Also called the escape rate, this is the slowest rate at which the pacemaker will allow the heart to beat. The minimum paced rate is calculated by the ventricular paced or sensed event to atrial paced output interval plus the programmed A–V delay measured in milliseconds and converted to rate by dividing 60,000 by that sum.
V–A interval: Also called the atrial escape interval, this is calculated by subtracting the paced AV interval from the minimum rate interval. It is initiated by a paced or sensed ventricular event and concludes with a paced atrial event or is interrupted by either an atrial or ventricular sensed event.
A–V delay: This programmed interval is initiated by an atrial sensed or paced event and is terminated with a ventricular paced stimulus unless interrupted by a ventricular sensed event (either a conducted beat through the AV node or a premature ventricular beat). Often AV delays initiated by sensed atrial events are programmed to be shorter than AV delays initiated by atrial paced events.
Upper rate limit: The fastest rate at which the ventricular channel can track intrinsic P waves or, in the case of rate-adaptive pacing on the basis of a sensor, the fastest rate at which the ventricular channel can track the sensor rate algorithm. The atrial tracking or upper rate limit is constrained by dividing 60,000 by the sum of the sensed AV delay and the postventricular atrial refractory period (PVARP).
PVARP: This is the Post-Ventricular Atrial Refractory Period. The PVARP is the timeframe during which the atrial channel is refractory after either a paced or sensed (R wave) ventricular event. Its purpose is to prevent atrial sensing and tracking of any V–A (retrograde) conduction of ventricular events to the atrium that would trigger a pacemaker-mediated tachycardia (see later).
Programming
Device programming has become more complex as dual- chamber pacing systems have become ubiquitous and biventricular pacing is becoming common place. It is important to note that the basic parameters discussed above can usually be derived with caliper measurements of the intervals observed on a 12-lead electrocardiogram (ECG) or rhythm strip. Following is a concise review of programming codes and timing cycles that provide the underpinnings of device programming.
Codes
A standard coding system has been adopted to delineate the basic settings of the device as follows: The first designation is the chamber paced, the second designation is the chamber sensed, the third designation is the device response to a sensed event, and the final designation reflects the rate-adaptive status of the device. There is a fifth position in the code that was originally used to indicate the antitachycardia response that the device will provide during a tachycardia, but now indicates whether multisite pacing is present or not. Table 31.2 reviews the mode codes in detail.
TABLE
31.2 Mode Codes for Cardiac Pacemakers
As an example, a DDIR pacemaker can pace in both the atrium and the ventricle, and sense activity in both the atrium and the ventricle. Further, it will inhibit upon sensing intrinsic activity, and it has rate-adaptive functionality as well. A VOO device will pace the ventricle asynchronously, without sensing intrinsic activity.
Timing Cycles
When a dual-chamber device paces the atrium, the ventricular channel is blanked for a period of 20 to 40 milliseconds as a safety feature to prevent inhibition of the ventricular channel by far-field (ventricular) sensing of the atrial paced output. The blanking period prevents “crosstalk inhibition,” which could cause, in patients with complete lack of AV conduction, a string of atrial paced events and ventricular asystole. After the blanking period, the ventricular channel is open to sensed events. Typically, the first 100 milliseconds is the safety alert period. If, during this alert period a sensed event occurs, then the AV interval is abbreviated, usually to 120 milliseconds. This abbreviated AV delay is designed to prevent pacing during the vulnerable period of the ventricle, for instance, when the sensed event is caused by a premature ventricular depolarization. During the remainder of the AV delay (after the blanking and safety alert period), any sensed ventricular event will cause the ventricular output to be inhibited and reinitiate the atrial escape interval. If by the end of the programmed AV interval no event has been sensed, the device will pace the ventricle. After every paced or sensed ventricular event, a PVARP is initiated. During this period, the atrial channel is refractory to detecting atrial activity. The purpose of the PVARP is to prevent detection of atrial activity produced by retrograde conduction through the AV node. Without making the atrium refractory to retrograde atrial events (V-to-A conducted beats) an endless loop cycle can be set up that continues until the retrograde conduction fails. This endless loop tachycardia is one of several types of pacemaker-mediated tachycardia (see Fig. 31.1 for a schematic of pacemaker timing cycles in comparison to the surface ECG).
Diagnostics
Modern pacing devices are capable of storing tremendous amounts of data and reporting data in a variety of usable formats. Following is a brief review of device diagnostics and their applications.
Histograms. Histograms are a statistical report of a parameter describing the relative frequency of an event relative to time, heart rate, or another parameter. Histograms do not correlate symptoms to specific events, and from them one can only infer cause and effect (Fig. 31.3).
