Chapter 59. Surgical Implantation of Pacemakers and Automatic Defibrillators

Pacemaker and defibrillator management is the subject of comprehensive reviews.1 In the author's experience with 2760 procedures, 23% of pacemaker recipients and 5% of implantable cardioverter defibrillator (ICD) recipients were octogenarians. The incidence of pacemaker insertion in patients older than 75 was 2.6% in a recent survey. The efficacy and cost-effectiveness of pacemakers are widely accepted, but the appropriate role for ICD insertion and biventricular pacing in older patients is still in evolution. Pacemaker and ICD technology is now applicable across the entire span of human age, with pacing for heart failure and ICD prophylaxis against lethal arrhythmias recent frontiers. Electrophysiologists now dominate these areas, reflecting decreased interest by thoracic surgeons and cardiology referrals. However, thoracic surgeons must maintain skills as implanters and consultants for complex or complicated cases. This chapter reviews practical information related to pacemaker and ICD insertion and management.

Device Description

A permanent pacemaker or ICD consists of leads2 and a generator. The generator contains a battery, a telemetry antenna, and integrated circuits. ICDs also include capacitors that store energy for high-output shocks. The power source is generally lithium iodide, but rechargeable and nuclear batteries have been used. The integrated circuits include programmable microprocessors, oscillators, amplifiers, and sensing circuits.3 The integrated circuits employ CMOS (complementary metal-oxide semiconductor) technology, which is subject to damage by ionizing radiation.4 Current pacemakers and ICDS monitor and report the status of internal components, external connections, programmed settings, recent activity, and notable arrhythmias. Unfortunately, each programmer controls only the devices of its manufacturer.

History

Early cardiac surgery was complicated by lethal iatrogenic heart block. Transthoracic pacing with Zoll cutaneous electrodes provided a solution.5 Percutaneous endocardial pacing (1959)6 and "permanent" pacemakers using epicardial electrodes (1960)7 followed. Advances in bioengineering and technology have dramatically improved the quality of life for recipients. Persistent problems include lead durability, inflammatory responses to pacemaker materials, infection, device size, programmer compatibility, and expense. Development of resynchronization therapy has made coronary sinus lead insertion an important technical skill.

Anatomy of Surgical Heart Block

The conduction system is vulnerable to injury during heart surgery. Complete heart block can result from suture placement during aortic, mitral, or tricuspid valve surgery or during closure of septal defects or during myotomy for idiopathic hypertrophic subaortic stenosis. These lesions are illustrated in Fig. 59-1. Infarction of the conduction system or inadequate myocardial protection can also result in surgical heart block.

International Pacemaker Code

A three-letter code describes the principal pacemaker functions (Table 59-1).8 The first letter is the chamber paced, the second the chamber sensed, and the third the algorithm integrating pacing and sensing functions. Fixed-rate ventricular and atrial pacemakers are VOO and AOO, respectively. Demand (rate-inhibited) pacers for the same chambers are VVI and AAI. VDD pacemakers pace only the ventricle, but sense both atrium and ventricle. DVI indicates atrial and the ventricular pacing, but only ventricular sensing. DDD is the most flexible of current designs. The suffix R after the three-letter code indicates rate responsiveness. Pacemakers capable of biventricular pacing or cardiac resynchronization therapy (CRT) are referred to as CRT-P. Biventricular pacemaker-defibrillators are called CRT-D.

Cellular Electrophysiology

Cell membrane depolarization and repolarization provide automaticity of the cardiac chambers and conduction system. The outside of the resting myocardial cell is positive, and the interior is negative. Unipolar pacing threshold is lowest when the negative terminal (cathode) of a pacemaker is connected to the heart and the positive terminal (anode) is connected to ground. Electrogram amplitude is unaffected by polarity.

Rhythm Disorders

Indications for Pacemaker Insertion

Pacemaker insertion guidelines (Table 59-2) are periodically updated. Documentation may be required for billing. Indication categories are "accepted," "controversial," or "not warranted." Profound sinus bradycardia or symptomatic second- or third-degree heart block indicates pacemaker insertion. Sinus bradycardia justifies pacemaker insertion if contemporaneous symptoms are documented by Holter or other means. Pacemakers may be appropriate for bradycardia less than 40 bpm if long-term necessary drug therapy is needed for supraventricular arrhythmias, ventricular tachycardia, hypertension, or angina. Although new indications for pacemaker therapy may be supported by clinical research, recognition of such advances by the Food and Drug Administration, insurance carriers, and regulatory agencies is often slow. Electrophysiology studies help define proper treatment.9 In 2008, an American College of Cardiology/American Heart Association/North American Society for Pacing and Electrophysiology (ACC/AHA/NASPE) task force released revised recommendations for pacemaker and ICD insertion. Any practitioner who implants arrhythmia control devices should be familiar with these guidelines.10

Atrioventricular Block

First-degree block is prolongation of the P-R interval beyond 200 milliseconds. First-degree block at a low atrial rate may progress to Wenckebach as the atrial rate increases. Second-degree block is incomplete dissociation of the atrial and ventricular rates, with increasing P-R intervals and dropped beats (Wenckebach and Mobitz I, usually atrioventricular nodal block), or frequently dropped beats without progression of the P-R interval (Mobitz II, usually in the His-Purkinje system).11 Third-degree block is complete atrioventricular (AV) dissociation, the atrial rate usually exceeding the ventricular rate. Left and right bundle-branch blocks and left anterior and posterior hemiblocks are partial conduction system blocks detected by electrocardiogram. Etiologies of AV block include ischemic injury, idiopathic fibrosis, cardiomyopathy, iatrogenic injury, AV node ablation, Lyme disease, bacterial endocarditis, systemic lupus erythematosus, and congenital lesions.

Sinus Node Dysfunction

Whether sinus node dysfunction warrants pacemaker insertion and whether bradycardia is caused by drugs required for ancillary conditions depend on the symptoms. Sinus node dysfunction is caused by coronary artery disease, cardiomyopathy, and reflex influences.

Reflex problems include carotid sinus hypersensitivity, vasovagal syncope, and oddities such as micturition-induced and deglutition syncope.12 Cardioinhibitory (asystole >3 seconds) and vasodepressor (marked fall in blood pressure despite adequate heart rate) components of reflex-mediated syncope are recognized. Medical therapy is favored for vasodepressor syncope.13 Tilt table testing provides objective data. Decisions on pacemaker insertion are based on symptoms and the duration of asystole. Pacemaker insertion is recommended for asystolic intervals greater than 3 seconds. Dual-chamber pacing (DDD or VDD) is favored for these patients, because AV synchrony increases stroke volume and decreases symptoms.

Features of Permanent Pacing

Dual-Chamber Pacing and Atrioventricular Synchrony

In the normal heart, stroke volume increases 5 to 15% by AV synchrony, versus asynchrony.14 Left ventricular hypertrophy, decreased diastolic compliance, and heart failure increase the quantitative importance of AV synchrony.14 Apical pacing of the right ventricle disrupts the normal sequence of activation, because depolarization spreads more slowly over the ventricular myocardium than through the conduction system.

Recent experience has emphasized clinical relevance of the sequence of activation. Right ventricular outflow tract pacing may improve stroke volume versus apical pacing because of favorable effects on the activation sequence.15 Disruption of activation sequence by DDD pacing reduces the ventricular-aortic gradient in some patients with idiopathic hypertrophic subaortic stenosis.16,17 Biventricular (RV apex and coronary sinus) pacing patients with advanced cardiomyopathy and an intraventricular conduction defect improves left ventricular function by restoring simultaneous contraction of the septum and free wall, so-called ventricular resynchronization.18,19 Single-site, epicardial left ventricular pacing may provide similar benefits. For temporary pacing in postoperative heart block, biventricular pacing is superior to right ventricular pacing19 and may be useful for left ventricular dysfunction after cardiac surgery. Clinical trials suggest that biventricular pacing is superior to standard pacing for heart block.20

Dual-Chamber Pacing Algorithm

DDD pacemaker programming includes a lower rate, an upper rate, and AV delay. When the intrinsic atrial rate is between the upper and lower rate limits, the pacemaker tracks the atrium to maintain a 1:1 response between the right atrium and right ventricle. If the atrial rate falls to the lower rate limit, the atrium is paced at the lower rate limit. If the atrial rate exceeds the upper rate limit, the ventricle is paced at the upper rate limit with loss of AV synchrony, resembling a Wenckebach effect.

Programmable AV delay defines the interval allowed between atrial and ventricular depolarization. Timing starts with the atrial electrogram or pacing stimulus and continues until the AV delay elapses. If no ventricular depolarization is detected during the delay, the ventricle is paced. Atrial latency, a varying delay between the atrial pacing artifact and the P wave, requires different AV delays for atrial sensing versus pacing.

