American College of Cardiology

GREGORATOS ET AL., ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices
http://www.acc.org/clinical/guidelines/pacemaker/index.htm; 2002

ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarhythmia Devices—Full Text

A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/NASPE Committee on Implantation)

This is a Guideline Update of the 1997 Implantation of Cardiac Pacemakers and Antiarrythmia Devices Guidelines. To highlight the changes, deleted text is indicated by strikeout, and revised text is presented in brown. A clean version of the document, with changes fully incorporated, is available for download and print.

I. Indications for Permanent Pacing

A. Pacing for Acquired Atrioventricular Block in Adults

Atrioventricular (AV) block is classified as first-, second-, or third-degree (complete) block; anatomically, it is defined as supra-, intra-, or infra-His. First-degree AV block is defined as abnormal prolongation of the PR interval. Second-degree AV block is subclassified as type I and type II. Type I seconddegree AV block is characterized by progressive prolongation of the PR interval before a blocked beat and is usually associated with a narrow QRS complex. Type II second-degree AV block is characterized by fixed PR intervals before and after blocked beats(no progressive prolongation of PR interval before a blocked beat) and is usually associated with a wide QRS complex. When AV conduction occurs in a 2:1 pattern, block cannot be unequivocally classified as type I or type II, although the width of the QRS can be suggestive as just described. Advanced second-degree AV block refers to the block of two or more consecutive P waves but with some conducted beats, indicating some preservation of AV coneduction. Third-degree AV block (complete heart block) is defined as absence of AV conduction.

Patients with abnormalities of AV conduction may be asymptomatic or may experience serious symptoms related to bradycardia, ventricular arrhythmias, or both. Decisions regarding the need for a pacemaker are influenced importantly by the presence or absence of symptoms directly attributable to bradycardia. Furthermore, many of the indications for pacing have evolved over 3040 years based on experience without the benefit of comparative randomized clinical trials, in part because no acceptable alternative options exist to treat most bradycardias.

Nonrandomized studies strongly suggest that permanent pacing does improve survival in patients with third-degree AV block, especially if syncope has occurred ^8-13. Although there is little evidence to suggest that pacemakers improve survival in patients with isolated first-degree AV block ¹⁴, it is now recognized that marked (PR more than 300 milliseconds) first-degree AV block can lead to symptoms even in the absence of higher degrees of AV block ¹⁵. Such marked first-degree AV block may follow catheter ablation of the fast AV nodal pathway with resultant slow pathway conduction. When marked first-degree AV block for any reason causes atrial systole in close proximity to the preceding ventricular systole and produces hemodynamic consequences usually associated with retrograde (ventriculoatrial) conduction, signs and symptoms similar to the pacemaker syndrome may occur ¹⁶. Marked first-degree AV block for any reason may also be associated with a pseudopacemaker syndrome ¹⁶ secondary to close proximity of atrial systole to the preceding ventricular systole that produces hemodynamic consequences similar to those associated with retrograde (ventriculoatrial) conduction. With marked first-degree AV block, atrial contraction occurs before complete atrial filling, ventricular filling is compromised, and an increase in pulmonary capillary wedge pressure and a decrease in cardiac output follow. Small, uncontrolled trials have suggested some symptomatic and functional improvement by pacing of patients with PR intervals more than 0.30 seconds by decreasing the time for AV conduction ¹⁵. Finally, a long PR interval may identify a subgroup of patients with left ventricular (LV) dysfunction, some of whom may benefit from dual-chamber pacing with a short(er) AV delay ¹⁷. These same principles also may be applied to patients with type I second-degree AV block who experience hemodynamic compromise due to loss of AV synchrony, even without bradycardia. Consideration should be given to demonstrating hemodynamic improvement by Although echocardiographic or invasive techniques may be used to assess hemodynamic improvement assessment before permanent pacemaker implantation, such studies are not required.

Type I second-degree AV block is usually due to delay in the AV node irrespective of QRS width. Because progression to advanced AV block in this situation patients with type I second-degree AV block, when due to delay in the AV node, is uncommon ^18-20, pacing is usually not indicated unless the patient is symptomatic. Nevertheless, controversy exists, and pacemaker implantation has been advocated for this finding ^21-23. On the other hand, type II second-degree AV block is usually infranodal (either intra- or infra-His), especially when the QRS is wide. In these patients, symptoms are frequent, prognosis is compromised, and progression to third-degree AV block is common ^1820,24. Thus, type II second-degree AV block and a wide QRS indicate diffuse conduction system disease and constitute an indication for pacing even in the absence of symptoms. However, it is not always possible to determine the site of AV block without electrophysiologic evaluation, because type I second- degree AV block can be infranodal even when the QRS is narrow ³⁴⁰. If type I second-degree AV block with a narrow or wide QRS is found to be intra- or infra-His at electrophysiologic study, pacing should be considered.

Because it may be difficult for both patients and their physicians to attribute ambiguous symptoms such as fatigue to bradycardia, special vigilance must be exercised to acknowledge the patient’s concerns that may be caused by a slow heart rate. Thus, in a patient with third-degree AV block, permanent pacing should be considered strongly even when the ventricular rate is more than 40 beats per minute (bpm), because the choice of a 40 bpm cutoff in these guidelines was not determined from clinical trial data. Indeed, it is not the escape rate that is necessarily critical for safety, but rather the site of origin of the escape rhythm (i.e., in the AV node, the His bundle, or infra-His).

AV block can sometimes be provoked by exercise. If not secondary to myocardial ischemia, AV block in this circumstance usually is due to disease in the His-Purkinje system and is associated with a poor prognosis. Thus, pacing is indicated ^343,344. Conversely, long sinus pauses and AV block can occur during sleep apnea. In the absence of symptoms, these abnormalities are reversible and do not require pacing ³⁴⁵. If symptoms are present, pacing is indicated as in other conditions.

Recommendations for permanent pacemaker implantation in patients with AV block in AMI, congenital AV block, and AV block associated with enhanced vagal tone are discussed in separate sections. Neurocardiogenic mechanisms etiologies in young patients with AV block should be assessed before proceeding with permanent pacing. Physiologic AV block in the presence of supraventricular tachyarrhythmias does not constitute an indication for pacemaker implantation except as specifically defined in the recommendations that follow.

In general, the decision regarding implantation of a pacemaker must be considered with respect to whether or not AV block will be permanent. Reversible causes of AV block, such as electrolyte abnormalities, should be corrected first. Some diseases may follow a natural history to resolution (e.g., Lyme disease), and some AV block can be expected to reverse (e.g., hypervagotonia due to recognizable and avoidable physiologic factorsstimuli, perioperative AV block due to hypothermia, or inflammation near the AV conduction system after surgery for arrhythmias in this region).

Conversely, some conditions may warrant pacemaker implantation owing to the possibility anticipated adverse consequencesof disease progression even if the AV block reverses transiently (e.g., sarcoidosis, amyloidosis, and neuromuscular diseases). Finally, permanent pacing for AV block after valve surgery follows a variable natural history, and therefore the decision for permanent pacing is at the physician’s discretion ³⁴⁶.

Recommendations for Permanent Pacing in Acquired Atrioventricular Block in Adults

Class I

1. Third-degree and advanced second-degree AV block at any anatomic level, associated with any one of the following conditions:
   1. Bradycardia with symptoms (including heart failure) presumed to be due to AV block. (Level of Evidence: C)
   2. Arrhythmias and other medical conditions that require drugs that result in symptomatic bradycardia. (Level of Evidence: C)
   3. Documented periods of asystole greater than or equal to 3.0 seconds ²⁵ or any escape rate less than 40 bpm in awake, symptom-free patients ^26,27. (Level of Evidence: B, C)
   4. After catheter ablation of the AV junction. (Level of Evidence: B, C) There are no trials to assess outcome without pacing, and pacing is virtually always planned in this situation unless the operative procedure is AV junction modification ^28,29.
   5. Postoperative AV block that is not expected to resolve after cardiac surgery. (Level of Evidence: C) ^30,30a,346
   6. Neuromuscular diseases with AV block, such as myotonic muscular dystrophy, Kearns-Sayre syndrome, Erb’s dystrophy (limb-girdle), and peroneal muscular atrophy, with or without symptoms, because there may be unpredictable progression of AV conduction disease. (Level of Evidence: B) ^31-37
2. Second-degree AV block regardless of type or site of block, with associated symptomatic bradycardia. (Level of Evidence: B) ¹⁹

Class IIa

1. Asymptomatic third-degree AV block at any anatomic site with average awake ventricular rates of 40 bpm or faster, especially if cardiomegaly or LV dysfunction is present. (Level of Evidence: B, C)
2. Asymptomatic type II second-degree AV block with a narrow QRS. When type II second-degree AV block occurs with a wide QRS, pacing becomes a Class I recommendation (see next section regarding Pacing for Chronic Bifascicular and Trifascicular Block).
   (Level of Evidence: B) ^21,23
3. Asymptomatic type I second-degree AV block at intra- or infra-His levels found incidentally at electrophysiologic study performed for other indications. (Level of Evidence: B) ^19,21-23
4. First- or second-degree AV block with symptoms suggestive similar to those of pacemaker syndrome and documented alleviation of symptoms with temporary AV pacing. (Level of Evidence: B) ^15,16

Class IIb

1. Marked first-degree AV block (more than 0.30 seconds) in patients with LV dysfunction and symptoms of congestive heart failure in whom a shorter AV interval results in hemodynamic improvement, presumably by decreasing left atrial filling pressure. (Level of Evidence: C) ¹⁷
2. Neuromuscular diseases such as myotonic muscular dystrophy, Kearns-Sayre syndrome, Erb’s dystrophy (limb-girdle), and peroneal muscular atrophy with any degree of AV block (including first-degree AV block), with or without symptoms, because there may be unpredictable progression of AV conduction disease. (Level of Evidence: B) ^31-37

Class III

1. Asymptomatic first-degree AV block. (Level of Evidence: B) ¹⁴ (See also “Pacing for Chronic Bifascicular and Trifascicular Block.”)
2. Asymptomatic type I second-degree AV block at the supra-His (AV node) level or not known to be intra- or infra-Hisian. (Level of Evidence: B, C) ¹⁹
3. AV block expected to resolve and/or unlikely to recur ³⁸ (e.g., drug toxicity, Lyme disease, or during hypoxia in sleep apnea syndrome in absence of symptoms). (Level of Evidence: B)

B. Pacing for Chronic Bifascicular and Trifascicular Block

Bifascicular and trifascicular block refers to electrocardiographic (ECG) evidence of impaired conduction below the AV node in two or three fascicles of the right and left bundles. Alternating bundle-branch block (also known as bilateral bundle-branch block) refers to situations in which clear ECG evidence for block in all three fascicles is seen on successive ECGs. Examples are right bundle-branch block and left bundle-branch block on successive ECGs, or right bundle-branch block with associated left anterior fascicular block on one ECG and associated left posterior fascicular block on another ECG. A strict definition of trifascicular block is block documented in all three fascicles whether simultaneously or at different times. Alternating bundlebranch block also fulfills this criterion. This term has also been used to describe first-degree AV block in association with bifascicular block. Patients with such ECG abnormalities there is convincing evidence thatand symptomatic, advanced AV block is associated with have a high mortality rate and a significant incidence of sudden death ^9,39. Although third-degree AV block is most often preceded by bifascicular block, there is impressive evidence that the rate of progression of bifascicular block to third-degree AV block is slow ³⁴⁷. Furthermore, no single clinical or laboratory variable, including bifascicular block, identifies patients at high risk of death from a future bradyarrhythmia due to bundle-branch block ⁴⁸.