FIGURE 31.3 Histogram: ventricular hysteresis.
Trends. Trends evaluate the progression of a parameter over time. They are not a statistical representation but instead describe the correlation of an activity over time with symptoms. Trends can document concurrence of patient and rhythm events if interrogated quickly after the event occurs. Trends require extrapolation to connect patient and rhythm events. (See Fig. 31.3).
Event monitoring: Event monitoring captures an exact record of an event as characterized by electrograms, marker channel, and intervals. These monitored records are not statistical reflections of data but the actual recordings. Therefore, they can capture the relationship of symptoms and objective data. They require neither inference nor extrapolation (see Fig. 31.3).
Troubleshooting and Complications
Device troubleshooting most often involves interrogation of the device and adjustment of the pacing mode or timing cycles in order to optimize device function. Further, device interrogation using a programmer can reveal diagnostic information about the integrity of the leads, status of the battery, and performance of the device’s algorithms, including the behavior of the rate-adaptive sensor function. A review of some specific device troubleshooting issues follows.
Endless Loop Tachycardia
Endless loop tachycardia is a type of pacemaker-mediated tachycardia specific to dual-chamber devices programmed to the VDD or DDD mode. Endless loop tachycardia is triggered by the atrial channel sensing retrograde conduction of a paced ventricular impulse. In response to the sensed event, the ventricle is paced again after the set AV interval, and retrograde conduction to the atrium recurs. As the atrium senses the retrograde V–A signal, the cycle begins again. The phenomenon is terminated by either applying a magnet to the device, thus reverting the device to nominal asynchronous pacing, or by reprogramming the device to lengthen the PVARP so that the atrial channel is refractory during the retrograde (V–A) conduction.
Pacemaker Syndrome
The so-called pacemaker syndrome is a constellation of physical symptoms and signs associated with loss of AV synchrony, most commonly associated with VVI pacing. Affected patients suffer weakness, dizziness, light-headedness, dyspnea on exertion, and sometimes even orthopnea and dyspnea at rest, independent of their underlying ventricular function. The symptoms result from ventricular pacing, typically with retrograde atrial conduction, which produces atrial contraction against a closed AV valve. The decrease in efficiency associated with loss of atrial kick as well as the increased back pressure within the pulmonary circuit both contribute to the symptomatology. Similar symptoms and physiology can result from atrial pacing with delayed AV conduction. The result is also related to atrial contraction against a closed AV valve. The treatment for pacemaker syndrome is device upgrade to a dual-chamber device. An atrial tracking ventricular pacemaker eliminates the physiologic underpinnings of pacemaker syndrome and typically alleviates the symptoms.
Lead Fracture/Failure
Lead fracture is the term used to describe failures in the integrity of the lead wires, insulation, and/or coil. Fractures often occur at the ingress of the lead into the thorax within the subclavian vein as it passes between the clavicle and the first rib, particularly at the suture sleeve, due to tight ligatures or a sharp angulation of the lead in the pacemaker pocket. Crush injuries and chronic abrasion at this site are common etiologies of lead fracture. Disruption of the insulation causes a reduction of the pacing impedance and is often manifest by intermittent oversensing and either failure to produce a paced output or failure to capture the heart. Disruption of the lead conductor causes an increase in the pacing impedance and can also be manifest by intermittent oversensing and failure either to produce a paced output or to capture the heart. After the ECG, the chest x-ray is often the first diagnostic modality to reveal a lead fracture. Device interrogation typically suggests the diagnosis (abnormally high or low lead impedance, as noted above).
Infection/Erosion
Device infection occurs most commonly from bacterial contamination at the time of device implantation. Most infections do not present within the first month after implantation but are manifest within the first 2 years after implantation. Some infections can be indolent and persist for years before becoming apparent. The most commonly responsible organisms are Staphylococcus species, with gram-negative organisms occurring predominantly in diabetic patients or those otherwise immunocompromised. Physical findings associated with device infection may range from normal pocket appearance to mildly erythematous overlying tissue, to a swollen, boggy pocket and incision line. Occasionally a device will erode through the skin in the setting of a chronic device infection. When the pocket appears normal, the infection is typically endovascular and is unmasked by the presence of fevers and positive blood cultures along with supportive findings from transesophageal echo or chest CT scan. Less commonly, device infection can occur secondary to bacterial endocarditis or other bloodstream infection. Vegetations can sometimes be observed on the leads, most commonly utilizing transesophageal echocardiography.