Rate Response to Increased Metabolic Demand

During high metabolic demand, cardiac output is augmented by increased ventricular contractility, venous return, and heart rate. In patients with heart block and a normal sinus exercise response, dual-chamber pacing maintains both AV synchrony and a physiologic rate response.21 However, sinus node incompetence (no atrial rate increase with exercise) or single-chamber ventricular (VVI) pacemakers, requires alternate mechanisms for rate response. The letter R after the three-letter pacemaker code indicates rate responsive capability. If a sensor detects increased metabolic demand, the lower rate of the pacemaker increases within a programmable range. Body vibration29 or respiratory rate22 is commonly employed to estimate demand. Other indicators are body temperature, venous oxygen saturation, QT interval, right ventricular systolic pressure, and right ventricular stroke volume. All such indicators can aberrantly increase heart rate, eg, during a bumpy car ride. Patients with sedentary life styles do not benefit from rate-responsive pacing. Adverse results of fast heart rates can include angina or infarction in patients with coronary disease.

Choice of Pacing Technique

Dual-chamber pacing has become the standard of care, except in chronic atrial fibrillation. Sinus rhythm appears to be better maintained by atrial than ventricular pacing,23 and paroxysmal atrial fibrillation, previously problematic for DDD pacemakers, is now well handled by mode switching. Advantages of dual-chamber pacing in reflex-mediated syncope with cardioinhibitory features have also been reported.13 Dual-chamber pacing may not be warranted in elderly patients, except when pacemaker syndrome, hypertension, or congestive heart failure is present. VVI or VVIR pacing is appropriate for patients with bradycardia and chronic atrial fibrillation. AAIR is useful for cardiac allograft recipients with sinus arrest or sinus bradycardia.24 Biventricular pacing using an endocardial coronary sinus lead is recommended for symptomatic heart failure.25

Pacemaker Technology

Epicardial versus Endocardial Leads

Epicardial leads are generally inferior to endocardial leads in electrical characteristics and are prone to conductor fractures.26 Steroid-eluting tips and small contact surfaces have improved epicardial leads. Epicardial pacing is more difficult at reoperative cardiac surgery, where epicardial fibrosis elevates pacing thresholds. Infected epicardial leads must be removed by thoracotomy. The epicardial approach is preferred in patients with congenital septal defects, single ventricle physiology, mechanical tricuspid valves, or venous thrombosis/occlusion. During thoracotomy, insertion of endocardial leads through an atrial pursestring is a useful option.27 Epicardial left ventricular pacing by the minimal access approach is increasing in importance, driven by a 5 to 10% technical failure rate for coronary sinus lead insertion.28

Approaches to DDD pacing during open-heart surgery are illustrated in Fig. 59-2. Fixed-screw, positive fixation leads are introduced via atrial pursestrings and fixed into position by axial rotation. When a tricuspid prosthesis or ring is to be inserted, a ventricular lead can be passed between the sutures securing the ring or the valve.

Unipolar versus Bipolar

Bipolar leads incorporate two insulated conductors. In unipolar systems, the patient's body is the second (anodal) conductor; a single conductor in the lead carries the negative current to the heart. Bipolar leads reduce electrical noise (oversensing) and adventitious pacing of the diaphragm or chest wall. These advantages are offset by increased engineering complexity. Bipolar leads historically were prone to breakdown of insulation or conductor fracture. This would compromise sensing, pacing, or both29 (Fig. 59-3). Recent bipolar leads are much improved and similar in dimensions and handling to unipolar leads (Figs. 59-4 and 59-5). The Medtronic Sprint Fidelis ICD lead is currently notable for an accelerating rate of lead fracture.

Lead Fixation

We prefer endocardial positive fixation leads, particularly in chambers with sparse trabeculation. A small wire spiral or screw holds these leads in place (Fig. 59-5G). In designs with fixed, extended screws, a soluble coating over the tip promotes venous passage. In retractable screw designs (Bisping), extension/retraction is accomplished by rotation of the pin at the lead tip. Axial clockwise rotation of screw-in leads during fixation provides tactile feedback on the firmness and security of attachment. Lead impedance provides feedback on the adequacy of tip extension and fixation.

Tined leads use miniature anchors to secure the lead tip within myocardial trabeculae (Fig. 59-5A). Tines require larger introducers than screw-in leads and are not secure in smooth-walled or dilated chambers. Nevertheless, some physicians prefer these leads.30

Temporary Pacing

Acute bradycardia can be treated with transthoracic pacing, temporary endocardial pacing, or chronotropic drugs, including atropine, dobutamine, or isoproterenol. Right ventricular perforation has become less prevalent with current temporary endocardial leads but must be borne in mind if hypotension develops acutely after removal of temporary wires.

Bradycardia after cardiac surgery is commonly treated with pacing via temporary atrial and ventricular epicardial wires. Problems include unfavorable evolution of right atrial or right ventricular thresholds and right atrial sensing. Atrial undersensing and pacemaker competition can precipitate atrial fibrillation or atrial flutter. If atrial sensing is not adequate, overdriving the atrium faster than the intrinsic rate can ameliorate competition. Reversing polarity or inserting a cutaneous ground wire under local anesthesia can improve pacing threshold. The output of the temporary pacemaker in volts or milliamps should be at least twice the threshold and should be measured daily. Ventricular undersensing in critically ill patients can result in pacing during the vulnerable period, which can precipitate ventricular tachycardia or fibrillation.

Cardiac Output and Pacing Rate

For patients with hemodynamic compromise following cardiac surgery, optimization of pacing rate and AV delay can affect hemodynamics by compensating for valve leaks or fixed stroke volume. Mean arterial pressure reflects cardiac output if systemic resistance is constant. Rate and timing adjustments over intervals of less than 20 seconds minimize reflex effects. Settings producing the highest sustained mean arterial pressure should also maximize cardiac output.

Pacemaker Insertion

Environment and Anesthesia

Pacemaker and ICD surgery is increasingly performed in electrophysiology (EP) laboratories.31 Whether in the operating room or EP lab, properly functioning equipment is essential. Infection control is critical32; operating room standards for air quality should be enforced. Problems calling for presence of an anesthesiologist include angina, transient cerebral ischemia, patient disorientation, lidocaine toxicity, dementia, myocardial ischemia, heart failure, anxiety, or ventricular tachycardia. If English is not the patient's first language, a translator is helpful. Vancomycin reactions (red man syndrome), pacing-induced ventricular fibrillation, air embolism, and Stokes-Adams attacks are rare emergencies that are less problematic if considered in advance. Intraoperative death can occur because of hemorrhage, pericardial tamponade, ventricular fibrillation (VF), heart failure, myocardial infarction, and other causes.

Monitoring

R-wave detection by an electrocardiogram (ECG) monitor is inadequate for pacemaker insertion, because pacing artifacts that trigger the monitor may fail to capture. Thus, subthreshold pacing can elicit regular beeping from the monitor in an asystolic patient. Oxygen saturation monitors are optimal, as they beep only during blood flow. Pulse oximeters should not be placed on the same extremity as the blood pressure cuff. When monitors are unreliable, palpation by an anesthesiologist or nurse of the temporal, facial, or radial artery pulse can detect asystole before the patient becomes symptomatic.

Venous Access Strategy

Choices include which side will be employed and cut down versus percutaneous venipuncture. Cut down approaches to the cephalic, subclavian, external jugular, and internal jugular system are described.33 Anatomical considerations29,34 may reduce the frequency of subclavian crush (Fig. 59-6). Subclavian puncture is associated with an apparently unavoidable but low incidence of pneumothorax, hemothorax, and major venous injury. Ultrasonic vessel locators may further reduce the frequency of injury.

Venous access is problematic in superior vena cava syndrome or subclavian/innominate vein obstruction or thrombosis (eg, with chronic dialysis, mediastinal tracheostomy, or multiple pacemaker leads). Access from below or transhepatically is possible,33,35 but potential for bleeding, venous thrombosis, and pulmonary embolism are concerns. Right parasternal mediastinotomy, exposure of the right atrium, and a Seldinger approach with small introducers and atrial pursestring sutures are useful in difficult cases36 (Fig. 59-7). Pacemaker lead extraction may permit reinsertion of leads via the extraction cannula.

Antibiotic Prophylaxis

Antibiotic prophylaxis is indicated for insertion of a prosthetic device.32 We prefer 1 g of intravenous cefazolin. We also irrigate the operative field with a solution of 1 g of cefazolin in a liter of warm saline. Patients who have a valve prosthesis or who have a penicillin or cefazolin allergy receive vancomycin (500 mg) and gentamicin (1 mg/kg).

Pacing Systems Analyzer

Pacing thresholds and electrogram amplitudes are measured with a pacing systems analyzer. Electrogram characteristics and slew rate can be assessed, and electrogram telemetry is available from most current pacemakers. Analyzers must be serviced and tested periodically, including batteries. Any discrepancies between measurements by the analyzer and the pacemaker should be noted and related functions of the analyzer rechecked.

Operator skill is needed to run the analyzer and record the results. The analyzer also can be placed in a sterile bag and operated by the surgeon. Manufacturer's representatives, an increasing intraoperative presence, are skilled in operation of these analyzers.

Cables

The cables connecting the analyzer to the leads are the patient's lifeline. Even with excellent quality control, cables with open or reversed connections may be delivered to the surgical field. Errors can also occur in connecting the cables to the analyzer. Routine testing of cables and the integrity of their connections is recommended. After passing the cables from the operating table, pacing is initiated from the analyzer at 5-V output. The connectors are then briefly touched to the subcutaneous tissue, with caution to minimize inhibition of the permanent pacemaker. Current measured in the analyzer should rise to 300 to 1000 ohms. If impedance is more than 5000 ohms, the circuit is faulty. The operation should not proceed until the problem is corrected, as the analyzer-cable circuit is defective. Connections to the analyzer should also be checked, because inadvertent reversal of polarity can make measured pacing thresholds inappropriately high. Even disposable leads with polarized connectors have been delivered to us with the connectors reversed.