Syncope is common in patients with bifascicular block. Although syncope may be recurrent, it is not associated with an increased incidence of sudden death ^40-52. Although pacing relieves the transient neurological symptoms, it does not reduce the frequencyoccurrence of sudden death ⁴⁶. Electrophysiologic study may be helpful to evaluate and direct the treatment of inducible ventricular arrhythmias ^53,54 that are common in patients with bifascicular and trifascicular block. There is convincing evidence that in the presence of permanent or transient third-degree AV block, syncope is associated with an increased incidence of sudden death regardless of the results of electrophysiologic study ^9,54,55. Finally, if the cause of syncope in the presence of bifascicular or trifascicular block cannot be determined with certainty or if treatments used (such as drugs) may exacerbate AV block, prophylactic permanent pacing is indicated, especially if syncope may have been due to transient thirddegree AV block ^40,52.

Of the many laboratory variables, the PR and HV intervals have been identified as possible predictors of third-degree AV block and sudden death. Although PR interval prolongation is common in patients with bifascicular block, the delay is often at the level of the AV node. There is no correlation between the PR and HV intervals or between the length of the PR interval, progression to third-degree AV block, and sudden death ^43,45,49. Although most patients with chronic or intermittent third-degree AV block demonstrate prolongation of the HV interval during anterograde conduction, some investigators ^50,51 have suggested that asymptomatic patients with bifascicular block and a prolonged HV interval should be considered for permanent pacing, especially if the HV interval is greater than or equal to 100 milliseconds ⁴⁹. The evidence indicates that although the prevalence of prolonged HV is high, the incidence of progression to third-degree AV block is low. Because HV prolongation accompanies advanced cardiac disease and is associated with increased mortality, death is often not sudden or due to AV block but rather due to the underlying heart disease itself and nonarrhythmic cardiac causes ^{43,46-49,51,54-56}.

Atrial pacing at electrophysiologic study in asymptomatic patients as a means of identifying patients at increased risk of future high- or third-degree AV block is probably not justifiedcontroversial. The probability of inducing block distal to the AV node (i.e., intra- or infra-His) with rapid atrial pacing is low ^{47,50,51,57-60}. Failure to induce distal block cannot be taken as evidence that the patient will not develop third-degree AV block in the future. However, if atrial pacing induces nonphysiologic infra-His block, some consider this an indication for pacing ⁵⁷.

Recommendations for Permanent Pacing in Chronic Bifascicular and Trifascicular Block

Class I

1. Intermittent third-degree AV block. (Level of Evidence: B) ^8-13,39
2. Type II second-degree AV block. (Level of Evidence: B) ^18,20,24,348
3. Alternating bundle-branch block. (Level of Evidence: C) ³⁴⁹

Class IIa

1. Syncope not proveddemonstrated to be due to AV block when other likely causes have been excluded, specifically ventricular tachycardia (VT). (Level of Evidence: B) ^40-51,53-58
2. Incidental finding at electrophysiologic study of markedly prolonged HV interval (greater than or equal to 100 milliseconds) in asymptomatic patients. (Level of Evidence: B) ⁴⁹
3. Incidental finding at electrophysiologic study of pacing- induced infra-His block that is not physiologic. (Level of Evidence: B) ⁵⁷

Class IIb

None. Neuromuscular diseases such as myotonic muscular dystrophy, Kearns-Sayre syndrome, Erb’s dystrophy (limb-girdle), and peroneal muscular atrophy with any degree of fascicular block, with or without symptoms, because there may be unpredictable progression of AV conduction disease. (Level of Evidence: C) ^31-37

Class III

1. Fascicular block without AV block or symptoms. (Level of Evidence: B) ^43,45,48,49
2. Fascicular block with first-degree AV block without symptoms. (Level of Evidence: B) ^43,45,48,49

C. Pacing for Atrioventricular Block Associated With Acute Myocardial Infarction

Indications for permanent pacing after myocardial infarction (MI) in patients experiencing AV block are related in large measure to the presence of intraventricular conduction defects. Unlike some other indications for permanent pacing, the criteria for patients with MI and AV block do not necessarily depend on the presence of symptoms. Furthermore, the requirement for temporary pacing in AMI does not by itself constitute an indication for permanent pacing [see ACC/AHA Guidelines for Management of Patients With Acute Myocardial Infarction ²³³⁵].

The long-term prognosis for survivors of AMI who have had AV block is related primarily to the extent of myocardial injury and the character of intraventricular conduction disturbancesdisturbances rather than the AV block itself ^11,61-64. Patients with AMI who have intraventricular conduction defects, with the exception of isolated left anterior fascicular block, have an unfavorable short- and long-term prognosis and an increased risk of sudden death ^11,24,61,63. This unfavorable prognosis is not necessarily due to development of highgrade AV block, although the incidence of such block is higher in postinfarction patients with abnormal intraventricular conduction ^61,65,350.

When AV or intraventricular conduction block complicates AMI, the type of conduction disturbance, location of infarction, and relation of electrical disturbance to infarction must be considered if permanent pacing is contemplated. Even with data available, the decision is not always straightforward, because the reported incidence and significance of various conduction disturbances vary widely ⁶⁶. Despite the use of thrombolytic therapy and primary angioplasty, which have decreased the incidence of AV block in AMI, mortality remains high if AV block occurs ^67-70.

Although more severe disturbances in conduction are in general associated with greater arrhythmic and nonarrhythmic mortality ^61-66>, the impact of pre-existing bundlebranch block on mortality after AMI is controversial ^52,66. A particularly ominous prognosis is associated with left bundle-branch block combined with advanced second- or thirddegree AV block and with right bundle-branch block combined with left anterior or left posterior fascicular block ^41,52,62,64. Irrespective of whether the infarction is anterior or inferior, the development of an intraventricular conduction delay reflects extensive myocardial damage rather than an electrical problem in isolation ⁶⁴. Although AV block that occurs during inferior MI can be associated with a favorable long-term clinical outcome, in-hospital survival is impaired, irrespective of temporary or permanent pacing in this situation ^67,68,71,72. Furthermore, pacemakers should not be implanted if the peri-infarctional AV block is expected to resolve or to not negatively impact long-term prognosis, as in the case of inferior MI ⁶⁹.

Recommendations for Permanent Pacing After the Acute Phase of Myocardial Infarction*

Class I

1. Persistent second-degree AV block in the His-Purkinje system with bilateral bundle-branch block or third degree AV block within or below the His-Purkinje system after AMI. (Level of Evidence: B) ^24,61-65
2. Transient advanced (second- or third-degree) infranodal AV block and associated bundle-branch block. If the site of block is uncertain, an electrophysiologic study may be necessary. (Level of Evidence: B) ^61,62
3. Persistent and symptomatic second- or third-degree AV block. (Level of Evidence: C)

Class IIb

Persistent second- or third-degree AV block at the AV node level. (Level of Evidence: B) ²³

Class III

1. Transient AV block in the absence of intraventricular conduction defects. (Level of Evidence: B) ⁶¹
2. Transient AV block in the presence of isolated left anterior fascicular block. (Level of Evidence: B) ⁶³
3. Acquired left anterior fascicular block in the absence of AV block. (Level of Evidence: B) ⁶¹
4. Persistent first-degree AV block in the presence of bundle-branch block that is old or age indeterminate. (Level of Evidence: B) ⁶¹*

*These recommendations generally follow the ACC/AHA Guidelines for the Management of Patients With Acute Myocardial Infarction ²³³⁵.

D. Pacing in Sinus Node Dysfunction

Sinus node dysfunction (sick sinus syndrome) constitutes a spectrum of cardiac arrhythmias, including sinus bradycardia, sinus arrest, sinoatrial block, and paroxysmal supraventricular tachyarrhythmias alternating with periods of bradycardia or even asystole. Patients with this condition may be symptomatic from paroxysmal tachycardia or bradycardia or both. Correlation of symptoms with the above arrhythmias by use of an ECG, ambulatory ECG monitoring, or an event recorder is essential. This correlation may be difficult because of the intermittent nature of the episodes. In the electrophysiology laboratory, abnormal sinus node function may be confirmed by demonstration of prolonged corrected sinus node recovery times or prolonged sinoatrial conduction times. However, utility of electrophysiologic studies for sinus node dysfunction is limited by issues of sensitivity and specificity.

Sinus node dysfunction may express itself as chronotropic incompetence in which there is an inadequate sinus response to exercise or stress. Rate-responsive pacemakers have clinically benefited patients by restoring physiologic heart rate during physical activity ^73-75.

Sinus bradycardia is accepted as a physiologic finding in trained athletes, who not uncommonly have a heart rate of 40 to 50 bpm while at rest and awake and may have a sleeping rate as slow as 30 bpm, with sinus pauses or type I seconddegree AV block producing asystolic intervals as long as 2.8 seconds ^76-78. These findings are due to increased vagal tone.

Although sinus node dysfunction is frequently the primary indication for implantation of permanent pacemakers ⁷³, permanent pacing in patients with sinus node dysfunction may not necessarily result in improved survival time ^26,79, although symptoms related to bradycardia may be relieved ^27,80 (see Section I, Selection of Pacemaker Devices). Nonrandomized observational studies suggest that dualchamber pacing may improve survival, improve quality of life and decrease morbidity (stroke and atrial fibrillation) compared with ventricular pacing ^81-83. However, one randomized prospective study 84 of 225 patients with sinus node disease and intact AV nodal conduction followed for a mean of 40 months demonstrated no difference in overall or cardiac mortality between the groups receiving atrial versus ventricular pacing. Multiple small studies suggest that dualchamber pacing improves quality of life and decreases morbidity (atrial fibrillation, stroke). Multiple prospective trials are ongoing to assess the superiority of dual-chamber versus ventricular-based pacing systems in this population (84a). During monitoring, pauses are sometimes observed during sleep. Duration of sinus pauses and their clinical significance is uncertain. If due to sleep apnea, apnea should be treated. A small retrospective trial of atrial overdrive pacing in the treatment of sleep apnea demonstrated a decrease “in episodes of central or obstructive sleep apnea without reducing the total sleep time” ⁴⁴⁷⁾. Although this initial trial is encouraging, it is premature to propose pacing guidelines until a larger body of data is available. Otherwise, there is not sufficient evidence to distinguish physiologic from pathologic nocturnal bradycardia.

