The Minnesota Code Manual of Electrocardiographic

Findings: Standards and Procedures for

Measurement and Classication
 The Minnesota Code Manual
of Electrocardiographic Findings
including measurement and comparison with the Novacode

Standards and Procedures

for
ECG Measurement
in
Epidemiologic and Clinical Trials

Second Edition
New and Enlarged

Ronald J. Prineas, MB, BS, PhD

Richard S. Crow, MD
Zhu-Ming Zhang, MD

Illustrated by Xueling Hu, M.S.

From the EPICARE, Division of Public Health Sciences,

Wake Forest University School of Medicine, and
The Division of Epidemiology, School of Public Health, University of Minnesota.
Ronald J. Prineas Richard S. Crow
Wake Forest University University of Minnesota
School of Medicine Minneapolis, MN
Winston-Salem, NC USA
USA crow@epi.umn.edu
rprineas@wfubmc.edu

Zhu-Ming Zhang
Wake Forest University
School of Medicine
Winston-Salem, NC
USA
zmzhang@wfubmc.edu

ISBN 978-1-84882-777-6 e-ISBN 978-1-84882-778-3

DOI 10.1007/978-1-84882-778-3
Springer London Dordrecht Heidelberg New York

Library of Congress Control Number: 2009937250

Springer-Verlag London Limited 2010

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright,
Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with
the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licences
issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers.
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such
names are exempt from the relevant laws and regulations and therefore free for general use.
Product liability: The publisher can give no guarantee for information about drug dosage and application thereof contained in this
book. In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

 Dedication:
To our mentors and colleagues, Henry Blackburn,
Pentti Rautaharju, and in memory of Geoffrey Rose
Contents
Preface for the second edition ............................................................................................................. ix
Preface for the rst edition .................................................................................................................. xi
Acknowledgments ............................................................................................................................... xiii

1 What is the Electrocardiogram or ECG? ............................................................................................. 1

The Electricity Part of the ECG
2 ECG Leads .......................................................................................................................................... 6
Bipolar Limb Leads(I, II, III) / Unipolar Limb Leads(aVR, aVL, aVF) / Chest Leads
(V1, V2, V3, V4, V5, V6)
3 Measuring Devices .............................................................................................................................. 10
Recording Paper Grid / Measuring Loupe / Plastic Ruler / Calibration Deection / Beats to
Be Measured / Mathematical Symbols
4 Q-QS Waves (1-Codes) ....................................................................................................................... 16
5 Frontal Plane QRS Axis (2-Codes) ..................................................................................................... 49
6 High R-Waves (3-Codes) .................................................................................................................... 55
7 ST Segment Depression (4-Codes) and Negative T-Waves (5-Codes) ............................................... 60
8 Atrioventricular (A-V) Conduction Defects (6-Codes) ....................................................................... 98
9 Intraventricular Conduction Defects (7-Codes) .................................................................................... 111
10 Arrhythmias, 8-Codes.............................................................................................................................134
11 Miscellaneous Codes (9-Codes) .......................................................................................................... 159
12 Exact Measurements ............................................................................................................................. 187
Frontal Plane QRS Axis / Amplitude Measurements / Q-X, Q-T Intervals
13 Coding the Whole ECG ....................................................................................................................... 203
Coding Hierarchy / Data Recording
14 ECG Data Acquisition Procedures and Maintenance of Recording Quality including Technician
Training ............................................................................................................................................... 206
Twelve-Lead Rest ECG Using Single Channel Recorder / Twelve-Lead ECG Using
Multichannel Recorder / Minimizing Biologic Variability
15 Criteria for Signicant Electrocardiographic Change ......................................................................... 226
16 ECG Indices That Add to Independent Prognostication for Cardiovascular Disease Outcomes ........ 263
17 Quality Control of Visual and Electronic Coding ............................................................................... 270
Appendix A Minnesota Code 2009 ....................................................................................................... 277
Q and QS Patterns / QRS Axis Deviation / High Amplitude R-Waves / ST Junction (J) and
Segment Depression / T-Wave Items / A-V Conduction Defect / Ventricular Conduction
Defect / Arrhythmias / ST Segment Elevation / Miscellaneous Items / Incompatible Codes
Appendix B Novacode and Minnesota Code Equivalents ................................................................... 287

Appendix C ..........................................................................................................................................325
Major and Minor ECG Abnormalities for Population Comparisons with Minnesota Code
and Novacode Equivalents
Index ..................................................................................................................................................... 327
vii
Preface to the Second Edition
The manual is suitable for training electrocardio-
graphers and technicians and can be accompanied
by sets of training ECGs already coded by trainers.
It is our expectation that the manual will serve as a
reference, guide, and training source for those con-
ducting studies that require objective evidence of
cardiac disease, both prevalent and incident, by non-
invasive, highly standardized, inexpensive record-
ing of the electrocardiogram. In our own ECG Read-
ing Center, this has included epidemiologic studies
among healthy populations, diabetics, psychiatric
patients, pregnant women, cohorts of patients with
clinical heart disease, populations exposed to envi-
ronmental contaminants such as arsenic, populations
exposed to Chagas disease, and in clinical trials of
HIV-infected participants, diabetics, hypertensives,
children, the aged, dietary intervention studies and
phase I and phase II drug studies.
It is 28 years since the publication of the rst
edition, which is now out of print. We have produced
a second edition because, in the interim, we have
received continuous requests over the years for
copies of the book that no longer existed and also
because there have been renements and extensions
to the Minnesota Code that allow a greater range
of abnormalities to be coded; there are even clearer
means of demonstrating correct and standardized
methods of measurements that are incorporated
into this second edition; some minor coding rules
have been changed; and now the use of the code
has been greatly expanded and is used in countless
epidemiologic studies and clinical trials worldwide.
Even as far back as 1981 the initial publication
describing the Minnesota Code was chosen as a
citation classic (CC/NUMBER 51 of SCI December
21, 1981:This Weeks Citation Classic :Blackburn
H, Keys A, Simonson E, Rautaharju P & Punsar
S. The electrocardiogram in population studies:
a classication system. Circulation. 21:1160-75;
1960). It had been cited more than 405 times in
published articles. Since then the bibliography has
grown many times largerat the time of writing,
over 700 citations were listed in Pub Med. The
introduction of digital ECG recordings and analysis
has only expanded the role of the Minnesota Code
now encompassed in computer programs to analyze
digital signals transferred over phone lines or directly
on solid digital recording platforms such as CDs.
The latter notwithstanding, archival paper tracings
are continually mined for data that were collected
ix
without digital recording and that are accompanied
by other uniquely rich data. Despite my expectations
during the 1960s that such archives would cease to
be used after the introduction of digital recording,
the tide of such treasures has hardly ebbed.
The changes included in this edition arise from
more than a quarter of a century of directing central
ECG reading and research centers and collectively
60+ large and small epidemiologic studies and mul-
ticenter national and international clinical trials. The
changes include the description of a new measuring
loupe in Chap. 3, developed over the past decade, to
better serve a more efcient and a more extensive
span for measurement of relevant durations, voltages,
and deviations from the isoelectric line. In Chap. 4,
the old code 1-2-6 has been removed because of lack
of prognostic value, and for a similar reason, code
1-2-8 has been down-coded to 1-3-8 to better repre-
sent its place in the hierarchy of Q-wave abnormali-
ties. In addition, a new code 1-3-7 has been added
to extend coding of inferior myocardial infarction.
In Chap. 7, newer more precise methods of measur-
ing ST-segment and T-wave voltages are presented.
Additions to conduction defects in Chap. 9 include
measurements for and classication of the Brugada
syndrome ECG pattern (code 7-9) and fragmented
QRS (code 7-10) both of the latter codes have
been associated with sudden death. The chapter on
arrhythmias has minor modications from the rst
edition, but, notably, premature beats need no longer
be frequent by the old denition to be coded in
a standard 12 lead ECG, where the presence of any
premature beats is signicant for prediction of future
cardiovascular disease. In Chap. 11, additional codes
have been added for lead reversal (with many exam-
ples), technical quality, left atrial enlargement (code
9-6), and early repolarization (code 9-7). More de-
tailed criteria are presented in Chap. 12 for the mea-
surement of QT interval, so important in testing all
new drugs. New coding forms are presented in Chap.
13, and Chap. 14 on ECG data acquisition has been
re-written and expanded to include training of ECG
recording technicians and maintenance of recording
quality. Chap. 15 on the criteria for signicant serial
change has been developed in a much more compre-
hensive manner and has added descriptive tables and
new codes for documenting serial change myocardial
infarction. Chap. 16 is a new addition on continuous
measurements, which can be derived from a standard
12-lead ECG that have independent prognostic value
and includes description of ultrashort heart rate vari-
ability. Chap. 17 on quality control is now greatly
expanded and includes quality control directions and
documentation for both paper (visual) and electronic
ECGs. Appendix A has all of the new Minnesota
Codes incorporated. Appendix B is new and details
NC, USA
x
the criteria and classication of the Novacode, in-
cluding signicant serial change, MI diagnosis, and
comparisons with the Minnesota Code. Finally,
Appendix C lists a summary of minor and major
code abnormalities that can be used in comparisons
of subgroups in experimental studies and analyses.
Ronald J. Prineas
May 2009
Preface to the First Edition
The electrocardiogram (ECG) is mainly used in
clinical and hospital practice for diagnosis and for
prognosis. But it is also used for systematic popula-
tion studies and clinical trials in and outside hospital,
where a repeatable, valid, and quantitative method is
required for classication of ECG ndings related
to disease. Useful classication depends, in turn,
on standardized methods of acquiring the data, on
mounting (sampling), and on reading and measure-
ment of the ECG.
In systematic studies the ECG is read centrally,
unbiased by clinical information. This blinded clas-
sication provides objective criteria for individual
events, group differences, and for sequential chang-
es in individuals and groups. Measurement classi-
cation criteria and procedural rules for standardized
ECG coding were devised and published from this
laboratory and became known as the Minnesota Code
(Blackburn H, Keys A, Simonson E, Rautaharju P,
Punsar S. Circulation. 1960;21:1160). Current updated
criteria and coding rules are found in the Appendix
to this manual.
Since 1960, these criteria and coding rules have
been tested and occasionally slightly modied
to improve validity and repeatability. The rules
are nevertheless continually subject to variation in
application because of different quality of recording,
baseline trace width, characteristics of the tracing,
and the number of beats to be measured. A set of
denitions and procedural rules has evolved in this
and other laboratories to dene more precisely wave
onset and offset and wave segments.
Other factors affecting standardized ECG coding
include ECG coder training, data acquisition, patient
preparation, technician training, and quality control.
These are presented in this manual along with unam-
biguous denitions and measurement procedures.
The current Minnesota Code criteria are found in the
Appendix, in sequence from 19-codes. In the body
of the manual, separate chapters are provided on the
exact measurement of continuous ECG variables
such as frontal plane axis and heart rate, on standard
ECG acquisition and mounting, and on quality
control of coding, as well as detailed presentation of
the wave classication system.
The codes in the Appendix do not need to be
learned by rote for this manual to be used as a
training and testing tool. Early in training as ECG
ndings are recognized, the detailed code may be
referred to. It is, however, necessary to develop an
xi
efcient personal system for scanning each ECG for
all codable ndings, and to learn thoroughly how to
measure the ndings detected. While the contents
of the coding chapters of this manual need not be
mastered in one course, the manual should be used
as reference when there is doubt how to measure a
particular wave form.
The ECG measurements described here are eas-
ily applied by intelligent, trained, and dedicated
medical, technical, or lay persons. The manual can
be used by electrocardiographers or experienced in-
vestigators to teach measurement and coding of the
ECG. This laboratory has for two decades trained
ordinary university students in coding skills as
part-time workers for periods of 13 years. Nurses,
physicians, and technicians have also been success-
fully trained. Adherence to specic rules and ongo-
ing quality control allow comparisons of results from
different observers and centers at different times.
Training requires intensive instruction for a full
10 day course, followed by continual experience.
An introductory lecture on electrocardiographic
history and physiology imparts understanding of
the reasons for the measurements and codes, and
is tailored to the sophistication of the students.
It explains the current setting of ECG coding for
population comparisons and clinical trials and their
different requirements from clinical diagnosis.
Within 3 months of initial training, further testing
for accuracy and speed is carried out.
The introductory lectures also explain the recording
of 12 lead ECGs and the expected patterns for each
lead, and identify P-, Q-, R-, S- and T-waves. Coders
are taken sequentially through the coding material in
each of Chapters 3 through 12. At the conclusion of
each chapter, sample electrocardiograms are coded
for the ndings and measurements described in that
chapter. The student codings are checked by the
instructor before proceeding to the next chapter and
remedial work is assigned where needed. Specic
codes are sought in each lead separately to recognize
the range of normal patterns in each lead.
At the conclusion of instruction with the text
material and practice ECGs, a separate test packet
of approximately 20 ECGs, as described in Chap.
13, and enriched with examples of major codable
ndings, is coded for the complete ECG. Results
are checked by the instructor. Duplicate coding of
actual unknown ECGs then starts, initially, with a
new coder against a senior experienced coder for
the rst three to six months of the program. In this
period, misunderstanding of the coding rules is
discovered and corrected.
After approximately three months of on-the-job
coding, new test packets with approximately 50 ECGs
per packet are coded and tabulated according to
standard tables of repeatability (Rose G, Blackburn H,
Gillum RF, Prineas RJ. Cardiovascular Survey Methods.
Was written in Minnesota
xii
Geneva: WHO; 1982). Coding rates (speed) and test
results (accuracy) are compared among coders so that
the suitability of coders, or the need for retraining,
is determined. Test packets are available from the
Director, ECG Coding Laboratory, Laboratory of
Physiological Hygiene, School of Public Health,
Stadium Gate 27, 611 Beacon Street SE, University
of Minnesota, MN 55455, USA.
Ronald J. Prineas MB, BS, PhD
January 1982
Acknowledgements
We thank many programmers, coders, and electrocardiographers, who over the past decades, have
contributed much to the process of ECG coding and renement of rules of application, and in particular,
we acknowledge the contributions of Dr. Yabing Li and Charles Campbell for unagging dedication to
their demand for precise denition of code items for both visual and electronic ECG records. We also
thank Dr. Elsayed Soliman for his specic renements in early repolarization denitions.