Fluoroscopy

Fluoroscopy is essential for transvenous device implantation, and operating room personnel must be familiar with the equipment. Distracting problems with image orientation, rebooting, timers, brakes, and locks can be avoided by a knowledgeable team. Sudden failure of fluoroscopy at a critical point can occur. If a backup unit is not available, the options include "blind" endocardial lead insertion, epicardial insertion, or postponing the procedure. The use of low-dose and pulsed image options can prevent overheating of the fluoroscope, although image quality is compromised. Radiation exposure should be controlled by monitoring fluoroscopy time.

Surgical Approach

We approach the patient from the left when feasible. The fluoroscope, on the patient's right, is positioned carefully to allow visualization of the apex, right atrium, and deltopectoral groove. The right arm is extended rightward on an arm board. The drapes are suspended from IV poles. The right-sided pole is caudad to the arm board. Careful positioning allows the left clavicular region to be exposed while leaving the patient adequate light and air. After skin preparation, towels are aligned with the deltopectoral groove and clavicle to define the essential landmarks. The region of the incision and generator is infiltrated with 1% lidocaine to produce a field block. A 5- to 6-cm horizontal incision is created 4 cm beneath the clavicle, the lateral extent of the incision just reaching the deltopectoral groove. This allows the generator to be positioned away from the deltopectoral groove and axilla, avoiding interference with motion of the left arm at the shoulder. An alternate incision directly over the deltopectoral groove facilitates exposure of the cephalic vein; this is particularly valuable in obese patients or elderly patients with atretic veins.

Venous Access

When the deltopectoral groove has been exposed, additional anesthesia is infiltrated into the lateral margin of the pectoralis and laterally into the deltoid. The dissection proceeds into the deltopectoral groove, following the lateral edge of the pectoralis, until the cephalic vein or another venous branch is exposed. Failure to find a vein may mean the incision is too far cephalad or caudad or not lateral enough. The incision can be deepened into the subpectoral fat if necessary. If the vein is too small to pass a pacemaker lead, the curved end of the guidewire for a 7-French introducer is passed centrally. The ability to manually stiffen and extend the curve by central manipulation of the tension in the guidewire is an important technical aid in tortuous veins. The method illustrated in Fig. 59-8 can be used to dilate the vein. If the guidewire will not pass centrally, a no. 18 Angiocath is advanced over the guidewire and used to inject a small amount of iodinated contrast to visualize the venous system fluoroscopically. If the cut down approach must be abandoned, visualizing the subclavian vein by venogram reduces the risk of injury during subclavian puncture. If dual-chamber pacing is planned, a guidewire should be reinserted through the introducer before the introducer is stripped away. This provides venous access for the duration of the procedure.36,37 A pursestring suture in the muscle usually provides adequate control of bleeding and allows stabilization of the lead(s).38 Ultrasonic localizers may be useful at this stage, with appropriate sterile precautions.

Right Ventricular Lead Insertion

From the patient's left, a gentle spiral in the distal 10 cm of the stylet will guide the lead toward the tricuspid valve, right ventricle, and pulmonary artery. Advancing the lead into the pulmonary artery outside the cardiac silhouette confirms that the lead is not in the coronary sinus. For fixed-screw positive fixation leads, withdrawing the stylet 3 to 5 cm minimizes the risk of apical perforation while screwing the lead tip into the myocardium with axial clockwise rotation. Reverse torque that develops as the lead is rotated should be noted; this torque is a guide to the security and safety of fixation. We fix the lead with three consecutive 360-degree clockwise rotations of the lead shaft, then release the torque. This fixation sequence is repeated if necessary, until the lead is secure, with substantial reverse torque after the first 360-degree rotation. No more than three complete axial rotations are employed in any sequence. A potential problem with screw-in leads is ventricular perforation. The lead tip should be fluoroscoped during fixation looking for extra-anatomical passage, following the edge of the cardiac silhouette around the apex, then cephalad. If this happens, the lead should be withdrawn and repositioned. An echocardiogram should be obtained and the patient monitored for tamponade.

When the lead tip has been properly positioned, the stylet is withdrawn and thresholds tested. The patient is asked to hyperventilate and cough to confirm fixation. The ventricular pacing threshold should be less than 0.7 V, with R-wave amplitude more than 5 mV, and impedance 400 to 1000 ohms, depending on lead design. There should be no diaphragmatic pacing at an output of 10 V.

If the lead is dislodged by hyperventilation and coughing, or thresholds are not adequate, the lead should be repositioned. A positive fixation lead can be unscrewed by counterclockwise axial rotation until it floats free. Positive fixation leads can be secured almost anywhere along the margins of the right ventricular silhouette (Fig. 59-9), including the right ventricular outflow tract (Fig. 59-10).39 In difficult cases, we have relocated leads as many as 15 times. The geographic center of the right ventricular silhouette is not a desirable location, as it can lead to entanglement of the lead in the chordae tendineae (see Fig. 59-9).

Coronary Sinus Lead Insertion

Left ventricular pacing via the coronary sinus (CS) is a valued skill since the MIRACLE trial and subsequent studies indicated that CRT reduces symptoms of heart failure and mortality of dilated cardiomyopathy.18,19,25 Electrophysiologists are familiar with CS entry for arrhythmia mapping; steerable mapping catheters and biplane fluoroscopy are important tools. However, the technical failure rate of CS lead insertion is 5 to 10%.18,25 The CS orifice is usually a posterior structure near the caudal aspect of the tricuspid valve (Fig. 59-11). Locating the os in heart failure patients is difficult because the CS may be angulated and distorted as a result of cardiac enlargement. Transesophageal echocardiography and venography may help locate the os. CRT candidates are prone to ventricular arrhythmias, making rapid defibrillation capability desirable. Even experienced operators may require hours to insert a CS lead with present methods, although technology for both endocardial and epicardial left ventricular lead placement is improving. Over-the-wire lead designs are the most successful. We prefer to insert the CS lead first via the cephalic approach, right atrial and right ventricular lead insertion after via subclavian puncture. CS lead insertion involves advancing an angled catheter into the CS. This is followed by CS venography and lead insertion through the cannula into a lateral branch of the CS. The CS cannula is then removed and stripped off without dislodging the lead. Positive fixation leads are not available. A large selection of angled CS cannulas, steerable probes, and lead designs testifies to the technical challenge. The difficulty of CS lead insertion in heart failure patients should not be underestimated; special training is desirable.

Length Adjustment

Lead length can be "too short," so that flattening of the diaphragm during a deep breath results in lead displacement, or "too long," so that a cough results in formation of an intracardiac loops that shorten the lead, also potentially causing displacement. Maximal inspiration and expiration by the patient during observation under fluoroscopy helps judge length. Vigorous coughing tests the security of lead tip fixation. This valuable feedback is lost if sedation is excessive. Minimizing sedation also promotes early detection of pacemaker syndrome.

Atrial Lead Insertion

For dual-chamber pacing, the atrial lead is introduced last. A J or S stylet shape is best for finding the atrial appendage from the left side.55 The S shape is also useful for passing a positive fixation lead to the right margin of the atrium near the junction of the atrium and inferior vena cava (Fig. 59-12). P-wave amplitude is often best in this location. The atrial pacing threshold should be less than 2 V. In the presence of complete heart block, it may be difficult to confirm atrial capture from the surface ECG. Pacing the atrium at 150 bpm results in rapid oscillation of the lead tip, which is visible fluoroscopically if mechanical function of the atrium is sufficient. Lead oscillation may thus be used to determine the atrial pacing threshold. This technique should only be used if the high atrial rate is not conducted to the ventricle. In patients with complete heart block, high-output atrial pacing can inhibit whatever temporary VVI pacing is supporting the patient, resulting in asystole.

The P-wave electrogram is the Achilles heel of dual-chamber pacing. If atrial sensing is not satisfactory, a DDD pacemaker will not function properly. The P-wave amplitude should ideally be greater than 2 mV. P-wave amplitude may vary during respiration, and the minimum value, not the maximum, determines the adequacy of sensing. P-wave amplitude is generally reduced during atrial fibrillation versus values in sinus rhythm.

Measurement of P-wave amplitude with unipolar leads can be confusing. Crosstalk or far-field sensing of ventricular depolarization from the atrial lead can occur, so that the signal measured through a pacemaker analyzer is not the P wave but the QRS complex. Simultaneous measurement and display of atrial and ventricular electrograms as well as the surface ECG can resolve this. An alternate solution involves programming the generator as a P-wave detector: The lower rate is set below the patient's intrinsic atrial rate, the AV delay is set shorter than the patient's P-R interval, and atrial sensitivity is set at 2 mV. When the generator is connected to the atrial and ventricular leads, every P wave will be followed immediately by a ventricular pacing spike if P-wave amplitude is greater than 2 mV. The pacemaker must be reprogrammed to clinically appropriate settings at the conclusion of surgery.