Recommendations for Permanent Pacing in Sinus Node Dysfunction

Class I

1. Sinus node dysfunction with documented symptomatic bradycardia, including frequent sinus pauses that produce symptoms. In some patients, bradycardia is iatrogenic and will occur as a consequence of essential long-term drug therapy of a type and dose for which there are no acceptable alternatives. (Level of Evidence: C) ^27,73,79
2. Symptomatic chronotropic incompetence. (Level of Evidence: C) ^27,73-75,79

Class IIa

1. Sinus node dysfunction occurring spontaneously or as a result of necessary drug therapy, with heart rate less than 40 bpm when a clear association between significant symptoms consistent with bradycardia and the actual presence of bradycardia has not been documented. (Level of Evidence: C) ^{26,27,73,78-80}
2. Syncope of unexplained origin when major abnormalities of sinus node function are discovered or provoked in electrophysiologic studies. (Level of Evidence: C) ^351,352

Class IIb

In minimally symptomatic patients, chronic heart rate less than 3040 bpm while awake. (Level of Evidence: C) ^{26,27,73,78-80}

Class III

1. Sinus node dysfunction in asymptomatic patients, including those in whom substantial sinus bradycardia (heart rate less than 40 bpm) is a consequence of long-term drug treatment.
2. Sinus node dysfunction in patients with symptoms suggestive of bradycardia that are clearly documented as not associated with a slow heart rate.
3. Sinus node dysfunction with symptomatic bradycardia due to nonessential drug therapy.

E. Prevention and Termination of Tachyarrhythmias by Pacing

Under certain circumstances, an implanted pacemaker may be useful for treating patients with recurrent symptomatic ventricular and supraventricular tachycardias ^85-94. Pacing can be useful in preventing and terminating arrhythmias. Reentrant rhythms including atrial flutter, paroxysmal reentrant supraventricular tachycardia, and VT may be terminated by a variety of pacing patterns, including programmed stimulation and short bursts of rapid pacing ^95,96. These antitachyarrhythmia devices may detect tachycardia and automatically activate a pacing sequence, or they may respond only to an external instruction (for example, application of a magnet).

Prevention of arrhythmias by pacing has been demonstrated in certain situations. In some patients with the long-QT syndrome, recurrent pause-dependent VT may be prevented by continuous pacing ⁹⁷. A combination of pacing and beta-blockade has been reported to shorten the QT interval and help prevent sudden cardiac death ^98,99. ICD therapy in combination with overdrive suppression pacing should be considered in high-risk patients.

Atrial synchronous ventricular pacing may prevent recurrences of re-entrant supraventricular tachycardia ¹⁰⁰ although this technique is rarely used given the availability of catheter ablation and other alternative therapies. Although ventricular ectopic activity may be suppressed by such pacing in other conditions, serious or symptomatic arrhythmias are rarely prevented ¹⁰¹. In some patients with bradycardia- dependent atrial fibrillation, atrial pacing may be effective in reducing the frequency of recurrences ⁹². In the Mode Selection Trial (MOST), 2010 patients with sinus node dysfunction were randomized between DDDR and VVIR pacing. After a mean follow-up of 33 months, there was a 21% lower risk of atrial fibrillation (p = 0.008) in the DDDR group than in the VVVIR group ³⁵³. Other trials are under way to assess the efficacy of atrial overdrive pacing algorithms and algorithms that react to premature atrial complexes in preventing atrial fibrillation, but data to date are sparse. Dual-site right atrial pacing or alternate single-site atrial pacing from nonconventional sites (e.g., septal or Bachmann’s bundle) may offer additional benefits to single-site right atrial pacing from the appendage in patients with symptomatic drug-refractory atrial fibrillation and concomitant bradyarrhythmias ⁹³. In patients with sick sinus syndrome and intra-atrial block (P wave more than 180 milliseconds), biatrial pacing may lower recurrence rates of atrial fibrillation ⁹⁴.

Potential recipients of antitachyarrhythmia devices that interrupt arrhythmias should undergo extensive testing before implantation to ensure that the devices safely and reliably terminate the ectopic mechanism without accelerating the tachycardia or inducing ventricular fibrillation (VF). Patients for whom an antitachycardia pacemaker has been prescribed have usually been unresponsive to antiarrhythmic drugs or were receiving agents that could not control their cardiac arrhythmias. When permanent antitachycardia pacemakers detect and interrupt supraventricular tachycardia, all pacing should be done in the atrium, because adverse interactions have been reported ^85,102 with use of ventricular pacing to interrupt supraventricular arrhythmias. Permanent antitachycardia pacing as monotherapy for VT is not appropriate given that antitachycardia pacing algorithms are available in tiered-therapy ICDs that have the capability of cardioversion and defibrillation in cases when antitachycardia pacing is ineffective or causes acceleration of the treated tachycardia.

Recommendations for Permanent Pacemakers That Automatically Detect and Pace to Terminate Tachycardias

Class I

None.

Class I

1. Symptomatic recurrent supraventricular tachycardia that is reproducibly terminated by pacing after drugs and catheter ablation fail to control the arrhythmia or produce intolerable side effects. (Level of evidence: C) (86-88, 90, 91)
2. Symptomatic recurrent sustained VT as part of an automatic defibrillator system. (Level of evidence: B) (103-105)

Class IIa

Symptomatic recurrent supraventricular tachycardia that is reproducibly terminated by pacing in the unlikely event that catheter ablation and/or drugs fail to control the arrhythmia or produce intolerable side effects. (Level of Evidence: C) ^86-88,90,91

Class IIb

Recurrent supraventricular tachycardia or atrial flutter that is reproducibly terminated by pacing as an alternative to drug therapy or ablation. (Level of Evidence: C) ^85-88,90,91

Class III

1. Tachycardias frequently accelerated or converted to fibrillation by pacing.
2. The presence of accessory pathways with the capacity for rapid anterograde conduction whether or not the pathways participate in the mechanism of the tachycardia.

Pacing Recommendations to Prevent Tachycardia

Class I

Sustained pause-dependent VT, with or without prolonged QT, in which the efficacy of pacing is thoroughly documented. (Level of Evidence: C) ^97,98

Class IIa

High-risk patients with congenital long-QT syndrome. (Level of Evidence: C) ^97,98

Class IIb

1. AV re-entrant or AV node re-entrant supraventricular tachycardia not responsive to medical or ablative therapy. (Level of Evidence: C) ^87,88,92
   2. Prevention of symptomatic, drug-refractory, recurrent atrial fibrillation. (Level of evidence: C) (93, 94)
2. Prevention of symptomatic, drug-refractory, recurrent atrial fibrillation in patients with coexisting sinus node dysfunction. (Level of Evidence: B) ^{93,94,354,355}

Class III

1. Frequent or complex ventricular ectopic activity without sustained VT in the absence of the long-QT syndrome.
2. Long QT syndrome Torsade de Pointes VT due to reversible causes.

F. Pacing in Hypersensitive Carotid Sinus and Neurocardiogenic SyncopeSyndromes

The hypersensitive carotid sinus syndrome is defined as syncope or presyncope resulting from an extreme reflex response to carotid sinus stimulation. It is an uncommon cause of syncope. There are two components of the reflex:

1. Cardioinhibitory, resulting from increased parasympathetic tone and manifested by slowing of the sinus rate or prolongation of the PR interval and advanced AV block, alone or in combination.
2. Vasodepressor, secondary to a reduction in sympathetic activity resulting in loss of vascular tone and hypotension. This effect is independent of heart rate changes.

Before concluding that permanent pacing is clinically indicated, the physician should determine the relative contribution of the two components of carotid sinus stimulation to the individual patient’s symptom complex. Hyperactive response to carotid sinus stimulation is defined as asystole due to either sinus arrest or AV block of more than 3 seconds, or a substantial symptomatic decrease in systolic blood pressure, or both ¹⁰⁶. Pauses up to 3 seconds during carotid sinus massage are considered to be within normal limits. Such heart rate and hemodynamic responses may occur in normal subjects and patients with coronary artery disease. The cause-and-effect relation between the hypersensitive carotid sinus and the patient’s symptoms must be made with great caution ¹⁰⁷. Spontaneous syncope reproduced by carotid sinus stimulation should alert the physician to the presence of this syndrome. Minimal pressure on the carotid sinus in elderly patients or patients receiving digitalis may result in marked changes in heart rate and blood pressure yet not be of clinical significance. Permanent pacing for patients with pure excessive cardioinhibitory response to carotid stimulation is effective in relieving symptoms ^108,109. Because 10% to 20% of patients with this syndrome may have an important vasodepressor component of their reflex response, it is desirable to define this component before concluding that all symptoms are related to asystole alone. Among patients whose reflex response includes both cardioinhibitory and vasodepressor components, attention to the latter is essential for effective therapy in patients undergoing pacing.

Evidence has emerged that suggests that elderly patients who have sustained otherwise unexplained falls may have carotid sinus hypersensitivity ³⁵⁶. In a subsequent study, 175 elderly patients who had fallen without loss of consciousness and had pauses greater than 3 seconds during carotid sinus massage (thus fulfilling the diagnosis of carotid sinus hypersensitivity) were randomized to pacing or nonpacing therapy. The paced group had a significantly lower likelihood of subsequent falling episodes during follow-up ³⁵⁷.

Neurally mediated syncope accounts for 10% to 40% of syncope patients. Neurocardiogenic syncope and neurocardiogenic syndromes refer to a variety of clinical scenarios in which triggering of a neural reflex results in a usually selflimited episode of systemic hypotension characterized by both bradycardia and peripheral vasodilation ¹¹⁰. Neurocardiogenic syncope accounts for 10% to 40% of syncope episodes. Vasovagal syncope is a term used to denote one of the most common clinical scenarios within the category of neurocardiogenic syncopal syndromes. Patients classically have a prodrome of nausea and diaphoresis (often absent in the elderly), and there may be a positive familial history of the condition. Spells may be triggered by pain, anxiety, stress, or crowded conditions. Typically, no evidence of structural heart disease is present. Other causes of syncope such as LV outflow obstruction, bradyarrhythmias, and tachyarrhythmias should be excluded. Head-up tilt-table testing may be diagnostic.

The role of permanent pacing in refractory neurocardiogenic syncope associated with significant bradycardia or asystole is controversial. Approximately 25% of patients have a predominant vasodepressor reaction without significant bradycardia ¹¹¹. An additional large percentage of patients will have a mixed vasodepressor/vasoinhibitory component of their symptoms. While one group of investigators have noted some benefit of pacing in these patients ^112,113, another study using a pacing rate 20% higher than the resting heart rate demonstrated that pacing did not prevent syncope any better than pharmacotherapy ¹⁰⁶. Because most individuals with neurocardiogenic syncope have a slowing of heart rate after the fall in blood pressure, pacing may be ineffective in most patients. Dual-chamber pacing, carefully prescribed on the basis of tilt-table test results, may be effective in reducing symptoms if the patient has a significant cardioinhibitory component to the cause of their symptoms ¹¹⁴. Results from a randomized trial (115)^358,359 in highly symptomatic patients with bradycardia demonstrated that permanent pacing increased the time to first syncopal event (P < .0007). In one of these trials ³⁵⁸, the actuarial rate of recurrent syncope at 1 year was 18.5% for pacemaker patients and 59.7% for control patients. The specific modality of pacing under these circumstances is under active investigation. One study demonstrated that DDD pacing with rate-drop response function was more effective than beta-blockade in preventing recurrent syncope in highly symptomatic patients with vasovagal syncope and relative bradycardia during tilt-table testing ³⁶⁰. Although spontaneous or provoked prolonged pauses are a concern in this population, the prognosis without pacing is excellent ¹¹⁶. Several investigators have concluded that some patients with syncope of undetermined origin may benefit from pacing if findings strongly suggestive of bradycardic etiology are discovered or provoked at electrophysiologic study ^117,118,361.

