Chapter 14. Conduction Disorders & Cardiac Pacing

Sinus node dysfunction (“sick sinus syndrome”)

• Sinus bradycardia: sinus rate of less than 60 bpm.
• Sinoatrial exit block, type I: progressively shorter P-P intervals, followed by failure of occurrence of a P wave.
• Sinoatrial exit block, type II: pauses in sinus rhythm that are multiples of basic sinus rate.
• Sinus arrest, sinus pauses: failure of occurrence of P waves at expected times.

Atrioventricular (AV) block

• First degree: prolonged PR interval more than 200 ms.
• Second degree
  • Type I: progressive increase in PR interval, followed by failure of atrioventricular (AV) conduction and nonoccurrence of a QRS complex.
  • Type II: abrupt failure of AV conduction not preceded by increasing PR intervals.
• “2:1 block:” due to lack of consecutive PR intervals; unable to assign either type I or type II block.
• “Advanced:” AV conduction ratio ≥ 3:1.
• Third degree (“complete”): independent atrial and ventricular rhythms, with failure of AV conduction despite temporal opportunity for it to occur.

The clinical presentation of patients with conduction system disease is determined by two underlying abnormal conditions: the inability to maintain or increase the sinus rate in response to metabolic need, and atrioventricular (AV) dyssynchrony (inappropriately timed atrial and ventricular depolarization and contraction sequences).

Sinus Node Dysfunction

Sinus node dysfunction (“sick sinus syndrome”) is usually due to a degenerative process that involves the sinus node and sinoatrial (SA) area (Table 14–1). Often, the degenerative process and associated fibrosis also involve the approaches to the AV node and the AV node itself, as well as the intraventricular conduction system; as many as 25–30% of patients with sinus node dysfunction have evidence of AV and bundle branch conduction delay or block.

Respiratory sinus arrhythmia, in which the sinus rate increases with inspiration and decreases with expiration, is not an abnormal rhythm and is most commonly seen in young healthy persons. Nonrespiratory sinus arrhythmia, in which phasic changes in sinus rate are not due to respiration, may be accentuated by the use of vagal agents, such as digitalis and morphine, and is more likely to be observed in patients who are older and who have underlying cardiac disease, although the arrhythmia is not itself a marker for structural heart disease; its mechanism is unknown. Ventriculophasic sinus arrhythmia is an unusual rhythm that occurs during advanced second-degree or complete AV block; it is characterized by shorter P-P intervals that enclose QRS complexes and longer P-P intervals that do not enclose QRS complexes. The mechanism is not known with certainty, but may be related to the effects of the mechanical ventricular systole. The ventricular contraction increases the blood supply to the sinus node, thereby transiently increasing its firing rate; the resulting increase in intra-atrial pressure causes subsequent inhibition of the sinus rate. Ventriculophasic sinus arrhythmia is not a pathologic arrhythmia and should not be confused with premature atrial depolarizations or SA block. None of the sinus arrhythmias indicate sinus node dysfunction.

Sinus node dysfunction is present when marked sinus bradycardia, pauses in sinus rhythm (sinus arrest), SA block, or a combination of these exist (Figures 14–1, 14–2, 14–3, 14–4, and Figure 14–5). Some clinically normal individuals who do not have structural heart disease can experience significant sinus bradycardia and prolonged sinus pauses under conditions of high vagal tone such as sleep. In some patients, a trigger, such as vomiting or coughing, can be identified; in other patients, high levels of acetylcholine may be responsible. Vagal stimulation, often from an identifiable trigger (Table 14–2), is commonly responsible for significant sinus bradyarrhythmias occurring in patients in an intensive care setting.

SA block may take the form of progressive delay in transmission of the sinus-generated impulse through the SA node to the atrium, finally resulting in a nonconducted sinus impulse and an absent P wave on the surface electrocardiogram (ECG) (Wenckebach, or type I second-degree exit block; see Figures 14–1 and 14–4), or abrupt failure of transmission of the sinus impulse to the atrium (type II second-degree exit block; see Figure 14–1). In type I second-degree exit block, there is less incremental delay with each successive impulse transmission through the SA nodal tissue (similar to type I AV nodal block); thus, the P-P intervals become progressively shorter until a P wave fails to occur. In type II second-degree exit block, abrupt failure of sinus impulse conduction to the atria can take the form of 2:1, 3:1 (and so on) SA block. Fixed 2:1 SA exit block cannot be distinguished from sinus bradycardia on the surface ECG.

Bradycardia-tachycardia syndrome is characterized by episodes of both bradycardias and supraventricular tachycardias (Figure 14–6). The bradycardia is due to sinus node dysfunction (sinus arrest or SA exit block) with associated junctional or ventricular escape rhythms. The supraventricular tachycardias may be atrial tachycardia, atrial flutter, atrial fibrillation, AV reciprocating tachycardia, or AV nodal reentry tachycardia (see Chapter 11); more than one type of tachycardia may occur in the same patient. Bradycardia-tachycardia syndrome often represents diffuse conduction system disease, but is not necessarily associated with structural heart disease.

Sinus bradycardia not uncommonly results from medications, particularly β-blockers, the rate-limiting calcium channel-blocking agents verapamil and diltiazem, and some commonly used antiarrhythmic agents such as sotalol, dronedarone, and amiodarone (see Figure 14–2). If these medications are necessary to treat the patient, permanent cardiac pacing is indicated.

The natural history of sinus node dysfunction is one of variable progression to an absence of identifiable sinus activity, with the process taking from 10 to 30 years. The condition itself is not associated with a high risk of arrhythmic death, although the morbidity caused by a sudden onset of bradycardia can be considerable. The ultimate prognosis for the patient with sinus node dysfunction typically depends on the presence and severity of underlying heart disease, rather than on the bradyarrhythmia itself.

Atrioventricular Block

Delay or block can occur anywhere along the course of the AV conducting system that is made up of the AV node, His bundle, and Purkinje fibers (left and right bundle branches). Like sinus node dysfunction, AV nodal-His block and bundle branch block (BBB) are often the result of sclerodegenerative processes. These processes can also involve the approaches to the AV node. Acquired AV nodal block is often due to acute ischemia and infarction (especially involving the inferior wall and right ventricle), infection, trauma, and medications (Table 14–3).

The AV node, or junction, is made up of three regions: atrionodal, central compact, and nodal-His. Cells of the atrionodal region have a relatively fast depolarization rate (45–60 bpm) and are responsive to autonomic nervous system input, whereas cells of the nodal-His region have a slower depolarization rate (about 40 bpm) and are generally unresponsive to autonomic influences. The site of origin of a junctional rhythm will therefore determine its rate, responsiveness to vagal and adrenergic input, and consequently the presence and severity of clinical symptoms.

The natural history of patients with AV block depends on whether or not there is an underlying cardiac condition with its own prognosis; however, the site of the block and the resulting rhythm disturbances themselves contribute to prognosis. First-degree AV block has little prognostic import. Persistent second-degree (types I and II), advanced second-degree (two or more consecutive nonconducted P waves), and complete AV block can all be associated with adverse outcomes, including death, unless the arrhythmias are vagally mediated or are due to other reversible causes.

Symptoms & Signs

The symptoms resulting from conduction disorders reflect cerebral hypoperfusion, low cardiac output at rest or during exercise, and rarely, hemodynamic collapse. Symptoms, which are often subtle, can be episodic or chronic and can change over time. Because a patient often adapts activity levels to compensate for the impairment in heart rate response, significant symptoms may not be evident unless the patient is closely questioned about specific activities and effort tolerance, or the clinician actually observes the patient during performance of activities of daily living such as walking or during formal treadmill exercise tests.