Telemetry of atrial and ventricular electrograms by pacemakers provides valuable EP data. For example, inability to pace the atrium or to measure atrial electrograms in the operating room may indicate low-amplitude atrial fibrillation or supraventricular tachycardia; this may be invisible on the surface ECG but detectable in electrograms. Electrogram telemetry can confirm proper DDD pacing (Fig. 59-13).

Generator Location

We use a watertight, three-layer skin closure for primary implants, three layers in generator replacements. Cosmetic appearance is important to many patients, and in others technique is critical to optimize healing. Past injury to the chest wall or surgery/radiotherapy for breast cancer can present a formidable technical problem. Bipolar systems facilitate generator placement behind the pectoralis or inside the rectus sheath. Diminutive generators are available, but battery life is reduced. Innovative locations for pacemaker generators include axillary, retromammary, intrathoracic, intra-abdominal, and preperitoneal sites. These approaches are rarely indicated with present generator designs.

Length of Stay after Pacemaker Implantation

Ambulatory Surgery

Same-day hospital discharge after pacemaker insertion is reasonable in patients who have an adequate escape rhythm and positive fixation leads. After monitoring and recovery from sedation, patients are ambulated and shown range of motion exercises for the shoulder. A chest x-ray documents lead position and rules out hemothorax, pneumothorax, or increasing heart size.

Pacemaker-Dependent Patients

Lead displacement can result from technical error, struggling of demented patients, and other factors. A small percentage of lead displacement is probably unavoidable.30,40 Patients who might suffer death or injury in the event of pacemaker failure should be observed in the hospital overnight on telemetry. However, lead displacement in our ambulatory patients has not been more frequent than in hospitalized patients.

Pacemaker Generator Replacement

Planning

Fifteen percent of functioning pacemakers were replacements in a recent survey. Complications of pacemaker generator replacement include infection, lead damage, connector problems, and asystole during the transition from the old generator to the new. Pacemaker independence at the time of initial pacemaker implant may progress to total pacemaker dependence by the time of generator replacement. Ambulatory surgery is common. As a practical matter, we no completely reverse warfarin for pacemaker generator replacement unless lead replacement is expected.41 Patients with leads more than 10 years old should be carefully evaluated for pacemaker dysfunction before generator replacement; a Holter monitor should be obtained if lead dysfunction is suspected. Rising pacing threshold may indicate impending lead failure. The possibility of unexpected lead replacement should be discussed with the patient in advance.

Backup Pacing

A backup temporary transvenous pacing wire can be inserted for pacemaker generator replacement, but this is rarely necessary. The output of the replacement unit must be higher than the threshold of the chronically implanted leads. This can be problematic if the expiring generator is an older type with a fixed output of 5.4 V. Lack of programmability prevents preoperative threshold testing, and the 5.4-V output is higher than possible for some current generators. The pacing threshold should be determined with a pacemaker analyzer and the replacement should be programmed to appropriate output. Before disconnecting the old generator, be sure that the pacemaker analyzer, cables, and connections are intact and personnel in the operating room are aware of the cables. Some place the analyzer in a sterile bag on the operative field. With many generators, an Allen wrench placed in the header establishes electrical continuity with the ventricular lead; the pacing threshold can then be established before disconnecting the old generator. The old generator should be kept within reach as a backup in case of trouble with the new generator, the analyzer, or the connectors. Replacement generators must be programmed unipolar or bipolar to match the indwelling lead(s).

Lead Sizes

The three common lead connectors for permanent pacing are 6, 5, and 3.5 mm (VS-1 or IS-1). If the new generator is not an exact match for the patient's lead size, a selection of step-up and step-down adapters can be helpful. The contacts do not always line up correctly across VS-1 and IS-1 connectors, even though the connector diameter is the same. It is important to determine in advance whether connections can safely be made across the VS-1/IS-1 interface.

Postoperative Care

Wound Care

Patients are instructed to keep implant wounds dry until an office visit 7 to 10 days postoperatively. Any wound drainage at the postoperative visit is cultured, and prophylactic antibiotics are started until culture results are available. We have abandoned aspiration of the rare postoperative hematoma in favor of close observation, unless infection is an issue or spontaneous drainage occurs or appears imminent.

Antibiotic Prophylaxis

Routine use of prophylactic antibiotics before dental work and other invasive procedures in pacemaker or ICD recipients is not recommended under AHA/ACC guidelines. We recommend prophylaxis for 3 months after device insertion, allowing time for the pacing leads to become endothelialized.

Testing and Follow-Up

Office/Clinic versus Telephone

Pacemakers require periodic testing to confirm sensing and pacing function and battery reserve. Currently these functions are tested at 1- to 3-month intervals. Whether follow-up should be done by transtelephonic monitoring or clinic or office visits is in dispute.61–63 Transtelephonic monitoring alleviates transportation issues for elderly patients, but some need help managing the process. In addition to reducing patient travel and strain on office resources, many commercial services provide emergency monitoring on a 24-hour basis, an advantage in managing apprehensive or incapacitated patients. Techniques for remote management of pacemakers and ICDs are available with some current devices.

Pacemaker Programming

Programming can be done in an office setting by trained, experienced personnel. In some situations, a manufacturer's representative may provide valuable help. Advanced techniques for pacemaker/ICD monitoring and programming are under development.

DDD pacemaker programming allows adjustment of electrogram sensitivity as well as pacing stimulus amplitude/pulse width for both the atrium and ventricle. Lower rate, upper rate, atrioventricular delay, and refractory periods for atrial and ventricular sensing are programmable. Rate responsiveness, unipolar/bipolar configuration, and many other options are also adjustable noninvasively.

We initially program newly inserted pacemakers to stimulation amplitude and pulse width higher than nominal. At the initial office visit, pacing thresholds are retested and amplitude and pulse width are adjusted to nominal levels if pacing thresholds are low. The use of high initial output is less important with steroid-eluting leads. Details of pacemaker programming have been described elsewhere.63,64 Some current pacemakers are capable of automated threshold adjustment.

Most problems detected by transtelephonic or Holter monitoring can be corrected by programming. The etiology of symptoms can often be elicited from real-time electrograms or stored data. Adjustments can include not only sensitivity or pacing output but also pacing mode for new-onset atrial fibrillation (Fig. 59-14A) or sinus node incompetence related to medication changes (Fig. 59-14B). Problems that may require reoperation include lead displacement, lead fracture (Fig. 59-15), insulation degradation (see Fig. 59-3), and exit block40 (Fig. 59-16).

Pacemaker interrogation should begin with a printout of the initial settings, an invaluable reference after involved programming. Telemetry defines time-related variation in heart rate, percentage of beats sensed and paced, the quality of the electrograms, lead impedance, and battery voltage.

Pacing amplitude and pulse width are finely tuned at a 1-year follow-up visit. At least 100% safety margin is programmed on the pulse width threshold. Parameters are adjusted to optimize patient comfort and battery life. Some pacemakers allow programmed reduction of a lower rate at night, eliminating unnecessary pacing during sleep. Very long atrioventricular delays can eliminate ventricular pacing in some patients with first-degree block, but a reduction in the upper rate limit to 105 bpm may be necessary to achieve this. Current pacemakers have a variety of algorithms to prevent excessive pacing in first-degree heart block, including automatic switching between AAI and DDD modes.

Complications of Pacemaker Insertion

Mortality

Death is a rare complication of pacemaker implantation.40,43 Lethal problems can include lead displacement, venous or cardiac perforation, air embolism, and ventricular tachycardia or fibrillation.40 A review of 650 pacemaker insertions by the author between January 1984 and April 1993 revealed only one perioperative death resulting from heart failure induced by general anesthesia in a child with congenital heart disease (Table 59-3).

Incidence of Complications

The incidence of early pacemaker complications in one recent series was 6.7%, 4.9% requiring reoperation.40 For patients older than 65, comparable figures were 6.1% and 4.4%, respectively. Lead displacement, pneumothorax, and cardiac perforation were the most common complications. The incidence of late complications was 7.2%.57 The incidence of reoperation in the author's review of 480 cases was 4.0% (see Table 59-3).

Lead Displacement

The incidence of endocardial lead displacement with early lead designs was more than 10%. With tined and positive fixation leads this has fallen to about 2%.2,30,40,43–45 The incidence of this complication was 1.5% for atrial and ventricular leads in our review (see Table 59-3). Relevant technical issues have been described in the preceding. We find that positive fixation leads can be applied in unique anatomical locations with essentially no increase in lead displacement (Table 59-4).

Myocardial Infarction

Pacemaker insertion may be appropriate as an adjunct to medical therapy of angina in patients with inoperable coronary artery disease. However, angina, myocardial infarction, or death can result from paced increases in heart rate of as little as 10 bpm.