The evaluation of patients with syncope of undetermined origin should take into account clinical status and not overlook other, more serious causes of syncope such as ventricular tachyarrhythmias.

Recommendations for Permanent Pacing in Hypersensitive Carotid Sinus Syndrome and Neurocardiogenic Syncope

Class I

Recurrent syncope caused by carotid sinus stimulation; minimal carotid sinus pressure induces ventricular asystole of more than 3 seconds’ duration in the absence of any medication that depresses the sinus node or AV conduction. (Level of Evidence: C) ^108,109

Class IIa

1. Recurrent syncope without clear, provocative events and with a hypersensitive cardioinhibitory response. (Level of Evidence: C) ^108,109
   2. Syncope of unexplained origin when major abnormalities of sinus node function or AV conduction are discovered or provoked in electrophysiologic studies. (Level of evidence: C)
2. Significantly symptomatic and recurrent neurocardiogenic syncope associated with bradycardia documented spontaneously or at the time of tilt-table testing. (Level of Evidence: B) ^358-360,362

Class IIb

Neurally mediated syncope with significant bradycardia reproduced by a head-up tilt with or without isoproterenol or other provocative maneuvers. (Level of evidence: B) (112-115)

Class III

1. A hyperactive cardioinhibitory response to carotid sinus stimulation in the absence of symptoms or in the presence of vague symptoms such as dizziness, lightheadedness, or both. (Level of Evidence: C)
   2. A hyperactive cardioinhibitory response to carotid sinus stimulation in the presence of vague symptoms such as dizziness, light-headedness, or both.
2. Recurrent syncope, lightheadedness, or dizziness in the absence of a hyperactive cardioinhibitory response. (Level of Evidence: C)
3. Situational vasovagal syncope in which avoidance behavior is effective. (Level of Evidence: C)

G. Pacing in Children, Adolescents, and Patients With Congenital Heart Disease

The indications for permanent cardiac pacemaker implantation in the child, adolescent, or young adult with congenital heart disease may be considered broadly as 1) symptomatic sinus bradycardia, 2) the recurrent bradycardia-tachycardia syndromes, 3) congenital third-degree AV block, and 4) advanced second- or third-degree AV block, either surgical or acquired. Although the general indications for pacemaker implantation in children are similar to those in adults, there are several important considerations in young patients. First, an increasing number of patients are surviving complex surgical procedures for congenital heart disease that result in palliation rather than correction of circulatory physiology. The residua of impaired ventricular function and abnormal physiology may result in symptomatic bradycardia at rates that do not produce symptoms in persons with normal cardiovascular physiology. Hence, the indications for pacemaker implantation in these patients need to be based on the correlation of symptoms with relative bradycardia rather than absolute heart rate criteria. Second, the clinical significance of bradycardia is age dependent; whereas a heart rate of 45 bpm may be a normal finding in an adolescent, the same rate in a newborn or infant indicates profound bradycardia.

Bradycardia and associated symptoms in children are often transient (e.g., paroxysmal AV block or sinus arrest) and difficult to document. Although sinus node dysfunction (sick sinus syndrome) is increasingly recognized in pediatric patients, it is not itself an indication for pacemaker implantation. In the young patient with sinus bradycardia, the primary criterion for a pacemaker is the concurrent observation of a symptom (e.g., syncope) with bradycardia (e.g., heart rate less than 35 to 40 bpm or asystole more than 3 seconds) ^25,27,119. In general, correlation of symptoms with bradycardia is determined by 24-hour ambulatory or transtelephonic electrocardiography. Symptomatic bradycardia (as defined) is considered an indication for pacemaker implantation, provided that other causes of the symptom(s) have been excluded. Alternative causes to be considered include seizures, breath holding, apnea, or neurocardiogenic mechanisms.

The bradycardia-tachycardia syndrome (sinus bradycardia alternating with atrial flutter or re-entrant atrial tachycardia) is an increasingly frequent problem in young patients following surgery for congenital heart disease. Substantial morbidity and mortality have been observed in young patients with recurrent or chronic atrial flutter, with the loss of sinus rhythm an independent risk factor for subsequent development of atrial flutter ^120,121. Thus, both long-term atrial pacing at physiologic rates as well as atrial antitachycardia pacing have been reported for treatment of sinus bradycardia and prevention or termination of recurrent episodes of tachycardia ^122,123. To date, the results of pacing for the bradycardia- tachycardia syndrome in children have been equivocal and the source of considerable controversy ^124,125. It is clear that long-term drug therapy (e.g., sotalolpropranolol or amiodarone) deemed essential for the control of atrial flutter may result in symptomatic bradycardia in some patients, whereas the use of other antiarrhythmic agents (e.g., quinidine) may potentially increase the risk of ventricular arrhythmias or sudden death in the presence of profound bradycardia. Thus, in young patients with recurrent arrhythmias associated with the bradycardia-tachycardia syndrome, permanent pacing should be considered as an adjunctive form of therapy. As an alternative therapy to antiarrhythmic medications that result in profound bradycardia and the need for pacemaker implantation, radiofrequency catheter ablation may modify the anatomic substrate of tachycardia in select patients with congenital heart disease.

Indications for permanent pacing in young patients with congenital complete AV block continue to evolve, based on improved definition of the natural history of the disease as well as advances in pacemaker technology and diagnostic methods. In several studies it has been observed that pacemaker implantation may improve long-term survival and prevent syncopal episodes among asymptomatic patients with congenital complete AV block ^126,127. Periodic evaluation of ventricular function is required in patients with congenital AV block, even after pacemaker implantation ³⁶³. Several criteria (average heart rate, pauses in the intrinsic rate, associated structural heart disease, prolonged QT interval, and exercise tolerance) must be considered in the asymptomatic patient with congenital complete AV block ^128-130.

The use of cardiac pacing with beta-blockade for prevention of symptoms in patients with the congenital long-QT syndrome is supported by observationalrecent studies ^98,131,364. The primary benefit of pacemaker therapy may be in patients with pause-dependent initiation of ventricular tachyarrhythmias ¹³² or those with sinus bradycardia or advanced AV block in association with the congenital long- QT syndrome ^133,134. Although pacemaker implantation may reduce the incidence of symptoms in these patients, long-term benefit on risk of sudden cardiac arrest remains to be determined ^98,131,133.

A poor prognosis has been established for patients with permanent postsurgical AV block who do not receive permanent pacemakers for rate support ¹³⁵. The presence of advanced second- or third-degree AV block persisting for 7 to 14 days after cardiac surgery is considered a Class I indication for pacemaker implantation ¹³⁶. The need for pacing in patients with transient advanced AV block with residual bifascicular block is less certain, whereas patients in whom AV conduction returns to normal generally have a favorable prognosis ¹³⁷.

Additional details that need to be considered in pacemaker implantation in young patients include risk of paradoxic embolism withdue to thrombus formation on an endocardial lead system in the presence of residual intracardiac defects and the lifelong need for permanent cardiac pacing ^138,139. Decisions about pacemaker implantation must also take into account implantation technique (transvenous versus epicardial) and long-term vascular access.

Recommendations for Permanent Pacing in Children, Adolescents, and Patients With Congenital Heart Disease

Class I

1. Advanced second- or third-degree AV block associated with symptomatic bradycardia, congestive heart failureventricular dysfunction, or low cardiac output. (Level of Evidence: C)
2. Sinus node dysfunction with correlation of symptoms during age-inappropriate bradycardia. The definition of bradycardia varies with the patient’s age and expected heart rate. (Level of Evidence: B) ^25,27,119
3. Postoperative advanced second- or third-degree AV block that is not expected to resolve or persists at least 7 days after cardiac surgery. (Level of Evidence: B, C) (135,136)^365,366
4. Congenital third-degree AV block with a wide QRS escape rhythm, complex ventricular ectopy, or ventricular dysfunction. (Level of Evidence: B) ^127,129,363
5. Congenital third-degree AV block in the infant with a ventricular rate less than 50 to 55 bpm or with congenital heart disease and a ventricular rate less than 70 bpm. (Level of Evidence: B, C) ^129,130
6. Sustained pause-dependent VT, with or without prolonged QT, in which the efficacy of pacing is thoroughly documented. (Level of Evidence: B) ^{97,98,131,132}

Class IIa

1. Bradycardia-tachycardia syndrome with the need for long-term antiarrhythmic treatment other than digitalis. (Level of Evidence: C) ^123,124
2. Congenital third-degree AV block beyond the first year of life with an average heart rate less than 50 bpm, abrupt pauses in ventricular rate that are two or three times the basic cycle length, or associated with symptoms due to chronotropic incompetence. (Level of Evidence: B) ¹²⁸
3. Long-QT syndrome with 2:1 AV or third-degree AV block. (Level of Evidence: B) ^133,134
4. Asymptomatic sinus bradycardia in the child with complex congenital heart disease with resting heart rate less than 3540 bpm or pauses in ventricular rate more than 3 seconds. (Level of Evidence: C)
5. Patients with congenital heart disease and impaired hemodynamics due to sinus bradycardia or loss of AV synchrony. (Level of Evidence: C)

Class IIb

1. Transient postoperative third-degree AV block that reverts to sinus rhythm with residual bifascicular block. (Level of Evidence: C) ¹³⁷
2. Congenital third-degree AV block in the asymptomatic infant, neonate,child, or adolescent, or young adult with an acceptable rate, narrow QRS complex, and normal ventricular function. (Level of Evidence: B) ^126,127
3. Asymptomatic sinus bradycardia in the adolescent with congenital heart disease with resting heart rate less than 3540 bpm or pauses in ventricular rate more than 3 seconds. (Level of Evidence: C)
4. Neuromuscular diseases with any degree of AV block (including first-degree AV block), with or without symptoms, because there may be unpredictable progression of AV conduction disease.