Syncope is the classic symptom of cerebral hypoperfusion due to bradycardia; however, symptoms of presyncope such as dizziness, lightheadedness, and confusion can reflect the same pathophysiology and warrant the same aggressive approach to diagnosis and management. It should be emphasized that patients with cerebral hypoperfusion often have impairment of memory surrounding the presyncopal or syncopal episodes and may therefore be unable to provide an adequate or accurate history surrounding the events; witnesses, if present, can contribute significant information in these cases.

Patients with sinus node dysfunction or AV block, in whom the escape pacemaker is unresponsive to autonomic nervous system input, cannot increase their heart rate in response to increases in oxygen demand. They are, therefore, intolerant of effort and will report symptoms of exercise-related breathlessness, weakness, and fatigue. These symptoms, which can be disabling, are often confused with other conditions such as hypothyroidism, medications, underlying heart disease, deconditioning, or simply older age.

During periods of AV block, the atria and ventricles often depolarize and contract asynchronously or with inappropriate and suboptimal timing between atrial and ventricular contractions. There are variable increases in atrial pressures and volumes depending on the degree to which the AV valves are open or closed at the onset of ventricular systole. The resulting atrial stretch and secretion of atrial natriuretic peptide produce reflex systemic hypotension and cerebral hypoperfusion. In addition, the increases in left atrial and pulmonary venous pressures can cause shortness of breath and pulmonary venous congestion, including frank pulmonary edema. The mistaken diagnosis of refractory left ventricular dysfunction is not infrequently made in this situation.

Patients who have the bradycardia-tachycardia syndrome (see Figure 14–6) have symptoms referable to both the bradycardia and the tachycardia. During tachycardia, the patient can experience uncomfortable palpitations, breathlessness, chest discomfort, and at times, symptoms of cerebral hypoperfusion from excessively rapid heart rates.

More rarely, bradycardias can lead to a potentially lethal form of polymorphic ventricular tachycardia known as bradycardia- or pause-dependent ventricular tachycardia (Figure 14–7); this tachycardia is often triggered in the setting of QT interval prolongation brought on by the longer R-R intervals associated with either bradycardia or pauses (torsade de pointes ventricular tachycardia). Symptoms in these patients can include not only palpitations, presyncope, and syncope, but also cardiac arrest.

Physical Examination

The physical examination of the patient with bradycardia reflects the origin of the QRS rhythm and the AV relationship more so than the heart rate per se. Junctional or ventricular escape rhythms resulting from atrial bradycardia or AV block produce AV dyssynchrony. This dyssynchrony results in varying degrees of atrial contribution to ventricular filling, as well as varying stroke outputs and systolic blood pressures. Because AV dyssynchrony causes changes in the positions of the mitral and tricuspid valves relative to their fully closed or open positions, the intensity of the first heart sound will vary, as will the audibility of atrial gallop (S4) sounds and the intensity of semilunar valve systolic ejection murmurs and AV valve regurgitant murmurs.

Examination of the venous pulse contour in the neck can reveal cannon a waves (due to atrial systole occurring against closed AV valves) and prominent cv waves (due to ventricular contraction occurring with the AV valves partially or even completely open), and the diagnosis of AV block can occasionally be made by recognizing these findings. Central venous pressure elevation (not to be confused with a and cv waves) is a common physical finding that is independent of the venous pulse contour.

The carotid pulse may vary in volume and upstroke velocity in patients with AV dyssynchrony. Examination of the chest may disclose rales, which reflect increased pulmonary venous pressure and valvular regurgitation rather than systolic or diastolic ventricular dysfunction. The liver may be enlarged and may pulsate because of transmitted a and cv waves. Peripheral edema may also be present if the AV dyssynchrony is chronic.

These same physical findings can also occur in patients who have sinus rhythm but develop AV dyssynchrony from being paced by a single-chamber ventricular system. In these patients, symptoms of weakness, fatigue, and congestive heart failure, together with physical findings indicating AV dyssynchrony, constitute the “pacemaker syndrome,” and this syndrome is treated by upgrading the single-chamber ventricular pacing system to a dual-chamber system in which sensing of the atrial rhythm triggers a paced ventricular response to restore AV synchrony (see later section, Permanent Pacing).

In addition to the findings described earlier, marked bradycardia (< 40 bpm) may result in ventricular dilatation in order to use the Frank-Starling effect to boost stroke volume and increase cardiac output despite the slow heart rate. This dilatation may lead to left ventricular gallop sounds and regurgitant murmurs, which disappear when normal heart rate is returned. Also, the enlarged ventricles may be palpable.

Diagnostic Studies

Sinus Node Dysfunction

Electrocardiography

The P waves inscribed on the surface ECG represent atrial depolarizations. Sinus node depolarization precedes atrial depolarization and is not seen on the surface ECG. The P waves that result from sinus-generated impulses must be inferred from their morphology and axis.

The sinus P wave has a mean frontal plane axis of +15 to +75° and is upright in leads I, II, and aVF, inverted in lead aVR, and variable in leads III and aVL. In the horizontal plane, the sinus P wave can be inverted in lead V1 but is upright in leads V3–V6. Respiratory variation in the sinus P-wave contour can be seen in the inferior leads and should not be confused with wandering atrial pacemaker, which is unrelated to breathing. Sinus arrhythmia is present when the P-wave morphology is normal and consistent and the P-P intervals vary by more than 160 ms. SA block exists when some impulses generated by the sinus pacemaking cells do not exit the SA node to depolarize the atria; in the absence of atrial depolarization, a P wave will not be inscribed on the surface ECG. High-degree (advanced) SA block, in which most of the sinus impulses fail to exit the SA node to the atrium, is inscribed on the surface ECG as pauses in sinus rhythm. These pauses often cannot be differentiated from sinus arrest caused by failure of impulse generation by the sinus node. If the pauses between sinus P waves are multiples of a basic P wave rate, however, the diagnosis of type II second-degree SA block can be made.

Electrophysiologic Studies

Sinus node function can be evaluated in the electrophysiology laboratory by means of simultaneous surface and intracardiac electrographic recordings made during basal conditions, physiologic and pharmacologic interventions, and atrial pacing. This evaluation can be undertaken in patients with either symptomatic sinus bradycardia or bradycardia-tachycardia syndrome. It can also be used in patients with recurrent syncope of unclear etiology, although the diagnostic yield and predictive value for the condition is limited. Measurements include the intrinsic heart rate, the sinus node recovery time (SNRT), the SA conduction time (SACT), and the response to parasympathetic (vagal) stimulation as assessed by carotid sinus massage.

The intrinsic heart rate (ie, the rate independent of autonomic influences) is the sinus rate during pharmacologic denervation of the sinus node using a β-blocker and atropine. The intrinsic heart rate is sometimes used to distinguish healthy persons from those with sinus node dysfunction.

Sinus node recovery time is the interval between the end of a period of pacing-induced overdrive suppression of sinus node activity and the return of sinus node function, manifested on the surface ECG by the postpacing sinus P-wave interval. The measured SNRT depends on several factors, among them the proximity of the pacing catheter to the sinus node, the SACT, the presence or absence of SA entrance block (in which atrial impulses fail to enter and depolarize the sinus node), and local neurohormonal influences. In normal persons, atrial pacing at rates of 120–130 bpm for 30 seconds or more is followed by a return of sinus node activity at a reproducible interval, with the basic sinus rate generally being achieved within three postpacing beats. The usual SNRT is less than 1.5 seconds, although considerable variation may exist depending on the prevailing autonomic tone. The corrected sinus node recovery time can be calculated by subtracting the basic sinus rate from the sinus node recovery time; a normal value is usually between 350 and 550 ms. In patients with sinus node dysfunction, sinus node recovery times are not reproducible and tend to be longer after more prolonged periods of pacing; return to the basic sinus rate within three postpacing beats is also inconstant and may be followed by additional (secondary) pauses in rate.

