Absence of Paced Complexes

The absence of paced P or QRS complexes when expected has two general causes: (1) failure of the device system to deliver pacing output and (2) pacing output being delivered but failing to stimulate the myocardium. Failure of pacing output is recognized by the absence of expected pacing spikes, whereas failure to capture is associated with the presence of pacing spikes that fail to capture the myocardium. True failure to capture must also be distinguished from apparent capture failure, in which the evoked response is of small amplitude, such as with atrial pacing. If the paced QRS amplitude is small, myocardial capture can be confirmed by reviewing multiple ECG leads or looking for the accompanying T wave. Failure to capture the myocardium could result from sub-threshold pacing output or circumstances, in which pacing occurs during the refractory period of the chamber (Figure 36-3).

Figure 36-3 Rhythm strip showing ventricular pacing spikes with failure of ventricular capture because pacing occurred within the ventricular refractory period.

Fused and Pseudo-Fused Pacing Complexes

The phenomena of fusion and pseudo-fusion, which are best appreciated in the context of ventricular pacing, result from differences between the surface QRS recording and the signal processed by an implanted device. The paced QRS complex of the ECG reflects global ventricular activation recorded from surface leads. However, the pacemaker’s sensing circuitry and pacing output use the local electrogram (EGM) at the pacing electrodes. In turn, local EGM timing relative to global ventricular activation will depend on the precise location of the ventricular electrode(s). With ventricular demand pacing (VVI mode), intrinsic beats should inhibit pacing output, but if the local EGM is relatively late, an intrinsic beat might generate a large part of the QRS complex before local ventricular activation can inhibit ventricular pacing output. A relatively early pacing spike may contribute to some portion of ventricular depolarization, and the resulting hybrid QRS complex will reflect fusion of both activation wavefronts (true fusion). If the pacing spike is late, pacing may not contribute to ventricular activation so that the QRS complex is identical to the intrinsic rhythm (pseudo-fusion) (Figure 36-4). Pseudo-fusion must be distinguished from device undersensing issues. A phenomenon termed pseudo-pseudo-fusion also occurs, during which noise artifacts resembling pacing stimuli are superimposed on an intrinsic QRS complex and give the appearance of pseudo-fusion (Figure 36-5).

Figure 36-4 Rhythm strip showing ventricular paced beats with varying degrees of ventricular fusion. Note the changing morphology of the beats; beat 1 is fully paced, beats 2 through 4 show fusion beats, beats 5 through 7 show pseudo-fusion beats, and beat 8 shows intrinsic sinus rhythm.

Figure 36-5 Pseudo-pacemaker stimuli (pseudo-pseudo-fusion) seen on the last four beats on multiple-lead rhythm strip in a patient with no existing pacing system.

Atrial Pacing

The conventional site of atrial pacing has been the right atrial (RA) appendage. The resulting paced P wave is positive in leads I, II, III, and aVF and is negative in aVR. In lead V1, the P-wave pattern is quite similar to sinus rhythm with a terminal negative deflection. Pacing from various alternative atrial sites is being actively explored.²² Pacing from the coronary sinus ostium or the low inter-atrial septum typically results in a biphasic or negative P wave in leads II, III, and aVF and a biphasic (initially positive and then negative deflection) P wave in lead V1.²³ The RA endocardium adjacent to Bachmann’s bundle at the confluence of the RA roof and the inter-atrial septum yields a positive P wave in leads I, II, and III; it starts immediately with the pacing spike, and the P-wave width may be noticeably shortened.^22–25

Right Ventricular Pacing

In describing ECG patterns observed with right ventricular (RV) pacing from various sites, one has to note some degree of variability caused by inter-individual differences in underlying cardiac pathology and by imprecision in the definition of various RV regions and in the confirmation of lead location. In reality, the RV septum, free wall, outflow tract, and apex are fairly large regions (rather than circumscribed points) that merge seamlessly into one another, and their boundaries are ill defined radiographically. Within those limitations, RV pacing generates certain basic QRS patterns.