xiii
1
What Is the Electrocardiogram or ECG1?
Electrocardiogram is a written record of a heart beat, while electrocardiograph is an
instrument with which it is recorded. The same is true for telegram, written record, and
telegraph, the instrument.
The abbreviation EKG is obsolete in this country. It comes from the German word, Das
Elektrokardiogramm as many early works are done in Germanic countries. The spelling and
abbreviation has been anglicized to ECG.
An Italian, Galvani, introduced in 1791 the concept that all living tissue can produce an elec-
tric current when adequately stimulated. He also showed that injured muscle generates current.
If a living nerve attached to a healthy frog leg muscle were allowed to touch the injured area
of another frogs muscle, the healthy nervemuscle frog preparation would twitch!
The Germans, Kolliker and Muller, showed, over a hundred years ago, with the same type
of nervemuscle preparation, that an electrical current was produced rhythmically with
each contraction of an animal heart.
An Englishman, Waller, in 1887 was the rst to demonstrate in his pet bulldog Jimmie that
the electrical action of the heart could be registered from the surface of the body. He is also
credited with making the rst human ECG, which he called a cardiogram. This was regis-
tered by light reected from a capillary tube of mercury, which oscillated with each heart
beat from the electrical potential differences the heartbeat causes between the right and left
hand.
Einthoven, a Dutchman, worked for some years at Leiden with ECG recording instru-
ments and nally in 1901 devised his own instrument a string galvanometer, so sensitive to
changes in intensity of electric currents and so rugged and stable in operation that a new
branch of medicine, electrocardiography, was made possible. His instrument consisted of
a stretched string of quartz lament coated with silver and suspended in a strong magnetic
eld. The secret of his success was this high-resistance tiny quartz string, which he rst
made by attaching fused quartz to the tail of an arrow, heating the quartz to a critical point
and ring the arrow, thus producing a very ne, uniform string.
The history of electrocardiography thereafter is too detailed to recount, but Einthovens
instrument brought it all about. It was early imported in Britain and the United States
and widely used here until after the Second World War. The tiny heart currents are now
picked up and amplied with transistorized ampliers, and instead of photographs of a
vibrating string we have an instant ECG, from a direct writing stylus on a moving paper
strip. In addition, we now transmit the ECG by FM radio, or by telephone. We can also re-
cord and store the ECG on digital recording media, and can now make ECG measurements
and even Minnesota Code classications with computers by converting the analog wave
form to digital, or numerical form. (see www.ecglibrary.com/ecghist.html.)

1
The development of the ECG wave generation and propagation is not meant to be comprehensive but as a guide to expected wave
patterns for coding. More comprehensive descriptions can be found in text books of clinical ECG interpretation.

1
The Electricity Part of the ECG
Heart is made up of many interwoven and interconnected bundles of muscle. Each indi-
vidual muscle cell has an electrical charge as we learned from Galvani and others. With
each heart beat, a wave of electrical excitement moves rapidly through the thousands of
linked heart muscle cells. There is at that moment an imbalance of the electrical charge at
the outer membrane of these cells caused by a rapid ux of charged ions through the cell
membrane. As the wave of excitation passes through the heart muscle, millions of indi-
vidual cell charge set up an electrical current in the chest. This current ows to the surface,
and at the skin produces differences in electrical voltage, which can be measured between
pairs of electrodes placed at any two points on the body.
At the beginning of each heart beat, excitation starts from the ring of the pacemaker sinus
node in the right atrium and passes in as wave through both atria, the upper thin-walled
chambers of the heart. The tiny differences in voltage between distant points on the skin
allow us to register a small deection on a meter, a galvanometer, named after Galvani.
Because the paper is moving at the time the meter needle deects, a little rounded wave is
produced. Einthoven named it the P-wave. This is followed by a delay as the impulse is re-
ceived at the upper part of the ventricular septum, in the A-V node, and this lag is recorded
on paper as a straight line.
The exciting electrical wave then spreads rapidly through the large muscle walls of the
ventricles over the special (His) bundle of conducting bers. The ventricular excitation
causes sharp and large deections (still only 1-4 mv), and these deections as registered on
the moving paper are called the QRS waves in the ECG.