SA conduction time reflects the time taken by a premature atrial-pacing stimulus delivered near the sinus node area to traverse the atrial tissue to reach the sinus node and prematurely depolarize it, the time to the formation of the next sinus impulse following the premature depolarization, and the return of the sinus-generated impulse through atrial tissue to the recording electrode. The SA conduction time is often prolonged in patients with clinical evidence of sinus node dysfunction other than sinus bradycardia alone.

Because the electrophysiologic tests of sinus node function are neither specific nor sensitive, they can sometimes show abnormal results in patients without sinus bradycardia and normal results in those with symptomatic sinus node disease. The test results should not, therefore, be relied on in isolation when making clinical decisions regarding either diagnosis or treatment.

Some of the problems of poor sensitivity and specificity can be overcome by recording the sinus node electrogram from an electrode catheter positioned in the vicinity of the sinus node (at the junction of the superior vena cava and the right atrium). The sinus node electrogram, however, cannot be reproducibly recorded in as many as 50% of patients. Sinus node electrography can distinguish between SA exit block and sinus arrest, and studies have shown that most pauses in sinus rhythm are due to SA exit block. Notwithstanding these technologic advances, the diagnosis of sinus node dysfunction generally remains a clinical one.

Exercise Testing

Treadmill exercise testing can be of substantial value in assessing chronotropic response (“chronotropic competence”) to increases in metabolic needs in patients with sinus bradycardia in whom sinus node dysfunction is suspected. The definition of chronotropic incompetence is not agreed upon, but it is reasonable to designate it as consisting of either an inability to achieve a heart rate exceeding 75% of age-predicted maximum (usually taken as 220 – age), or 100–120 bpm at maximum effort. Irregular (and nonreproducible) increases, and even decreases, in sinus rate during exercise can also occur, but are rare. Similarly rare are abrupt changes in rate occurring during the postexercise recovery period. Chronotropic incompetence can result from medications (see Table 14–1) and should be distinguished from intrinsic sinus node dysfunction.

The Bruce treadmill exercise protocol, which is usually used to diagnose the presence and severity of coronary artery disease, is generally inappropriate for patients with sinus node dysfunction, in whom the goal is to assess heart rate at lower workloads expected to be encountered during average daily activities. Specific protocols, such as the chronotropic assessment exercise protocol (CAEP), have therefore been developed for this purpose. In addition to documenting chronotropic incompetence, treadmill exercise testing can be used to aid in optimal programming of rate-adaptive cardiac pacemakers that are usually required in these patients.

Atrioventricular Block

Electrography and Electrocardiography

The advent of intracardiac His bundle electrography has provided important information regarding normal and abnormal AV conduction in humans. The technique involves positioning a multipolar electrode catheter across the tricuspid valve in proximity to the AV nodal-His bundle area to record electrical activity as it passes through these structures. Because of its location, the catheter records electrical activity at the level of the low right atrium, His bundle, and proximal right bundle branch in addition to ventricular electrical activity. The sinus node pacemaker cells normally initiate the cardiac impulse, but they are not registered on either the surface ECG or the His bundle electrogram. The onset of the P wave on the surface ECG signifies the beginning of atrial depolarization; because the intracardiac electrode catheter lies at the level of the low right atrium, the early regions of atrial depolarization will not be detected by it; as the atrial depolarization wavefront passes through the region in which the catheter is located (the low right atrium), a deflection is registered (A). As the impulse traverses the His bundle, another deflection is registered, representing its depolarization (H). The His bundle deflection is followed by a ventricular deflection (V), which is registered at the time the wavefront of ventricular depolarization reaches the electrodes and often follows the onset of inscription of the QRS complex on the surface ECG.

His bundle electrography is useful in indicating the site of AV conduction delay or block. Normally, the conduction time through the AV node is 90–150 ms, and the conduction time through the His-Purkinje system is 35–55 ms. In a patient with a prolonged PR interval, a prolonged AH interval signifies delayed impulse conduction within the AV node, and a prolonged HV time represents delayed impulse conduction within the His bundle or the bundle branches. Conduction delay within the His bundle itself manifests as more than one His deflection (“split” His deflections).

In first-degree AV block (a delay in conduction between the atria and the ventricles), all atrial impulses are conducted to the ventricles; it is characterized by a prolonged PR interval that exceeds 200 ms (Figure 14–8). The components of the PR interval are interatrial conduction (10–50 ms), AV nodal conduction (90–150 ms), and intra-His and His-Purkinje conduction (35–55 ms). The conduction delay in first-degree AV block can thus represent prolonged intra-atrial, AV nodal, intra-His, or His-Purkinje conduction, and the recordings help clarify the location of the delay.

In patients with a QRS complex that is narrow and normal-appearing, first-degree AV block is AV nodal in more than 85% and is intra-His in less than 15%. In patients with a wide QRS complex, first-degree AV block is AV nodal in less than 25%, infranodal in about 45%, and at more than one site in about 33%.

In second-degree AV block, not all atrial impulses are conducted to the ventricles. The ratio of P waves to QRS complexes describes the AV conduction ratio. Type I (Wenckebach) second-degree AV block is present when the conduction of atrial impulses to the ventricles is progressively delayed because of AV (generally AV nodal) refractoriness, with eventual failure of conduction of an atrial impulse to the ventricles. The AV conduction ratio in type I second-degree AV block can be 3:2, 4:3, 5:4, and so on; this ratio is also referred to as a Wenckebach period. AV conduction ratios in type I second-degree AV block need not be constant, and therefore, the Wenckebach period may not be reproducible. Because type I second-degree AV block usually occurs within the AV node, the PR interval of the first conducted P wave of the Wenckebach period is often prolonged, and because this conduction disturbance does not involve the bundle branches, the QRS complexes are expected to be narrow and normal-appearing unless there is preexisting bundle branch disease.

In a typical, or classic Wenckebach period, the PR intervals progressively lengthen, the R-R intervals progressively shorten, and the R-R interval encompassing the nonconducted P wave is less than twice the preceding R-R interval. Typical Wenckebach periods are usually seen with low AV conduction ratios (3:2, 4:3, and 5:4), but as the AV conduction ratio increases (exceeding 6:5), more and more Wenckebach sequences are atypical and do not follow the rules. If the sinus rate is not constant, for example in vagal bradycardias, sequences that resemble Wenckebach conduction (referred to as “vagotonic block”) often occur; they should, however, not be considered type I second-degree AV block, in which the sinus rate must be constant for the diagnosis to be made.

In type II second-degree AV block, atrial impulses intermittently fail to be transmitted to the ventricles, but progressive conduction delay prior to the conduction failure does not occur. Because prior conduction delay from the atria is not present, the failure of antegrade conduction is often abrupt and unpredictable and may be advanced. In contrast to type I second-degree AV block, in which the conduction delay is usually in the AV node, the conduction delay in type II second-degree AV block can be within the His bundle or, more commonly, distal to the His bundle in the bundle branches. If the block is within the His bundle, the QRS complexes will be narrow and normal-appearing, or only mildly aberrant, unless preexisting BBB is present. If the block is infra-His, the QRS complexes will show a BBB pattern. In contrast to type I second-degree AV block, the PR interval of the conducted P waves is constant and often (but not always) normal (Figure 14–9).