The treatment for device infection with or without endocarditis is with antibiotics and device and lead extraction. The replacement device cannot be reimplanted at the time of device and lead extraction but should be delayed for few days or longer until it is deemed appropriate from the infectious disease stand point. Not all patients require immediate reimplant after extraction and the indication should be reinvestigated prior to reimplant. The reimplant is usually performed on the contralateral side.
Extraction
The most compelling reason for lead extraction is device–related infection, either localized to the pocket or endovascular with associated bacteremia or endocarditis. Multiple leads can compromise the venous flow, risking subclavian or superior vena cava (SVC) occlusion with symptoms, or prevent the addition of leads for upgrade to an ICD or BiV system. Lead extraction can range from simply applying traction to a recently implanted lead, to the use of mechanical, electrosurgical, or excimer laser extraction sheaths to facilitate the removal of fibrosed leads from the endovas cular surface. Lead extraction using these devices can be complicated by serious bleeding complications l eading to tamponade and even death, and are thus best relegated to experienced operators in large volume centers.
IMPLANTABLE CARDIOVERTER- DEFIBRILLATORS
ICDs initiated a new era in the treatment of ventricular tachyarrhythmias. In contrast to modern devices, the early ICD systems were implanted via thoracotomy with epicardial placement of defibrillating patches and sensing electrodes and abdominal implantation of the device can. The devices themselves had no programmability, no significant diagnostics, and an abbreviated battery life of about 1 year. Patients requiring permanent pacing had to have a separate pacing device implanted. Over time, ICD components have been reduced in size, allowing for prepectoral implantation with transvenous leads, possess bradycardia and antitachycardia pacing (ATP), and hundreds of programmable parameters, diagnostics, and event storage, all while device longevity has expanded to 5 to 7 years. The devices now have full pacing and resynchronization capabilities as well.
Indications
Indications for implantation of ICDs have expanded greatly over the past decade, based on data collected from the major ICD trials and are summarized in Table 31.3. It is very important to note that all these indications require a reasonable expectation of survival with an acceptable functional status for at least 1 year. ICDs are not indicated for patients with incessant ventricular tachycardia (VT) or ventricular fibrillation (VF), NYHA Class IV patients with drug refractory heart failure who are not transplant candidates or candidates for a CRT device, patients with significant psychiatric illnesses, when VF or VT is amenable to surgical or catheter ablation (e.g., atrial arrhythmias associated with Wolf–Parkinson–White (WPW) syndrome, right ventricular outflow tract (RVOT) or left ventricular outflow tract (LVOT), idiopathic VT in the absence of structural heart disease), or patients with ventricular arrhythmias due to a completely reversible disorder in the absence of structural heart disease.
TABLE
31.3 Indications for ICDs
VF, ventricular fibrillation; VT, ventricular tachycardia; EPS, electrophysiology study; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; HCM, hypertrophic cardiomyopathy; SCD, sudden cardiac death; DCM, dilated cardiomyopathy.
The Center for Medicare and Medicaid Services (CMS) published updated guidelines in 2005 for reimbursement for ICD implantation, recognizing the data from the primary and secondary prevention trials of ICDs with and without capacity for cardiac resynchronization. The practical (reimbursed) indications for ICD implantation are listed in Table 31.4.
TABLE
31.4 CMS-Approved (Reimbursed) Indications for ICD Therapy
VF, ventricular fibrillation; VT, ventricular tachycardia; EPS, electrophysiology study; MI, myocardial infarction; HCM, hypertrophic cardiomyopathy; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention; DCM, dilated cardiomyopathy.
Providers must be able to justify the medical necessity of devices other than single-lead devices.
Devices
The current generation of ICDs includes single-chamber VVI devices, dual-chamber devices with fully programmable pacing capabilities, and CRT devices capable of biventricular pacing and antitachycardia therapies. As technology has progressed, these devices have become smaller, with better longevity and greater programmability. The diagnostics available allow for extensive troubleshooting and event monitoring.
Lead Systems
Typically, single-chamber ICD systems are implanted using active- or passive-fixation multipolar leads with shocking coils that lie in the RV apposed to the endocardium as well as the SVC. With this configuration, the device can deliver energy from the RV (+) coil to the SVC (-) coil or vice versa, or from either coil (+) to the ICD can (-) itself or vice versa. Various investigations have demonstrated that defibrillation thresholds (DFTs) can be reduced using optimal polarity and an “active can” configuration.
ICDs can be attached to epicardial leads or patches implanted during surgery. These leads are typically placed using minimally invasive techniques for patients with high DFTs or during open heart surgery performed for other indications. Subcutaneous arrays and even azygous vein leads can be placed. Virtually all ICDs are implanted with transvenous in lieu of surgically placed leads.