Summary Fig. 1-1

SA
NODE R

LEFT
ATRIUM

RIGHT T
ATRIUM
P

A-V
LEFT
NODE
RIGHT VENTRICLE Q S
VENTRICLE
LEFT
BUNDLE

RIGHT
BUNDLE

FIGURE 1.1. Normal heart beat is initiated by spontaneous ring of the sinus (SA) node in the right
upper chamber of the heart (right atrium)

2
 As the excitation wave goes through the ventricles, the activated cardiac muscle contracts
(excitationcontraction coupling) and ejects the blood into the systemic and pulmonary cir-
culation. Then the electrical charges at the muscle cell return in a slowly receding wave to
the original resting, electrical state. This slower wave of electrical recovery of the ventricu-
lar muscles is inscribed on the moving paper as another rounded wave called the T-wave.
The shape and direction of the QRS and T-waves depend on the sequence of depolariza-
tion and repolarization, the balance and direction of the individual electrical forces of the
wave of excitement through the heart, and the location of the electrodes on the skin. For
illustration, consider two poles in water with a pressure gauge on each. One hooks up the

R Fig. 1-2

T
P

Q S

FIGURE 1.2. The excitation wave passes through the muscles of both atria, activating them to con-
tract. This activation produces electrical currents in the chest, which are measured as
differences of potential between the electrodes on the body surface. A moving paper
strip records these as the small rounded P-wave of the normal sinus beat

R
Fig. 1-3

T
P

P-R S
Q

FIGURE 1.3. The electrical wave of activation reaches the atrioventricular (A-V) node between the
atria and ventricles and there is a brief delay. The P-R interval includes the P-wave and
the period of delay in the A-V node
3
two pressure gauges so that the meter reads positive if the righthand pole registers a high-
er pressure than does the left. Scooping up a wave in the middle and shoving it toward this
right-hand pole makes the pressure higher there and the meter registers an upward deec-
tion. A wave toward the left pole makes the pressure lower at the righthand pole and the
meter, and therefore, registers a downward deection. If a wave starts in the middle and its
force travels equally toward and reaches each pole at the same time, we get no difference
in pressure, and hence no deection at all.

R
Fig. 1-4

T
P

Q S

FIGURE 1.4. The activation rapidly descends the bundle of His in the muscular septum between the
two ventricles, and activates those muscles from left to right. This septal activation
produces the rst ventricular deection of the ECG, the Q-wave

R
Fig. 1-5

T
P

Q S

FIGURE 1.5. The activation then spreads rapidly through the special conducting tissues of the ven-
tricles and the wave progresses, in a generally right to left direction, producing the
major ventricular deection, the R-wave. All regions of the ventricles are eventually
activated, the entire QRS complex is recorded, and the ventricles contract and pump
blood

4
 R
Fig. 1-6

T
P

Q S
ST
SEGMENT

FIGURE 1.6. There follows a short period of relative inactivity recorded as the ST segment

R
Fig. 1-7

T
P

Q S

R
Fig. 1-8

T
P

Q S

FIGURES 1.7 and 1.8. Then the recovery wave spreads in reverse of depolarization (from the epi-
cardium through the ventricular wall) over the same pathway repolarizing
the heart, producing a broad blunt wave, the T-wave that is in the same direc-
tion as the R wave
5
2
ECG Leads
The body acts as a large conductor of electrical currents generated by heart. To record, these
currents require that only any two points on the body be connected to the electrocardiograph.
This establishes the necessary completion of an electrical circuit and is done by means of
electrodes attached to the limbs or the chest, each pair of attachments being one lead. The
ECG leads generally used are I, II, III, aVR, aVL, aVF, V1, V2, V3, V4, V5, and V6.

Bipolar Limb Leads (I, II, III)

The major direction of the electrical force wave through the heart ventricles goes from the
right to left in a downward direction. Consequently, if we attach the ECG electrodes on the
arms with the positive pole of the galvanometer as the left arm, then as the excitation wave
approaches it, there is a positive or upright reection. Actually, the forces from instant to
instant form a loop in space, initially and briey toward the right arm, giving the small Q,
then sweeping in a broad orbit toward the left arm, giving the R, and back to the center
giving the small S.

FIGURE 2.1.
6
 The right arm/left arm lead is lead I and is usually registered as a predominantly upward
wave because the average and major direction of the wave force is toward the positive left
arm (see Fig. 2.1).
The voltage difference between the right arm and the left leg electrodes is measured by lead
II. The major direction of the electrical force wave goes parallel, or almost so, to this lead,
from above downward, and so the ECG again registers upright, as a mainly positive wave.
In lead III, potential differences are reected between the left hand and left leg. Here the
average major force of the wave rolls over the line of the lead at right angles, and we get a
low, absent, or approximately equal positivenegative wave (see Fig. 2.1).
Einthoven devised this triangle (Fig. 2.2) (right arm, left arm, left leg) and calculated
that if we record these three leads at exactly the same time, the height of waves in I and
III always adds up to those in lead II. He taught us how to calculate the direction of the
major wave force from the voltage values in any two... of these... leads. This is called
measuring the electrical axis, which we take up in Chap. 5. The predominant deection
of the QRS waves usually points upward, is positive in I and II, and may be up, down, or
in-between in III.

Fig. 2-2

FIGURE 2.2.

Unipolar Limb Leads (aVR, aVL, aVF)

The potential differences in the right arm, left arm and left leg are also recorded between
an electrode from each of these sites and a neutral or zero potential by connections from all
limb electrodes within the electrocardiograph.
The unipolar leads then reect potential values from the right arm (aVR), left arm (aVL),
and left leg (aVF) and are useful in determining the electrical position of the heart.
The ECG waves in aVR are generally negative or downward deections; those in aVL
and aVF may be upright or of intermediate position depending on the anatomic and
electrical position of the heart (see Fig. 2.3).

7
FIGURE 2.3.

Chest Leads (V1, V2, V3, V4, V5, V6)

We owe to Drs. Wilson and Johnston from the University of Michigan at Ann Arbor the
development of chest leads. Much of our information about heart attacks and other heart
muscle problems is obtained from these six leads that have been widely used for the last 60+
years. They are often called Wilson leads or V leads or precordial leads, and are named V1,
V2, V3, V4, V5, V6. Six is standard but more may be taken. The chest leads are also unipolar
leads, reecting potential differences between six points on the chest and a combined poten-
tials lead inside the electrocardiograph from the three extremity electrodes.
For reference purposes we will dene these positions here, but it requires practice to
locate the landmarks on a real chest:
Subscript 1 (V1) shall be used for a lead from the right sternal margin at the fourth
intercostal space; subscript 2 (V2) for a lead from the left sternal margin at the fourth
intercostal space; subscript 3 (V3) for a lead midway between 2 (V2) and 4 (V4);
subscript 4 (V4) for a lead from the fth intercostal space where it is crossed by the
midclavicular line; subscript 5 (V5) for a lead from the junction of the left anterior
axillary, line with the horizontal position of position 4; subscript 6 (V6) for a lead on
the same horizontal level but at the left midaxillary line (see also Chap. 14).
As we look at the body and the heart from the front, or in the frontal plane, we nd that
the major wave force is directed from bodys right to left and down, which accounts for the
direction of deection in frontal plane limb leads I, II, and III. For the chest leads, we look
at the heart and body from above, i.e., at the horizontally oriented plane. We nd that the
major QRS force is directed to the left and somewhat toward the back.
8
 Each of these chest electrodes register positive when the major wave force sweeps toward
it. The main wave force is largely away from the positive electrode at V1 and V2, and so
those leads will register predominantly downward deections. It is toward V5 and V6 so
they will register predominantly upward deections, while V3 and V4 will be somewhere
in-between or equiphasic. In detail, the loop starts out toward V1 and away from V6, regis-
tering a small R in V1 and Q in V6. The broad mass of the loop then creates the main force
described above, away from V1 and V2, toward V5 and V6 (see Fig. 2.4).

FIGURE 2.4.

This is the pattern that should be remembered now, usually negative QRS waves in V1
and V2 and positive waves in V5 and V6 with a transitional zone in between. One should be
alerted if the wave directions are different from this pattern.
The T-wave recovery force also moves in a slower and smaller loop in space and generally
follows the orientation of the QRS forces. With some small exceptions that will be learned,
the direction of the T-wave in the limb and chest leads is in the same direction as that of
the predominant QRS wave. One should quickly detect whether the T-wave is opposite in
direction to the main direction of the QRS wave.

9
3
Measuring Devices
The amplitude (distance of positive peaks and negative nadirs from the baseline) and dura-
tion (width from beginning or onset to end or offset) of ECG waves are measured by visual
reference to the grid lines on the ECG recording paper, or by use of devices including a
magnied measuring loupe or a clear plastic ruler on calipers. Use of such devices has been
demonstrated to improve coding repeatability.

Recording Paper Grid

ECG recording paper is divided into a grid of heavier lines 5 mm apart and lighter lines 1
mm apart. When, as in the majority, ECGs are recorded at a paper speed of 25 mm/second,
this means that each millimeter mark on the horizontal axis of the grid represents 1/25
second (0.04 second), 0.25 mm represents 0.04/4 = 0.01 second; 0.5mm = 0.04/2 = 0.02
second; and 0.75 mm = 3 0.04/4 = 0.03 second (see Fig. 3.1). Amplitude of waves
and points of wave onset and offset are measured in millimeter deviations from the
baseline (see Fig. 3.2). For durations, (onset to offset 1 mm = 40 ms, and for amplitude
1 mm = 100 V = 0.1 mV.

1mm.

5mm.

a a = 0.25 mm.= 0.01 sec. = 10 ms

b b = 0.5 mm.= 0.02 sec. = 20 ms
c c = 1 mm.= 0.04 sec. = 40 ms

Fig. 3-1

FIGURE 3.1.

10
Measuring Loupe
For small waves and for wave duration obscured by a grid line, a magnied measur-
ing device must be used. The loupe used in this ECG reading center has improved the
repeatability of measurements between different coders and on different occasions. This
anastigmatic Loupe 10 has high resolving power and wide visual eld, with a grid pro-
tected by a coverplate on the bottom (see Fig. 3.4). It has three precision-constructed ach-
romatic lenses (to enable observers to simultaneously inspect the whole picture area, i.e.,
at objects less than 32 mm in diameter) and with specially designed scaled reticle for
ECG measurement. It is placed at on the tracing in a good light, positioned in front of
the coder. The top lens system may be focused by a screw. With 20 mm effective aperture
of the loupe, an observer can inspect the whole image eld by merely moving his/her
eyeball, without moving his/her face. This is an advantageous feature for a QT interval
measurement. The loupe with special reticle may be obtained from our ECG Reading Cen-
ter, at cost, by writing to the Director, Epidemiological Cardiology Research Center (EPI-
CARE), Department of Epidemiology and Prevention, Division of Public Health Sciences,
Wake Forest University School of Medicine, 2000 West First Street Suite 505, Winston-
Salem, NC 27104, USA.