Second-degree AV block with a 2:1 AV conduction ratio may represent either AV nodal or His-Purkinje block (Figures 14–10 and 14–11). Two consecutive PR intervals are not recorded in 2:1 AV conduction; therefore, the presence or absence of progressive PR prolongation cannot be ascertained, and distinguishing the diagnoses may be difficult. If the PR interval of the conducted P waves is prolonged and the QRS complexes are narrow and normal-appearing, AV nodal block is probably present. If the PR interval of the conducted P waves is normal and the QRS complexes have a BBB pattern, His-Purkinje block is probably present. If the PR interval of the conducted P wave is prolonged and the QRS complexes have a BBB pattern, or if the PR interval of the conducted P wave is normal and the QRS complexes appear normal, it may not be possible to distinguish between the two types, and more than one site of AV block may also be present. Altering the AV conduction ratio by means of carotid sinus massage (to produce sinus and AV nodal slowing) or intravenous atropine (to enhance sinus rate and AV nodal conduction) will often allow identification of the nature of the AV block and thus its location (see Figure 14–9).

In advanced second-degree AV block the AV conduction ratio is 3:1 or greater, and atrial impulses are only occasionally conducted to the ventricles. In contrast, in complete AV block, no atrial impulses are conducted to the ventricles despite temporal opportunity for this to occur, and the atria and ventricles are depolarized by their respective pacemakers that are independent of each other (Figure 14–12). The atrial rate in complete AV block is almost always faster than the ventricular rate. The QRS rhythm, or the escape rhythm, originates distal to the site of block and may be in the AV junction, His bundle, bundle branches, or distal Purkinje system. The morphology of the QRS complexes and their rate will depend on their site of origin. A narrow QRS complex escape rhythm originates from the AV junction. On the other hand, a wide QRS complex rhythm is not a reliable guide to the origin of the rhythm because rhythms originating in the longitudinally separated predivisional region of the His bundle can have a wide QRS complex.

If the atrial rate is not sinus, the existence of advanced or complete AV block is diagnosed by the presence of a slow ventricular rate with varying intervals between the QRS complexes or by a slow and regular QRS rate (Figure 14–13), respectively. Atrial fibrillation and flutter are commonly associated with advanced AV block and slow QRS rates. The rate of the ventricular rhythm, as well as the QRS-complex morphology, will depend on the site of origin of the rhythm. A regular rhythm in a patient with atrial fibrillation confirms the presence of complete AV block, with a QRS pacemaker originating below the level of conduction block that is independent of the atrial rhythm.

In vagotonic block, a high degree of vagal tone, such as occurs with sympathetic withdrawal during sleep or in highly conditioned athletes, may be associated with slowing of the sinus rate, pauses in sinus rhythm, variable degrees of delay in AV conduction manifested by prolongation of PR intervals (often irregular), and failure of conduction of P waves resembling type I or II second-degree AV block (Table 14–4). It is important to recognize vagotonic block because it often occurs in normal individuals as well as in patients with inferior or right ventricular myocardial infarction, or any other clinical condition in which hypervagotonia is present (see Table 14–2). It not uncommonly accompanies the use of certain medications, notably β-blocking agents, some antihypertensive drugs, and occasionally, digitalis. It can also be seen during swallowing (deglutition bradycardia), coughing (tussive bradycardia), and yawning. In the critical care setting, vagotonic block (and significant bradycardia) can occur during endotracheal suctioning or esophagogastric intubation, and in patients with elevated intracranial pressure.

Exercise Testing

Unlike the value of exercise testing in sinus node dysfunction for both diagnosis and evaluation of chronotropic competence, exercise testing in patients with AV block is generally not useful. AV node conduction is enhanced by the vagolysis and increased sympathetic drive that occurs with exercise. Thus, patients with first-degree AV block and type I second-degree AV block are expected to have shorter PR intervals during exercise; in patients with type I second-degree AV block, a longer Wenckebach period can develop (eg, 3:2 at rest becoming 6:5 during exercise). Patients with 2:1 AV conduction, in whom the site of conduction block may be uncertain, can benefit from exercise testing by observing whether the AV conduction ratio increases in a Wenckebach-like manner (eg, to 3:2 or 4:3) or decreases (eg, to 3:1 or 4:1) (see Figure 14–9). In the latter case, the increase in the sinus rate finds the His-Purkinje system refractory, causing the higher degrees of block. This response is always abnormal because it indicates intra- or infra-His block, which will require permanent cardiac pacing.

The major reversible causes of conduction system disturbances are high vagal tone and medications. High vagal tone, whether or not it is accompanied by withdrawal of sympathetic tone, can cause or contribute to both atrial and ventricular bradycardia. Vagally mediated bradycardias are usually transient and not accompanied by symptoms of presyncope or frank syncope, and no treatment is needed. If necessary, intravenous atropine can be used to facilitate AV nodal conduction to avoid ventricular bradycardia; however, the atropine-induced increase in atrial rate can lead to a paradoxical slowing of ventricular rate as a result of more rapid stimulation of, and encroachment on, the refractory period of the AV conduction system (“time-dependent” refractoriness). Moreover, the effects of intravenous atropine are short-lived, and its long-term use is accompanied by significant side effects.

In contrast to the majority of vagally mediated bradycardias, “vasovagal” episodes (hypotension with variable degrees of bradycardia or asystole) can be frequent, abrupt, unpredictable, and disabling. These highly symptomatic episodes, also referred to as neurocardiogenic or neurally mediated syncopal syndromes, can require heart rate support with oral theophylline or ephedrine. In some circumstances, they can also require permanent dual-chamber cardiac pacing, using special algorithms to detect an abrupt fall in heart rate and respond with tachypacing until the spontaneous heart rate increases. Intravascular volume support with fluids, support hosiery, and even mineralocorticoids can also be helpful. Because left ventricular baroreceptor stimulation (from vigorous systolic ventricular contraction) and its consequent reflex peripheral vasodilation play a role in this syndrome, drugs having negative inotropic effects (eg, β-blockers, disopyramide) have been used in management, although there is evidence that these agents have limited, if any, efficacy. α-Agonists, such as midodrine, have also been used with some success. In addition, because a central effect is recognized, anticholinergic agents and serotonin reuptake inhibitors have been used with variable success.

Commonly used medications that cause or contribute to bradycardia do so by enhancing vagal tone (eg, digitalis), reducing the facilitation of AV conduction that results from sympathetic tone (eg, β-blockers and antiarrhythmic agents with β-blocking properties such as sotalol and propafenone), or direct action on SA and AV conduction tissue (eg, verapamil and diltiazem). Simple withdrawal of these medications will reverse the bradycardia, although the process may require several days. If the offending medications are necessary to treat other conditions such as angina pectoris or heart failure and cannot be discontinued, permanent cardiac pacing will be required (Table 14–5).

It is important to exclude AV nodal blocking medications, such as digitalis preparations, β-blockers, and rate-limiting calcium channel blockers, as a cause of, or contributor to, slow ventricular rates in atrial fibrillation and flutter; withdrawal of these drugs or reduction in dosage can result in reversal of the AV block and permanent cardiac pacing may be avoided. If the ventricular rhythm is slow in the absence of these agents, intrinsic AV conduction system disease is likely to be present, and permanent cardiac pacing will usually be indicated. Electrical cardioversion of the atrial arrhythmia should be undertaken with caution, if at all, in patients with slow ventricular rates in the absence of medication; because of diffuse underlying conduction disease, postcardioversion sinus bradycardia or even asystole can occur.

In bradycardia-tachycardia syndrome, prolonged pauses in sinus rhythm frequently occur following an abrupt termination of the tachycardia. These pauses are often associated with symptoms of cerebral insufficiency, and cardiac pacing is frequently required. Alternatively, elimination of the tachycardia by radiofrequency ablation can potentially avert the need for cardiac pacing; severity of the underlying sinus node dysfunction, however, may still make cardiac pacing necessary even after ablation.