Dual-chamber devices possess an atrial lead in addition to the ventricular shocking coil lead. The atrial lead is typically a standard bipolar pacing lead without a shocking coil and plays no role in defibrillation.
Implantation
The most common site for device implantation is the left prepectoral space. This site gives access to the left subclavian vein for transvenous lead placement. Especially in “active can” configurations, this site of implantation allows for lower DFTs as the path for energy transmission from can to coil or vice versa traverses the LV myocardium. In the case of prior device infection, scarring, subclavian stenosis, or mastectomy on the left side, the right prepectoral space may be used. Lead implantation technique is much like that used for standard pacing lead implantation; however, there is an impetus to implant the lead tip at the RV apex so that the RV shocking coil rests completely within the right ventricle.
Device Function
Detection
ICDs have a variety of programmed routines designed to aid in the detection and verification of VT and VF, and to minimize the number of inappropriately delivered therapies. The device must be able to sense low-amplitude high-rate signals in VF, while not oversensing far-field atrial activity or ventricular repolarization. Appropriate sensing thresholds must be achieved at the time of implantation, or the device cannot be relied on to appropriately detect and treat malignant ventricular arrhythmias.
Detection algorithms utilize counters, and detection criteria are based on signal counts registered faster than the tachycardia threshold criterion programmed into the device. For example, if an ICD is programmed to detect VT at cycle lengths of less than 400 milliseconds and the device senses consecutive R waves with a cycle length of 390 milliseconds, it begins to count consecutive R waves until it reaches the programmed detection criterion, perhaps 15 beats. If the device detects 15 consecutive R waves with cycle length <400 milliseconds, it registers a VT event and administers therapy. VF is a more unstable arrhythmia, with shorter cycle lengths, and smaller and more variable wavelet amplitudes. The device cannot be assured of sensing consecutive signals to meet the VF criterion, so the criterion is often programmed to detect VF if 15 of 20 R waves are detected with a cycle length below the VF threshold cycle length.
Single-chamber devices use these cycle length criteria in addition to analyzing the suddenness of arrhythmia onset, the duration and persistence of the arrhythmia, and the morphology of the sensed R waves. Dual-chamber devices have the advantage of being able to compare ventricular sensed activity to atrial sensed activity, so rates and relationship of A to V can be compared in the detection criteria. Further, dual-chamber devices can detect AV dissociation. So the detection algorithms can be more sophisticated and potentially more accurate in the detection of VT requiring therapy and the discrimination of SVT or AF not requiring ICD therapy (Fig. 31.4).
FIGURE 31.4 Device diagnostics: histograms, trends, and event monitoring.
For example, a dual-chamber device set to apply VT therapies at ventricular cycle lengths of 400 milliseconds or less may detect VT as in the previous example while at the same time detecting atrial signals with a cycle length of 200 milliseconds. Recognizing the atrial tachycardia (flutter) and the 2:1 ventricular response, the device monitors but does not “detect and treat” VT. If, in the same example, the device senses atrial signals with a cycle length of 400 milliseconds, it monitors the sinus tachycardia or SVT but does not deliver therapy for VT.
Therapies
The therapy for VF is defibrillation upon detection with consecutive high-energy shocks pending redetection until the arrhythmia is terminated. Upon meeting the detection criteria, the device begins to charge its capacitor to the programmed output for the initial shock. Upon completion of capacitor charging, the device then rechecks for the presence of the arrhythmia and if present, it delivers the shock. After the initial shock, the device monitors for arrhythmia meeting criteria and if present, it charges again, typically to a higher or maximum output. If after charging the arrhythmia persists, the device again delivers therapy. The cycle continues until the arrhythmia is terminated.
Therapies for VT include low-energy synchronized cardioversion as well as ATP. The advantage of ATP is that it is painless and is not often perceived by the patient. When an ATP device detects VT, it can deliver a programmed burst of pacing impulses at a cycle length just shorter than the detected arrhythmia in an attempt to interrupt the reentrant VT circuit. After the burst, the device monitors for persistence of the arrhythmia. If VT persists, further bursts of ATP can be delivered, followed if necessary by low- or high- output cardioversion. ATP bursts can be programmed at a fixed cycle length representing a percentage of the VT cycle length, or at a progressively shorter (accelerating) cycle length for a programmed number of pulses. The number of ATP attempts prior to administration of shocks can be programmed too. Low-output cardioversion shocks are typically synchronized to the intrinsic R wave of the VT. Based on a variety of investigations, ATP is not inferior to low-energy cardioversion in terms of efficacy, and because it is painless, it has become the preferred therapy for “slow VT.” In addition, recent data have documented that faster tachyarrhythmias, between 200 and 250 bpm, can be successfully pace terminated approximately 50% of the time.
