HLC 2691 1–14

Heart, Lung and Circulation (2018) xx, 1–14 REVIEW

1443-9506/04/$36.00
https://doi.org/10.1016/j.hlc.2018.06.1046

1
2 Interpreting the Normal Pacemaker
3 Electrocardiograph
4 Harry G. Mond, OAM, MD, PhD, FRACP, FACC, FCSANZ, FHRS a[5_TD$IF],b*
5 Q1 a
[7_TD$IF]
Department of Cardiology, The Royal Melbourne Hospital, and University of Melbourne, Melbourne, Vic, Australia
6 b
Cardioscan Pty Ltd, Melbourne, Vic, Australia
7
8 Received 28 January 2018; received in revised form 6 May 2018; accepted 17 June 2018; online published-ahead-of-print xxx

Modern cardiac pacing systems have sophisticated software to document, evaluate and record intrinsic and
paced rhythms as well as correct pacing abnormalities and rhythm disturbances by applying algorithms,
which are generally company specific. To the cardiologist and technologist, these algorithms may be
difficult to interpret on both the 12-lead electrocardiograph (ECG) and Holter ambulatory monitoring
recordings, which are usually performed because of patient symptoms or physician concern. The tracings
may appear bewildering and mimic pacemaker malfunction, thus leading to unnecessary tests or even
surgery. This review will define the common programmed pacemaker modes and describe a range of ECG
appearances of normal pacemaker function during the application of testing, correcting or therapy
algorithms.
Keywords Electrocardiography  Holter monitoring  Cardiac pacemakers

9
10
11 Introduction
This review of the normal pacemaker ECG will revisit the 30
12 Q2
Q3 The implanted cardiac pacemaker has evolved from simple, different modes of cardiac pacing, how to recognise paced 31
13 non-programmable, asynchronous ventricular pacing to from intrinsic rhythms and will also review the ECG appear- 32
14 sophisticated multi-programmable dual chamber and biven- ances of the more common pacemaker company algorithms. 33
15 tricular models, all of which now incorporate company spe- The illustrations and discussion will be generic and levelled 34
16 cific testing, correcting and therapy algorithms. A at a general cardiology reader. 35
17 consequence of this technological evolution has been the
18 appearance of bizarre pacing rhythms on both 12-lead elec-
19 trocardiograph (ECG) and ambulatory Holter monitor
20 recordings. Even to the experienced ECG reporter and car-
Pacing the Myocardium 36

21 diac technologist, these appearances may suggest pacing By definition, all electrical circuits are bipolar requiring a
22 malfunction, which may lead to patient anxiety, needless cathode and an anode [1]. When applied to a conventional 37
23 investigations, inappropriate programming and even unnec- cardiac pacing lead, the terms unipolar and bipolar indicate 38
24 essary surgery. Today, 12-lead ECGs and Holter recordings the number of electrodes on the lead. A unipolar lead has a 39
25 are frequently performed in patients with implanted pace- single pole; the cathode at its tip, whereas a bipolar lead has 40
26 makers. They are indicated to confirm correct pacemaker both poles very close to each other at its distal end and 41
27 programming, evaluate palpitations, dizziness or syncope within the chamber being paced. For unipolar pacing, the 42
28 and to determine if there is an appropriate rate response anode lies on the titanium can of the implanted pulse 43
29 to physiologic demand. generator. Q4 44

*Corresponding author at: Department of Cardiology, The Royal Melbourne Hospital, 3050, Victoria, Australia. Tel.: +613 9527 9243; Fax: +613 9527 9262.,
Email: hmond@bigpond.net.au
© 2018 Published by Elsevier B.V. on behalf of Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia
and New Zealand (CSANZ).

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

2 H.G. Mond

Figure 1 Pacemaker stimulus artefact in 12-lead ECGs.

Left: Oscilloscopic appearance of a unipolar stimulus artefact. Initial deflection is followed by an exponential decay curve.
Right A: Unipolar dual chamber pacing. Ventricular pacing is from the apex with an extreme axis deviation and no R waves
across the chest leads. A magnified V4 highlights the large stimulus artefacts.
Right B: Same patient, bipolar pacing. No atrial and attenuated ventricular stimulus artefacts best seen in magnified V4.