FIGURE 3.4. Fig. 3-4

ALE LUPE
SC

3
Lenses

10
8
6

0 1 2 3 4 5
10
9
8
7
6
5
4
3
2
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16MM

1
2
3
SEC 25MM 0.04 .12 .22
4
.44
SEC 50MM 0.02 0.04 .22
1 2 3 5 5 11
MM
6
7
8

9
10

12
Plastic Ruler

For measuring longer wave durations and intervals and higher waves, a clear exible plas-
tic ruler with 1 mm interval is useful. A transparent ruler enables the coder to see the ECG
trace beneath so that it can be positioned most accurately. The thinner the ruler, the less
will be parallax error.

Calibration Deection

The rst deection usually seen in the ECG is a square calibration wave (see Fig. 3.5).
This should be exactly 10 mm high from the top margin of the baseline to the top of the
square wave.

Fig. 3-5

calibration wave

FIGURE 3.5.

Beats to Be Measured
The rst beat in a lead is dened as a beat with a complete P-wave, QRS complex, and
T-wave. If part of the P-wave is missing, that beat is not included for coding measurements
(see Fig. 3.6). The last beat in a lead to be included for coding measurement must include
the T-wave, at least to its peak (see Fig. 3.7).
13
 1st complete codable beat
Fig. 3-6

1st complete codable beat

FIGURE 3.6.

Fig. 3-7

last complete codable beat

last complete codable beat

FIGURE 3.7.
14
Mathematical Symbols

To conserve space and to make precise denitions, mathematical symbols have been used
throughout this text for the following:

> = greater than

< = less than
= equal to or greater than
= equal to or less than

So that, 0.06 second < Q duration < 0.07 s, means a Q-wave duration greater than 0.06
second but less than 0.07 second.

Differences in Measurement between Visual and Electronic Measurements

There are times when checks (over-reading) need to be made for visual conrmation of com-
puter coding of digital ECG data. At such times, it is important to recognize that there are dif-
ferences in measurement precision between the two modes of coding. First, most electronic
programs use either an average or median beat, whereas visual coding generally requires
accepting the ndings in the majority of beats. Second, the electronic measurement starts
from an isolelectric line of virtually no width, whereas the paper record has to contend with an
isolelectric line (baseline) of nite width. The electronic signals at the time of publication can
measure at a sampling rate of 500 second, so that the electronic measurements applied in com-
puter coding are often a shade longer in duration when measuring specic intervals and a shade
greater in measuring voltage deviations from the isoelectric line.1

Reference

1. Rautaharju PM, Seale D, Prineas RJ, Wolf H, Crow R, Warren J. Changing electrocardio-
graphic recording technology and diagnostic accuracy of myocardial infarction criteria:
improved standards for evaluation of ECG measurement precision. J Electrocardiography.
1978;11(4):322-330.

15
4
Q-QS Waves (1-Codes)
Injured regions of the heart may become electrically inactive. Myocardial infarction is the
most frequent cause of this. The normal excitation wave may be altered by this nonfunctioning
part of the heart, thus changing the appearance of the QRS complex. In this situation, the
early part of the QRS complex appears as a deep, wide negative Q- or QS-wave in certain
leads. Smaller areas of injury cause lesser Q-waves.
Ideally, one would measure the amplitude and duration of all Q-waves and refer to standard
values for classication. Practically, this is too tedious for visual-manual coding. Instead, the
code provides classes that generally reect degrees of Q-QS abnormality according to lead.

R
Fig. 4-1

R
Fig. 4-2

FIGURES 4.1 and 4.2. The 1-codes classify Q- and QS-waves, which also depend on the type of
R-wave present. The earliest positive deection in a QRS complex is the
R-wave. Any negative deflection that precedes the R-wave is a Q-wave
(see Fig. 4.1). Any negative deection that follows the R-wave is an S-wave
(see Fig. 4.2)

16
FIGURE 4.3. A Q-wave and an S-wave may be present in the same complex

FIGURE 4.4. There may be no Q- or S-wave, in which case the QRS complex would consist only
of an R-wave

17
FIGURE 4.5. If there is no R-wave, then by denition there can be no Q-wave (because a Q is the
rst negative wave to precede an R-wave and an S the rst to follow). The whole QRS
complex is negative and is called a QS-wave

FIGURE 4.6. A special form of the QS-wave is a W pattern. Here the negative QRS complex is
notched with a central deection. However, the peak fails to reach the reference baseline
(the upper margin of the baseline at the onset of QRS) and the W pattern is classied
as a QS-wave
18
 Fig. 4-7

Q wave amplitude

measure vertically
from the onset of
the QRS at the lower
margin of the baseline
to nadir

Fig. 4-8

QS wave amplitude

measure vertically
from the onset of
the QRS at the lower
margin of the baseline
to nadir

FIGURES 4.7 and 4.8. The presence or absence of codable Q- or QS-waves depends on the amplitude
of the Q- or QS-wave, which must be 1 mm in the majority of beats in any lead
(with two exceptions for codes 7-7 and 7-8, see Chap. 9), the duration of the
Q-wave, which must be 0.02 second in the majority of beats in any lead,
the amplitude of the accompanying R-wave, and the lead location of the Q- or
QS-wave

19
 Fig. 4-9

no codable Q

codable QS

FIGURE 4.9. If the amplitude of the Q- or QS-wave is <1 mm in 50% or more of beats in a lead, it
is not codable

Fig. 4-10

Q begins at
this point

FIGURE 4.10. The duration of the Q-wave is measured horizontally from the rst sharp downward
deection from the baseline to the point that intersects the ascending trace of the
R-wave

20
 Fig. 4-11

actual Q wave (ab)

a b

c d

theoretical Q wave (cd)

FIGURE 4.11. The baseline trace of the P- and T-wave is often much thicker than the trace of the
QRS complex, so that the Q-wave duration is practically measured from the upper
margin of the baseline at the beginning of the QRS complex horizontally to a point
that intersects the inside edge of the ascending limb of the R-wave

Fig. 4-12a

Q begins at
this point

21
 Fig. 4-12b

Q begins at
this point

FIGURE 4.12a,b. At the onset of Q, in the presence of a thick baseline trace, a stepped shoulder
often precedes the steeper descent of the Q-wave. This shoulder is included in the
measurement of the Q-wave and illustrated in (a) or there may be smaller negative
deections before the main negative Q. These initial negative deections count as
the start of the Q-wave and are illustrated in (b)

Fig. 4-13

Q duration

measure on the top of the

baseline from the beginning
of the Q to where the Q
intersects the scale

0 1 2

FIGURES 4.134.15. The use of a loupe to measure Q-wave duration

22
 Fig. 4-14

0.05 sec. < Q duration < 0.06 sec.

Fig. 4-15

0.03 sec. < Q duration < 0.04 sec.

Q-waves to be coded must be 0.02s duration in >50% (a majority) of beats. It is not

necessary to measure the duration of QS-waves for coding
23
 Fig. 4-16

measure vertically
from the onset of
the QRS to the
peak of the R

Fig. 4-17

measure vertically
R
from the onset of
the QRS to the
peak of the R

P T

FIGURES 4.16 and 4.17. R-wave amplitude determines the presence or absence of Q- or QS-waves.
R-wave amplitude is measured from the upper margin of the P-R baseline
at the onset of the QRS complex vertically to the peak of the R-wave (see
Fig. 4.16), even if the R-wave is preceded by a Q-wave (see Fig. 4.17)

24
 Fig. 4-18
initial R

T
R
P

FIGURE 4.18. R-wave amplitude determines if a Q- or QS-wave is present. If there is no negative

wave of 1 mm amplitude preceding the R-wave, then the R-wave is called an initial
R-wave

Fig. 4-19
initial R

P T
R

Fig. 4-20
no initial R T
P

QS

FIGURES 4.19 and 4.20. An initial R-wave must be 0.25 mm in amplitude to be classied

25
 Fig. 4-21

QS

initial R<0.25 mm.

QS

FIGURE 4.21. If an initial positive QRS deection does not reach 0.25 mm amplitude and no other
R-waves occur in the complex, a QS-wave is coded

Fig. 4-22

initial R

a amplitude 0.25 mm.

b reaches peak 0.02 sec.

FIGURE 4.22. Small initial R-waves (0.250.5 mm) are difcult to discern when the baseline is
wide. For this reason, we try to dene wave sharpness; they need to reach their
peak within 0.02 second of the onset of QRS

26
 Fig. 4-23
QS pattern

QS pattern Fig. 4-24

R R

QS
initial R0.25 mm Fig. 4-25
R peak duration >0.02 sec.

R R

FIGURES 4.234.25. A QS pattern is considered to be present when the R-waves present are
<0.25 mm (see Figs. 4.23 and 4.24) or reach their peak in >0.02 second (0.5
mm) (see Fig. 4.25). If the initial R is > 0.5 mm, then there is no restriction
on the R peak time

27
FIGURE 4.26. An initial R-wave is an exception to the majority rule. If an initial r is present in any
beat in a lead, no q- or qs-code is made for that lead except in V1. That is, in V1, an
initial R must be present in the majority of beats to be classied an RS pattern. In any
other lead, no 1-code is made if there is a single beat having an initial R-wave, even
if the majority of beats have Q- or QS-waves

28
 Fig. 4-27
code as QS

terminal R>1mm. R<1mm. R<1mm.

Fig. 4-28

code as QR

terminal R>1mm. R<1mm. terminal R>1mm.