RV pacing almost always produces a QRS pattern resembling left bundle branch block (LBBB) with predominantly negative QS complex in leads V1 and V2, and left ventricular (LV) stimulation will result in a right bundle branch block (RBBB) pattern.^1–15 By altering the pattern of ventricular depolarization, ventricular pacing alters ventricular repolarization so that the T-wave vector is opposite to the QRS complex; diffuse T-wave inversion that can be mistaken for other causes of repolarization abnormalities is also seen.

The RV apex has traditionally been the site of ventricular pacing, and the electrodes are commonly placed along the diaphragmatic wall. The resulting superior QRS vector (mean frontal axis more commonly to the left superior quadrant) produces negative QRS complexes in the inferior leads (Figure 36-6). Uncomplicated RV apical pacing can produce RBBB-like QRS morphology in lead V1 in 8% to 10% of patients and can be manufactured by malposition of leads V1 and V2 electrodes at the level of the second or third intercostal space (Table 36-1).² When seen, R-wave regression is the rule so that the presence of tall R waves in V3 and V4 is suggestive of an LV pacing site once fusion with intrinsic beats has been excluded.^1,2,6,7 The pacing lead is likely in the RV apex or distal septum if moving the V1 and V2 electrodes down to the fifth intercostal space restores the expected the QRS morphology. However, a dominant R wave may not always be eliminated by this maneuver if RV pacing originates from the midseptal region.²

Figure 36-6 A, Twelve-lead ECG showing typical features of right ventricular apical pacing with LBBB-like morphology in lead V1 and predominantly negative QS complex in lead V1 and in the inferior leads (II, III, and aVF). Atrial pacing is also present. B, Twelve-lead ECG showing typical features of right ventricular septal pacing with LBBB-like QRS morphology in lead V1, predominantly negative QS complex in lead V1, and positive QRS complexes in the inferior leads (II, III, and aVF).

Table 36-1 Causes of a Predominant R Wave in Lead V1 During Ventricular Pacing

Pacing alternate RV sites, predominantly in the right ventricular outflow tract (RVOT) or midseptum as a way of avoiding detrimental effects on LV performance has created interest, although the optimal RV pacing site remains to be defined.^26–32 RVOT pacing shifts the frontal plane QRS axis to the left inferior quadrant so that the QRS is positive in the inferior leads 1-10 (see Figure 36-6).³³ As pacing moves superiorly toward the pulmonic valve, the QRS axis shifts rightward.^1,2 In general, QRS duration with septal pacing is shorter compared with apical pacing.^1–3,6–12

Left Ventricular Endocardial and Epicardial Pacing

The QRS complex during LV pacing will show an RBBB-like pattern in lead V1, and this persists when leads V1 and V2 are recorded one intercostal space lower. The frontal plane axis will vary, depending on the precise LV site. A lead could be inadvertently placed into the LV cavity if the lead passes through a patent foramen ovale or atrial septal defect, perforates the interventricular septum, or is mistakenly inserted into a subclavian artery. The diagnosis of a malpositioned lead into the LV chamber is easily missed on a single-lead ECG.1–3,6–12 However, it is difficult to differentiate between endocardial and epicardial LV pacing.^1–3,6–12 LV epicardial pacing can be achieved by accessing the LV coronary venous tributaries via the coronary sinus or by direct epicardial lead placement through a thoracotomy. The main venous branches used for LV pacing are the middle cardiac, posterolateral, lateral, and anterior interventricular veins. However, there is considerable anatomic variability.^{1,2,6,7,12,14} Pacing from branches intermediate in location produces a QRS morphology that can be deduced from the patterns typical with pacing from the main veins (Figure 36-7).

Figure 36-7 A, Twelve-lead electrocardiogram (ECG) showing sinus rhythm with left bundle branch block (QRS ~150 ms) and a single premature ventricular contraction in a patient with symptomatic heart failure (NYHA class III functional class) and ejection fraction of 20% on recent imaging. B, Twelve-lead ECG in same patient, after insertion of a cardiac resynchronization pacing therapy (CRT)–implantable cardioverter defibrillator (ICD) with the left ventricular lead placed in a posterolateral vein. Note the dominant R wave in V1, negative QRS complexes in the inferior leads (II, III, and aVF), and qR complexes in leads I and V6. C, Twelve-lead ECG in a patient with a CRT-ICD showing the left ventricular lead in the anterolateral branch of the coronary sinus. Unipolar pacing spikes are seen with dominant R wave in V1, positive QRS complexes in the inferior leads (II, III, and aVF), and negative QRS complexes in leads I and aVL.