repetition rate. On occasion, fusion beats may occur when 70

45 The ECG of Normal Pacemaker there is insufficient time for the pacemaker lead to sense a 71
46 Rhythms propagated intrinsic QRS (Figure 3A). With intact atrio-ven- 72
tricular conduction, the ventricular paced QRS may demon- 73
On the ECG tracing, delivery of pacemaker energy to the
strate a retrograde P wave due to ventriculo-atrial 74
47 myocardium can be seen as a stimulus artefact immediately
conduction (Figure 3B). 75
48 prior to atrial or ventricular depolarisation. It represents about
The 12-lead ECG is helpful in confirming the ventricular 76
49 two to five volts delivered for about 0.5 ms and on an oscillo-
pacing site, which traditionally has been the right ventricular 77
50 scope appears as an exponential decay curve (Figure 1). With
apex. There is a left bundle branch block appearance with 78
51 unipolar pacing, the stimulus artefact is large and represents
poor or absent R waves from V4 to V6 (Figure 1). The apical 79
52 the dissipation of energy through the chest to the implanted
impulse is propagated through both ventricles in a retro- 80
53 pulse generator. Bipolar pacing creates very small or absent
grade manner, resulting in a left or extreme axis deviation 81
54 stimulus artefacts on the ECG. Figure 1 demonstrates both
(Figure 1). The right ventricle can also be paced from non- 82
55 unipolar and bipolar pacing. The size of both the unipolar
apical sites. R waves develop in V4 to V6, as the lead position 83
56 and bipolar stimulus artefact is dependent on the programmed
progresses more superior in the right ventricular cavity, 84
57 voltage and varies with the respiratory cycle (Figure 2A) [2].
whereas the axis becomes more normal (Figure 4) [3]. 85
Today, a well-established non-bradyarrhythmic indication
58 Atrial Pacing
for cardiac pacing in patients with congestive cardiomyopathy 86
59 Confirmation of atrial pacing, particularly bipolar, can be
is cardiac resynchronisation therapy using biventricular pac- 87
60 difficult to confirm on the ECG. Because the lead is usually
ing. Clinically, for maximum benefit, the left ventricle is usu- 88
61 positioned within the appendage, close to the sinus node, the
ally paced about 40 to 80 ms before the right ventricle, resulting 89
62 P wave is often similar in appearance to the sinus P wave
in a right bundle branch block appearance on the ECG (Figure 90
63 (Figure 2A). Pacing from a low atrial position close to the
5) [4]. Two ventricular stimulus artefacts can often be identi- 91
64 tricuspid valve may mimic junctional rhythm and is very
fied, although today, multipolar left ventricular pacing may 92
65 different to the sinus wave (Figure 2B).
result in more than one left ventricular stimulus artefact. 93

66 Ventricular Pacing
67 Unlike atrial pacing, recognition of ventricular pacing is Modes of Cardiac Pacing 94
68 generally easy, provided a stimulus artefact can be recog- With the development of atrial and dual chamber pacing 95
69 nised (Figure 1). The paced QRS is wide with a regular systems, it became necessary to create a simple and easily 96

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 3

Figure 2 Atrial sensing (As) and bipolar pacing (Ap). Holter monitoring leads V5, V1 and V3.
A. Three channel Holter recordings demonstrating Ap from the atrial appendage similar to sinus P waves (As). The stimulus
Q5 artefact size varies with respiratory cycle (red arrows demonstrating progressively smaller spikes.
B. Ap adjacent to coronary sinus ostium. Paced P waves are inverted.

97 recognised classification of the different modes of cardiac (Figure 7). This ECG appearance can be confusing and lead 125
98 pacing (Table 1). Although, there are many theoretical com- to misinterpretation of otherwise normal function. 126
99 binations, only five are used clinically (Table 2).
Dual [1_TD$IF]Chamber DDD 127
100 Single [9_TD$IF]Chamber Ventricular This is the most widely used pacing mode [6]. There is atrial
101 The original cardiac pacemaker had no sensing (VOO) and and ventricular pacing and sensing as well as atrial and ven- 128
102 asynchronous pacing continued unabated, irrespective of the tricular synchronisation with an electronic PR interval, called 129
103 underlying rhythm (Figure 6A). As sensing circuits devel- the atrio-ventricular delay (Figure 8A). In patients with heart 130
104 oped, pacemakers were able to recognise the intrinsic QRS block and a normal sinus mechanism, there will be predomi- 131
105 and respond appropriately. The first synchronous designs nantly atrial sensing followed by ventricular pacing (Figure 8A 132
106 delivered the full energy of the pacing impulse into the left), whereas patients with sick sinus syndrome generally 133
107 intrinsic QRS (VVT) (Figure 6B). With improvements in sens- have fixed atrial pacing followed by ventricular sensing. With 134
108 ing circuits, ventricular inhibited (VVI) pacing became estab- DDD pacing, there is a low rate, usually programmed 60 or 70 135
109 lished (Figure 6C). Ventricular pacing continues at a beats per minute (bpm) and if the sinus rate falls below this 136
110 minimum programmed pulse repetition rate, unless the level, atrial pacing commences (Figure 8A right). 137
111 pacemaker output is inhibited by intrinsic ventricular
112 rhythms. Dual [1_TD$IF]Chamber VDD 138
There is sensing in both chambers, but only ventricular 139
pacing (Figure 8B). This is only suitable for patients with 140
113 Single [10_TD$IF]Chamber Atrial complete heart block and a normal sinus mechanism. Unlike 141
114 Although single chamber atrial pacing would appear appro- DDD pacing, if the sinus rate falls below a programmed 142
115 priate in patients with sick sinus syndrome and normal atrio- lower rate, there will be no atrial pacing, but rather ventric- 143
116 ventricular conduction, it has never been widely accepted, ular pacing and loss of atrio-ventricular synchrony. The 144
117 because of the concern of development of atrio-ventricular absence of atrial pacing in the presence of atrial sensing 145
118 block which is about 1% per year with a freedom rate of may suggest pacemaker malfunction. 146
119 88.6% at 10 years [5]. Consequently, when atrial pacing alone
120 is indicated, a dual chamber pacemaker is implanted and Dual [1_TD$IF]Chamber DDI 147
121 programmed appropriately. The most commonly used single There is normal pacing and sensing in both chambers, but a 148
122 chamber atrial mode is AAI pacing (Figure 2), where sensing fixed ventricular rate, irrespective of the intrinsic atrial rate. 149
123 is only available in the atrium and ventricular ectopics will The ECG appearances are confusing as the rhythm appears to 150
124 not be sensed unless there is retrograde conduction be single chamber VVI with intermittent atrial pacing 151