Although cardiac pacing using AAIR or DDDR devices (see later section, Permanent Pacing) may serve to suppress the tachycardias to some degree, treatment with antiarrhythmic agents is usually required. When the tachycardia occurs despite antiarrhythmic drug therapy, control of the ventricular rate also becomes necessary, but AV nodal blocking agents are often only partially effective in achieving this control; moreover, their use can be associated with significant side effects. Radiofrequency AV node ablation, together with dual-chamber cardiac pacing, has thus emerged as a useful and cost-effective technique in the management of this rhythm disturbance. When treating bradycardia-tachycardia syndrome, current pacemakers use a special algorithm to switch from a DDD or DDDR mode of operation to a VVI, VVIR, DDI, or DDIR mode on sensing an atrial tachyarrhythmia, and back again to DDD or DDDR mode when a normal atrial rate is sensed; this helps maintain an even pulse rate whether in or out of tachycardia.

Cardiac Pacing

Temporary or permanent cardiac pacing, in which an electrical stimulus depolarizes cardiac tissue, is indicated when bradycardia causes symptoms of cerebral hypoperfusion or hemodynamic decompensation. Occasionally, patients with bradycardia-dependent ventricular tachycardia require pacing to prevent the pauses in rhythm that lead to the tachyarrhythmia (Tables 14–5 and 14–6; see Figure 14–7). Although emergency pacing can be accomplished temporarily by transcutaneous pacing systems, in all but the most critical situations, stable temporary pacing is best ensured by the transvenous insertion of electrodes into the right atrium, right ventricle, or both. Permanent cardiac pacing is also usually performed through the transvenous route; in some circumstances, however, epicardial placement of electrodes via thoracotomy or a subxiphoid approach is still used.

Temporary Pacing

Transmyocardial Pacing

Transmyocardial pacing involves the percutaneous placement of cardiac pacing wires into the ventricular cavity or onto the ventricular wall through a transthoracic needle. The reliability of this technique is poor, and it is a highly invasive procedure with significant potential morbidity. Transmyocardial pacing is performed only in an emergency setting, usually during cardiac arrest, when transvenous pacing cannot be accomplished rapidly or when transcutaneous pacing is unavailable or unsuccessful. The reported incidence of successful capture with transthoracic pacing varies from 5% to 90%; typically, it is 21–40%.

Often, however, because of the clinical circumstances in which this type of pacing is used, electrical capture is not followed by mechanical systole. The major complications of transthoracic pacing include myocardial or coronary artery laceration, pericardial tamponade, pneumothorax, and hepatic or gastric damage. Transthoracic pacing should therefore be reserved for situations of the utmost gravity where no other pacing system is feasible or available. External (transcutaneous) pacing should always be tried first, because it is probably as efficacious (if not more so) and is associated with significantly less morbidity. Transthoracic pacing should never be used in awake or stable patients. The technique is now all but obsolete.

Transcutaneous Pacing

This method, in which electrical current is delivered to the heart through the skin via large surface electrodes, is usually reserved for standby prophylaxis in patients recognized to be at high risk for bradycardia, for example during inferior and large anterior wall acute myocardial infarctions (Table 14–7), and in some patients with suspected sinus node dysfunction who are undergoing elective cardioversion. Because of its ease of use and relative efficacy, this pacing modality has virtually eliminated the need for transmyocardial pacing in emergency situations.

The transcutaneous pacing system uses two large, low-impedance surface electrodes placed on the anterior and posterior chest walls. A long duration pacing stimulus output of 20–40 ms (not programmable by the operator) and current output of more than 100 milliamperes (mA) (programmable by the operator) are often necessary to overcome the impedance offered by the chest wall, muscle and bone, and intrathoracic structures. The transcutaneous pacemaker paces the ventricle and inhibits its output when it senses spontaneous ventricular electrical activity, thus functioning in VVI (demand) mode (see later section, Permanent Pacing). Because the pacing pulses are 20–40 ms in duration and the current output is large, they create a deflection of high amplitude on the surface ECG recording that should not be confused with QRS complexes. If ventricular depolarization (capture) is occurring, the pacer output pulse will be followed by a QRS complex that is best seen on the pacemaker generator's oscilloscope and strip-chart recording. Significant distortion, or total obscuration, of the paced QRS complex can exist on the bedside rhythm monitor or surface ECG recording. Ventricular capture should always be verified by confirming the presence of a pulse, either through palpation or through visualizing a proper waveform with pulse oximetry or arterial pressure monitoring. Skeletal muscle twitching occurs at a stimulus output of 30 mA, but ventricular capture does not usually occur until 35–80 mA; sedation of the awake patient is usually required to mitigate the painful muscle contractions.

Transcutaneous cardiac pacing can be effective in up to 70% of patients and has its best use in an emergency situation when pacing of short duration is required or as a bridge to permanent cardiac pacemaker implantation. The majority of pacing failures (specifically, failure to capture) occur in patients during the advanced stages of cardiopulmonary arrest. The likelihood of successful transcutaneous pacing in patients with cardiac arrest of more than 15 minutes in duration is approximately 33–45%. Failure to capture can also occur after prolonged (hours to days) pacing and likely represents increases in impedance; repositioning of the electrodes can restore pacing capability.

Temporary Transvenous Pacing

Although transcutaneous pacing offers ease of use, rapid initiation of pacing therapy, and very low complication rates, transvenous pacing is far more stable and better tolerated if pacing is needed for longer than 20–30 minutes. Transvenous pacing is usually performed by placing an electrode catheter in the right ventricle. In rare cases where temporary atrial pacing is also required, catheters can be positioned in the right atrium or in the proximal portion of the coronary sinus.

Venous access can be obtained by several approaches. The internal jugular, subclavian, and femoral veins are all potential sites for introduction of the pacing catheter into the right heart, although the femoral vein is the least desirable due to the potential for pacing lead dislodgement and infection. The median cubital and basilic veins can also be used, but these sites are also associated with a high incidence of lead dislodgement (because of arm motion) and are rarely, if ever, used today.

Prior to obtaining venous access, the existence of a bleeding diathesis or coagulopathy should be excluded or corrected if possible. If this is not possible, the femoral vein should be considered as the initial access site because it is easier to apply pressure and achieve hemostasis in this region if a complication occurs. Other factors, such as the patient's pulmonary status, location of dialysis shunts, previous neck surgery, or radiation therapy should be taken into account when considering the appropriate site for venous access. The presence of a prosthetic tricuspid valve is a contraindication to right ventricular pacing; in this circumstance, left ventricular pacing can be performed by positioning the pacing catheter in the left ventricular veins via the coronary sinus.

There are two main types of transvenous pacing catheters. The flexible balloon-tipped catheter is advanced into the heart in a similar way as a Swan-Ganz catheter, using blood flow to guide it into the right ventricle; it is important to note that during a cardiac arrest, the inflation of the balloon will be useless because of the lack of circulation. The position of the catheter tip within the heart is confirmed either through assessing the electrogram recorded from the pacing catheter tip (negative pole connected to any precordial lead on an ECG machine) or by observing capture of ventricular tissue. Once the catheter tip crosses the tricuspid valve, the balloon should be deflated to allow advancement into the ideal right ventricular apical position. Balloon-tipped pacing catheters can be inserted at the bedside if necessary. The nonfloating, rigid, fixed-curve catheters are easier to manipulate and are more stable once positioned in the right ventricle; because of their rigid design they are typically placed only under fluoroscopic guidance. In general, temporary transvenous pacing lead positioning should be accomplished using fluoroscopy whenever feasible.

Ventricular capture thresholds should be < 2 mA (or < 2 volts [V] with some pulse generators) and ideally < 1 mA (or < 1 V) in stable lead positions and should not change with coughing or deep breathing. Atrial leads are typically less stable, and capture thresholds around 2 mA (or 1–2 V) are acceptable. The presence of myocardial infarction, ischemia, antiarrhythmic drug therapy, hyperkalemia, and other metabolic derangements can increase capture thresholds. Current or voltage output of the pulse generator should be programmed to at least twice capture threshold.