Class III

1. Transient postoperative AV block with return of normal AV conduction within 7 days. (Level of Evidence: B) ^136,137
2. Asymptomatic postoperative bifascicular block with or without first-degree AV block. (Level of Evidence: C)
3. Asymptomatic type I second-degree AV block. (Level of Evidence: C)
4. Asymptomatic sinus bradycardia in the adolescent with longest RR interval less than 3 seconds and minimum heart rate more than 40 bpm. (Level of Evidence: C) ¹⁴⁰

H. Pacing in Specific Conditions

1. Hypertrophic Obstructive Cardiomyopathy
Early observational studies suggested that pacing the right ventricular apex would reduce the LV outflow gradient. In patients with severely symptomatic hypertrophic cardiomyopathy, implantation of a dual-chamber pacemaker with a short AV delay has been shown to decrease the magnitude of LV outflow obstruction and alleviate symptoms (141-143). These findings come from nonrandomized unblinded studies. The mechanisms by which pacing improves the LV outflow gradient are not completely understood. Pacing therapy can change the ventricular contraction pattern by prematurely activating part of the ventricle, creating a regional dyssynchrony. This early paced portion of the ventricle faces low chamber pressure and stress and contracts against a lower afterload (144). Altered LV activation causes disordered ventricular contractility with late septal activation, increases LV systolic dimension, and reduces systolic anterior motion of the mitral valve. Thus, LV outflow obstruction is reduced and the atrial contribution to LV filling is maintained. Selection of an optimal AV delay appears to be critical in achieving an optimal hemodynamic result (142,145). The optimal AV delay appears to be the longest AV interval that consistently results in a completely paced QRS morphology (146). Some patients with too short a native AV delay may benefit from AV junction ablation so that the paced and sensed AV delay can be optimized (147). Pacing may cause thinning of the LV wall and decrease outflow obstruction (142, 148). Two recent observational studies have suggested that a decrease in LV outflow gradient produced by temporary dual-chamber pacing may have adverse effects on ventricular filling and cardiac output (149, 150). Another small observational study of dual-chamber pacing in hypertrophic cardiomyopathy patients without outflow obstruction failed to show significant hemodynamic or short-term benefit (151).

One study (142) demonstrated that dual-chamber pacing eliminated or ameliorated symptoms in 74 of 88 patients. Patients in this study were not required to have a beneficial hemodynamic response to temporary pacing as a selection criterion for permanent pacing. A recent randomized study (152) demonstrated that DDD pacing reduced outflow tract gradient and improved New York Heart Association (NYHA) functional class. Early nonrandomized studies demonstrated a fall in the LV outflow gradient with dual-chamber pacing and a short AV delay and symptomatic improvement in some patients with hypertrophic obstructive cardiomyopathy ^141-143,154. One long-term study ¹⁵³ in eight patients supported the long-term benefit of dual-chamber pacing in this group of patients. The outflow gradient was reduced even after cessation of pacing, suggesting that some ventricular remodeling had occurred consequent to pacing. Two randomized trials ^152,154 demonstrated subjective improvement in approximately 50% of study participants but there was no correlation with gradient reduction, and a significant placebo effect was present. A third randomized trial ³⁶⁷ failed to demonstrate any overall improvement in quality of life with pacing, although there was a suggestion that elderly patients (aged more than 65 years) may derive more benefit from pacing. Although these data are encouraging, a recent randomized, double-blind crossover study (154) of 19 patients demonstrated no significant subjective or exercise capacity improvement in the paced versus nonpaced group at 2 to 3 months of follow-up, despite a significant decrease in LV outflow gradient.

In a small group of patients with symptomatic, hypertensive cardiac hypertrophy with cavity obliteration, VDD pacing with premature excitation statistically improved exercise capacity, cardiac reserve, and clinical symptoms ³⁶⁸. However, several individual patients in this study demonstrated symptomatic and hemodynamic improvement from dual-chamber pacing. Dual-chamber pacing may improve symptoms and LV outflow gradient in pediatric patients. However, rapid atrial rates, rapid AV conduction, and congenital mitral valve abnormalities may preclude effective pacing in some patients ¹⁵⁵.

The lack of large, prospective, placebo-controlled data makes this indication for permanent pacing controversial. There are currently no data available to support the contention that pacing alters the clinical course of the disease or bivenimproves survival or quality of life. Therefore, routine implantation of dual-chamber pacemakers should not be advocated in all patients with symptomatic hypertrophic obstructive cardiomyopathy. Patients who may benefit the most are those with significant gradients (more than 30 mm Hg at rest or more than 50 mm Hg provoked ^154,369-371. In highly symptomatic patients, septal myectomy or percutaneous septal alcohol ablation should be considered instead of dual-chamber pacing ³⁷². For the patient with hypertrophic obstructive cardiomyopathy who is at high risk for sudden death and has a definite indication for pacemaker implantation, the clinician should weigh the long-term advantages of implantation instead an ICD, even if the patient’s condition might not warrant an ICD implant at that point in time (see Section II-E).

Pacing Recommendations for Hypertrophic Cardiomyopathy

Class I

Class I indications for sinus node dysfunction or AV block as described previously. (Level of Evidence: C)

Class IIb

Medically refractory, symptomatic hypertrophic cardiomyopathy with significant resting or provoked LV outflow obstruction. (Level of Evidence: CA) ^{142,145,146,152,154,367}

Class III

1. Patients who are asymptomatic or medically controlled.
2. Symptomatic patients without evidence of LV outflow obstruction.

2. Idiopathic Dilated Cardiomyopathy
Several observational studies have shown limited improvement in patients who have symptomatic dilated cardiomyopathy refractory to medical therapy with dual-chamber pacing with a short AV delay ^156-159. Theoretically, a short AV delay may optimize the timing of mechanical AV synchrony and ventricular filling time. In patients with prolonged PR intervals more than 200 milliseconds, diastolic filling time may be improved by dual-chamber pacing with a short AV delay ¹⁷. In one study ¹⁵⁷, cardiac output was increased 38% by shortening AV delay when the average PR interval was 283 milliseconds before pacing. When the PR interval was shorter, no benefit of pacing was noted. Permanent pacing in symptomatic patients with drug-refractory dilated cardiomyopathy and a prolonged PR interval may be useful if short-term benefit is demonstrated in acute studies. At this time no long-term data are available, and there is no consensus for this indication. The mechanisms by which dual-chamber pacing might benefit patients with dilated cardiomyopathy are poorly understood. One hypothesis is that a well-timed atrial contraction primes the ventricles and decreases mitral regurgitation, thus augmenting stroke volume and arterial pressure. Several studies have not demonstrated improvement in cardiac output with dual-chamber pacing in patients with congestive heart failure (160, 161). One randomized controlled trial of 12 patients showed no significant benefit of VDD pacing through a range of PR intervals despite the presence of both tricuspid and mitral regurgitation (160). One study (162) in 89 patients with LV dysfunction suggested that VVI pacing in the right ventricular outflow tract (simulating a normal high to low ventricular activation) improved cardiac output by 18.8% when compared with pacing the right ventricular apex.

Thirty to fifty percent of patients with congestive heart failure have intraventricular conduction defects ^373,374. These conduction abnormalities progress over time, lead to discoordinated contraction of an already hemodynamically compromised ventricle, and are an independent predictor of mortality ³⁷⁵. Delayed activation of the LV during right ventricular pacing also leads to significant dyssynchrony in both LV contraction and relaxation. Biventricular pacing can provide a more coordinated pattern of ventricular contraction, reduce the QRS duration, and reduce intraventricular and interventricular asynchrony. Biventricular pacing was initially demonstrated to improve cardiac index acutely, decrease systemic vascular resistance and pulmonary capillary wedge pressure, increase systolic blood pressure, and lower V-wave amplitude compared with right ventricular or AAI pacing in several trials ^376-378. Advances in lead design have allowed the insertion of endocardial leads into distal branches of the coronary sinus to pace the LV. These advances have led to several small and large prospective trials to study the efficacy of biventricular pacing in patients with congestive heart failure and intraventricular conduction defects. Auricchio et al. ³⁷⁹ demonstrated that pacing in the mid-lateral LV augments positive pulse pressure changes more than pacing other areas. In the Pacing Therapies for Congestive Heart Failure trial ³⁸⁰, increases in LV dP/dt and pulse pressure were significantly better with biventricular than with right ventricular pacing. ConvincingRecent data from several prospective, randomized trials ^381-384 support the hemodynamic and subjective improvement that was noted in multiple previous anecdotal and smaller trials ^385-387. These trials demonstrate that in patients with New York Heart Association (NYHA) class III or IV congestive heart failure, decreased ejection fraction, and prolonged QRS duration, biventricular pacing decreases QRS duration and improves 6-minute walk distance, NYHA class, and quality-of-life scores. In the Multisite Stimulation in Cardiac Insufficiency trial ³⁸¹, rehospitalizations from congestive heart failure were also reduced. No data exist demonstrating that biventricular pacing improves survival, although early data suggest trends in improvement in decreasing spontaneous ventricular ectopy and ICD shocks ^388-390. Ongoing studies will determine whether a combination of biventricular pacing with an ICD will result in an improvement in subjective symptoms plus improved survival. Preliminary data (163, 164) suggest that simultaneous bivenimproves tricular pacing may improve cardiac hemodynamics and thus lead to subjective and objective symptom improvement. Prospective controlled trials are under way to confirm these initial findings and further define the benefit of biventricular pacing in patients with symptomatic, drug-refractory dilated cardiomyopathy. Overall there are sparse long-term data to show improvement in hemodynamics, symptom relief, or survival for pacing in dilated cardiomyopathy. Even less data exist in patients with ischemic cardiomyopathy.

Pacing Recommendations for Dilated Cardiomyopathy

Class I

Class I indications for sinus node dysfunction or AV block as described previously. (Level of Evidence: C)

Class IIa

None. Biventricular pacing in medically refractory, symptomatic NYHA class III or IV patients with idiopathic dilated or ischemic cardiomyopathy, prolonged QRS interval (greater than or equal to 130 milliseconds), LV end-diastolic diameter greater than or equal to 55 mm, and ejection fraction less than or equal to 35%. (Level of Evidence: A) ^381,383

Class IIb

Symptomatic, drug-refractory dilated cardiomyopathy with prolonged PR interval when acute hemodynamic studies have demonstrated hemodynamic benefit of pacing. (Level of evidence: C)(17, 156-158)

Class III

1. Asymptomatic dilated cardiomyopathy.
2. Symptomatic dilated cardiomyopathy when patients are rendered asymptomatic by drug therapy.
3. Symptomatic ischemic cardiomyopathy when the ischemia is amenable to intervention.

3. Cardiac Transplantation
The incidence of bradyarrhythmias after cardiac transplantation varies from 8% to 23% ^165-167. The majority of bradyarrhythmias are associated with sinus node dysfunction. Because of symptoms and impaired recovery and rehabilitation, some transplant programs recommend more liberal use of cardiac pacing for persistent postoperative bradycardia. About 50% of patients show improvementresolution of the bradyarrhythmia within 6 to 12 months, and long-term pacing is often unnecessary in a large number of patients ^168-170. Significant bradyarrhythmias and asystole have been associated with reported cases of sudden death ¹⁷¹. No predictive factors have been identified to indicate which patients will develop post-transplantation bradyarrhythmias. In some patients, the need for pacing may be transient. The benefits of the atrial contribution to cardiac output and chronotropic competence may optimize the patient’s functional status. Attempts to treat the bradycardia temporarily with measures such as theophylline ¹⁷² may minimize the need for pacing. Post-transplant patients who have irreversible sinus node dysfunction or AV block with previously stated Class I indications should have permanent pacemakers.