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

4 H.G. Mond

Figure 3 A[1_TD$IF]. Unipolar ventricular inhibited pacing with fusion. The first two beats are paced, the last two sinus. Middle
complexes are fusion beats. The sinus rate is gradually increasing, but too early to be sensed by the apical ventricular pacing
lead and a stimulus artefact is present (red vertical arrows). The first complex is mainly sinus, whereas the second
predominantly pacing. The contribution of each can be judged by the T wave size.
B. Bipolar ventricular pacing with retrograde conduction. Retrograde P waves in the ST segment (red vertical arrow) result in
the atria contracting against closed atrio-ventricular valves. There is a sinus beat followed by ventricular fusion and no
retrograde conduction.

152 (Figure 8C). DDI pacing was originally indicated for parox- activity, whereas minute ventilation and closed loop stimu- 170
153 ysmal atrial tachyarrhythmias to prevent rapid ventricular lation are more physiologic sensors which can be used in 171
154 pacing [7]. Modern pacemakers now have automatic mode specific circumstances. When a rate adaptive sensor is pro- 172
155 switching algorithms which have a more physiologic grammed ON, the letter R is added to the code although 173
156 response, although, DDI pacing may be indicated if mode when the sensor is programmed OFF, it will appear in 174
157 switching is unreliable. parenthesis (Tables 1 and 2). 175

158 Rate Adaptive Pacing

159 Pacemaker recipients with sick sinus syndrome manifesting
160 as atrial bradyarrhythmias, are unable to increase their pac-
Pacemaker Algorithms 176

161 ing rate with physiologic demand. This has been corrected by As the complexity of both single and dual chamber pacing 177
162 sensors that emulate changes in cardiac output in a linear evolved, a number of intrinsic rhythm and pacemaker inter- 178
163 fashion in response to physical exertion (Figure 9). Although actions emerged, which were corrected by an array of both 179
164 many sensors have been developed, only three have stood generic and company specific algorithms, and although 180
165 the test of time; activity, minute ventilation and closed loop familiar to the pacemaker specialist, nevertheless produce 181
166 stimulation. These sensors are now routinely incorporated confusing patterns on the ECG and particularly Holter 182
167 into otherwise conventional pacemaker systems with stan- recordings, mimicking pacemaker malfunction. The ECG 183
168 dard bipolar leads. By far, the most popular sensor is an features of a number of the more common algorithms will 184
169 accelerometer within the pulse generator which detects be discussed, but beforehand, it behoves us to review the 185