FIGURES 4.27 and 4.28. Any positive QRS deection that follows an initial negative deection of
1 mm amplitude is classied a terminal R. If a terminal R is <1 mm in
amplitude and no other R-waves are present, a QS pattern is coded instead
of a QR pattern. A terminal R must be 1 mm in amplitude in the majority
of beats in a lead

29
 Fig. 4-29
QS pattern

a>1mm. b

b=0.04 sec. c

c<0.25 mm. b

a c

FIGURE 4.29. Sometimes the QRS complex ends in an elevated ST segment. With a wide baseline
trace, it may be difcult to decide whether a terminal R is present. Therefore, a limit
is placed on the peak time required. The peak of the terminal R-wave must 1 mm
in amplitude, and fall by at least 0.25 mm or more within 0.04s (1 mm)

Q waves Fig. 4-30

amplitude must be 1mm.

duration must be 0.02 sec. in the PLUS

majority no initial R
the terminal R must be 1mm. of beats wave in any
beat in a lead
the peak of the R must fall 0.25mm. except V1
in 0.04 sec.

FIGURE 4.30. Required Q-wave characteristics are summarized

30
 Fig. 4-31

W pattern QS W pattern QRS

R < 1 mm. R 1 mm.

Q 1 mm. Q 1 mm.

FIGURE 4.31. If the central peak of the W (with an initial negative deection 1 mm) does not rise
to 1 mm above the upper margin of the P-R baseline, then the W pattern is classied
as a QS. If the peak of the W rises to 1 mm or more above the baseline, the pattern
is classied as a QRS

Fig. 4-32

W pattern RS W pattern QS

Q < 1 mm. Q < 1 mm.

R 0.25 mm. R < 0.25 mm.

a a
b b

FIGURE 4.32. If the initial negative deection of the QRS is less than 1 mm and if a W pattern is
present having a peak rising above the baseline, that positive peak is regarded as an
initial R-wave. Therefore, it has to be only 0.25mm in amplitude and reach its peak
in 0.02 second (measured from the isoelectric line after the initial negative deec-
tion) to be classied as an initial R
31
 Fig. 4-33
W Pattern QS

Fig. 4-34
W Pattern QS

QS initial R < 0.25 mm. terminal R < 1 mm.

Fig. 4-35

FIGURES 4.334.35. Other forms of QS-waves

32
 no QS waves Fig. 4-36

negative component terminal R 1 mm. initial R 0.25 mm.

< 1mm.
Q wave no Q or QS
no Q or QS

FIGURE 4.36. shows other forms of W patterns that are not codable as QS-waves

Fig. 4-37
Q/R ratio

R peak = 10 mm.

Q nadir = 5 mm.

Q 5 1
= =
R 10 2

FIGURE 4.37. The ratio of the Q- to R-wave amplitude is also important in coding. Remember that
the amplitude of the R-wave is measured from the upper margin of the preceding P-R
baseline, and the Q-wave amplitude is measured from the lower margin

33
 Fig. 4-38

1-1-1 R=15 mm

lead I, II, V2-V6

Q duration 0.03 sec.

Q/R ratio 1/3

Q=5mm

Fig. 4-39

no 1-code
lead aVR

FIGURES 4.38 and 4.39. Q- or QS-waves are never coded in lead aVR. The Q-QS wave code
depends on the leads in which they occur, the duration, the amplitude of
the Q-wave, and the Q/R ratio. To qualify as Code 1-1-1, the Q-wave must
be 0.03 second duration in the majority of beats, plus a Q/R ratio 1/3 in
majority of beats, in any of leads I, II, V2-V6

34
 Fig. 4-40

1-1-2 1-1-1

lead I, II, V1-V6 lead II

Q 2 1 Q 2 1 1
= = = = >
R 12 6 R 4 2 3

FIGURE 4.40. Code 1-1-2 requires a Q 0.04 second duration in the majority of beats in any of
leads I, II, V1-V6. Note that if the Q/R ratio is 1/3 this would be coded 1-1-1 (e.g.,
illustrated by lead II). This is also the only Q-code applicable in lead V1. It is important
to remember the special rules for classication of an initial and terminal R in V1

Fig. 4-41

1-1-3
lead aVL
Q duration 0.04 sec.
Q amplitude 1 mm.
R amplitude 3 mm.

FIGURE 4.41. Code 1-1-3 in lead aVL requires a Q 0.04 second duration plus R amplitude 3 mm
in the majority of beats. Because of this special R amplitude requirement in a aVL, it
is particularly important to check the calibration pulse before assigning this code
35
 1-1-4 Fig. 4-42

lead III
Q duration 0.05 sec.
Q amplitude 1 mm.

PLUS

lead aVF
Q amplitude 1 mm.

FIGURE 4.42. Code 1-1-4 depends on Q-waves in III and aVF considered together. The Q duration
in III must be 0.05 second in the majority of recorded beats, plus a Q-wave
amplitude 1 mm in the majority of beats in aVF. The Q-waves in aVF do not need
to be 0.02 second duration for this code

Fig. 4-43

1-1-5
lead aVF
Q duration 0.05 sec.
Q amplitude 1 mm.

FIGURE 4.43. Code 1-1-5 in lead aVF requires a Q-wave of 0.05 second duration in the majority
of beats. If Q-waves are 0.05 second duration in the majority of beats of both III
and aVF, code 1-1-4
36
 Fig. 4-44
1-1-6

V2 V3 V4 V5

or or or

V3 V4 V5 V6

FIGURE 4.44. Code 1-1-6 requires ndings in two chest leads. A QS pattern must be present in
majority of beats in any of leads V3-V6 when an initial R is present in any beat in
the immediately preceding chest lead (lead adjacent to the right on the chest). For
example, an initial R in V3 and a QS in V4

37
 Fig. 4-45

1-1-6

V1

V2

Fig. 4-46
no 1-1-6

V1

V2

FIGURES 4.454.48. A code 1-1-6 between V1 and V2 requires that the initial R 0.25 mm in V1 be
present in the majority of beats (see Figs 4.454.47). A code 1-1-6 between
other chest leads requires only that the initial R 0.25 mm in the rst lead be
present in at least one beat (see Fig. 4.48)

38
 Fig. 4-47
no 1-1-6

V1

V2

Fig. 4-48

1-1-6

V3

V4

39
 1-1-7 Fig. 4-49

V1

V2

V3

V4

FIGURE 4.49. Code 1-1-7 requires a QS pattern in the majority of beats of all of leads V1, V2, V3,
and V4. If V5 or V6 also has a QS, still code 1-1-7

40
 Fig. 4-50

1-2-1
lead I, II, V2-V6
Q
ratio 1/3
R
0.02 sec. Q duration < 0.03 sec.
Q amplitude 1 mm.

FIGURE 4.50. Code 1-2-1 is a lesser Q-QS code in leads I, II, V2-V6. That is, the Q/R ratio is re-
quired to be 1/3 in the majority of beats and the Q amplitude 1 mm, has to Q
duration has to be 0.02 second and < 0.03 second in the majority of beats

Fig. 4-51

1-2-2
lead I, II, V2-V6
Q
ratio < 1/3
R
0.03 sec. Q duration < 0.04 sec.

FIGURE 4.51. Code 1-2-2 is a lesser form of a 1-1-2 in leads I, II, V2-V6. It requires a Q duration
0.03 second and < 0.04 second in the majority of beats and Q/R ratio < 1/3 in the
majority of beats
41
 Fig. 4-52

1-2-3
QS pattern in lead II

Fig. 4-53

1-2-3
QS pattern in lead I

terminal R<1 mm. initial R<0.25 mm.

FIGURES 4.52 and 4.53. Code 1-2-3 requires a QS pattern in the majority of beats of lead I or
lead II

42
 1-2-4 Fig. 4-54

lead III

0.04 sec. Q duration < 0.05 sec.

Q amplitude 1mm.

lead aVF

Q amplitude 1mm.

Q duration < 0.05 sec.

FIGURE 4.54. Code 1-2-4 is a lesser form of 1-1-4 in leads III and aVF together. The Q duration in
lead III must be 0.04 and < 0.05 second in the majority of beats, and there must be
a Q-wave amplitude 1 mm in the majority of beats of aVF. The Q-waves in aVF do
not need to be 0.02 second duration for this code

1-2-5 Fig. 4-55

lead aVF

0.04 sec. Q duration < 0.05 sec.

Q amplitude 1mm.

1-2-4 Fig. 4-56

lead III

0.04 sec. Q duration < 0.05 sec.

lead aVF

Q amplitude 1mm.

FIGURES 4.55 and 4.56. Code 1-2-5 is a lesser form of 1-1-5 in aVF with Q duration 0.04 but
< 0.05 second in the majority of beats (see Fig. 4.55). Again, if Q-waves
of amplitude 1 mm and 0.04 and < 0.05s duration in the majority of
beats of both III and aVF, code 1-2-4 (see Fig. 4.56).
43
 Fig. 4-57
1-2-7

V1

V2

V3

V4

FIGURE 4.57. Code 1-2-71 is a lesser form of 1-1-7. It requires a QS pattern in the majority of leads
V1 plus V2 plus V3

1
1-2-6 code from past editions is now removed and the previous 1-2-8 is now 1-3-8, but the same codes have been used
for all other codes in previous editions.

44
 Fig. 4-58

1-3-1
lead I, II, V2-V6
0.02 sec. Q duration < 0.03 sec.
1 Q 1
>
3 R 5
Q amplitude 1mm.