Pacing Recommendations After Cardiac Transplantation

Class I

Symptomatic bradyarrhythmias/chronotropic incompetence not expected to resolve and other Class I indications for permanent pacing. (Level of Evidence: C)

Class IIb

Symptomatic bradyarrhythmias/chronotropic incompetence that, although transient, may persist for months and require intervention. (Level of Evidence: C)

Class III

Asymptomatic bradyarrhythmias after cardiac transplantation.

I. Selection of Pacemaker Device

Once the decision has been made to implant a pacemaker in a given patient, the clinician must decide among a large number of available pacemaker generators and leads. Generator choices include single- versus dual-chamber devices, unipolar versus bipolar configuration, presence and type of sensor for rate response, advanced features such as automatic mode switching, size, battery capacity, and cost. Lead choices include polarity, type of insulation material, fixation mechanism (active versus passive), and presence of steroid elution, and typical pacing impedance. Some lead models typically show low (300-500 Ohms) and some high (greater than 1000 Ohms) pacing impedance, and this can have implications with regard to the generator’s battery longevity. Other factors that importantly influence the choice of pacemaker system components include the capabilities of the pacemaker programmer, which provides the link between the pacemaker system and the physician, and local availability of technical support.

Even after selecting and implanting the pacing system, the physician has a number of options for programming the device. In modern single-chamber pacemakers, programmable features include pacing mode, lower rate, pulse width and amplitude, sensitivity, and refractory period. Dual-chamber pacemakers have the same programmable features, as well as maximum tracking rate, AV delay, and others. Rate-responsive pacemakers require programmable features to regulate the relation between sensor output and pacing rate and to limit the maximum sensor-driven pacing rate. These programmable parameters must be individually adjusted for each patient, and the choice of one programmable parameter will often depend on the availability of another parameter. For example, in a patient with complete AV block and paroxreaysmal atrial fibrillation, a dual-chamber pacemaker without mode-switching capability most appropriately might be programmed to DDIR* mode, whereas in the same patient, a pacemaker with mode-switching capability most appropriately might be programmed to DDDR mode with mode switching. In recent years, wWith the advent of more sophisticated pacemaker generators, optimal programming of pacemakers has become increasingly complex and device-specific and requires specialized knowledge on the part of the physician.

Many of these considerations are beyond the scope of this document. The discussion below focuses on the most fundamental choice the clinician has with respect to the pacemaker prescription: those that have the greatest impact on procedural time and complexity, follow-up, patient outcome, and cost: (1) the choice among single-chamber ventricular pacing, single-chamber atrial pacing, and dual-chamber pacing. and (2) whether or not to use a generator that incorporates a sensor for rate-responsive pacing.

Table 1 gives brief guidelines on the appropriateness of different pacemakers for the most commonly encountered indications for pacing. Figure 1 is a decision tree for selecting a pacing system in a patient with AV block. Figure 2 is a decision tree for selecting a pacing system in a patient with sinus node dysfunction.

An important challenge in selecting a pacemaker system is anticipating progression of abnormalities of automaticity and conduction and selecting a system that will best accommodate these developments. Thus, it is reasonable to select a pacemaker with more extensive capabilities than needed at the time of implantation but that may prove useful in the future. Some patients with sinus node dysfunction and paroxysmal atrial fibrillation, for example, may develop AV block in the future (as a result of natural progression of disease, drug therapy, or catheter ablation) and may ultimately benefit from a dual-chamber pacemaker with mode-switching capability. Patients who are likely to develop ventricular tachyarrhythmias, for which an ICD would be warranted, should receive a pacemaker that is compatible with ICDs.

1. Newer Technical Innovations

Rate-Responsive Pacemakers
An increasing percentage of pacemakers implanted in the United States incorporate sensors to detect states of exercise and trigger accelerations in pacing rate. Approximately 97% of all generators implanted in the United States in 2000 had rate response as a programmable option (Morgan Stanley Dean Witter. Equity Research, North America. Healthcare: Pharmaceuticals, Hospital Supplies & Medical. July 26, 2000). An industrywide survey in 1996 indicates that 83% of all generators implanted in 1996 in the United States had rate response as a programmable option. In pacemaker patients who are chronotropically incompetent (i.e., unable to increase sinus node rate appropriately with exercise), rateresponsive pacemakers allow for increases in pacing rates with exercise and have been shown to improve exercise capacity and quality of life.

In the United States, the vast majority of sensors incorporated into rate-responsive implantable pacemakers are piezoelectric crystals or accelerometers that detect motion, vibration, pressure, or acceleration. Other technologies using sensors that measure minute ventilation or QT interval may provide a heart rate response more proportional to exercise than piezoelectric sensors or accelerometers. An advantage of all of these sensor technologies is that they do not require specialized pacemaker leads, although minute ventilation sensing requires a bipolar lead. An older technique that measured circulating blood temperature has largely been abandoned.

Pacemaker generators that incorporate two rate-responsive sensors are now available. These dual-sensor pacemakers incorporate one sensor that rapidly responds to exercise (i.e., piezoelectric crystal or accelerometer) and one that responds more proportionately to increasing levels of exercise (i.e., minute ventilation or QT interval). Studies have shown more appropriate rate response to various types of exercise when information from two sensors is used than when a single sensor is used ^391,392. Studies are lacking that demonstrate an improvement in quality of life resulting from dual-sensor pacemakers compared with single-sensor pacemakers.

The challenge of adjusting the response of these generators to exercise appropriately in individual patients is increasingly becoming recognized. To facilitate optimal programming of rate-response capability, recently introduced many current generators incorporate procedures for initial programming of rate-response parameters, subsequent automatic adjustment of these parameters, and retrievable diagnostic data (such as heart rate histograms or heart rate plots) to assess the appropriateness of the rate response.

Single-Lead VDD Pacemaker Systems
Despite advances in rate-responsive pacemakers, it is widely appreciated that the best signal to guide heart rate response to exercise (and other forms of physiologic stress) is a normally functioning sinus node. Most commonly, dual-chamber pacemakers incorporating separate atrial and ventricular leads are used to detect atrial depolarization. Single-lead transvenous pacing systems that are capable of sensing atrial depolarization are available. The distal end of the lead is positioned in the right ventricle for ventricular pacing and sensing; a pair of electrodes is incorporated in the more proximal portion of the lead body lying within the right atrial cavity for atrial sensing. With current technology, single-lead VDD* pacing systems are not capable of atrial pacing. The atrial signal sensed by single-lead VDD pacemakers has a less consistent amplitude than that typically sensed by conventional dual-chamber pacemakers and varies significantly with posture, but sensing performance is generally satisfactory ¹⁷⁴. Single-lead VDD pacemaker systems are an reasonablealternative to dual-chamberlead pacemakers in patients with AV block in whom atrial pacing is not required and in whom simplicity of implantation or avoidance of two leads is desired.

*This and other three- or four-letter notations conform to the NASPE/BPEG generic pacemaker code ¹⁷³.

Automatic Mode Switching
When nonphysiologic atrial tachyarrhythmias, such as atrial fibrillation or flutter, occur paroxysmally in a patient with a dual-chamber pacemaker programmed to conventional DDD or DDDR mode, the tachyarrhythmia will generally be tracked near the programmed maximum tracking rate, leading to an undesirable acceleration of ventricular pacing rate. Newer dual-chamber generators incorporate algorithms for detecting rapid, nonphysiologic atrial rates and automatically switch modes to one that does not track atrial activity, such as DDI or DDIR. When the atrial tachyarrhythmia terminates, the pacemaker automatically reverts back to the DDD or DDDR mode. This automatic mode switch feature is especially helpful in patients with AV block and paroxysmal atrial fibrillation and expands the usefulness of dual-chamber pacemakers in such patients. Virtually all dual-chamber pacemakers now implanted in the United States incorporate automatic mode switching as a programmable option.

Rate-Drop Response for Neurocardiogenic Syncope
A programmable feature of some dual-pacemaker generators is automatic acceleration of pacing rate (e.g., to 100 bpm) for up to several minutes after detection of a sudden fall in intrinsic heart rate, such as would typically occur in patients with neurocardiogenic syncope. Of the two randomized studies showing a benefit of implanted pacemakers in patients with neurocardiogenic syncope in decreasing the recurrence of syncope in such patients, one of them ³⁵⁸ used pacemakers with rate-drop response, whereas the other ³⁵⁹ used pacemakers programmed to a more conventional hysteresis function. A large prospective, randomized multicenter trial is under way that will assess the specific benefit of rate-drop response in patients with neurocardiogenic syncope ³⁹³.

Pacemaker Leads
The vast majority of implanted pacemakers use transvenous endocardial leads, with the remainder using epicardial leads. Transvenous leads may be bipolar or unipolar in configuration. Bipolar configurations have the advantage of avoiding myopotential inhibition and skeletal muscle stimulation, and an increasingly important advantage is that unlike most unipolar pacing systems, they are compatible with concomitantly implanted ICDs. However, some manufacturers’ bipolar leads have higher failure rates than their unipolar leads.

The insulation material used in pacemaker leads is either silicone rubber or polyurethane. Polyurethane-insulated leads have a thinner diameter and better handling characteristics than silicone-insulated leads. However, Historically, some bipolar lead models with polyurethane insulation have shown unacceptably high failure rates due to degradation of the insulation. More recently introduced Many current polyurethane leads, using different polymers and different manufacturing processes, appear to be avoiding these unacceptably high failure rates.

Active fixation leads, in which the distal tip of the lead incorporates a small helical screw for fixation to the endocardium, are an alternative to passive fixation leads. Active fixation leads allow for more alternatives in the site of endocardial attachment. For instance, whereas a passive fixation ventricular lead generally must be positioned in the right ventricular apex, an active fixation lead may be positioned in the apex, outflow tract, or inflow tract of the right ventricle. Active fixation leads have an additional advantage of greater ease of extraction after long-term implantation. A disadvantage of active fixation leads is that they generally have higher chronic capture thresholds than do passive fixation leads, although this difference is minimized with the incorporation of steroid elution (see below).

An important advance in pacemaker leads is the development of leads with lower capture thresholds, which result in reduced battery consumption during pacing. Steroid-eluting leads incorporate at their distal tip a small reservoir of corticosteroid that slowly elutes into the interface between the lead electrode and the endocardium, reducing the inflammation and fibrosis that normally occur at this interface. As a result, steroid-eluting leads have significantly lower longterm capture thresholds than leads not incorporating steroid. The benefit of steroid elution was originally demonstrated in passive fixation transvenous leads ¹⁷⁵; the benefit has also been demonstrated in active fixation transvenous leads ¹⁷⁶ and epicardial leads ¹⁷⁷. Similar improvements in capture thresholds have been achieved with modification in electrode shape, size, and composition (178)³⁹⁴.

2. Methodology of Comparing Different Pacing Modesemaker Generators and Configurations

Two or more pacemaker modes can be compared with respect to exercise capacity, quality of life, clinical end points (such as death, heart failure, atrial fibrillation, and stroke), and cost. For end points such as exercise capacity or quality of life, pacemaker modes can be compared using a randomized crossover study design, provided that the patients have pacing systems that can be programmed to each of these modes. (For example, dual-chamber, rate-responsive pacemakers can be crossed over between VVIR and DDDR pacing.)