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 5

well as terminology differ between companies and a com- 190

prehensive description lies well outside the confines of this 191
manuscript. The most basic dual chamber timing cycles are 192
demonstrated in Figure 10. There is a low rate interval, 193
divided into two zones; the atrio-ventricular delay and 194
the ventriculo-atrial interval. The atrio-ventricular delay 195
duplicates the PR interval of the unpaced ECG and may 196
also be terminated prematurely by a conducted ventricular 197
sensed event. The ventriculo-atrial interval is divided into 198
the post ventricular atrial refractory period (PVARP) and 199
the atrial tracking interval. Because of the potential for 200
inappropriate sensing in both atria and ventricles, a number 201
of refractory periods are mandated within the cycle of 202
which the PVARP is both the longest and programmable. 203
Figure 4 ECGs with ventricular pacing from two sites in It is designed to prevent inappropriate sensing of retrograde 204
the right ventricle.
P waves or large far-field R waves by the atrial channel. The 205
Left: Postero-anterior (PA) fluoroscopic image of two
atrium is, therefore, refractory from the time of atrial pacing 206
bipolar leads in the right ventricle. Right: A. The top
or sensing to the end of the PVARP, with this period known 207
lead is in the right ventricular outflow tract. There is a
near iso-electric lead 1 and tall R waves in leads II and III. as the total atrial refractory period. Actual sensing in these 208
There is a tall R wave in V6. zones is governed by either total refractoriness (blanking 209
B. The bottom lead is at the right ventricular apex. There periods) or if ‘‘atrial sensing” occurs, specific algorithms 210
is a left axis deviation and no R waves in the V leads. may come into play. 211
At the end of the PVARP is the atrial tracking interval, 212
where there is normal atrial sensing. At rest, this period is 213
186 basic pacemaker dual chamber timing cycles in order to generous, allowing sensing of a normal sinus mechanism or 214
187 understand the complexities of these algorithms. ectopy. With exertion, the zone gradually disappears until an 215
upper atrial sensed rate is achieved. This upper rate can be 216
188 Pacemaker Timing Cycles extended by programming both a rate adaptive atrio-ven- 217
Timing cycles comprise a complex set of rules which govern tricular delay and a rate adaptive PVARP, which allows these 218
189 pacemaker sensed and paced events. The specific rules as intervals to shorten in a physiologic manner (Figure 9). 219

Figure 5 12-lead ECG. Bipolar, dual chamber biventricular pacing.

A. Right bundle branch block indicating left ventricular pacing before right.
B. Magnified view of leads V1 to V3 confirming left ventricular stimulus artefact (LV) followed 80 ms later by the right
ventricular stimulus artefact (RV).

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

6 H.G. Mond

Q6 Table 1 Four letter pacemaker identification code.

First letter Second letter Third Letter

(chamber paced) (chamber sensed) (chamber response)

[6_TD$IF]A — atrium A — atrium O no response

V — ventricle V — ventricle I — inhibited
D — atrium and ventricle D — atrium and ventricle T — triggered
O no pacing or sensing O no pacing or sensing D — atrium and ventricle

Table 2 Pacemaker modes most commonly used clinically.

Single chamber:

VVI(R) Single chamber. Ventricular pacing (V), ventricular sensing (V), inhibited by intrinsic ventricular rhythms (I) and
rate adaptive function available (R).
AAI(R) Single chamber. Atrial pacing (A), atrial sensing (A), inhibited by intrinsic ventricular rhythms (I) and rate adaptive
function available (R).

Dual chamber:
DDD(R) Dual chamber. Atrial and ventricular pacing (D), atrial and ventricular sensing (D), inhibited by intrinsic atrial and
ventricular rhythms as well as atrium triggering ventricle after a programmable atrio-ventricular delay (D) and rate
adaptive function available (R).
VDD Dual chamber. Ventricular pacing (V), atrial and ventricular sensing as well as atrium triggering ventricle after a
programmed atrio-ventricular delay (D). Rate adaptive function (R) rarely required as it results in VVIR pacing.
DDI(R) Dual chamber. Atrial and ventricular pacing (D), atrial and ventricular sensing (D) with limited atrio-ventricular
synchrony, resulting in lower rate ventricular pacing irrespective of the intrinsic atrial rate (I) and rate adaptive
function available (R).

Figure 6 ECGs lead II; unipolar ventricular pacing modes from the same patient.
[2_TD$IF]Above. VOO (asynchronous). No sensing. Stimulus artefacts are delivered irrespective of intrinsic rhythm and may fall within
the refractory periods, thus not contributing to ventricular depolarisation.
Middle. VVT (triggered). Full output is delivered into the sensed intrinsic beat. Fusion beats are present.
Below: VVI (inhibited). Intrinsic QRS complexes are sensed and inhibit the next stimulus artefact.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 7