Q 2 1
= =
R 8 4

FIGURE 4.58. Code 1-3-1 is a lesser form of 1-1-1 and 1-2-1. If the Q/R ratio is < 1/3, but 1/5 in
the majority of beats and the Q-wave is 0.02 but < 0.03 second duration in the ma-
jority of beats in any of leads I, II, V2-V6, then a 1-3-1 code is present

Fig. 4-59

1-3-2

V1

V2

V3

FIGURE 4.59. Code 1-3-2 is a lesser form of 1-1-7 and 1-2-7. Here, QS-waves are required in
the majority of beats in V1 and V2, and there must be no initial R-waves in any beats
in lead V2

45
 Fig. 4-60

1-3-3
lead aVL
0.03 sec. Q duratoin < 0.04 sec.
R 3 mm.
in the majority of beats

FIGURE 4.60. Code 1-3-3 is a lesser form of 1-1-3. It requires a Q-wave 0.03 but < 0.04
second duration in the majority of beats plus an R amplitude 3 mm in the
majority of beats of lead aVL

Fig. 4-61

1-3-4

III

aVF

FIGURE 4.61. Code 1-3-4 is a lesser form of 1-1-4 and 1-2-4. It also requires a Q-wave amplitude
1 mm in the majority of beats of aVF plus a Q-wave 0.03 second but < 0.04 second
duration in the majority of beats of lead III. The Q-waves in lead aVF do not need to
be 0.02 second duration for this code
46
 Fig. 4-62

1-3-5
lead aVF
0.03 sec. Q duration < 0.04 sec.
Q amplitude 1mm.

FIGURE 4.62. Code 1-3-5 is a lesser form of 1-1-5 and 1-2-5. It requires a Q-wave of amplitude 1
mm and 0.03 but < 0.04 second duration in the majority of beats of lead aVF. If the
conditions of 1-3-4 are met, then 1-3-4 and not 1-3-5 is coded

Fig. 4-63
1-3-6
QS pattern in lead III and aVF

III

aVF

FIGURE 4.63. Code 1-3-6 is the only 1-code for QS waves in leads III and aVF. It requires
QS-waves in the majority of beats of leads III and aVF. And note the initial R in III
is < 0.25 mm
47
FIGURE 4.64. Code 1-3-7 is the only 1-Code for QS waves in lead aVF alone. It requires QS-waves
in the majority of beats of leads aVF and no QS in lead III

FIGURE 4.65. Code 1-3-8 is a lesser form of 1-1-6. It is also an exception to the majority rule.
Instead of a QS pattern being required to the left (on the chest) of a chest lead with
an initial R-wave, this code requires that the initial R-waves in the lead to the left be
smaller than those in the lead to the right. Here, the R-wave amplitude in all beats
in the lead to the right must be > 2 mm and in all beats in the lead to the left must be
2 mm. Such a pattern is not codable between V1 and V2
48
5
Frontal Plane QRS Axis (2-Codes)
The main electrical forces of the QRS ventricular wave go in a leftward and downward
direction when looking at the body face on. The measurement of angle of this direction,
projected to the frontal plane, is useful because it may change with age and with size
and condition of the heart muscle. If the detailed axis, as described in Chap. 12, is not
measured, axis patterns are coded.

incomplete Fig. 5-1 incomplete

beat beat

next to last last complete,

complete, normal beat
normal
beat

Fig. 5-2

next to last last abnormal incomplete

complete, complete, beat beat
normal normal
beat beat

FIGURES 5.1 and 5.2. All QRS axis measurements are taken on the next to last complete normal
beat in leads I, II, and III dened as that normal beat which immediately
precedes the last normal codable beat. If there are only two beats in a lead,
the measurements should be taken on the last complete normal beat

49
 2-1 Fig. 5-3

I
8

-3

II 2

-7

III 2

-9

FIGURE 5.3. Code 2-1 is made for left axis deviation more negative than 30. The sum of the amplitude
of the principal positive wave (R) and the principal negative wave ( Q or S) must be
positive in lead I, zero or negative in lead II, and negative in lead III. To determine
the amplitude of the main positive wave (whether R or R), measure from the upper
baseline margin at the onset of QRS to the peak of the R- or R- wave. For the amplitude
of negative waves, measure form the lower margin of the baseline at QRS onset to
the nadir of the largest wave. R is a second R-wave (Fig. 9.25). In Fig. 5.3 the value of
the positive wave in lead I is +8.0 mm and of the negative wave, 3.0 mm. The algebraic
sum is therefore +5.0 mm. Measurements of positive or negative waves are rounded
up to the nearest 0.25 mm. If the sum of the positive and negative waves is not a whole
number, it should be rounded up to the nearest 0.5 mm. For example, a sum of 5.25 mm
would become 5.5 mm; a sum of 7.75 mm would become 8.0 mm. Remember to use
the next to last complete normal beat for these measurements

50
 2-2 Fig. 5-4

I 3

-6

II

III 6

FIGURE 5.4. Code 2-2 is made for right axis deviation more positive than +110. The sum of the
positive and negative waves must be negative in lead I and positive in lead III.
In addition, the absolute value of lead I must be at least half of the absolute value in
lead III

51
 2-3 Fig. 5-5

2
I

-6

II

-1

III

FIGURE 5.5. Code 2-3 is made for a less marked right axis deviation than 2-2, between +90 and
+119. The sum of the positive and negative waves must be negative or zero in lead
I and positive in leads II and III

52
 Fig. 5-6

2-4 extreme axis deviation

I
1

-4

II negative

III
0

-5

FIGURE 5.6. Code 2-4 is made for extreme axis deviation. The sum of the positive and negative
waves must be negative in leads I, II, and III

53
 2-5 Fig. 5-7

I 4

-4

II 4

-4

III 5

-5

FIGURE 5.7. Code 2-5 is made for an indeterminate axis. The sum of the positive and negative
wave must be near zero in leads I, II, and III

Frontal Plane T-Wave Axis

The frontal plane T-wave axis follows similar principles to those establishing QRS frontal
plane axis. This calculation is not for separate coding but for calculation of the frontal
QRS/T angle (see Chap. 16). Simple directions for measurement of T-wave and QRS axis
has been described with the use of a hand-held protractor.1

Reference

1. Prineas RJ. New device for determining direction of cardiac vectors. Lancet. 1967; July 8:80-81.
54
6
High R-Waves (3-Codes)
When the QRS amplitude is high, it is usually due to one or more of several factors: an
unusually large heart muscle mass (a big or thick heart), an unbalanced size of a ventricle,
or a large heart in relation to the chest size or closeness to the chest wall. The lead direction
is maximally parallel to the average direction of the QRS forces in that person.
Limits for normal size R-waves are based on distributions of the values for the amplitudes
of QRS-waves (by age, sex, and body build). Codable classes represent various degrees of
cardiac abnormality or various degrees of association with heart size.
Codes 3-1 through 3-4 categorize high amplitude QRS voltage relevant to ventricular
hypertrophy patterns. Codes 3-1 and 3-3 are high left R amplitude patterns as measured on the
next to last complete normal beat, and 3-2 high right R amplitudes.

Fig. 6-1

3-1
lead V5 or V6
R amplitude > 26 mm.

FIGURES 6.16.3. Code 3-1 is coded if any of the following criteria are present: R amplitude >26 mm in
either lead V5 or V6 (see Fig. 6.1); R amplitude >20 mm in any of leads I, II, III,
or aVF (see Fig. 6.2); R amplitude >12 mm in lead aVL (see Fig. 6.3)

55
 Fig. 6-2

3-1
lead I, II, or aVF
R amplitude > 20 mm.

Fig. 6-3

3-1
lead aVL
R amplitude > 12 mm.

56
FIGURES 6.46.6. Code 3-2 requires that all three conditions be met. First, V1 must have an RS
or QRS pattern. The R-wave amplitude in V1 must be 5 mm in the majority of
beats (see Fig. 6.4). In addition, the amplitude of the R-waves must be equal to
or greater than the amplitude of the S-waves, in the majority of beats in lead V1
(see Fig. 6.5). Finally, Second amplitude must be greater than R amplitude in
the majority of beats in any one of leads V2V6 (see Fig. 6.6). All three of these
conditions must exist to code 3-2
57
 Fig. 6-7
3-3
lead I
15 mm.<R amplitude 20 mm.

Fig. 6-8
3-3

V1

S=11 mm.

V5

S=25 mm.

FIGURES 6.76.9. Code 3-3, an intermediate left pattern, is coded if one or both of the following
two criteria is present: R-wave amplitude >15 mm but 20 mm in lead I (see
Fig. 6.7); R-wave amplitude in V5 or V6 plus Second or QS amplitude in V1 >35
mm (see Figs. 6.8 and 6.9). Code 3-3 also requires that there is no 3-1 code
58
 Fig. 6-9
3-3

V1

QS=13 mm.

V6

R=24 mm.