Sensing thresholds can also be affected by myocardial ischemia or infarction, hyperkalemia, and class I antiarrhythmic agents, leading to undersensing (“failure” to sense). Ectopic ventricular depolarizations are often undersensed because of poor intracardiac signal quality. These considerations need to be borne in mind when setting the sensitivity of the pacemaker; inappropriate pacing occurring because of an undersensed QRS complex can initiate ventricular tachyarrhythmias if the pacing stimulus falls on the terminal portion of the QRS complex or T wave.

A daily chest radiograph and paced 12-lead ECG should be obtained and compared with prior studies to check for possible lead migration. Pacing and sensing thresholds should be checked at least daily, with any significant changes being investigated for possible lead migration, lead disconnection from the pulse generator, or change in the patient's clinical status. Pulse generator battery status should be monitored by the appropriate biomedical personnel, and batteries replaced as needed. Temporary leads and access sites should be changed at least every 3 or 4 days to decrease the risk of infection and venous thrombosis.

Although temporary transvenous pacing is relatively low risk, there are potentially serious complications. Complication rates range from 4% to 20% and include pneumothorax, hemothorax, arterial puncture, air embolism, serious bleeding, myocardial perforation, cardiac tamponade, nerve injury, thoracic duct injury, catheter-related arrhythmias, infection, and thromboembolism. The risk of complications is increased if pacing is initiated in emergent situations. To minimize risk, transvenous pacing should be accomplished when the patient is relatively hemodynamically stable. It is important to remember that if the patient has left bundle branch block (LBBB), pacing catheter manipulation in the area of the right bundle branch can result in complete AV block; in these circumstances, transcutaneous pacing should be in place should rate support be required.

Permanent Pacing

Because of the complexity of pacing system design, an identification code has been developed that describes the function of currently available pacemaker generators. The “mode” code consists of three primary letters. The first letter stands for the chamber in which pacing is occurring: A for atrium, V for ventricle, and D for dual, or both. The second letter stands for the chamber in which sensing of the electrical signal occurs: A, V, D, or O for neither. The third letter refers to the mode of response of the generator to the sensed signal: I for inhibited output, D for both inhibited and triggered output that is delivered in response to a sensed signal (eg, a paced ventricular complex delivered in response to a sensed P wave), and O for not applicable. Currently available pacing systems incorporate one or two sensors that allow the pacing rate to increase and decrease with changes in metabolic need; sensor-based pacing systems thus adapt the pacing rate to the activities of daily living. Pacing systems with this feature add an R following the three primary letters (eg, AAIR, DDDR), indicating the existence of rate-adaptive capability. Current pacemakers have numerous functions that can be altered noninvasively by a programmer; such units are described as having multifunction programmability (Table 14–8). Several of the newer temporary pulse generators also have such features.

Modes of Pacing

Asynchronous Pacing (VOO, AOO, DOO)

In the asynchronous mode of pacemaker function, no electrical signals are sensed, and the pulse generator delivers output pulses without regard to any electrical activity occurring spontaneously within the heart (Figure 14–14). Because the native cardiac rhythm is not sensed, competitive rhythms (paced and native depolarizations) can result. Asynchronous pacemaker generators are no longer manufactured; however, the asynchronous pacing mode can be programmed. Asynchronous pacing also occurs whenever a magnet is placed over an implanted generator in order to evaluate capture function. With a magnet in place, asynchronous pacing and concomitant occurrence of the spontaneous rhythm result in iatrogenic parasystole (Figure 14–15). At the energy output of today's generators (1.5–7.5 V), induction of repetitive ventricular or atrial rhythms is usually not observed, although this possibility exists, especially if myocardial ischemia or electrolyte imbalance is present.

Single-Chamber Demand Pacing [VVI(R), AAI(R)]

Both sensing and pacing circuits are present in these units. When a spontaneous intracardiac signal is sensed, VVI and AAI pulse generators will inhibit their output and no pacemaker stimulus artifact will appear. Electrical signals sensed by demand pacemaker generators can originate not only from the heart but also from the environment (electrocautery, cellular telephones, electronic article surveillance systems, tasers), from the patient (muscle potentials), or from the pacing system itself (lead fracture or insulation breaks). Such sensed signals may cause inhibition of output, leading to pauses in paced rhythm; this phenomenon is termed “oversensing” (Tables 14–9 and 14–10), a problem that can generally be corrected by noninvasive programming or, in the case of lead fracture or insulation break, by replacement of the lead. Current generator design and programming capability have not only helped to reduce problems of oversensing but have also simplified their correction.

The pacing function of a demand pulse generator cannot be evaluated if the patient's spontaneous rhythm exceeds the programmed standby (base) rate of the generator. Applying a magnet over the pulse generator converts it to an asynchronous mode of function, and capture (stimulation) of the atria or ventricles by the pacemaker can be confirmed, provided that the pacing stimuli fall outside the refractory period of the cardiac tissue. Conversely, if the patient's rhythm is continually paced, the sensing function of the generator cannot be evaluated. Programming the device to a lower rate may allow the emergence of a spontaneous cardiac rhythm, which should then be sensed, resulting in inhibition of pacemaker output.

Continuous single-chamber ventricular pacing will result in AV dyssynchrony (unless the atrial rhythm is fibrillation). Although the symptoms attributable to bradycardia are alleviated through ventricular rate support from the pacemaker, these may be replaced by symptoms due to AV dyssynchrony, or pacemaker syndrome. Either single-chamber atrial pacing (with intact AV conduction) or dual-chamber pacing is a solution to avoid pacemaker syndrome.

Single-Lead P-Synchronous Pacing (VDD)

These are systems in which electrodes for both the atrium and the ventricle are located on a single lead. The lead is positioned in the right ventricle where the tip electrodes sense and pace the ventricle; atrial electrodes are located on the lead body, at the level of the atrium and only sense, at least at the present time. When the atrial electrodes sense an electrical signal, a ventricular pacing stimulus is delivered after a programmable AV delay that corresponds to the PR interval. If a spontaneous QRS complex occurs, the ventricular output is inhibited. Tracking of the atrial rhythm in a 1:1 relationship allows the ventricular paced rate to change with the sinus rate. The programmed upper rate is the maximum ventricular paced rate that can occur in a 1:1 relationship to atrial activity, and prevents rapid ventricular paced rates should the atrial rate become too fast. If the atrial rate exceeds the programmed upper rate limit, the paced ventricular rate can become irregular because of an electronic Wenckebach protection, can slow to one-half the programmed upper rate limit, or can fall back gradually until 1:1 tracking can resume. This last feature results in disengagement of the tracking function, which causes transient AV dyssynchrony (Figure 14–16).

If no atrial activity is sensed, as occurs in sinus bradycardia, these pulse generators pace the ventricles on demand at the programmed standby rate; atrial pacing does not occur, at least at the present time. Thus, at slow atrial rates, the pacing system behaves as though it were a VVI system, and AV synchrony is lost (see Figure 14–14). Although this pacing system seems ideal for patients with normal sinus rhythm and AV block, atrial bradycardia, which occurs commonly over ensuing years, either spontaneously or as a result of medications, makes the VDD device ultimately suboptimal for most patients.

Dual-Chamber Pacing [DDD(R)]

These pacing systems are capable of sensing and pacing in both the atrium and the ventricle on demand (see Figure 14–14). They therefore approach the physiology of normal AV conduction. The ability to sense retrograde atrial depolarizations can lead to ventricular stimulus delivery and ventricular pacing in response; if the paced ventricular depolarization travels retrograde to the atrium to depolarize it, the process can become repetitive. This event creates an artificial extra-AV-nodal bypass tract, causing a “pacemaker-mediated tachycardia.” Specific algorithms have been designed to terminate these tachycardias and are automatic once they have been programmed.