own rhythm (Figure 11) [8]. This may result in confusing 233
pauses being present before ventricular pacing occurs. 234
In order to encourage intrinsic rhythms during rest or 235
sleep, the pacing rate can be programmed lower at these 236
times to mimic the physiologic bradycardia of the normal 237
heart. Called sleep or rest rate, the function is either clock or 238
sensor based and the rate of onset and offset may be gradual 239
[3_TD$IF]Figure 7 ECG lead II. Unipolar AAI pacing with a pro- or rapid. Even if clock based, the sensor is used as a cross- 240
longed atrio-ventricular delay (horizontal red arrow). check if the patient becomes roused or active during sleep 241
Within two ventricular ectopics (vertical red arrows) hours. In turn, sensor based algorithms are also time based in 242
are atrial stimulus artefacts. Although normal pace- that they require a certain period of rest before the sleep 243
maker function, atrial pacing appears asynchronous as function is activated. An example of a sensor based sleep rate 244
there is no retrograde conduction. 24-hour Holter monitor is shown in Figure 12. Because the 245
programmed low rate is violated, this may be interpreted as 246
pacemaker malfunction. 247
220 The Recognition of Pacemaker
221 Algorithms on the ECG Rapid [13_TD$IF]Pacing Rhythms 248
When interpreting pacemaker Holter monitor reports, brief 249
The array of algorithms required to correct inappropriate dual
‘‘inappropriate” rapid ventricular pacing is frequently seen 250
222 chamber intrinsic rhythm and pacemaker interactions are com-
and difficult to interpret. Most of these are short runs of a 251
223 plex and usually company specific. The features are in contin-
supraventricular tachycardia, which the ventricular channel 252
224 ual development and differ between companies and even
follows until the programmed upper limit. Another is pace- 253
225 between models from the same company. The presentation
maker mediated tachycardia, which is an endless loop re- 254
226 of these algorithms as well as the perceived ECG abnormalities
entry tachycardia that occurs in patients with dual chamber 255
227 of conventional pacing modes can be categorised into groups
pacemakers who have retrograde conduction. The pace- 256
228 which correspond to the way they present on the ECG.
maker acts as the anterograde limb of an accessory pathway 257
229 Pacing [12_TD$IF]Repetition Rate Below the Programmed Low Rate and, in order to create the substrate for a pacemaker medi- 258
In theory, with single chamber pacing, the low programmed ated tachycardia, retrograde conduction must be sensed out- 259
230 pulse repetition rate cannot be violated. There is, however, a side the PVARP which then triggers ventricular pacing, close 260
231 programmable function called hysteresis which allows the to or at the upper rate. Triggers for retrograde conduction 261
232 sensing window to be extended so as to encourage the patient’s include atrial or ventricular ectopics (Figure 13A), 262

Figure 8 ECG leads II. Dual chamber pacing.

A: DDD. Atrial pacing (Ap), sensing (As) and ventricular pacing (Vp), sensing (Vs).
Left panel: Following a Vs (ventricular ectopic), As-Vp occurs after 860 ms.
Right panel: Following a Vs, Ap-Vp occurs after 940 ms which is the low rate.
B: VDD. As-Vp. Carotid sinus pressure demonstrates sinus slowing, resulting in VVI pacing as the sensed sinus rate falls
below the low rate limit.
C: DDI. As-Vp with a limited ventricular tracking response to atrial sensing. Vp cannot violate the programmed ventricular rate
of 65 bpm (920 ms). Although P waves are sensed, there may be a significant delay before the next Vp (red vertical arrow). Atrial
pacing (Ap) occurs, if no P waves are sensed. Although the rhythm suggests VVI pacing, atrial pacing is also present.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

8 H.G. Mond

Figure 9 DDDR pacing during exercise. At rest, there is atrial pacing (Ap) and ventricular (V) fusion. As the Ap rate
increases to 125 bpm, there is both ventricular pacing (Vp) and ventricular sensing (Vs). The atrio-ventricular delay shortens
from 280 ms to 200 ms during exercise.

263 intermittent loss of atrial capture or sensing (Figure 13B), of sensed tachycardia sequences, so that the pacemaker can 270
264 generally coupled with a prolonged atrio-ventricular delay exclude intrinsic tachyarrhythmias. Ventricular pacing is 271
265 and/or short PVARP. then withheld or the PVARP is extended for one beat, to 272
266 The presence of a true endless loop tachycardia is rare block the retrograde conduction and break the incessant 273
267 today as each pacemaker company has one or more algo- tachycardia cycle (Figure 13B). 274
268 rithms designed to prevent, recognise and treat pacemaker In patients with proven neuro-cardiogenic syncope, dual 275
269 mediated tachycardia. Once detected, there are a set number chamber pacing has been shown to be beneficial in those with 276

Figure 10 The basic dual chamber timing cycles at rest (above) and upper rate (below). See text for details.
Abbreviations: Ap, atrial pacing; Vp, ventricular pacing; AV, atrioventricular delay

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 9

Figure 11 Three channel Holter recording demonstrating hysteresis with ventricular pacing (Vp) and atrial fibrillation (Vs).
The pacing rate is 50 bpm (1200 ms). Following a Vs, the rate falls to 40 bpm (1500 ms) to encourage the intrinsic rhythm.
Ventricular fusion is frequently seen with hysteresis.