Fig. 6-10
3-4

aVL V1 V3

R amplitude >12 mm. R amplitude >5 mm. R<S amplitude

R>S amplitude

FIGURE 6.10. Code 3-4 is coded if criteria for both codes 3-1 and 3-2 are met

59
7
ST Segment Depression (4-Codes) and Negative T-Waves (5-Codes)
Toward the end of ventricular excitation (QRS), the earliest part of the heart to be excited
begins to recover or recharge to produce the T-wave. If the heart muscle is injured or
short of oxygen, the electrical recovery starts prematurely, and this may produce a sizable
current. The trace between QRS and T is then displaced up or down, depending on the
location of the injury and on the lead. The ST segment is continuous with the T-wave and
both are often coded together; thus, they are learned together.
Moreover, certain damage or inadequate blood supply to the heart ventricles may reverse
the usual sequence of recovery. This may result in distinctly negative T-waves in leads
where they are usually positive. The lead orientation may display, however, low or diphasic
(partly positive, partly negative) T-waves and these, along with ST depression, may be
indistinct and cause coding problems.
ST segment depression coding requires:
1. Identication of the J-point, the junction between QRS and ST
2. Determination of the slope and shape of the ST segment
3. Measurement of the amplitude of ST segment depression

Fig. 7-1
R
R

J J
Point T Point T
P P

FIGURE 7.1 and 7.2. The end of QRS is called J-point, a break in direction of the tracing at the end
of QRS. In the accompanying gures, the J-point is marked with an arrow.
Correct placement of the J-point is the rst requirement for determining ST
segment depression and slope
60
 Fig. 7-2

R
R

J J
Point Point
P P

T
T
Q

Fig. 7-3

R R

P J P J
Point Point

T
T

FIGURE 7.3. Notice that the J-point is not always level with the preceding P-R baseline and here
is below it

61
 R
R Fig. 7-4

J
Point
J T
P T P
Point

FIGURE 7.4. It shows one example of a depressed J-point and another with J-point level with the
P-R baseline

Fig. 7-5

J J

FIGURE 7.5. Sometimes the J-point is not such a sharp angle at the end of the QRS. Though a break
in the curve between the QRS and ST segment can be seen, the exact point is indeter-
minate. In such cases, a tangent is drawn along the straightest section to be found (one at
least 2 mm in length) of the lower margin of the ST segment at the beginning of ST. The
J-point is on the upper margin opposite the point the tangent and the ECG separate

R Fig. 7-6
R

perfect curves

no J points

P
P

FIGURE 7.6. Where the J-Point is not easily located or where there is no straight part of the ST
segment at least 2 mm long for constructing a tangent, a truly curved ST segment
exists and no J-point is measured
62
 R
Fig. 7-7

two possible J points

P

Fig. 7-8

two possible J points

Fig. 7-9

end of QRS

FIGURES 7.77.9. It is not uncommon to observe two separate breaks in the curve at the end of the
QRS. When in doubt as to which is the real J-point, take the second or rightward
point to determine the J-point for 4-codes (see Fig. 7.7 and 7.8). In Fig. 7.9, the
J-point dees assignment. The S-wave continues in a smooth curve to the peak of
the T-wave. The end of the QRS is determined from where a horizontal tangent
from the upper margin of the baseline at the beginning of the previous QRS
complex intersects the upward curve. In this example, no J-point depression and
no ST segment depression is coded (see Code 9-2, Chap.11)
63
 Fig. 7-10

ST segment

ST segment

FIGURE 7.10. The J-point marks the end of the QRS and the beginning of the ST segment. The end
of the ST segment is sometimes distinct but more often merges imperceptibly with
the T-wave. The ST segment is then considered to extend to the peak of the T-wave.
In the rst example, the ST is exaggeratedly represented as a distinct segment. In the
second, the ST segment merges with the T-wave and extends to the peak of T. Both
T-waves are positive

Fig. 7-11

ST segment
ST segment

FIGURE 7.11. It shows similar types of ST segment to Fig. 7.10, but with both T-waves negative
64
 Fig. 7-12

ST segment
ST segment

FIGURE 7.12. The J-point is depressed in the rst beat with the ST segment merging into a positive
T-wave. The second beat shows the J-point level with the preceding P-R baseline and
the ST segment merging into the T-wave

Fig. 7-13

ST ST
segment segment

FIGURE 7.13. The J-point is depressed with the ST segment merging in different forms into a
negative T-wave

65
 Fig. 7-14

R
ST segment
flat T wave

J point
P

FIGURE 7.14. An ST segment can be associated with a at T-wave, giving no measurable segment,
even though the J-point can be distinguished

Fig. 7-15

perfect curve
no clearly defined ST segment
no J point

FIGURE 7.15. An example of indeterminate ST segment is illustrated. When a near-perfect curve

is present (no straight ST segment for at least 2 mm), there is no clearly dened ST
segment. With this type of ST-T wave, there is no measurable J-point

66
FIGURES 7.167.18. Different levels of J-point displacement are illustrated. The reference for deter-
mining J-point displacement is the upper margin of the baseline immediately
preceding the QRS complex. In Fig. 7.16, the J-point is depressed; in Fig.7.17,
it is elevated. In Fig. 7.18, there is neither elevation nor depression

67
 Fig. 7-19
upward sloping ST segment

Fig. 7-20
upward sloping ST segment

Fig. 7-21
upward sloping ST segment

FIGURES 7.197.21. The slope of the ST segment is measured from the J-point to the start of the
T-wave or to the peak of the T-wave when there is no distinction between the end
of the ST segment and the start of the T-wave. The ST segment is upward sloping
in all of these gures. The level of the J-point does not determine the slope
68
 Fig. 7-22

upward sloping ST segment

J
J

FIGURE 7.22. Remember, where there are two possible J-points, the depression and slope of the ST
segment are determined from the second

Fig. 7-23

upward sloping ST segment

elevated J point

FIGURE 7.23. It shows the J-point elevated from the P-R baseline and having an upward sloping
ST segment

69
 Fig. 7-24

U shaped curve - considered

upward sloping ST segment an upward sloping ST segment

FIGURE 7.24. The direction of the ST segment is classied as horizontal, downward, or upward
sloping based on the following rule: The segment is not horizontal or downward
sloping if any part of it slopes upward. This also applies to curved or U-shaped ST
segments. Both ST segments illustrated in this gure are considered upward sloping

Fig. 7-25

downward sloping ST segment

FIGURE 7.25. The exception to this rule sometimes occurs when the ST segment is convex upwards.
If the amplitude at the end of the ST segment is lower than the J-point, this convex
pattern is considered equivalent to a downward sloping ST segment and is so coded
70
 upward sloping ST segment Fig. 7-26

FIGURE 7.26. The convex ST segment is not considered downward sloping because the end of the
ST segment is higher or at the same level as the J-point

horizontal ST segment Fig. 7-27

FIGURE 7.27. Both the J-point and ST segment slope are important in determining 4-codes. The J-point
is level with the preceding P-R baseline, and there is a horizontal ST segment

horizontal ST segment Fig. 7-28

FIGURE 7.28. The J-point is depressed and there is a horizontal ST segment

71
 horizontal ST segment Fig. 7-29

FIGURE 7.29. The J-point can also be elevated with a horizontal ST segment

horizontal ST segment Fig. 7-30

J
J

FIGURE 7.30. Always remember to use the second of the two possible J-points for determining
J-point depression and ST segment slope. In the gure there are two apparent J-points
with a horizontal ST segment because the second of two possible J-points is taken as
the correct J-point

downward sloping ST segment Fig. 7-31

no J point depression

FIGURE 7.31. The J-point is at the same level as the P-R baseline. Thus, there is a downward sloping
ST segment and no J-point depression
72
 downward sloping ST Fig. 7-32
segment

FIGURE 7.32. The J-point can also be depressed and be followed by a downward sloping ST segment

Fig. 7-33

J
downward sloping ST
segment J

J point depressed

FIGURE 7.33. Again, remember where two possible J-points appear, always use the second J-point
to determine depression and ST segment slope

downward sloping ST Fig. 7-34

segment

FIGURE 7.34. An elevated J-point accompanied by a downward sloping ST segment is shown

73
 measurement of J point depression Fig. 7-35

FIGURE 7.35. The two most important criteria for measuring 4-codes are correct J-point location and
ST slope determination. The amplitude of J-point depression is measured vertically
to the intersection of a tangent drawn horizontally from the upper margin of the P-R
baseline at the onset of QRS

measurement of J point depression Fig. 7-36

FIGURE 7.36. Locate the J-point by constructing a tangent to the ST segment and measuring
vertically to the level of the baseline, upper margin

measurement of J point depression Fig. 7-37

FIGURE 7.37. The J-point is depressed and the ST segment is upward sloping
74
FIGURES 7.38 and 7.39. J-point depression is measured in millimeters, the code depending on the
degree of depression. The procedure is to code conservatively in case of doubt.
The majority of beats in the lead (>50%) should qualify to be coded in that
category. Leads III and aVR are ignored in coding ST segments and T-waves
because their waves are too variable to be reliable. The rst and major 4-code
is 4-1-1. To qualify, the J-point must be depressed 2mm in the majority of
beats in any of leads I, II, aVL, aVF, V1V6. The slope of the ST segment must
also be downward sloping as in Fig. 7.38 or horizontal as in Fig. 7.39
75
 Fig. 7-40

4-1-2
in any of leads I, II, aVL, aVF, V1-V6
1 mm. J point depression < 2 mm.
horizontal ST segment

Fig. 7-41
4-1-2
in any of leads I, II, aVL, aVF, V1-V6
1 mm. J point depression < 2 mm.
downward sloping ST segment

FIGURES 7.40 and 7.41. For code 4-1-2, the J-point depression must be 1 mm but < 2 mm with
the ST segment horizontal or downward sloping in the majority of beats
in any of the leads I, II, aVL, V1V6

76
 Fig. 7-42

4-2
in any of leads I, II, aVL, aVF, V1-V6
0.5 mm. J point depression < 1 mm.
downward sloping ST segment

Fig. 7-43
4-2
in any of leads I, II, aVL, aVF, V1-V6
0.5 mm. J point depression < 1 mm.
horizontal ST segment

FIGURES 7.42 and 7.43. Code 4-2 is similar to the previous codes, only less marked. To qualify, the
J-point must be depressed 0.5 mm but <1.0 mm and the ST segment must
be either downward sloping (see Fig. 7.42) or horizontal in the majority of
beats in any of the leads I, II, aVL, aVF, V1V6 (see Fig. 7.43)

77
 Fig. 7-44

4-3
any of leads I, II, aVL, V1-V6
end of ST segment
depressed 0.5 mm.
J point depression
< 0.5 mm.
downward sloping
ST segment