Hemopneumothorax

Hemopneumothorax and pericardial tamponade can result from injury to the heart, lungs, arteries, or veins. Errors with the Seldinger technique can cause such injuries. In patients older than 65, pneumothorax has been related to subclavian puncture.45 In our experience with more than 1000 pacemaker insertions by cephalic cut down, hemopneumothorax did not occur. In contrast, a recent review of 1088 consecutive implants by subclavian puncture revealed a 1.8% incidence of pneumothorax.45

Pacemaker Syndrome

Loss of atrioventricular synchrony produces symptoms related to reflex effects or contraction of the atria against closed AV valves. The resulting constellation of symptoms is known as pacemaker syndrome.46 The symptoms are quite variable, but severely affected patients may refuse pacemaker magnet testing. Symptoms are relieved immediately by conversion from VVI to dual-chamber pacing.

Lead Entrapment

Fixed-screw pacemaker leads can become firmly entangled in chordae tendineae beneath the tricuspid valve. Possible responses include further escalation of force, lead extraction,34,47 or an open procedure. Our experience with this involved three firmly entangled leads in 1000 lead implants. The leads were capped and abandoned rather than escalate risk. There were no untoward consequences of lead abandonment. Consequently, we now avoid the anatomic center of the right ventricle when implanting these leads. This problem has not recurred in more than 750 subsequent implants.

Infection and Erosion

Pacemaker infection can appear as frank sepsis, intermittent fever with vegetations/ inflammation, and/or purulence or drainage at the pacemaker pocket. Indolent generator erosion is another presentation. Antibiotic suppression may temporarily abolish signs of infection, but the problem usually recurs weeks or months later.46–48 Negative cultures at an erosion may suggest moving the device to a fresh, adjacent site, but this is usually futile. Clinical resolution of recurrent device infection almost always requires removal of all hardware and insertion of a new device in a fresh site, optimally after a device-free interval.48,49 The incidence of erosion, infection, hematoma, and lead displacement early after pacemaker implantation is reduced by operator experience.44

Pacemaker Dysfunction

Mechanical defects in leads, lead displacement, or connection errors can cause pacemaker dysfunction. Lead dysfunction usually represents scarring at the lead—myocardial interface, changes in myocardial properties owing to tissue necrosis or drug effects, or a poor choice of lead position. Insulation erosion can cause oversensing or, in bipolar leads, pacing failure owing to short-circuiting between the two conductors.

Generator Dysfunction

Electrical component failures are rare. Three pacemaker or ICD generator failures have required urgent device replacement over the past 10 years at our center. New pacemaker and lead designs may contain flaws that do not become apparent for many years.2,50

Undersensing

Undersensing is failure to detect atrial or ventricular electrograms. The result is an atrial or ventricular pacing that should have been inhibited by the unsensed beat. In a dual-chamber pacemaker, undersensing may also cause failure to pace the ventricle after the P wave. Undersensing is often correctable by programming increased generator sensitivity, but this can lead to oversensing. The latitude for reprogramming can be estimated by examining telemetered electrograms42 (Figs. 59-17 and 59-18).

Oversensing

Inappropriate pacemaker inhibition or triggering may result from detection of myopotentials (muscular activity). This is most common in unipolar systems and may be correctable by programming reduced pacemaker sensitivity. External insulation erosion within the pacemaker pocket is a cause of oversensing that can be repaired. Internal insulation defects in bipolar leads (see Fig. 59-3) cannot be repaired.

Crosstalk and Far-Field Sensing

A deflection on the ventricular lead immediately after the atrial pacing artifact could be either premature ventricular depolarization (see Fig. 59-17) or far-field sensing of an atrial depolarization (see Fig. 59-13). Many pacemakers deal with this ambiguity by pacing the ventricle at a short (100 millisecond) atrioventricular delay, known as safety pacing.20

Complexities of DDD programming involve blanking and refractory periods used to compensate for crosstalk or prevent retrograde AV conduction from causing pacemaker-mediated tachycardia (see the following). Crosstalk is ameliorated by bipolar lead systems.

Exit Block

Exit block is rising pacing threshold due to edema or scarring at the lead tip. Pacing threshold tends to increase over 7 to 14 days after lead insertion and stabilizes at about 6 weeks. This phenomenon is related to inflammation at the lead tip and is ameliorated by steroid-eluting leads.2,51 Exit block may be overcome by programming increased amplitude or pulse width, but this shortens battery life. In unipolar systems, pacing of the chest wall and/or diaphragm may result from high generator output.

Lead Fracture

Fracture of lead insulation or conductors may be demonstrable by chest x-ray (see Fig. 59-15). Lead impedance less than 300 ohms suggests an insulation break, whereas impedance more than 1000 ohms suggests conductor problems, a loose set screw, or improper connection. High impedance also can indicate incomplete extension of the fixation coil in a Bisping lead. Telemetry may detect impending lead fracture as electrical noise during hyperventilation, coughing, bending, or arm swinging. Oversensing of this type usually mandates lead replacement or repair.

At reoperation, dysfunctional leads can be capped or removed by lead extraction techniques. Extraction of chronically implanted leads is potentially hazardous and results in endothelial venous damage even when successful. Accordingly, our practice is to limit lead extraction to infection or mechanical problems (Figs. 59-19 and 59-20). Lead fracture has been promoted historically by design errors, bipolar construction, certain forms of polyurethane insulation, and epicardial insertion.64 Technical factors in fracture include ties applied to the lead without an anchoring sleeve, kinking, lead angulation, vigorous exercise programs, and subclavian crush.2,29,35

Subclavian Crush

Subclavian crush is believed to be caused by lead entrapment between the clavicle and first rib, in the costoclavicular ligament. Stress during body movement is then thought to cause early lead failure. This pertains primarily to leads implanted by percutaneous puncture of the subclavian vein and seems to be minimized by cephalic cut down. Techniques to minimize this problem have been described (see Fig. 59-6).2,29,35

Pacemaker-Mediated Tachycardia

DDD pacemakers can propagate a reentrant arrhythmia, pacemaker-mediated tachycardia. This involves retrograde conduction through the atrioventricular node, initially triggered by a premature ventricular depolarization. If the pacemaker senses the retrograde atrial depolarization and paces the ventricle, a cycle is set up that can continue indefinitely at the upper rate limit of the pacemaker. This problem can be mitigated by avoiding high upper rate limits and adjusting the postventricular atrial refractory period so that the pacemaker ignores atrial depolarizations for 300 to 350 milliseconds after the QRS complex. Current pacemakers also attempt to break reentrant arrhythmias by periodic interruption of continuous high-rate pacing. Pacemaker telemetry provides notification of high-rate pacing suspicious of pacemaker-mediated tachycardia.

Innovations and Special Problems

Lead Repair

Repair can extend the useful life of implanted leads for years. Most repairs are done within the device pocket or in surrounding areas. Fracture or erosion of conductors or insulation can result from normal wear or active life styles. We have seen lead dysfunction resulting from obsessive sit ups (abdominal insulation erosion), handball (insulation/conductor erosion), and wood chopping (conductor damage).

Repair kits contain silicone glue and silicone tubing or unipolar tip replacements. Possible repairs differ for unipolar and bipolar leads. Unipolar leads consist of a single conductor surrounded by insulation. Insulation breaks can be overlaid with glue and tubing. Conductor fracture can be repaired by splicing a new lead tip onto the functional segment. Bipolar leads contain two conductors with two levels of insulation, one to prevent short-circuiting between the conductors and the other to prevent external current leaks and conductor-generator contact. An example is shown in Fig. 57-2. External insulation repair in such leads is similar to unipolar lead repair. However, internal insulation faults require converting the lead to unipolar function and splicing on a new lead tip. Conductor fracture also can be repaired this way. The repair involves exposing and baring about 10 mm of the conductor to be preserved. The conductor to be excluded is cut back 5 to 10 mm to avoid short-circuiting. A new tip is attached to the bared conductor, using an internal set-screw, silicone glue, and ties. These repairs have been robust in our experience but are not recommended in pacemaker-dependent patients.

Unipolar and bipolar epicardial leads can also be repaired, if the break is accessible. Unipolar lead repair is similar to repair of unipolar endocardial leads. Bipolar epicardial leads generally consist of two unipolar leads connected by a Y to coaxial segments for connection. The simplest repair involves locating the unipolar segment of the good lead and splicing on a new tip. Particularly challenging is a fracture near one of the cardiac electrodes, at a point of metal fatigue. If the broken conductor is not the tip electrode, function is restored by reprogramming the generator to a unipolar configuration. If the cathode segment is fractured, most generators cannot be programmed to use the anode, and pacing is lost. However, function can be restored by exposing the lead, splicing a new tip on the anode, connecting that lead to generator, and capping the cathode.52

When exposing leads within the pocket for repairs, the electrocautery should be set as low as practical, to avoid melting external lead insulation. If cautery contacts a bare conductor, myocardial injury can render the lead useless because of exit block. Also, conduction of cautery to the myocardium can induce VF. Sharp dissection is preferred when close to friction points or angulation that promote conductor exposure.

Pacemaker Lead Extraction

Indications for lead extraction include chronic infection or life-threatening mechanical defects.53 Some recommend that any dysfunctional pacemaker lead should be removed, but there is little objective data to support this. Until recently, extraction of transvenous leads required external traction or thoracotomy/cardiotomy with inflow occlusion or cardiopulmonary bypass. Chronically implanted leads can be densely fibrosed to the right ventricular myocardium, vena cava, innominate vein, or subclavian vein.