Figure 12 Holter monitor heart rate histogram over 24 hours, demonstrating rest rate function of a pacemaker programmed
DDD (St Jude Medical/Abbott, Sylmar CA, USA). The low rate is 70 bpm, whereas rest rate is 60 bpm. During daytime, the
heart rate ranges between 70 and 110 bpm with occasional short runs of supraventricular tachycardia. At 23:00 hours until
01:30, the patient is restless. From then until 06:00, rest rate is activated with occasional movement. At 06:00 hours, the patient
wakes and rest rate is terminated.

Figure 13 Pacemaker mediated tachycardia.

A. ECG lead II, DDD pacing with endless loop tachycardia. There is atrial sensing (As) and ventricular pacing (Vp) with a
prolonged atrio-ventricular delay of 240 ms (red horizontal arrow). A sensed non-conducted junctional ectopic (red vertical
arrow) lies outside the PVARP, followed by ventricular pacing which sets up the endless loop tachycardia at 120 bpm.
B. DDD pacing, lead II demonstrating an automatic algorithm to terminate pacemaker mediated tachycardia. There is
initially atrial (Ap) and ventricular (Vp) pacing. Following atrial failure to capture (red downward vertical arrow), there is a
retrograde P wave (red circle) which initiates a pacemaker mediated tachycardia at 120 bpm. The pacemaker determines that
this is not an intrinsic tachycardia and extends the PVARP to 500 ms for one beat, terminating the tachycardia.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

10 H.G. Mond

Figure 14 Three channel Holter recording to demonstrate pacemaker therapy to correct sudden bradycardia (Medtronic Inc.
Minneapolis MN, USA).
Left: Full disclosure demonstrating abrupt onset of dual chamber pacing as intervention therapy for sinus bradycardia. After
two minutes, there is step down pacing to 85 bpm followed by the reestablishment of sinus rhythm.
Right: Sinus bradycardia with abrupt rapid dual chamber intervention pacing (Ap-Vp) with step-down pacing in the bottom
panel.

Figure 15 Single channel, Holter monitor recordings illustrating algorithms to minimise ventricular pacing.
A. Managed Ventricular Pacing [MVPTM AAI(R) DDD(R), Medtronic Inc.]. Atrial pacing (Ap) and ventricular sensing (Vs)
with continual ventricular monitoring in the atrio-ventricular delay. Failure of atrio-ventricular conduction (1 in red), results
in Ap-Vp with an atrio-ventricular delay of 80 ms. If two out of four paced or sensed P waves fail to conduct (1 and 4 in red),
then DDD(R) pacing (Ap-Vp) is established at the programmed atrio-ventricular delay of 240 ms.
B. Successful atrio-ventricular delay extension algorithm. There is Ap-Vp with a 240 ms atrio-ventricular delay. Extension to
at least 320 ms results in atrio-ventricular conduction (Ap-Vs) (St Jude Medical/Abbott).
C. Unsuccessful atrio-ventricular delay extension. With As-Vp, the atrio-ventricular delay is extended from 280 ms to 440 ms
and Vp remains. As-Vp then continues (St Jude Medical/Abbott).

277 predominant bradycardia [9]. The algorithm common to all characteristic appearances are those of accelerated pacing 280
278 manufacturers is based on bradycardia recognition and in the presence of a bradycardia and then gradual rate reduc- 281
279 appropriate rate intervention. On the Holter monitor, the tion (Figure 14). 282

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 11

283 Pacing [14_TD$IF]Pauses of chronic ventricular pacing on left ventricular function [10], 288
284 The presence of pauses in pacing on the Holter monitor there has, in recent years, been a concerted effort to design 289
285 usually denotes pacemaker malfunction such as a high algorithms to minimise ventricular pacing in patients with 290
286 threshold exit block or over-sensing of intrinsic or extracor- intact atrio-ventricular conduction. One group, programmed 291
287 poreal signals. Because of the potentially deleterious effects ‘‘AAI(R)-DDD(R)” is effectively AAI(R) pacing with 292

Figure 16 ECG (electrocardiograph) lead II, ventricular safety pacing. Atrial (Ap) and ventricular (Vp) pacing with atrio-
ventricular delay 160 ms (first set of red vertical arrows). A late ventricular ectopic is sensed in the ventricular safety pacing
window during the atrio-ventricular delay, resulting in shortening to 110 ms (second set of red vertical arrows).

Figure 17 Three channel Holter recordings. Auto-threshold ventricular testing algorithms.