Dual-chamber devices depend on a stable atrial rhythm for optimum function. Because of their potential for rapid paced ventricular rates, these systems should not be used with atrial arrhythmias such as chronic fibrillation or flutter, multifocal tachycardia, or refractory automatic tachycardia; single-chamber VVI(R) devices should be used instead.

If the atrial tachyarrhythmias are paroxysmal, however, a programmed “mode switch” feature should be programmed on that automatically changes the mode from DDD(R) to either DDI(R) or VVI(R) modes when a rapid atrial rate is detected; this removes the ability to track the atrial rate as long as the atrial tachyarrhythmia persists (Figure 14–17). Studies have established that in patients with bradycardia, compared with ventricular pacing, atrial-based pacing reduces the incidence of atrial fibrillation and may reduce stroke.

Rate-adaptive pacing systems are appropriate for patients with permanent atrial arrhythmias with slow ventricular response and for patients whose sinus node dysfunction prevents rate acceleration, but who would benefit from an increase in paced ventricular or atrial rates, respectively, in response to increases in metabolic demand. Current sensors measure body motion and acceleration, minute ventilation, QT interval, or intracardiac impedance; sensors to measure other parameters (eg, right ventricular dP/dt [rate of rise of pressure within the right ventricle]) are under development. Changes within the sensor's established parameters, designed to reflect physiologic needs, result in changes in paced rates. Sensor-based pacing rate depends on the individual sensor used, however; for example, if an activity sensor is being used, the paced rate can increase in response to body vibrations that are unrelated to actual physical activity, such as shivering or tremor. This can cause problems in hospitalized patients, especially those in intensive care units. Several manufacturers, therefore, currently incorporate two sensors into their pacemakers to confirm the need for appropriate changes in pacing rate. For example, the activity sensor input can be confirmed by a more physiologic sensor such as minute ventilation, resulting in a more specific and accurate response to the change in pacing rate for the particular change in metabolic need. Recognition of sensor-based pacing and changing pacing rates is necessary to avoid erroneous diagnoses of pacemaker malfunction.

A growing number of studies have reported that right ventricular pacing can be detrimental, especially in patients with left ventricular systolic dysfunction. The left ventricular dyssynchrony caused by pacing from the right ventricle is thought to induce further cardiomyopathy and systolic heart failure due to functional LBBB. Dyssynchrony-induced cardiomyopathy has also been observed in patients with spontaneous LBBB, where abnormal electrical activation can initiate electrical remodeling that leads to myocardial remodeling; this form of cardiomyopathy can be reversed with cardiac resynchronization therapy through biventricular pacing. When there is preexisting cardiomyopathy, biventricular pacing can address the dyssynchrony caused by unavoidable right ventricular pacing.

In patients who do not need continuous ventricular pacing, a new development in dual-chamber pacing design provides various parameters that can be programmed to minimize unnecessary ventricular pacing. The AV delay can be programmed to automatically extend to allow native AV conduction as much as possible. Another programmable solution is managed ventricular pacing (MVP) mode (Medtronic, Inc., Minneapolis, MN) that maintains AAI(R) mode until AV block is detected, at which time it automatically switches to DDD(R) mode; AAI(R) mode is restored when the pacemaker detects resumption of native AV conduction.

The type of pacing system implanted is indicated on an identification card supplied to the patient by the manufacturer; patients should carry these cards with them at all times. It is important to note, however, that such information does not guarantee the operation of a particular mode of function, rate, or any parameter that can be programmed by the patient's pacemaker physician. As pulse-generator design and function increase in complexity, it is best to assume that the pacemaker is performing normally until proved otherwise (there are, of course, malfunctions and “pseudo” malfunctions; these are addressed in the following sections). Similarly, ECGs in paced patients should be considered to reflect normal device function unless they are interpreted otherwise by personnel experienced in pacemaker ECG.

Unipolar and Bipolar Pacing

Unipolar pacing systems have the cathode (stimulating electrode) in the heart and the anode at the generator. The distance between the cathode and anode in these systems results in the inscription of large pacing artifacts whose direction (pacing-artifact axis) in the frontal plane points toward the anode in analog ECG recorders, but not in digital recorders in which stimulus artifact amplitudes and axes vary due to digital sampling of the pacing stimulus (see below).

Bipolar pacing systems have both lead electrodes within the heart, usually less than 1 cm apart, in either or both atrium and ventricle. Either the distal (tip) electrode or the proximal (ring) electrode of the lead can serve as the cathode. Because of the small interelectrode distance, the pacing artifacts are small and their direction in the frontal plane reflects the direction of current flow (Figure 14–18). It is common for the small pacing artifacts to appear to be absent on ECGs and ECG monitoring strips; even computer-interpreted ECGs may fail to indicate that a pacemaker is present. Identification of paced (as opposed to spontaneous) P waves and QRS complexes must therefore be undertaken; often, magnet application with comparison of paced P and QRS morphologies with the initially recorded complexes is necessary to accomplish this.

In sum, ECGs recorded on digital rather than analog machines can show marked variations in both amplitude and polarity of pacing artifacts. Because the digital equipment samples the pacing stimuli at specific time intervals and then recreates them on paper, the inscribed stimulus artifacts are not seen in real time. In some ECG leads, the pacing stimuli may not be visible at all, raising the questions of spontaneous wide QRS complex rhythms or even failure of generator output. It is important to recognize this recording artifact in patients with pacemakers to avoid an erroneous diagnosis of pacemaker malfunction. It is equally important to document the morphology of paced complexes so that when the pacing stimuli cannot be seen, normal pacemaker function can be assumed until an accurate evaluation can be made.

Some permanent bipolar pacing systems offer lead polarity that can be programmed to unipolar; therefore, the presence of a bipolar lead on chest radiograph does not ensure bipolar lead function, and the ECG appearance of the pacing artifacts may differ from what is expected.

Electrocardiographic Patterns of Paced Complexes

These patterns depend on how the myocardium is depolarized. Paced atrial complexes reflect the sequence of atrial activation initiated by the pacing impulse and thus, in part, the site of the pacing electrode(s). Because the atrial electrodes can be located in the atrial appendage or screwed into any portion of atrial tissue, paced P wave contours and axes will vary.

Pacing from the right ventricular apex produces paced QRS complexes that have a LBBB configuration (reflecting right ventricular myocardial depolarization before left ventricular depolarization) and a superior mean frontal plane axis (the apex of the heart is depolarized before the base; Figure 14–19). Paced QRS complexes usually have a duration of 120–180 ms; if they are substantially longer, intrinsic myocardial disease, hyperkalemia, or antiarrhythmic drug therapy (eg, amiodarone) should be suspected.

Pacing from the right ventricular outflow tract also results in QRS complexes that have a LBBB pattern, but the mean frontal plane axis is inferiorly directed (the base of the heart is depolarized before the apex). Occasionally, pacing from the interventricular septum can result in paced QRS complexes that show an indeterminate conduction delay pattern; they can even be narrow and relatively normal-appearing. This reflects almost simultaneous activation of both the right and left sides of the interventricular septum.

Pacing from the left ventricular epicardium produces paced QRS complexes having a right BBB pattern, reflecting left ventricular myocardial activation in advance of right ventricular activation. The mean frontal-plane QRS axis will depend on the location of the epicardial electrodes relative to each other (bipolar system) or to the pulse generator serving as anode (unipolar system).