Waveforms
The shock waveform for VT and VF is the same: a prolonged (relative to a pacing impulse) biphasic waveform lasting 5 to 20 milliseconds. The waveform is a truncated exponential decay voltage wave with a drop from the initial voltage to the trailing-edge voltage, called the tilt. A typical tilt is a 65% reduction of the voltage at the end of the first phase of the biphasic pulse. Then the capacitor polarity is reversed, producing a leading-edge negative voltage for the second phase equal to the trailing-edge positive voltage of the first phase. The second phase also has a tilt and is truncated after a few milliseconds. Biphasic waveforms with second phases equal to or shorter in duration than the first phase have been associated with significantly lower DFTs as compared with monophasic waveforms. Thus, all current production ICDs utilize a biphasic waveform (Fig. 31.5).
FIGURE 31.5 ICD detection: dual-chamber discrimination and event report.
Polarity
Modern ICDs have the programmability to add or subtract various electrodes from the circuit (can or SVC electrode) or change the initial (positive or negative) polarity of the leads and the can, as well as the polarity of the waveform. Polarity changes are sometimes undertaken to reduce DFTs in patients with high DFTs.
Programming
Device programming for ICDs involves programming pacing modes as well as detection parameters and therapies for VT and VF. Modern ICDs have full pacing capabilities and, depending on the device and indication for implantation, may be programmed for ventricular backup pacing, dual-chamber tracking and pacing, and even resynchronization pacing. The following review focuses on the antiarrhythmia features of ICDs. Refer to the pacing section of the chapter for more on pacemaker programming.
Detection Criteria
Therapies for VT and VF are administered only after detection criteria are met. Therefore, the criteria programming is critical for optimal device function. The concepts of arrhythmia detection have been presented previously. What remains is the actual interface programming between the device and the physician. As a practical matter, the detection criteria sets are typically divided into VT parameters and VF parameters. The VT parameters may further be broken down into “slow” VT and “fast” VT zones, based on the premise that slower VT may be more amenable to painless therapies (ATP) and faster VT may be more prone to acceleration or failure of ATP to convert the rhythm back to the baseline rhythm. Furthermore, “monitor zones” can be established so that rhythms in a given rate zone can be recorded as “events” and retrieved later for analysis. So the task of the device programmer is to match the detection criteria and therapies to the anticipated needs of the patient.
For example, if an 80-year-old patient with prior myocardial infarction and LV dysfunction has a device implanted after a documented VT episode with a VT cycle length of 380 milliseconds, and the patient is now being treated with amiodarone and long-acting beta-blockers, the physician programmer may opt for VVI backup pacing at 40 bpm, a “slow VT” zone of 400 to 320 milliseconds, and a “fast VT” zone of 319 to 290 milliseconds, with a VF zone of anything <290 milliseconds. Therapies may include three attempts at ATP at 81% of the VT cycle length, followed by 20, 30 J, maximum output shocks if ATP fails to convert the “slow” VT. The fast VT zone may be programmed for 20, 30 J, maximum output for six shocks, and the VF zone may be programmed likewise, 20, 30 J, maximum output for six shocks.
Now suppose the patient goes home and comes back to the Emergency Department with “weakness” but has received no shocks as far as he can tell. Telemetry reveals NSR 60 bpm. Interrogation of the device reveals normal function and no recorded events. Perhaps the patient is having VT that has now slowed below the detection criteria as a result of the addition of negative chronotropes and antiarrhythmic drugs.
Another patient with a similar profile but no history of arrhythmia may have a single-chamber device implanted for primary prevention. In this case, the physician programming the device may simply try to protect the patient from any arrhythmia reasonably anticipated to be inappropriately fast and hemodynamically unstable, and set a single zone below 320 milliseconds with six maximum output shocks.
Therapies—Antitachycardia Pacing
ATP as described previously is typically programmed to be administered in a burst of constant cycle length impulses or as a “ramp” of decreasing cycle length impulses. Typically, the device is programmed to initial ATP at a cycle length of 80% to 85% of the arrhythmia cycle length. No benefit has been demonstrated of “ramp” ATP over static cycle length ATP, and both achieve termination of VT in up to 90% of attempts. Typical bursts are 8 to 12 impulses, with reapplication of detection criteria after the burst to redetect persistent arrhythmias. Posttherapy criteria are often less stringent than initial detection criteria so as to shorten time to redetection and retreatment of persistent arrhythmias. The programmer decides the number of attempts at ATP prior to reverting to a cardioversion strategy, but typically several attempts at ATP are made before administering shocks in an initial program. Rates of acceleration of VT are low, in the 1% to 3% range, but are quite variable among patients, among cycle lengths, and among morphologies of tachycardias within a single patient.