A. Atrial pacing (Ap), ventricular sensing (Vs) with 240 ms atrio-ventricular delay. At 1:00 AM, there is an abrupt onset of
ventricular pacing (Vp), when the atrio-ventricular delay shortens to 180 ms. The condensed ECG strip below (red vertical
arrows) shows the period of testing (Medtronic Inc).
B. Atrial sensing and pacing (As-Vs). There is an abrupt onset of As-Vp, when the atrio-ventricular delay shortens to 30 ms.
After six beats, a back-up pulse occurs 100 ms after loss of ventricular capture (1.5 volts greater than threshold). As-Vp
follows at the programmed atrio-ventricular delay (St Jude Medical/Abbott).
C. Magnified view of panel B at the time of loss of ventricular capture.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

12 H.G. Mond

293 continual monitoring in the atrio-ventricular delay. Failure of discussed under timing cycles, dual chamber upper rate 300
294 atrio-ventricular conduction may result in an can be extended by programming a rate adaptive atrio-ven- 301
295 ‘‘inappropriate” pause before re-establishment of DDD(R) tricular delay and, as the atrial pacing or sensing rate 302
296 pacing (Figure 15A). increases, the atrio-ventricular delay shortens in a physio- 303
logic manner (Figure 9). 304
297 Varying [15_TD$IF]Atrio-ventricular Delay Another means of minimising ventricular pacing in 305
298 This is a common finding on pacemaker Holter monitor patients with intact atrio-ventricular conduction, is an 306
299 recordings and may be due to a number of causes. As atrio-ventricular delay extension algorithm. If dual chamber 307

Figure 18 Two channel Holter recording with DDD pacing, demonstrating upper rate behaviour during exercise. There is
atrial sensing (As), ventricular pacing (Vp) with an upper rate of 150 bpm. When the sinus rate exceeds 150 bpm, the ventricle
cannot follow 1:1 and Exhibits 2:1 block with Vp falling abruptly to 75 bpm. When the exercise is terminated, the sinus rate
falls and atrio-ventricular synchrony is re-established.

Figure 19 Mode switching. DDD pacing with atrial fibrillation.

A. DDD with upper rate ventricular pacing (Vp) at rest (120 bpm).
B. VVI 30 bpm. Same patient. The rhythm is atrial fibrillation with complete heart block.
C. Retrieved saved electrogram demonstrating onset of atrial fibrillation and mode switching (St Jude Medical/Abbott). The
marker channel above and atrial and ventricular intracardiac electrograms below. There is atrial (A) and ventricular pacing
(VR). Atrial fibrillation potentials (P) are chaotic within and outside the refractory period. There is accelerated ventricular
pacing for five beats and mode switching (AMS1) then occurs.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

Normal Pacemaker Electrocardiography 13

308 ventricular pacing is present, then, at regular scheduled the ECG as there is forced ventricular pacing generally using 330
309 intervals, the atrio-ventricular delay is extended and, if ven- very short atrio-ventricular delays (Figure 17A). With failure 331
310 tricular sensing occurs, then the atrio-ventricular extension to capture, there is a back-up stimulus artefact which may be 332
311 remains (Figure 15B). If ventricular pacing continues, then recognised on the ECG (Figure 17B,C). This back-up pulse 333
312 the algorithm is deactivated and the programmed atrio-ven- may occur every time the voltage is reduced or only when 334
313 tricular delay re-established (Figure 15C). failure to capture is recognised. The delay and polarity of this 335
314 At a time when only unipolar leads were available and back-up pulse varies between manufacturers. Once the 336
315 skeletal muscle or extracorporeal sensing was common, inap- thresholds are determined, there is automatic readjustment 337
316 propriate sensing of the atrial stimulus artefact or noise in the of the voltage output. 338
317 atrio-ventricular interval could result in ventricular inhibi-
318 tion with asystolic pauses, referred to as crosstalk. Today, Loss of [16_TD$IF]Atrio-ventricular Synchrony 339
319 most pacemaker manufacturers have a safety pacing window During exercise, the sensed sinus rate may reach and then 340
320 within the atrio-ventricular delay, where sensing, particu- exceed the upper rate limit as discussed in the timing cycles. 341
321 larly of ventricular ectopics results in shortening of the atrio- The result is a loss of atrio-ventricular synchrony, seen ini- 342
322 ventricular delay to a value between 95 to 120 ms, although tially as an atypical Wenckebach sequence and then a 2:1 343
323 potentially shorter with faster pacing rates or very short block (Figure 18) [11]. Clinically, there can be almost instan- 344
324 programmed atrio-ventricular delays (Figure 16). taneous halving of the ventricular rate, which the patient 345
325 Capture threshold is the minimum amount of voltage describes as like running into a brick wall and may be 346
326 required to consistently capture the paced chamber. In recent interpreted on the ECG as failure of atrial sensing. 347
327 years, pacemaker companies have developed specific auto- One of the limitations of DDD(R) pacing is the sensing of 348
328 threshold algorithms that test for ventricular capture thresh- atrial tachyarrhythmias resulting in ventricular pacing at the 349
329 olds. The threshold testing algorithms can be recognised on upper rate limit (Figure 19A). The atrial tachyarrhythmia can 350

Figure 20 Persistent ECG artefact from rate adaptive sensors.