Over the last decade, left ventricular pacing has been accomplished from the coronary veins approached via the coronary sinus. This technique is used along with simultaneous, or near simultaneous, pacing from the right ventricle in patients with advanced heart failure and wide QRS complexes (≥ 120 ms) in order to “resynchronize” ventricular depolarization-contraction. The biventricular-paced QRS complexes can be narrower and more normal-appearing than the patient's spontaneous QRS complexes, reflecting electrical resynchronization and suggesting a beneficial result of this therapy; electrical synchrony, however, does not necessarily correlate with mechanical synchrony, and a persistently wide QRS complex does not predict failure of therapy. The mean frontal plane axis of the paced complexes will vary with the location of the electrodes. Because the timing of the two ventricular leads can now be programmed separately, and either lead can be activated first by up to 80 ms, biventricular paced QRS complexes can be preceded by tightly coupled double-pacing artifacts (Figure 14–20).

Spontaneous QRS complexes occurring in patients with pacemakers often show marked T-wave inversion (Figure 14–21). This phenomenon has been explained as “T-wave memory” or the temporary persistence of abnormal repolarization “learned” by the myocardium during pacing. The ECG abnormality should not be interpreted as acute or chronic myocardial disease (including ischemia and infarction) in the absence of clinical indications.

Pacemakers and Magnetic Resonance Imaging

Traditionally, magnetic resonance imaging (MRI) has been contraindicated in patients with pacemakers. This is based on a number of observed deleterious effects that the strong magnetic field can have on a pacemaker and on a handful of reported deaths possibly associated with MRI scanning. Potential effects range from activation of rate-adaptive sensors causing rapid pacing rates to reverting of the pacemaker generator to a default mode and settings. There is also concern that the magnetic field can induce movement of current in the pacemaker lead, leading to heating at the lead tip-tissue interface with acute rises in capture threshold; in a pacemaker-dependent patient, critical loss of capture could then ensue.

There are instances where the benefits of MRI scanning (eg, early detection of cancer, imaging of the brain) need to be weighed against the potential risks to the patient with a pacemaker. Protocols have been established by different institutions that allow for non–pacemaker-dependent patients to undergo MRI scanning under close observation after making temporary programming changes that will minimize interaction with the magnetic field. Pacemakers implanted in the last 10 years have advanced electromagnetic interference protection that decreases the likelihood of encountering a problem. Pacemakers and leads that are Food and Drug Administration–approved to be used with MRI have recently become available, and MRI compatibility is projected to eventually become a standard feature of pacemakers in the future. MRI compatibility should be part of the discussion with patients prior to pacemaker implantation, and the choice of pacemaker type should be a factor, especially if there are underlying conditions that will likely require future MRI scans (eg, history of cancer, orthopedic problems, neurologic abnormalities).

Pacing System Malfunctions

Pacing system malfunctions fall into four general categories: (1) undersensing, or “failure” to sense; (2) oversensing, or the sensing of unwanted signals; (3) failure to capture and stimulate myocardial tissue; and (4) failure of output. Undersensing of cardiac electrical signals because of poor intrinsic signal quality does not represent sensing failure as such, but rather the inability to detect the suboptimal signal itself; undersensed P waves and QRS complexes are not rare. Premature ventricular complexes (see Table 14–9; Figure 14–22) generate suboptimal signals because they originate from within the myocardium, away from the normal conduction apparatus; they can occur in patients without structural heart disease, during acute myocardial ischemia and infarction, or as a result of drug toxicity and electrolyte imbalance. Undersensed P waves can be caused by changes in atrial volume, ectopic atrial rhythms, or retrograde atrial depolarizations. Undersensing of spontaneous complexes and consequent failure to inhibit output results in the delivery of an earlier-than-expected pacing stimulus, which can, on occasion, induce repetitive rhythms.

Ventricular pacing artifacts can sometimes occur after the onset of spontaneous QRS complexes that have a right BBB configuration, raising concern of undersensing. This happens because of delayed conduction in the right bundle branch: the wavefront of ventricular depolarization does not reach the lead electrode in the right ventricle (especially the apex) in time to inhibit the output of the pacing stimulus. This phenomenon of stimulus delivery within a QRS complex, called “pseudofusion” (as opposed to true fusion), may also be observed in patients with inferior and right ventricular myocardial infarction and is probably due to the conduction delay resulting from ventricular scarring. The same principles apply to patients who have a left ventricular epicardial electrode and either underlying LBBB or ventricular scarring and to patients who have a right atrial electrode and an intra-atrial conduction delay. Undersensing in these cases is due to intrinsic conduction system disease rather than to a malfunctioning unit. The problem is managed by extending the programmed AV delay (in the case of ventricular pseudofusion), programming a higher sensitivity, or if necessary, increasing the pacing rate to overdrive the native rhythm.

Oversensing refers to sensing of unwanted electrical signals such as T waves, myopotentials, and environmental signals (eg, electrocautery; see Table 14–10). Programming the pulse generator to sense only electrical signals of larger magnitude will often solve the problem. When a programmer is not immediately available, placing a magnet over the pulse generator will temporarily eliminate the oversensing by converting the generator to a nonsensing asynchronous mode. Because competitive rhythms with the magnet in place can induce ventricular tachyarrhythmias, these patients should be in a monitored unit.

Failure to capture exists when pacing stimuli do not depolarize nonrefractory myocardium (Figure 14–23; Table 14–11). This condition may result from poor electrode position; a subthreshold programmed output; output reduction due to battery depletion; or an increase in myocardial stimulation threshold that usually results from acute myocardial infarction, drug toxicity, electrolyte imbalance, cardiopulmonary resuscitation, or fibrosis at the electrode-tissue interface. Pacemaker noncapture can be managed by noninvasive programming of the generator's energy output (voltage and pulse duration), surgical repositioning of the lead, or generator exchange, depending on the underlying problem.

The difference between failure to capture (stimulus artifact is present) and failure of output (lack of stimulus output when such output is expected and indicated) should be recognized. If the pacing stimulus has not been delivered (Table 14–12), capture cannot be ascertained. Applying a magnet will aid in determining the cause for the lack of stimulus output. Asynchronous pacing will result with magnet application if the lack of pacing stimulus output results from inhibition due to oversensing. If there is a true problem with delivery of a stimulus output, no stimulus output and therefore no paced complexes will be seen despite magnet application. The main causes of failure of output are battery depletion and generator component failure. This is to be distinguished from the situation when pulse generator output is occurring normally but the current is not reaching body tissues to depolarize it, due to lead fracture or insulation break, and loose set screw (loss of connection between the lead connector pin and the pacemaker generator). Management of pacemaker failure of output will require replacement of the generator, whereas lead problems usually require lead replacement, and loose set screws are addressed merely by restoring a good connection between the lead connector pin and the generator.

Assessment of Pacing System Function

All patients should have a 12-lead ECG, with and without a magnet applied, to allow identification of spontaneous (where present), purely paced, and fusion P waves and QRS complexes, as well as sensing and pacing functions in both atria and ventricles. The (nonprogrammable) pacing rate with a magnet in place is not the same as the programmed rate, and the mode of function with the magnet in place may differ from the programmed mode (eg, a DDD system may have a magnet mode that is VOO, although this is currently a rare event; Figure 14–24). The magnet mode, rate, and AV interval for each device is manufacturer and model specific, and may vary, with a number of stimuli delivered at one rate and AV interval, followed by another rate and AV interval.

In addition, all patients should have highly penetrated posteroanterior and lateral chest radiographs in order to assess the leads and, when possible, to identify the pacemaker's manufacturer and model number. The number and type (unipolar, bipolar, MRI-compatible) of leads can be ascertained, as well as the positions of the lead tips and pulse generator. Lead tips lying outside the cardiac silhouette suggest the possibility of myocardial perforation. Occasionally, lead insulation degradation, wire fracture, or improper connections between the lead and generator can be seen.

More sophisticated evaluation techniques, such as interrogation of the programmed parameters of the pulse generator, sensing and pacing threshold determination, and recording of intracardiac electrograms, can be necessary to detect and then determine the cause of pacemaker malfunction; these evaluations should be performed by a pacemaker specialist.