Lead extraction was developed by Byrd.34,47 A locking stylet is passed inside the central channel to the tip of the lead where it uncoils, allowing traction to be applied to the lead tip. Telescoping Teflon, plastic, or metal sheaths fitting the lead are passed along the lead to mobilize it. When the long sheath reaches the lead tip, countertraction is applied to the myocardium with the sheath while traction is applied to the lead tip with the locking stylet (see Fig. 59-19). Success with this technique has been greater than 90%, with a 3% chance of serious morbidity or death. Laser or radiofrequency energy can also be transmitted through specially constructed sheaths to ablate adhesions.80 Technical details have been described.34,47 Extraction of leads more than 10 years old is difficult and tedious. Complete removal of lead tips is inversely related to the age of the lead (Fig. 59-21).55

Accufix Lead

An unusual fracture affects the Telectronics Accufix lead, a bipolar, Bisping-type atrial screw-in lead.53 A J-shape near the tip directs the lead to the atrial appendage. A curved retention wire welded to the indifferent ring electrode near the tip and bonded to the lead body with polyurethane maintains the J-shape. Fracture and extrusion of this retention wire (see Fig. 59-20) was associated with deaths from cardiac tamponade, related to punctures of the atrium or aorta by fractured wire. More than 45,000 of these leads were implanted, and many have been surgically extracted. Because some morbidity and mortality occurred during extraction of this lead, the manufacturer recommended conservative management. Recently, conservative management has also been recommended for a fracture tendency in the Sprint Fidelis ICD lead.

Atrial Fibrillation and Mode Switching

Sinoatrial node dysfunction can involve both sinus bradycardia and paroxysmal atrial fibrillation. Early DDD pacemakers responded to atrial fibrillation by pacing at the upper rate limit. Initially DDD pacing was felt to be contraindicated in atrial fibrillation for this reason. The current view that atrial pacing decreases the frequency of paroxysmal atrial fibrillation, and mode switching allows DDD pacing to be used despite a history of paroxysmal atrial fibrillation. Mode switching is triggered when a programmed upper rate limit is exceeded. The pacemaker then switches to VVIR mode until the atrial rate returns to the physiologic range. Successful mode switching requires bipolar leads and high sensitivity in patients with low-amplitude atrial fibrillation. Management of atrial fibrillation in elderly people may involve fewer medications and interventions than in younger patients.

Database Support

Pacemaker and ICD data are needed for billing, operative notes, device tracking, programming, and follow-up. Data should be available in real time in the event of an emergency room visit involving device malfunction. Commercial and home-grown software packages are available for this. Security, audit trails, 24/7 availability, and multiuser wireless capability are important characteristics for such systems.

General Surgery and Pacemakers

General surgery in a pacemaker-dependent patient raises important questions,56 especially when surgery requires unipolar cautery. The following must be documented: (1) model and manufacturer of the pacemaker, from pacemaker ID card, monitoring service, medical record, or x-ray appearance57,58; (2) magnet mode behavior and any peculiarities related to impedance sensing monitors; (3) successful testing of programmer on the pacemaker; (4) programmed parameters, polarity, battery life, and lead characteristics; (5) degree of pacemaker dependence57; (6) a backup plan transthoracic pacing or chronotropic agents if the pacemaker fails; (7) reprogramming to VOO, DOO, or VVT mode intraoperatively with rate response off to prevent inhibition56 or pacemaker acceleration84 by electromagnetic interference (EMI); (8) a physician to deal with any intraoperative pacemaker problem; and (9) restored pacemaker program, thresholds, and function postoperatively. Regarding pacemaker dependence, if the preoperative electrocardiogram reveals 100% pacing, the pacemaker should be reprogrammed while monitoring to determine the presence and rate of an escape rhythm. Pacemaker dependence may increase during anesthesia, with withdrawal of sympathetic stimulation.

Electrocautery

Manufacturers recommend against using electrocautery in pacemaker patients because of possible electromagnetic interference (EMI) or pacemaker damage. If electrocautery must be used, unipolar cautery is likely to cause EMI, whereas bipolar cautery is not. Unipolar pacemakers are also more susceptible to EMI than bipolar units. EMI effects include: (1) Pacemaker oversensing of EMI as a rapid heart rate. Pacemaker inhibition results that reverses when the EMI stops. (2) Pacemaker reprogramming. (3) Acceleration of impedance-sensing pacemakers to the upper rate limit.57 (4) Reversion to a "backup mode" or "magnet mode." (5) Permanent loss of pacing, fortunately rare.56

EMI pacemaker inhibition can be minimized by programming sensing off and increasing the pacing rate above the intrinsic heart rate anticipated during surgery. However, if competition with spontaneous beats does occur, there is a risk of inducing of atrial fibrillation or ventricular tachycardia. In view of this, the pacemaker should be returned to an appropriate sensing mode as soon as possible after the completion of surgery.

Magnet Mode

A permanent magnet placed over a pacemaker closes a magnetic reed switch and initiates "magnet mode." In some pacemakers, magnet mode is VOO, eliminating all sensing. Other pacemakers convert to VOO for a few beats, then revert to the programmed function. A magnet may also induce a threshold margin test; this assesses the adequacy of the pacing margin by decreasing the pulse width in a predictable pattern.

Steroid-Eluting Leads

Fibrosis at the lead tip can be limited by incorporating a dexamethasone pellet that dissolves over months. This improves early pacing thresholds versus conventional leads51 and has been particularly advantageous in epicardial leads.

Adults with Congenital Heart Disease

Congenital heart disease may include a persistent left superior vena cava draining to the coronary sinus. This favors a right-sided surgical approach, although the left side can be used.38 Preoperative echo-Doppler or angiography can define caval and coronary sinus anatomy. A left superior vena cava may dislocate the subclavian vein, increasing the risk of subclavian vein puncture and favoring cephalic cut down.54 Situs inversus and corrected transposition are disorienting, if undetected prior to pacemaker insertion.

Positive fixation leads are particularly useful for atrial pacing after a Mustard operation or a caval-pulmonary anastomosis and for pacing the smooth-walled "right" ventricle in corrected transposition of the great arteries.38 A coronary sinus lead can provide ventricular pacing in some patients after Fontan surgery (Fig. 59-22). Pacing via the coronary sinus is useful in patients with a mechanical tricuspid valve.

Pacing in Infants and Children

Transvenous leads in children should include an intracardiac loop to allow for growth (see Fig. 59-4). Unipolar, positive fixation leads were ideal for this purpose,38 but other approaches have been described. We prefer cephalic cut down, with optical magnification, if needed. A flexible guidewire is passed centrally and a 7-French introducer introduces the lead. A longitudinal split of the cephalic vein facilitates advancing the introducer (see Fig. 59-8). In very small infants, the external jugular vein at the thoracic inlet may be useful. Subclavian vein puncture can be guided by a catheter introduced via the femoral vein. Thoracotomy is a third option.27 Infants less than 6 months of age are suboptimal candidates for transvenous pacing because of limited long-term lead utility. Subpectoral generator placement in children less than 6 years old reduces infection risk.

Dementia

Dementia is problematic for surgery under local anesthesia. Sedation can make dementia worse and exaggerates bradycardia. The demented patient's arms should be secured to prevent groping for the surgical wound. Postoperative confusion and thrashing can cause pacemaker lead displacement. An intracardiac loop my reduce this hazard. A family member at the bedside may alleviate this problem. Every effort should be made to anticipate and avoid such problems.

Atrioventricular Node Ablation

AV node ablation controls ventricular rate in refractory atrial fibrillation but leads to permanent heart block. Historically, the AV node was ablated, then temporary pacing supported heart rate until a permanent pacemaker was inserted. Ventricular escape rhythm was often poor, favoring positive fixation leads, overnight observation on telemetry, and high pacemaker output. Alternatively, the pacemaker can be inserted and allowed to heal before ablation.

Transplant Recipients

The usual indication for pacemaker insertion in cardiac transplant recipients is sinus bradycardia or sinus arrest, managed with AAIR pacing.24,59 In our experience, most outgrow the need for pacing within 2 years.24,59 The surface ECG may be confusing, with P waves from atria of both the donor and the recipient and AV dissociation of the recipient atrium and donor ventricle. Need for ventricular pacing can be evaluated by pacing the atrium at a rate of 150 bpm. If a 1:1 ventricular response is observed, the AV node is essentially normal. The location of the atrial appendage is more medial than usual in these patients, and positive fixation leads are preferable (Fig. 59-23).

Implantable Cardioverter Defibrillator Recipients

Before integrated devices were available, crosstalk between pacemakers and ICDs could lead to inappropriate ICD shocks or ICD undersensing of ventricular fibrillation. Successful independent implantation of pacemakers and ICDs has been described,60 but availability of ICDs with integrated DDD pacemakers renders this technique superfluous.

Long QT Syndrome

Long QT is a genetically determined repolarization abnormality that is associated with sudden death. Recommended therapy includes stellate ganglionectomy and/or adrenergic blockade.61 In severe cases, the pacing threshold may be too high for ventricular pacing, and atrial pacing may be preferable. ICD therapy is now commonly used.