Above: Minute ventilation sensor. The sensor measures changes in transthoracic impedance which reflect changes in minute
ventilation. A very small subthreshold current is delivered across the chest wall every 40 ms resulting in marked artefact
throughout the cardiac cycle (Boston Scientific, St Paul, MN, USA).
Below: Closed loop stimulation. The sensor delivers 16 bipolar subthreshold impulses commencing 50 ms after the onset of
the sensed or paced QRS. The intracardiac impedance measurements reflect changes in the cardiac contraction dynamics
(Biotronik, Berlin, Germany).

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046
 HLC 2691 1–14

14 H.G. Mond

351 be diagnosed by programming VVI pacing with a slow pulse No external financial support. 379
352 repetition rate (Figure 19B). To correct this pacemaker tachy- No acknowledgements. 380
353 cardia, an algorithm called mode switching counts the num-
354 ber of sensed atrial events, both in the PVARP and the atrial
355 tracking interval and following the application of company References 381
356 specific formulae, the pacemaker reverts to VVI(R) or DDI(R) [1] Mond HG. Unipolar versus bipolar pacing — poles apart. Pacing Clin 382
357 pacing (Figure 19C). The ventricular rate is now dependent Electrophysiol 1991;14:1411–24. 383
358 on the sensor to increase the ventricular rate. Atrial monitor- [2] Mond HG. The Cardiac Pacemaker. Function and Malfunction. Clinical 384
Cardiology Monographs. New York: Grune and Stratton; 198393–8. 385
359 ing, however, continues during the tachyarrhythmia and [3] Rosso R, Medi C, Teh AW, Thuy TH, Feldman A, Lee G, et al. Right 386
360 once the arrhythmia reverts, DDD(R) pacing recommences. ventricular septal pacing: A comparative study of outflow tract and mid 387
ventricular sites. Pacing Clin Electrophysiol 2010;33:1169–73. 388
361 Persistent ECG [17_TD$IF]Artefact [4] Barold SS, Herweg B. Usefulness of the 12-lead electrocardiogram in the 389
362 Minute ventilation and closed loop stimulation are highly follow-up of patients with cardiac resynchronization devices. Part II. 390
Cardiol J 2011;18:610–24. 391
363 physiologic pacemaker sensors used with rate adaptive car- [5] Hiroshi M, Yuichi U, Rinya K, Akihiko U, Takashi M, Yasushi T, et al. 392
364 diac pacing. They both require the generation of frequent Long-term clinical performance of AAI pacing in patients with sick sinus 393
365 sub-threshold currents, either transthoracic or intracardiac, syndrome: a comparison with dual-chamber pacing. Europace 394
2004;6:444–50. 395
366 which may be detected on the ECG as artefact (Figure 20). [6] Mond HG, Proclemer A. The 11th world survey of cardiac pacing and 396
implantable cardioverter-defibrillators: calendar year 2009. Pacing Clin 397
Electrophysiol 2011;34:1013–27. 398
[7] Barold SS. The DDI mode of cardiac pacing. Pacing Clin Electrophysiol 399
367 Conclusion 1987;10:480–4. 400
[8] Rosenqvist M, Vallin HO, Edhag KO. Rate hysteresis pacing: How 401
368 Electrocardiographs and Holter recordings are frequently valuable is it? A comparison of the stimulation rates of 70 and 50 beats 402
369 performed on patients with implanted cardiac pacemakers per minute in patients with sick sinus syndrome. Pacing Clin Electro- 403
370 and may demonstrate an array of poorly understood appear- physiol 1984;7:332–40. 404
[9] Baron-Esquivias G, Morillo CA, Moya-Mitjans A, Martinez-Alday J, 405
371 ances, often mimicking pacemaker malfunction. The objec- Ruiz-Granell R, Lacunza-Ruiz J, et al. Dual-chamber pacing with closed 406
372 tive of this review was to provide a window into the ECG loop stimulation in recurrent reflex vasovagal syncope. The SPAIN 407
373 patterns of normal pacemaker function and outline the range study. JACC 2017;70:1720–8. 408
[10] McGavigan AD, Mond HG. Selective site ventricular pacing. Curr Opin 409
374 of common company specific algorithms that test, correct or Cardiol 2006;21:7–14. 410
375 provide therapy. [11] Mond HG, Vohra J. The electrocardiographic footprints of Wenckebach 411
block. Heart Lung Circ 2017;26:1252–66. 412

376 Disclosures
377 Associate Professor Harry G Mond is Medical Director of
378 Cardioscan Pty Ltd.

Please cite this article in press as: Mond HG. Interpreting the Normal Pacemaker Electrocardiograph. Heart, Lung and
Circulation (2018), https://doi.org/10.1016/j.hlc.2018.06.1046