Therapies—Cardioversion and Defibrillation
Low-energy synchronized cardioversion may be programmed for detected VT with outputs typically between 5 and 20 J. These synchronized therapies are preferred because they are effective, have shorter charge times, and are less likely to cause VF via a R-on-T mechanism, thus preventing some syncopal events. If the ATP or low-energy shocks are unsuccessful or accelerate the rhythm to VF, then the device delivers high-energy synchronized shocks. Devices can be programmed to deliver five or six distinct therapies in sequence, each often more aggressive than the previous therapy. Therapies delivered at or above the DFT have a high probability of converting the rhythm back to baseline, with repeat therapies sometimes necessary to convert successfully.
Troubleshooting
High Defibrillation Thresholds
Because delivered therapies convert the malignant rhythm as a probability function based on delivered energy in excess of the defibrillation threshold or DFT, it is important to estimate the DFT at the time of device implantation. Initial device therapies are typically programmed with a margin of safety above the estimated DFT to increase the chance of conversion with the first shock delivered. A variety of scenarios can lead to high DFTs, but they can be divided into device-related and patient-related categories.
Device-related causes of high DFTs may include inappropriate lead positioning at implantation or subsequent dislodgement of the lead. Loose header screw or lead failure/fracture may result in high-impedance failure of defibrillation. Inappropriate shocking vector, as in the case of an active can system implanted in the right prepectoral pocket, may result in unacceptably high DFTs. In this case, a subcutaneous shocking electrode on the left side or an azygous vein shocking coil could be implanted and the system reprogrammed to shock from RV to azygous or vice versa or from RV to subcutaneous array or vice versa. Waveform morphology, polarity, and tilt may also be reprogrammed if the device nominal setting leads to high DFTs.
Patient-related characteristics that may lead to higher DFTs include the use of drugs that may increase the DFT, including Class I agents, nonselective beta-blockers, nondihydropyridine calcium antagonists, and especially amiodar-one. Hypoxia and ischemia may both lead to refractory VF and failure of internal and even high-energy external shocks, so it is imperative that these clinical parameters be treated and optimized prior to elective DFT testing. A potential procedural complication, pneumothorax, may affect DFTs in active can configurations when air is interposed between the heart and the device. Recognition of this phenomenon is critical so that the situation is remedied prior to repositioning of leads or addition of extra coils or arrays. Finally, multiple prior attempts at defibrillation may raise DFTs during subsequent attempts within a brief period of time. Retesting hours to days after implantation may reveal lower DFTs than initially observed at implantation.
Evaluating Inappropriate Shocks
The evaluation of an ICD shock begins with an appropriate history and physical examination focusing on the circumstances of the discharge in question and the integrity of the device implantation and the patient’s cardiopulmonary status. The history should assess for antecedent anginal and presyncopal symptoms, dyspnea, nausea/vomiting/diarrhea, and other potential causes of electrolyte imbalances as well as external factors such as proximity to sources of electromagnetic interference (EMI). The physical examination should assess for decompensated heart failure, rate and regularity of rhythm, trauma to the device, or the anatomic location of the leads (often beneath the clavicle).
Interrogation of the device is paramount, and evaluation of stored events and electrograms should reveal the nature of the episode during which the therapy was administered. If the therapy was appropriate, one should determine whether it was successful and assess potential reasons why subsequent therapies may have been required. If the therapy was inappropriate, a determination should be made as to whether it was in response to a conducted supraventricular arrhythmia, a far-field oversensing of myopotentials or atrial arrhythmias, or noise from lead fracture or failure.
Failure to Detect
The most common cause of failure to detect ventricular arrhythmias is inappropriate programming of tachycardia zones and detection criteria. Very often a device is implanted and antiarrhythmic drug therapy is initiated at the same time. Subsequent symptomatic arrhythmias may occur at rates lower than observed prior to implantation and initiation of drug therapy, as a consequence of drug therapy. The resulting scenario is that of an implanted device blinded to the culprit arrhythmia because of the programmed tachycardia zone. Ventricular arrhythmias can therefore be slowed into a rate range where physiologically normal tachycardia may occur. An example might be an athlete with an ICD implanted for symptomatic Brugada syndrome. The patient could potentially have physiologic sinus tachycardia below the 400- to 380-millisecond range, but could also develop VT with a similar cycle length. Detection algorithms utilizing atrial channel activity would be imperative in discriminating malignant ventricular arrhythmias in this patient.