Idiopathic Hypertrophic Subaortic Stenosis

Idiopathic hypertrophic subaortic stenosis (obstructive cardiomyopathy), with severe left ventricular outflow obstruction, causes angina and/or syncope. Right ventricular pacing with a short atrioventricular delay pre-excites the ventricle and decreases outflow gradients in some patients16,17 (Fig. 59-24). ICD therapy is increasingly common in this population.

Permanent Biventricular Pacing

End-stage cardiomyopathy and heart failure progress with time. Cardiac transplantation or left ventricular assist devices are effective in end-stage heart failure. CRT is a lower cost option for class III or IV failure. Clinical trials demonstrate modest subjective and objective benefits of CRT in dilated cardiomyopathy with left ventricular ejection fraction less than 36% and QRS intervals greater than 120 milliseconds.18,19,62 Mortality benefits are suggested by COMPANION and other recent trials.63 Marked clinical improvement is seen in some patients (see Fig. 59-11), but up to 40% of patients are nonresponders. Endocardial LV lead insertion fails in 5 to 10% of candidates, usually because of difficulty cannulating the CS. These failures often result in referrals to thoracic surgeons for epicardial lead insertion. More than 100,000 patients undergo CRT annually in the United States, which could lead to 5000 referrals yearly for epicardial lead insertion. Although minimal access64 and robotic65 LV lead insertion have been developed, referrals might increase if better techniques for epicardial LV lead insertion were available.

Temporary Biventricular Pacing

Clinical success with CRT and the potential to increase cardiac output while reducing myocardial oxygen consumption make temporary biventricular pacing attractive for management of low-output states after cardiac surgery. Preliminary results66 indicate that this approach is promising. Temporary biventricular pacing can be recommended and is readily achievable for patients in low-output states with second- or third-degree block after cardiac surgery.19,67

Atrial and Ventricular Tachyarrhythmias

Overdrive pacing can be effective for ventricular tachyarrhythmias, Wolff-Parkinson-White syndrome, or atrial flutter. Implantable defibrillators are under development for atrial fibrillation. Anti-tachycardia ventricular pacing for ventricular tachyarrhythmia has been integrated into ICD therapy.

Environmental Issues

Electromagnetic Interference

EMI56 can be caused by electrocautery, cellular telephones, magnetic resonance imagers,68 microwaves, diathermy, arc welders, powerful radar or radio transmitters, and theft detectors in retail stores. Any defective, sparking electrical appliance or motor, electric razor, lawn mower, or electric light can be problematic. The importance of EMI is related to pacemaker dependence. Pacemaker recipients who are not pacemaker dependent will not be distressed by brief periods of pacemaker inhibition, but pacemaker-dependent patients can lose consciousness in 5 to 15 seconds. Bipolar pacing systems provide added protection against EMI. Cellular telephones should be separated by several inches from pacemaker generators, preferably on the contralateral side.69 ICD circuits are insulated against inappropriate firing precipitated by EMI. During surgery using unipolar cautery, the defibrillation circuit of an ICD should be temporarily disabled.

Mechanical Interference

Lithotripsy, trauma, dental equipment, and even bumpy roads can affect pacemakers. Automobile accidents have caused pacemaker damage and disruption of pacemaker wounds.70 Vibration causes inappropriately high heart rates in rate-responsive units. Patients with poor escape rhythms should be discouraged from exposure to deceleration injury in contact sports, basketball, handball, downhill skiing, surfing, diving, mountain climbing, and gymnastics. Participants in these activities should realize that abrupt pacemaker failure could occur in the event of lead displacement related to trauma.

Radioactivity

The integrated circuits of current pacemakers can be damaged by radiotherapy.4 If the pacemaker cannot be adequately shielded from the radiation field, it may be necessary to remove and replace it or move the pacing system to a remote site.

Quality of Life

Quality of life is not a major concern for most pacemaker recipients. Although many clinics require periodic visits, the model of transtelephonic monitoring with office visits only for problems is acceptable to most patients. This latter system involves a preoperative visit, a 10-day postoperative visit, a 1-year visit to adjust output, and no additional visits unless functional problems or impending battery depletion are detected. Some recipients are never happy with their pacemakers because of body image problems, vague symptoms, or concern that life will be artificially prolonged. The value of generator replacement in patients with advanced debilitation has been a subject of ethical concern.71

Background

More than 400,000 deaths in the United States each year are classified as sudden and likely to be caused by arrhythmias.72 Michel Mirowski conceptualized the implantable defibrillator in the late 1960s. Overcoming theoretical, engineering, and financial obstacles, he participated in a successful clinical trial of his device in the early 1980s.73 Today's implantable cardioverter defibrillator (ICD) reflects dramatic and expensive growth in technology. The efficacy of the ICD in prevention of sudden death is well established. Clinical trials, experience, and the passage of time emphasize survival advantages of the ICD over other modalities, including antiarrhythmic drugs and subendocardial resection. The ICD is associated with the lowest sudden death mortality (1 to 2% per year) of any known form of therapy.74–77 The ICD is expensive ($12,000 to $20,000 for the generator and $2000 to $8000 for the lead system), with discomfort and lifestyle issues. Prophylactic ICD insertion for primary prevention is increasing.

Clinical trials including AVID, MUSST, MADIT I and II, SCD-HeFT, and COMPANION demonstrated benefits of ICD therapy.78–89 The CABG Patch Trial,90 which compared coronary artery bypass graft (CABG) to CABG+ICD, and DINAMIT,91 which studied ICD implantation early after myocardial infarction, failed to demonstrate advantages of ICD insertion. Trials now focus on prophylactic ICD therapy, CRT-Ds and cost-effectiveness. Accumulating evidence supports prophylactic use of CRT-Ds in patients with coronary disease or dilated cardiomyopathy.92 The appropriate role for prophylactic CRT-Ds is still evolving. The Heart Rhythm Society will not certify surgeons to implant ICDs unless they have passed its certification exam. Patients with prophylactic device insertion are being followed in a national registry. Information about this can be found at http://www.accncdr.com/webncdr/ICD. ICD insertion for primary prevention has doubled or tripled the number of candidates for these devices.

ICD battery life is greater than 5 years. Today's devices are highly programmable. Size and weight are similar to pacemakers of the 1970s. Lead systems have evolved from epicardial patches requiring thoracotomy to endocardial systems.74–77 Defibrillation thresholds (DFTs) are reduced by biphasic shocks93 and "hot can" technology.94 Pectoral implantation is the current standard, some deep to the pectoralis major. Abdominal implantation is now reserved for special cases (Figs. 59-25 and 59-26). Electrograms can be downloaded to expedite decisions about antiarrhythmics, prevent inappropriate shocks, and detect oversensing.95 Remote monitoring of ICD function is available from some manufacturers. VVI, DDD, biventricular, and antitachycardia pacing have been integrated into ICDs.96 Accelerated development has pressured the Food and Drug Administration to rapidly approve new technology. When the U.S. Health Care Finance Administration refused in 1995 to allow Medicare reimbursement to support device development, ICD development shifted overseas.

Physiology

Ventricular tachycardia (VT) associated with ischemic cardiomyopathy is commonly a reentrant arrhythmia that may be prevented by drugs, catheter ablation, or surgical maneuvers that alter the timing and electrical attributes of the reentrant circuit. Myocardial infarction creates the areas of scarring and slow conduction needed for reentry. Other forms of VT and ventricular fibrillation (VF) involve aberrancies of automaticity related to acute myocardial ischemia, increased ventricular wall stress, and myopathic cellular injury. Some class I antiarrhythmics have been shown to increase postinfarction mortality,97 possibly owing to proarrhythmic effects.

Indications

Candidates for ICD insertion for secondary prevention therapy have suffered documented VT or VF in the absence of acute myocardial infarction and have been proved unsuitable for antiarrhythmic drug or surgical therapy, based on programmed electrical stimulation studies in the EP laboratory. However, many patients who suffer cardiac arrest do not have inducible VT at EP study, and many patients with a history of syncope and presyncope have inducible VT but no history of a clinical arrhythmia. Many antiarrhythmics have negative inotropic and proarrhythmic effects. Serial EP studies of drug efficacy have been discredited in clinical trials.93

In mid-1996 the Food and Drug Administration approved an indication for prophylactic ICD insertion based on early termination of the MADIT Trial.83 If EP studies demonstrate inducible VT in patients with nonsustained VT and a history of myocardial infarction, an ICD is indicated. MADIT II results support ICD insertion in all patients with a history of myocardial infarction and left ventricular ejection fraction less than 30%. SCD-HeFT supports ICD insertion for patients with LVEF less than 36% and class II or III heart failure and either ischemic or dilated cardiomyopathy. Indications for ICD insertion for primary prevention therapy have recently been reviewed.98,99

Device Description

ICDs usually employ one bipolar lead for ventricular pacing/rate sensing and another to deliver the defibrillation current (Fig. 59-27). These are integrated into a single lead body for endocardial insertion. A unipolar ventricular lead paired with an "active can" delivers the defibrillation current in some designs. Rate alone currently identifies malignant rhythms for treatment; waveform is no longer monitored. Up to 30% of ICD shocks are inappropriate for sinus tachycardia or other supraventricular tachyarrhythmias.74,100 A bipolar atrial lead can help differentiate supraventricular arrhythmias from VT. Subcutaneous leads can be added for patients with high DFTs. Most implants are just caudad to the clavicle, but long lead lengths allow abdominal locations. Positive fixation leads are preferred.