Studies that compare clinical end points require long-term follow-up without crossover. In long-term studies, patients can be randomly assigned to receive different types of pacemakers (e.g., hardware randomization, single-chamber ventricular pacemakers versus single-chamber atrial pacemakers), or all patients may receive a single type of pacemaker system (e.g., dual-chamber, rate-responsive) and be randomly assigned to different modes (e.g., software randomization, VVIR versus DDDR).

Quality-of-life measures have been emphasized as important end points when comparing different modes of pacing, and there are important considerations in the choice of the instrument used to measure quality of life ^179-181,395. Although the quality of life experienced with different modes of pacing may be compared using short-term crossover studies, long-term studies that include quality-of-life end points may reflect effects of chronic adaptation to stimulation not detectable in short-term comparisons. Several reported or ongoing long-term randomized comparisons of pacing modes have quality-of-life end points ^83,395,396.

An important consideration in the assessment of trials that compare pacing modes is the percent of pacing among the study patients. For example, a patient who is paced only for very infrequent sinus pauses will probably have a similar outcome with ventricular pacing compared with dual-chamber pacing, regardless of the potential physiologic benefits of dual-chamber pacing. Several older studies comparing pacing modes do not include data on the frequency of pacing ^{84,84a,397,398}, but newer studies do ^355,399.

3. Pacing in Sinus Node Dysfunction

Short-Term Outcomes
Short-term crossover studies in patients with sinus node dysfunction have shown improved quality of life with atrialbased (i.e., atrial or dual chamber) pacing versus ventricular pacing, with or without rate response ^180,182,395. There are conflicting data regarding any improvement in maximum exercise performance in rate-responsive atrial-based pacing compared with rate-responsive ventricular pacing in patients with sinus node dysfunction ^182,183,395.

Long-Term Outcomes
Over the past decade, a number of nonrandomized observational studies have been published comparing atrial-based pacing (either atrial pacemakers or dual-chamber pacemakers) with ventricular pacing in patients with sinus node dysfunction. These studies have been reviewed extensively ^{83,184,185,395,396}. The incidences of atrial fibrillation, stroke, heart failure, and total mortality appear to be consistently lower in patients receiving atrial-based pacing than in those receiving ventricular pacing ³⁹⁵. A consistent finding is that the incidence of atrial fibrillation is lower in patients receiving atrial-based pacemakers than in those receiving ventricular pacemakers; atrial-based pacing is associated with a reduction in risk of atrial fibrillation averaging 74% (185). The findings of the studies were mixed with regard to mortality end points: some studies showed a lower mortality in atrial-based pacemaker patients and some showed no significant difference. The studies suffer from limitations common to all nonrandomized studies (most importantly, uncertainty as to the clinical equivalence of the patient groups). In some of these studies, the patient groups appear to be well matched, whereas in others, there is insufficient information to assess their comparability.

Andersen et al. ^84,84a published the first randomized study comparing pacemaker modes with long-term follow-up in patients with sinus node dysfunction. Two hundred twenty-five patients were assigned randomly to atrial and ventricular pacing. After mean follow-up of 5.5 years, the patients assigned to atrial pacing had significantly lower incidences of atrial fibrillation, thromboembolic events, heart failure, cardiovascular mortality, and total mortality compared with the ventricular paced patients. During a mean of 40 months of follow-up, there were significantly fewer thromboembolic events in the atrial paced patients. There was a trend toward less atrial fibrillation in the atrial paced group, but it did not reach statistical significance. The study was not powered to detect a mortality difference between the two patient groups. However, when follow-up was extended to 8 years, atrial pacing was associated with significantly decreased “allcause” and “cardiovascular” mortality compared to ventricular pacing (84a).

Connolly et al. ³⁹⁸ published a randomized comparison of ventricular pacemaker implantation versus atrial-based pacemaker implantation in patients with a variety of indications for pacing. Among all patients, there was no significant difference in the combined incidence of stroke or death or in the likelihood of a heart failure hospitalization between the two treatment groups. There was, however, a statistically significant decrease in the incidence of atrial fibrillation in the patients with atrial-based pacemakers compared with ventricular pacemakers, which became apparent only after 2-year follow-up. The reduction in atrial fibrillation was seen both in patients paced for sinus node dysfunction and in those paced for AV block ³⁴⁵. A subgroup analysis of patients paced for sinus node dysfunction did not show any trends toward particular mortality or stroke benefit from atrial- based pacing in these patients.

The results of the MOST trial, in which 2010 patients with sinus node dysfunction were randomized between DDDR and VVIR pacing ^355,395,396, have been reported ³⁵³. After mean follow-up of 33 months, there was no significant difference in the incidence of death or stroke between the two groups, but there was a 21% lower risk of atrial fibrillation (p = 0.008), lower heart failure scores (p = 0.0001), 27% lower risk of heart failure hospitalizations (p = 0.02), and improved quality of life in the DDDR group compared with the VVIR group. Of the patients randomized to VVIR pacing, 37.7% crossed over to DDDR, most commonly for pacemaker syndrome.

In summary, studies consistently demonstrate that in patients with sinus node dysfunction, the incidence of atrial fibrillation in patients receiving atrial or dual-chamber pacemakers is lower than in patients receiving ventricular pacemakers. Published data are mixed regarding stroke, heart failure, and mortality benefit with atrial-based pacing compared with ventricular pacing. Pacemaker syndrome is common in patients with sinus node dysfunction treated with ventricular pacing. In summary, available data suggest that in patients with sinus node dysfunction, the incidence of atrial fibrillation in patients receiving atrial or dual chamber pacemakers may be lower than in patients receiving ventricular pacemakers. Published studies do not adequately address the issues of other clinical end points, such as heart failure, mortality, or quality of life.

Role of Single-Chamber Atrial Pacemakers
Single-chamber atrial pacemakers offer the advantages of atrial-based pacing without the added complexity and cost of dual-chamber pacemakers. Such pacing systems, with rateresponsive capability if appropriate, have been advocated for patients with sinus node dysfunction but no evidence of AV block ^{21,179,186-188}. Use of single-chamber atrial pacemakers is limited by concerns about subsequent development of AV block. The risk of developing significant AV block after atrial pacemaker implantation for sinus node dysfunction has been estimated to be 0.6% to 3.05.0% per year ^186,188,189. Pre-existing bundle-branch block, but not AV Wenckebach rate, is predictive of a higher likelihood of subsequent development of AV block ^{186,187,189,190}. In selected patients with sinus node dysfunction, use of singlechamber atrial pacemakers is an acceptable approach that maintains normal AV synchrony, without the added cost and extra lead of a dual-chamber pacemaker system but there is a small risk of subsequent development of AV block requiring pacemaker revision. With rate-responsive atrial pacemakers, the risk of developing hemodynamically significant firstdegree AV block during rate accelerations has not been extensively studied but may be important ¹⁹¹.

A randomized study of DDDR versus VVIR pacing in patients with sinus node dysfunction is ongoing, with end points of total mortality, atrial fibrillation, stroke, heart failure, quality of life, and cost (83).

4. Pacing in Atrioventricular Block

Short-Term Outcomes
A number of short-term crossover studies have compared pacing modes in patients with AV block with respect to quality of life and exercise capacity. These studies have been reviewed in depth ^{83,180,395,396}. Studies comparing dual-chamber pacing with non–rate-responsive ventricular pacing have shown improved exercise capacity and symptomatology with dual-chamber pacing. Studies comparing rate-responsive ventricular pacing with non–rate-responsive ventricular pacing have shown similar advantages with rateresponsive ventricular pacing. However, studies comparing dual-chamber pacing (with or without rate response) with rate-responsive ventricular pacing have shown modest or no significant difference in exercise capacity; with respect to symptoms, most but not all of the studies have shown an advantage of dual-chamber pacing. It is likely that the symptomatic advantage of dual-chamber pacing over rate-responsive ventricular pacing is derived from the maintenance of AV association during rest and low-level activity.

Gillis et al. ³⁹⁹ examined atrial fibrillation recurrences in patients with paroxysmal atrial fibrillation who received pacemakers after AV ablation. Patients were crossed over between 6-month periods of DDDR pacing and VDD pacing; the principal difference between the two pacing modes was the lack of atrial pacing support and the possibility of asynchronous ventricular pacing during VDD pacing. No significant difference was seen in the time to atrial fibrillation recurrences between the two modes of pacing. However, nearly 15% of patients originally randomized to VDD pacing prematurely crossed over to DDDR pacing because of symptoms presumably related either to asynchronous ventricular pacing or atrial fibrillation.

Long-Term Outcomes
Two nonrandomized observational studies comparing patients with AV block who received dual-chamber pacemakers or ventricular pacemakers have shown improved survival associated with implantation of dual-chamber pacemakers among those patients with heart failure but no difference in survival between the two pacing modes among patients without heart failure ^74,192. In the randomized studies of Connolly et al. ³⁹⁸ and Skanes et al. ⁴⁰⁰, the subgroup of patients implanted for AV block showed trends toward lower cardiovascular mortality or stroke and atrial fibrillation with physiologic pacing over ventricular pacing, but these did not reach statistical significance. In an ongoing study, patients with AV block are randomly assigned to receive a ventricular pacemaker or a dual-chamber pacemaker; the primary end point is total mortality (83).

5. Pacing in the Elderly

More than 85% of pacemaker recipients are at least 64 years old ¹⁹³. Elderly pacemaker patients are the rule, not the exception.

It has been suggested that elderly patients requiring pacing should be considered for less sophisticated devices, e.g., single- chamber ventricular pacemakers or non–rate-responsive pacemakers. However, studies in elderly patients show improved exercise capacity and alleviated symptoms with rate-responsive ventricular pacing or dual-chamber pacing compared with non–rate-responsive ventricular pacing ^75,194. A retrospective analysis of 36,312 elderly Medicare patients receiving pacemakers suggested that dualchamber pacing is associated with improved survival compared with ventricular pacing, even after correction for confounding variables ¹⁹⁵.

Lamas et al. conducted a prospective, randomized longterm comparison of rate-responsive ventricular pacing and rate-responsive dual-chamber pacing in elderly patients aged 65 years or older, with the principal end points being quality-of-life parameters ³⁵⁵. Overall, pacemaker implantation regardless of pacing mode was associated with large improvements in quality of life. In the intention-to-treat analysis, dual-chamber pacing showed modest quality-of-life advantages over ventricular pacing, primarily in the patients paced for sinus node dysfunction. However, the most striking finding of the study was that 26% of patients randomized to ventricular pacing crossed over to dual-chamber pacing because of symptoms of pacemaker syndrome, with improvement in quality-of-life indices after crossing over. has recently been completed G.A. Lamas, PACE Study, unpublished data, 1997). The primary end point of the trial was quality-of-life measures; only transient improvement in a minority of the quality-of-life measures was found to be associated with rate-responsive dual-chamber pacing compared with rate-responsive ventricular pacing.

On the basis of these studies, rate-responsive ventricular pacing and dual-chamber pacing appears to offer benefits over fixed-rate ventricular pacing with respect to quality of life in elderly patients, but there may not be any benefit of dual-chamber pacing over rate-responsive ventricular demand pacing. It does not appear appropriate to withhold use of dual-chamber or rate-responsive pacemakers uniformly in the elderly, although such a decision may be appropriate in any patient who is extremely sedentary or has a limited life expectancy.