ICD Management during Surgical Procedures and MRI Scanning
The ICD detection algorithms can be “spoofed” by high- frequency signals such as those delivered during electrocautery use intraoperatively. As a result, inappropriate therapies could be delivered by the device during a surgical procedure. Device detection can be turned off through the use of a device programmer or a magnet applied to the implant site, so long as telemetry monitoring and external defibrillation are available during the procedure.
MRI scanners can be a source of EMI in addition to inactivating device therapies while the patient is inside the scanner. This vulnerability of the devices to MRI effects makes them a contraindication to perform these studies. However, sometimes the importance of MRI could outweigh the risks, but this should be assessed on a case-by-case basis and under controlled and monitored situations. New pacemaker devices that are MRI compatible have been recently approved for clinical use and postmarketing evaluation of these devices is of utmost importance.
SUMMARY OF IMPORTANT ISSUES FOR THE BOARD EXAMINATION
The issues discussed in this chapter are the technical aspects of pacemaker and defibrillator therapy. All of the issues discussed are important for patient care, but for the examination, the indications for pacemaker and defibrillator implantation and the supporting multicenter clinical trials are of primary importance. In addition, pacemaker electrocardiography, which depends on understanding the basic timing cycles and intervals, is likely to be both important to the cardiologist in practice as well as in test material. In a parallel way, evaluation of the appropriateness of defibrillator therapy, ATP and shock therapies, determining the presence or absence of a ventricular or supraventricular arrhythmia, lead dysfunction, and the appropriateness of the programmed parameters, as well as the effectiveness of the therapy, are central to patient care and for the examination. A list of essential facts is provided as bulleted points.
Essential Facts
Ohm’s law: Voltage = current × resistance (V = IR).
All pacemaker intervals are initiated and terminated by a sensed event (usually silent to the ECG) or by a pacemaker output (usually apparent on the ECG).
Pacemakers make decisions on a beat-to-beat basis, on the basis of the current interval and not as a result of the heart rate. Therefore each beat and each interval must be analyzed separately.
Conversion of intervals to rate equivalents or back are done as follows:
Pacemaker magnets close the “reed switch”. Closing the “reed switch” will almost always disable sensing for pacemakers and cause the pacemaker to function at a fixed rate regardless of the intrinsic rhythm.
Pacemaker magnets, when placed over an ICD, will disable detection and therapy of tachyarrhythmias. This is most commonly only temporary, i.e., while the magnet is over the “reed switch”. However, in some devices, depending on the programming of the device, it can turn off arrhythmia detection and therapy until the device is reprogrammed.
SUGGESTED READINGS
Ellenbogen KA, Kay GN, Lau CP, et al. Clinical Cardiac Pacing and Defibrillation and resynchronization. 3rd ed. Philadelphia: Saunders; 2007.
Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices). Circulation. 2008;117:e350–e408.
QUESTION AND ANSWERS
Questions
1. The AV delay is 200 milliseconds and the time from a paced QRS to the next atrial paced event is 800 milliseconds. To what basic or lower rate has the pacemaker been programmed?
a. 100 beats/min (bpm)
b. 90 bpm
c. 80 bpm
d. 70 bpm
e. 60 bpm
2. The sensed AV interval is 150 milliseconds and the PVARP (postventricular atrial refractory period) is 350 milliseconds. What is the most rapid atrial rate that the pacemaker can track 1:1?
a. 300 bpm
b. 250 bpm
c. 200 bpm
d. 150 bpm
e. 120 bpm
3. A pacemaker-dependent 65-year-old woman says that her activity-sensing pacemaker programmed to the DDDR mode (unipolar) makes her heart race every time she sweeps the floor. Which of the following could remedy the patient’s situation?
a. Increase the rate response slope.
b. Program the atrial channel to bipolar paced configuration.
c. Program the ventricular channel to bipolar sensed configuration.
d. Decrease the atrial sensitivity from 1 to 3 mV.
e. Decrease the ventricular sensitivity from 1 to 4 mV.