The ICD contains a high-energy battery and a capacitor that steps up the output voltage to 600 to 800 V at 35 to 40 J. Shocks with positive and negative phases reduce DFTs. ICDs incorporate integrated circuits and a telemetry antenna. A broad range of programmable diagnostic and therapeutic functions are supported.

Surgical Procedure

Patient Preparation

Most ICD recipients are at increased mortality/morbidity risk for any surgery. Preoperative optimization of therapy for ischemia, heart failure, and systemic illnesses is critical. The mortality of ICD implantation in our program has been low since 1983, with ischemia more lethal than severe cardiomyopathy. Ischemia on stress testing mandates, CABG, coronary angioplasty, intra-aortic balloon pump, or deferral of DFT testing.

Surgical Approach

Epicardial patch leads101,102 are now rarely used. They fell into disfavor when the CABG Patch Trial demonstrated that infectious complications were more common in ICD recipients than control patients.90 Extrapericardial ICD patches cause less fibrosis, less impairment of diastolic properties,103,104 and less potential for graft impingement than intrapericardial patches. Biphasic waveforms, improved leads, high-output "hot can" generators, and subcutaneous leads have made the endocardial approach successful.

Manufacturer's Representatives

The complexity of current devices and a large number of spare parts, leads, and catheters have legitimized the presence of manufacturer's representatives during ICD implants. This increases influence of the manufacturer on the implant and warrants oversight. CRS regulations mandate electrophysiologist oversight of DFT measurement.

Technique

We prefer local anesthesia and unipolar cautery for ICD insertion, with invasive, radial artery pressure monitoring during DFT measurement. Positioning and draping are similar to that for pacemaker insertion, with adhesive R2 electrodes added over the right breast and beneath the left scapula. R2 electrodes should not be placed directly over the site of a possible subcutaneous ICD patch or array in the lateral axilla; this can result in serious equipment damage from arcing if external defibrillation is necessary.

Currently, deep sedation is used for induction and reversal of VF for DFT measurement. If multiple shocks are required, endotracheal intubation may be needed. We use intravenous cefazolin or combined vancomycin/gentamicin for antibiotic prophylaxis.

We prefer a cephalic vein cut down via a 6-cm incision over the deltopectoral groove. If the vein is small, a guidewire is used to position a 7- or 9-French introducer. Enlargement of the cephalic venotomy may be helpful if the vein is too small for the introducer (see Fig. 59-8). The guidewire can be left in place if a multiple-lead system is to be employed. Introducer kinking can lead to difficulty passing ICD leads. High central pressures can lead to bleeding until the introducer is removed. Introducer kinking is minimized by aligning the introducer with the course of the subclavian vein before removing the obturator.

As the lead enters the ventricle, VT or VF may be triggered. This is less disruptive if an external defibrillator is connected and precharged, with a capable OR nurse or electrophysiologist at the controls. The ventricular lead should be advanced gently as far as possible to the apex. Positive fixation ICD leads are preferred for both atrium and ventricle. Fluoroscopic and impedance checks for lead displacement are warranted during DFTs, particularly if external defibrillation is necessary. Intravenous norepinephrine is useful for blood pressure support. Transesophageal echocardiography is useful for monitoring patients under general anesthesia during DFTs. We do not use Swan-Ganz catheters for monitoring, because their removal can dislodge the ICD leads.

When a CRT-D (Fig. 59-28) is indicated, we insert the LV lead via cephalic cut down. The RV and RA leads are inserted next via subclavian puncture. Coronary sinus localization can be expedited by steerable electrophysiology catheters, dye injection to detect blood flow out of the coronary sinus, transesophageal echocardiography, and other techniques.

Left upper quadrant abdominal generator placement is reserved for cosmetic insertions, which may also situate the generator in the posterior rectus sheath (see Fig. 59-25). With general anesthesia or sedation/local anesthesia, leads are tunneled to the abdomen from the subclavian incision. Caution is exercised with the tunneler; it must pass anterior to the costal margin. The lead is looped in the shoulder and firmly secured to minimize the possibility that a tug on the abdominal end would displace the lead (see Fig. 59-25).

Defibrillation Threshold

The DFT test is the point of maximum risk in ICD insertion. VF is usually induced twice and reversed by the ICD. Backup defibrillation is available through cutaneous electrodes. The DFT should be at least 10 J less than the maximum ICD output. If the DFT is too high after optimizing RV lead location, an subcutaneous axillary patch or array is used to distribute the defibrillation current over the posterolateral left ventricle. High-output generators and reversed shock polarity may also help. DFT testing can depress cardiac function83–85,105,106 and cause a low-output state. The mortality of ICD insertion is about 1%.106 Complications include myocardial infarction, heart failure, infection, lead displacement, venous occlusion, tamponade, hemopneumothorax, and pocket hematoma.

Refractory Ventricular Fibrillation

An iatrogenic left pneumothorax can raise the DFT so that VF is refractory to external defibrillation. The differential diagnosis includes myocardial ischemia, electromechanical dissociation, and inappropriate ICD lead location. Refractory VF rarely may require open cardiac massage or cardiopulmonary bypass until a solution can be found.

Postoperative Care

Telemetry and an overnight stay are desirable after ICD insertion for VT or VF. Pacing and sensing properties confirm lead stability before discharge. In some cases, DFTs are also retested. An ICD magnet should be available for treatment of patients with freshly implanted ICDs. The magnet is used to inhibit inappropriate shocks caused by ventricular lead displacement or supraventricular arrhythmias. Patients undergoing ICD insertion for primary prevention may be discharged on the day of surgery. We administer ciprofloxacin for 5 days after discharge to complete prophylaxis based on a historical precedent. A postoperative office visit is scheduled 7 to 14 days after implant.

Surgical Follow-Up

At the postoperative office visit, lead position, patient symptoms, EP follow-up appointments, and the surgical site are assessed. If drainage is present, the wound is cultured and treated. For sterile drainage, ciprofloxacin or trimethoprim-sulfamethoxazole is administered for 10 days, and the patient is asked to keep the site dry until healing is complete. Refractory infection requires hospitalization and ICD explantation.

Device Follow-Up

ICDs require outpatient EP evaluation at 1- to 3-month intervals to cycle the capacitors, confirm battery life, test pacing thresholds, and download electrograms. Data are reviewed for aborted charging cycles arrhythmias. Programming is adjusted accordingly. For any shocks reported, telemetry can confirm proper function, detect inappropriate shocks, and demonstrate electrical noise or oversensing that requires lead revision. Wireless ICD monitoring from home is now a clinical reality for some ICDs.

Late Follow-Up and Generator Replacement

Battery Depletion

ICD battery life is now typically more than 60 months. Complications of generator replacement include infection, myocardial infarction, and death. Progression of heart failure and/or coronary artery disease increases the risk of replacement. Replacement is done under local anesthesia, with sedation during DFTs. DFTs may be deferred if the risk is excessive. Patients are discharged on the same day with antibiotic coverage, unless leads are replaced.

Lead Dysfunction

The incidence of failure increases with time. About 50% of ICD recipients have some lead revision by 10 years of follow-up.107–109 Problems include lead fracture, high DFTs, oversensing, undersensing, and exit block. The incidence of transvenous lead displacement is 7% and of fracture is 6% at 2-year follow-up.109 Oversensing can be caused by insulation damage. Oversensing can be corrected with a new transvenous rate-sensing lead, preferably of the positive-fixation type. Visible insulation damage inside the defibrillator pocket may be repairable, as described in the following.110

Patch leads can fail because of conductor fracture or distortion by fibrosis. With endocardial leads, DFTs may increase as a result of cardiac enlargement, which shifts the left ventricle laterally, away from the right ventricular lead. Insertion of a subcutaneous patch and/or a high-output generator can correct high DFTs.

Additional Issues

Bridge to Transplant

The benefits and liabilities of CRT-D therapy for VT/VF in patients who are awaiting cardiac allografting are under investigation, 98,99,111 with cost an important concern.

Quality of Life

Issues include the discomfort and distress of shocks and inconvenience of outpatient visits. These are particularly trying for elderly patients. ICD generators, though shrinking in size and weight, are bulky compared with pacemakers. Many patients are elated to be rescued by their ICD from a malignant arrhythmia, but others find this distressing.112-114 Many ICD patients do not comply with recommended limitations on driving automobiles. 115 Fortunately, the rate of accidents in ICD recipients is reported to be low.116

Cost-Effectiveness

ICD therapy is expensive, but VT/VF management in the absence of an ICD is also costly. The incremental cost per year of life added is estimated at $10,000 to $200,000.92,117,118 Reduced cost of ICD generators and leads would increase the economic appeal of ICD prophylaxis.

• Open surgery is now rarely required for insertion of arrhythmia control devices.
• Thoracic surgery remains essential for epicardial lead insertion and extraction and venous access in difficult patients.
• Resynchronization therapy is an evolving area in which the versatility of the thoracic surgeon can still play a vital role.
• Skills of thoracic surgeons with endocardial and epicardial leads remain valuable in specific clinical situations and warrant attention and renewal, lest they be lost.