Pacing from the middle cardiac vein results in a superior QRS vector, with sharp negative deflection in leads II, III, and aVF and varying degrees of R wave. Lead I may be isoelectric or positive, depending on how septal the pacing site is. Pacing from the rightward branches of the middle cardiac vein could capture the RV myocardium, producing an LBBB-like pattern in lead V1 and a sharply negative deflection in lead III.

As the LV pacing site shifts to veins that are more lateral, the QRS complexes in lead I become increasingly more negative and those in lead III become positive. Pacing from the anterior interventricular vein produces an inferiorly directed QRS vector with predominant R waves in the inferior leads. If the leads reside in a vein that overrides the septum and paces the RV epimyocardium, an LBBB pattern may be seen.¹⁴ With apical locations in the target vein, leads V4, V5, and V6 are typically negative, and lead aVR is positive; the converse is true with more basal locations.¹⁴

Biventricular Pacing

With biventricular or CRT pacing, the final QRS morphology reflects a composite of paced left and right ventricular activation as well as any contribution from intrinsic activation, since such patients usually have intact atrioventricular (AV) nodal conduction (with sinus rhythm or atrial fibrillation) or premature ventricular contraction (PVCs).^{1,2,6,7,9,12,14} Maximizing the potential benefit of resynchronization pacing therapy requires careful inspection of the QRS morphology while adjusting the AV interval to minimize intrinsic activation. Furthermore, contemporary CRT devices allow independent programming of right and left ventricular pacing output timing to optimize LV hemodynamics.^{1,2,6,7,9,12,14} These factors will necessarily dictate the final paced QRS morphology. If RV and LV pacing spikes are sufficiently separated in time, careful inspection of surface ECG may identify up to three pacing spikes (RA, RV, and LV) (Figure 36-8). The QRS complex during biventricular pacing should be compared with pacing from individual sites. The close similarity of QRS vector and morphology to either RV or LV pacing alone suggests unbalanced contribution to global activation and might warrant adjustment of the pacing parameters.^14,34

Figure 36-8 Surface electrocardiogram lead II showing two different ventricular pacing stimuli artifacts (left ventricular pacing stimulus [LVP] and right ventricular pacing stimulus [RVP]) in a patient with a cardiac resynchronisation therapy–implantable cardioverter defibrillator. This difference was noted when the interventricular pacing output timing (V-V timing) was programmed so that the LV pacing stimulus preceded RV pacing output by 50 ms. When the V-V interval was 0 ms, only a single ventricular pacing stimulus was seen.

QRS duration is typically, but not always, shortened compared with pacing any single ventricular site. Paced QRS duration remains stable over time unless LV lead dislodgement occurs. Importantly, a very narrow paced QRS complex may reflect an undesirable contribution by intrinsic activation rather than optimal resynchronization pacing. Thus, studies have shown a poor correlation between the degree of shortening of paced QRS duration and the clinical response to resynchronization pacing therapy.^{1,2,6,7,9,12,14} In contrast, Ricci and colleagues suggested that variation of the intrinsic QRS duration over time might be useful if correlated with remodeling of the ventricles by echocardiography.³⁵ They have suggested periodic review of the intrinsic QRS complex to confirm the continued presence of intraventricular conduction abnormalities.

In older CRT systems, the unipolar LV lead (cathode) could be coupled with the RV ring anode for LV pacing. At lower pacing outputs, cathodal stimulation would capture the LV myocardium; however, at higher pacing outputs, anodal stimulation could result in simultaneous RV capture. Detection of anodal stimulation requires awareness of the possibility and careful inspection of multiple surface ECG leads for potentially subtle changes in QRS morphology as the pacing output is varied^{1,2,6,7,9,12,14,36} (see Figure 36-2). Currently, biventricular ICDs restrict the coupling of the unipolar LV lead with the RV defibrillation coil electrode, whose larger surface area reduces the likelihood of anodal stimulation. This and the increased reliance on bipolar LV leads have decreased the incidence of anodal stimulation. CRT pacemakers lack a coil electrode, so anodal stimulation is still possible.