In an ongoing large, multicenter, randomized study of elderly (aged 70 years or older) patients undergoing pacemaker implantation for AV block (UKPACE, United Kingdom Pacing and Cardiovascular Events), patients are randomly assigned to receive a ventricular pacemaker or a dual-chamber pacemaker. The study’s primary end point is total mortality, and secondary end points are morbidity, exercise capacity, and quality of life ^83,395,396.

6. Optimizing Pacemaker Technology and Cost

The cost of a pacemaker system increases with its degree of complexity and sophistication. For example, the cost of a dual-chamber pacemaker system exceeds that of a singlechamber system with respect to the cost of the generator (additional $1000), the second lead (approximately $900), additional implantation time and supplies, and additional follow-up. The implant costs and complication rate of dualchamber pacemakers also exceed those of single-chamber pacemakers ³⁹⁸. Some pacemaker models offer more extensive diagnostics than others, at a premium cost. Similarly, the cost of a rate-responsive generator exceeds that of a non-rate-responsive generator by $500 to $1000. Against these additional costs are the potential benefits of the more sophisticated systems with respect to quality of life, morbidity, and mortality. Furthermore, the additional diagnostic information available from more sophisticated generators may eliminate the need for additional diagnostic testing, such as ambulatory monitoring or exercise electrocardiography, that would otherwise be necessary occasionally in selected pacemaker patients. Little is known about the cost-effectiveness of the additional features of more complex pacemaker systems. Several ongoing trials assessing the clinical benefits of dual-chamber or rate-responsive pacing include economic analyses to estimate the incremental cost-effectiveness of these features ^83,395,396.

Approximately 16% of pacemaker implantations are for replacement of generators; of those, 76% are replaced because their batteries have reached end of service ¹⁹³. Hardware and software (i.e., programming) features of pacemaker systems that prolong useful battery longevity may improve the cost-effectiveness of pacing. Optimal programming of output voltages, pulse widths, and AV delays can markedly decrease battery drain; a study showed that expert programming of pacemaker generators can have a major impact on longevity, prolonging it by an average of 4.2 years compared with nominal settings ¹⁹⁶. Extensive diagnostic capabilities, which typically add $500 to $1000 to the cost of a pacemaker generator, may allow for optimal programming by the experienced physician with regard to improved device longevity. Newer lead designs, such as those incorporating steroid elution or high pacing impedance, allow for less current drain; the cost of such leads is approximately $125 greater than that of conventional leads. Generators that automatically determine whether a pacing impulse results in capture allow for programming outputs closer to threshold values than conventional generators, and this new technology may also have a major impact on device longevity. Although all of these features arguably should prolong generator life, there are other constraints on the useful life of a pacemaker generator, including battery drain not directly related to pulse generation and the limited life expectancy of many pacemaker recipients; rigorous studies supporting the overall cost-effectiveness of advanced pacing features are lacking.

The cost of pacemaker implantation may vary between different locations within a hospital (e.g., cardiac catheterization laboratory versus operating room); costs can be minimized by selecting the most economical site for implantation that preserves excellent patient outcome. One study comparing hospital charges for pacemaker implants in the catheterization laboratory versus the operating room within a single hospital showed a 45% lower charge for implants in the catheterization laboratory ⁴⁰¹. There has been a trend to shorter hospital stays for pacemaker implantations, and some implantations are now being performed on an outpatient basis.

Reuse of explanted pacemakers, not currently performed to any extent in the United States, may eventually add significantly to the cost-effectiveness of cardiac pacing ¹⁹⁷. Reports from Canada and Sweden indicate the safety and cost savings of selective pacemaker reuse ^197,402.

J. Pacemaker Follow-up

After implantation of a pacemaker, careful follow-up and continuity of care are required. The committee considered the advisability of extending the scope of these guidelines to include recommendations for follow-up and device replacement but deferred this decision given other published statements and guidelines on this topic. These are listed below as a matter of information. No endorsement is implied. In general, follow-up is dictated by the patient’s disease substrate, the device used, and evolving technology. NASPE has published a comprehensive series of reports on antibradycardia pacemaker follow-up ^198-200. The Canadian Working Group in Cardiac Pacing has also published a consensus statement on pacemaker follow-up ⁴⁰³. In addition, the Health Care Financing Administration (HCFA; currently the Centers for Medicare and Medicaid Services) has established guidelines for monitoring of patients covered by Medicare who have antibradycardia pacemakers ²⁰¹. These documents are endorsed by this writing group.

Many of the same considerations are relevant to both pacemaker and ICD follow-up. Programming undertaken at implantation should be reviewed before discharge and changed accordingly at subsequent follow-up visits as indicated by interrogation, testing, and patient needs. With careful attention to programming pacing amplitude, pulse width, and diagnostic functions, battery life can be enhanced significantly without compromising patient safety. Taking advantage of programmable options also allows optimization of pacemaker function for the individual patient.

The frequency and method of follow-up is dictated by multiple factors, including other cardiovascular or medical problems managed by the physician involved, the age of the pacemaker, and geographic accessibility of the patient to medical care. and the results of transtelephonic testing. Some centers may prefer to use transtelephonic monitoring (TTM) as the mainstay of follow-up, with intermittent clinic evaluations. Others may prefer to do the majority or all of the patient follow- up in clinic. Automatic features such as automatic threshold assessment have been incorporated increasingly into newer devices and facilitate follow-up for patients who live far from follow-up clinics ⁴⁰⁴. These automatic functions are not, however, universal and need not and cannot supplant the benefits of direct patient contact, particularly with regard to history taking and physical examination. Clinic In patients who are pacemaker dependent, clinical evaluation may be more frequent than for those who are not pacemaker dependent. In general, follow-up usually includes assessment of the clinical status of the patient, battery status, pacing threshold and pulse width, sensing function, and lead integrity, as well as optimization of sensor-driven rate response. The schedule for clinic follow-up should be at the discretion of the caregivers providing pacemaker follow-up. As a guideline, the 1984 HCFA document suggests the following: for single-chamber pacemakers, twice in the first 6 months after implant and then once every 12 months; for dual-chamber pacemakers, twice in the first 6 months, then once every 6 months.

Guidelines for TTM of permanent pacemakers have evolved clinically based on evolution of pacemaker and transtelephonic technology. Legislation regarding TTM has not been revised since 1984 (MED-MANUAL, MEDGUIDE ¶27,201, Coverage Issue Manual §50-1 Cardiac Pacemaker Evaluation Services [effective date: October 1, 1984]). Clearly stated guidelines are necessary and should be written in such a way as to mandate TTM that achieves specific clinical goals.

Appropriate clinical goals of TTM should be divided into those pieces of information obtainable during nonmagnet (free-running) ECG assessment and assessment of the ECG tracing obtained during magnet application. The same goals should be achieved whether the service is being provided by a commercial or noncommercial monitoring service.

Goals of TTM nonmagnet ECG assessment are as follows:

• Determine whether the patient displays intrinsic rhythm or is intermittently or continuously pacing at the programmed settings.
• Characterize the patient’s underlying atrial mechanism, e.g., sinus versus atrial fibrillation.
• If intrinsic rhythm is displayed, determine that normal (appropriate) sensing is present for one or both chambers depending on whether it is a single- or dual-chamber pacemaker and programmed pacing mode.

Goals of TTM ECG assessment during magnet application are as follows:

• Verify effective capture of the appropriate chamber(s) depending on whether it is a single- or dual-chamber pacemaker and programmed pacing mode.
• Assess magnet rate. Once magnet rate is determined, the value should be compared to value(s) obtained on previous transmissions to determine whether any change has occurred. The person assessing the TTM should also be aware of the magnet rate that indicates elective replacement indicators for that pacemaker.
• If the pacemaker is one in which pulse width is an indicator of elective replacement indicators, the pulse width should also be assessed and compared to previous values.
• If the pacemaker has some mechanism to allow transtelephonic assessment of threshold, i.e., Threshold Margin Test (TMT™), and that function is programmed “on,” the results of this test should be demonstrated and analyzed.
• If a dual-chamber pacemaker is being assessed and magnet application results in a change in AV interval during magnet application, that change should be demonstrated and verified.

1. Amount of Time Necessary for Verification

The amount of time spent during TTM should not be specified in terms of actual seconds or minutes but in terms of verification of the above clinical parameters. Whatever length of time is necessary to verify the points listed above during free running” and “magnet,” obtained ECG assessment should be sufficient.

2. Length of Electrocardiographic Samples for Storage

It is important that the caregiver(s) providing TTM assessment be able to refer to a paper copy or computer-archived copy of the transtelephonic assessment for subsequent care. The length of the ECG sample saved should, once again, be predicated based on the clinical information that is required, e., the points listed above. It is the experience of personnel trained in TTM that an ECG sample of 6 to 9 seconds that is quality and is carefully selected can demonstrate all of the points for each of the categories listed above, i.e., a 6- to 9-second strip of nonmagnet and 6- to 9-second strip of magnet-applied ECG tracing.

Historically, there may have been some reason to collect another nonmagnet ECG example after the sample had been obtained during magnet application. In very early pacemakers, it may have been that magnet application would rarely result in a “stuck” reed switch, and a subsequent nonmagnet tracing would verify that function was indeed normal. However, with contemporary pacemakers, there is no logical clinical reason to obtain another nonmagnet tracing.

Because the indications for device implantation are evolving and some of the original indications for a particular patient may have been controversial, future replacement decisions may be more or less certain and must be individualized.

3. Frequency of Transtelephonic Monitoring

The follow-up schedule for TTM varies between centers, and there is no absolute schedule that need be mandated. Regardless of any of the following schedules to which the center may adhere, TTM may be necessary at unscheduled times if, for example, the patient experiences symptoms potentially reflecting an alteration in rhythm or device function.

The majority of centers with TTM services follow the schedule allowed by HCFA. In the 1984 HCFA guidelines, there are two broad categories for follow-up (as shown in Table 2): guideline I, which was thought to apply to the majority of pacemakers in use at that time, and guideline II, which would apply to pacemaker systems for which sufficient long-term clinical information exists to ensure that they meet the standards of the Inter-Society Commission for Heart Disease Resources for longevity and end-of-life decay. The standards to which they referred are 90% cumulative survival at 5 years after implant and an end-of-life decay of less than a 50% drop of output voltage and less than a 20% deviation of magnet rate, or a drop of 5 bpm or less, over a period of 3 months or more. As of 2000, it would appear that most pacemakers would meet the specifications in guideline II.

There is no federal or clinical mandate that these TTM guidelines be followed. The ACC, AHA, and NASPE have not officially endorsed these HCFA guidelines. Nevertheless, they may be useful as a framework for TTM. An experienced center may choose to do less frequent TTM and supplement it with in-clinic evaluations as stated previously.

© 2002 by the American College of Cardiology Foundation, American Heart Association, Inc., and North American Society for Pacing and Electrophysiology