4. The VVI pacemaker is programmed to a lower rate of 80 bpm. There are PVCs and usually the heart rate is paced at 80 bpm, but intermittently there are intervals between intrinsic R waves of 960 milliseconds. Which of the following could be the explanation for the electrocardiographic findings?
a. Paced bipolar impedance of 300 W
b. Hysteresis rate of 60 bpm
c. Sleep rate of 55 bpm
d. Paced unipolar impedance of 250 W
e. PVC response
5. Which of the following scenarios is NOT an indication for an implantable cardioverter-defibrillator (ICD) insertion?
a. A 45-year-old man with history of myocardial infarction 2 years ago, left ventricular ejection fraction (LVEF) 25%, New York Heart Association (NYHA) Class I
b. A 33-year-old woman, with nonischemic cardiomyopathy, LVEF 35%, NYHA Class III
c. A 65-year-old man with history of prior MI, who presents with syncope and found to be in incessant VT requiring IV lidocaine and multiple external shocks
d. A 69-year-old man with history of MI, LVEF 45%, who survived VF cardiac arrest that required external shock
e. A 55-year-old man with history of MI, nonsus- tained VT, LVEF 37%, with inducible sustained VT during EPS
Answers
1. Answer E: The cycle length consists of the AV interval + VA interval. These two intervals added together and converted to a heart rate yield the lower rate or base rate programmed for this pacemaker patient.
200 ms + 800 ms = 1,000 ms = cycle length
Base heart rate = 60,000/1,000 ms = 60 bpm
Note that the AV interval can be dynamically shortened (based on the sensor and/or the atrial rate) and there can be a shortening of the AV interval if there is a sensed P wave instead of an atrial paced beat. In addition, the rate of the pacemaker can be increased with apparent increases in the base rate if the sensor detects a need to increase the paced rate.
2. Answer E: The maximal rate at which a DDD pacemaker can track an atrial rhythm is limited by the shortest interval in which the atrium can be detected. By adding together the PV interval (the AV interval initiated by a P wave and terminated with a ventricular pacemaker output) and the PVARP (the time during which the atrium cannot sense another P wave after a ventricular sensed or paced event), the total refractory period can be calculated. That interval converted to a heart rate is the maximal rate at which the pacemaker can participate in producing a paced rhythm. Both the PV delay and the PVARP can vary on the basis of atrial rate and sensor rate, but in this example the intervals are fixed. Thus,
AV interval + PVARP = 150 ms + 350 ms = 500 ms
Converting this to heart rate, we see that the maximal paced rate = 60,000/500 milliseconds = 120 bpm. To increase the upper tracking rate it would be necessary to shorten the sensed AV interval, the PVARP, or enable rate-adaptive shortening of these intervals.
3. Answer D:.This woman is using her upper body, which has the potential to activate her activity sensor and to produce myopotentials from use of the pectoralis major muscle. Increasing the rate response slope will increase the heart rate related to her activity. Programming the atrium to bipolar paced configuration would not impact her heart rate but could help if her complaint was secondary to stimulation of the pectoral muscles. Programming the ventricle to bipolar sensed configuration would reduce the likelihood that the ventricular channel would detect myopotentials, because the muscle would no longer be within the antennae being sensed by the ventricle. Sensed events on the ventricular channel would inhibit ventricular output and could be responsible for syncope due to bradycardia. Decreasing the ventricular sensitivity would potentially decrease the likelihood that myopotentials would be sensed, but this would cause inhibition of the ventricular output and a decreased heart rate. Decreasing the atrial sensitivity from 1 to 3 mV will likely decrease the probability of sensing myopotentials on the atrial channel. These atrial sensed events would have been tracked to the ventricle and cause the patient to perceive a tachycardia rhythm.
4. Answer B: The rate of 80 bpm needs to be converted to an interval. Cycle length = 60,000/80 = 750 milliseconds. The interval of 960 is longer than the 750 (80 bpm). Longer intervals than the lower rate (base rate) of the pacemaker suggest (a) inhibition, (b) failure of output by the pacemaker (lead or generator related), or (c) an algorithm that explains the particular circumstance. A paced bipolar impedance of 300 Q is relatively low, but normal. If this represented a marked drop from previous measurements, then it is possible that there is a short within the pacing lead. The hysteresis rate of 60 bpm translates to a sensed escape interval of 1,000 milliseconds. This could explain the ECG as long as the paced intervals between ventricular events were 750 milliseconds. If the ECG findings occurred only at night, then intervals of 960 milliseconds would be normal, but would not explain the other paced intervals at 750 milliseconds. A unipolar pacing impedance cannot be too low to work. This is low but does not explain the findings. PVC responses can extend the PVARP to avoid initiating a pacemaker-mediated tachycardia but would not change the escape interval of the pacemaker.
5. Answer C: All the scenarios mentioned represent Class I indication for insertion of an ICD. Incessant VT is a contraindication to insertion of an ICD as this will lead to multiple ICD shocks. Ruling out any reversible causes (ischemia, electrolyte imbalances) and controlling the VT with medication or ablation procedure should be performed first prior to insertion of an ICD.