Fig. 7-45

no 4 - code

FIGURES 7.44 and 7.45. To qualify as a 4-3 code, the J depression must be < 0.5 mm but the ST
segment must be sloping downwards (see Fig. 7.44). A 4-3 code is coded
in leads I, II, aVL, V2V6 if present in the majority (>50%) of beats in any
of these leads. This code is different from the previous codes in which it
requires that the end of the ST segment must be at least 0.5 mm below the
P-R baseline. In Fig. 7.45, even though the baseline is downward sloping,
this is not a 4-3 code because the end of the ST segment (here the same as
the T-wave nadir) is depressed less than 0.5 mm below the P-R baseline
78
 Fig. 7-46

4-4
any of leads I, II, aVL, V1-V6
J point depression 1 mm.
upward sloping ST segment

FIGURE 7.46. A 4-4 code is different in concept from the other 4-codes in which the ST segment is
upward sloping from a J-point depressed 1 mm or more from the P-R baseline

Fig. 7-47
4-4
any of leads I, II, aVL, V1-V6
J point depression 1 mm.
upward sloping ST segment

FIGURE 7.47. Use the second J-point to determine the amount of ST depression

79
 Fig. 7-48
4-4

FIGURE 7.48. Another example of a 4-4 code is a curved, concave ST segment where no straight
part of the segment is 2 mm in length. When the trough of such a curve is depressed
1 mm from the P-R baseline, it should be coded 4-4. To measure the ST depression,
measure from the upper margin of the lower point vertically to the upper margin of
the P-R baseline

Fig. 7-49
4 codes
lead V1

4-1-1 4-1-2 4-2 4-4

FIGURE 7.49. 4-1-1, 4-1-2, 4-2, and 4-4 may be coded in V1, but not a 4-3. In lead aVF, 4-1-1, 4-1-2,
and 4-2 are the only possible 4-codes and there are no 4-codes for III or aVR. Most
4-1 through 4-3 coding requires that a 5-code or negative T-wave code be present.
This is because in all 4-1 to 4-3 codes the end of the ST segment is depressed and
the end of the ST segment is the beginning of the T-wave. Therefore, the onset of
the T-wave must be negative and codable. The exceptions are that 4-codes in V1 are
coded without 5-codes, in aVF when the QRS is mostly negative and in aVL when
the R-wave is < 5 mm
Figures 7.507.56 show different forms of the T-wave that need to be recognized to code 5-codes
correctly
80
 Fig. 7-50

negative T wave

FIGURE 7.50. The negative T-wave is distinct from the level (isoelectric) ST segment and is
1.0 mm below the P-R baseline. It is important to note in the following illustrations
that the presence of a negative T-wave is determined by its relation to the top of the
preceding P-R segment

diphasic T waves Fig. 7-51

positive - negative type negative - positive type

FIGURE 7.51. Illustration of the diphasic T-wave, called diphasic because there is both a nega-
tive and a positive component to the T-wave. The diphasic T-wave is categorized as
positivenegative or negativepositive depending on which component occurs rst
81
 Fig. 7-52

negative T wave negative - positive diphasic T wave

FIGURE 7.52. Difference between a simple negative T-wave and a negativepositive diphasic
T-wave is shown

negative T waves Fig. 7-53

R
R

P P

T
T

FIGURE 7.53. Further examples of negative T-waves are shown. Here, the S-wave is followed by a
negative T-wave. In the rst beat, the ST segment merges with the T-wave, and in the
second, the ST segment is distinct from the T-wave

82
 negative T wave Fig. 7-54

J
ST

FIGURE 7.54. An example of a T-wave in which the ST segment is distinct from the T-wave

flat T wave Fig. 7-55

P P

FIGURE 7.55. Illustration of a at T-wave that neither rises nor falls from the baseline

measurement of negative T wave Fig. 7-56

FIGURE 7.56. Having recognized the negative T-wave, its amplitude is determined. To measure
a negative T-wave, nd the attest segment of the baseline in the following T-P
interval. Measure from this point on the top margin of the baseline vertically to the
top of T-wave nadir
83
 Fig. 7-57
5-1
any of leads I, II, V2-V6
negative T wave amplitude
5 mm.

5-1
Fig. 7-58

R height must be 5 mm. in aVL in aVF the QRS must be mainly positive

FIGURES 7.57 and 7.58. The negative T-wave amplitude is always measured in millimeters, and is
measured conservatively. A majority of beats in a lead have to meet the
criteria for each code. The 5-codes are also listed in the order of magnitude.
The negative T-wave is coded to the most severe or lowest numbered category
for which it qualies in any lead of the set. The greatest negative T-wave code
is 5-1. To qualify as 5-1, the T-wave must be 5 mm negative in any of leads
I, II, V2-V6 (see Fig. 7.57) or in aVL when the R-wave is 5 mm, or in aVF
when the QRS is mainly positive (see Fig. 7.58), that is, the sum of R-wave
amplitudes is larger than the sum of negative deections (Q- or S-waves)

84
 Fig. 7-59
5-1
any of leads I, II, V2-V6

FIGURE 7.59. The 5-1 criteria are met, even though the ST segment is upward sloping. The T-wave
is separate and negative

Fig. 7-60

5-1
any of leads I, II, V2-V6

FIGURE 7.60. Illustration of another possible 5-1 pattern. Here the ST-J depression is prominent
with the ST segment horizontal and depressed 5 mm because the end of the ST
segment is the beginning of the T-wave, this point is taken to measure the negative
T-wave; amplitude, i.e., 5 mm to give a 5-1 code

85
 Fig. 7-61
5-1
any of leads I, II, V2-V6
diphasic T wave

5mm

FIGURE 7.61. Negative T-wave amplitude in a diphasic T-wave is measured from the top of the
level part of the T-P interval to the top of the nadir of the T-wave

Fig. 7-62

5-1
any of leads I, II, V2-V6
diphasic T wave

5mm

FIGURE 7.62. This procedure is used for measuring either type of diphasic T-wave, but it is only the
negative component of the T-wave that is measured for coding

86
 Fig. 7-63

5-2
any of leads I, II, V2-V6
1 mm T amplitude < 5 mm.

FIGURE 7.63. A 5-2 code is less prominent than 5-1. To qualify as 5-2, the negative T-wave
amplitude must be 1 mm but <5 mm in any leads I, II, V2V6

Fig. 7-64

5-2

R height must be 5 mm. in aVL in aVF the QRS must be mainly

positive

FIGURE 7.64. In lead aVL when the R-wave is 5 mm, or in lead aVF when the QRS is mainly
positive, 5-2 is coded when the T-wave is negative, 1 mm but <5 mm

87
 Fig. 7-65

5-2
1 mm T amplitude < 5 mm.

FIGURE 7.65. The same forms of 5-1 patterns are applicable to 5-2 patterns, the only difference
being their degree. Remember, for a 5-2 code, the negative T-wave amplitude must
be 1 mm but <5 mm

Fig. 7-66

5-2

FIGURE 7.66. An ST-J depression with a horizontal ST segment is pictured. The ST-J depression is
equal to the depth of the T-wave, that is 1 mm but <5 mm

88
 Fig. 7-67

5-2
diphasic T wave
negative - positive

FIGURE 7.67. A diphasic T-wave (negativepositive type) with negative phase 1 mm but <5 mm
negative

Fig. 7-68

5-2
diphasic T wave
positive - negative

FIGURE 7.68. Another diphasic T-wave, a positivenegative type with a negative phase 1 mm but
<5 mm negative

89
 Fig. 7-69

5-3
lead I, II, aVL, V3-V6
T wave negative but < 1mm

FIGURE 7.69. Some of the patterns for 5-3 codes are similar to 5-1 second and 5-2 second, but special
differences should be noted. To qualify as a 5-3, the T-wave may be negative or at.
The only kind of diphasic T-wave that can qualify is the negativepositive type. The
negative T-wave amplitude must be <1 mm and cannot be coded in aVF or V2, unlike 5-1
and 5-2. In the Figure, negative T-wave is clearly recognized. As in 5-1 and 5-2 codes,
5-3 codes cannot be coded in V1. To be coded in aVL the R-wave must be 5 mm

Fig. 7-70

5-3
lead I, II, aVL, V3-V6

negative-positive T wave

FIGURE 7.70. Illustration of a negativepositive T-wave that qualies as a 5-3

90
 no 5-3 Fig. 7-71

T wave is positive-negative

FIGURE 7.71. Illustration of a positivenegative T-wave that cannot be coded as 5-3 and a 5-code
is not made

5-3 Fig. 7-72

lead I, II, aVL, V3-V6

FIGURE 7.72. A pattern with an upward sloping ST segment and distinctly negative T-wave <1 mm
in amplitude

5-3 Fig. 7-73

lead I, II, aVL, V3-V6

flat T wave

FIGURE 7.73. Illustration of a 5-3 code for a at T-wave

91
 Fig. 7-74

5-4
any of leads I, II, aVL, V3-V6
R wave positive and the
T/R ratio < 1/20
R wave must be 10 mm.

FIGURE 7.74. Illustration of the lowest order T-wave code (5-4), different in concept from other
5-codes in that the T-waves are positive but of low amplitude relative to R-wave
amplitude. To qualify as a 5-4 , the T-wave must be positive and the T amplitude
must be less than 1/20 of R amplitude in any of leads I, II, aVL, V3-V6, but R-wave
amplitude must be 10 mm high. Note that the positive T-wave is measured from
the upper margin of the T-wave to the upper margin of the following T-P baseline

Fig. 7-75

no code
T 1
>
R 20

T P T P T P T P

measure the T wave height to the previous P-R interval

FIGURE 7.75. When a rapid heart rate causes the T-wave to fall on the following P-wave, use the
previous P-R interval as the reference point for measuring positive T-wave. Using this
procedure, the T/R ratio in this would not meet the criteria for a 5-4 code. No 5-code
92
 only lead V1 Fig. 7-76

no 5-codes

4-1-1 4-1-2 4-2

FIGURE 7.76. Illustration of how in lead V1, codes 4-1-1, 4-1-2, and 4-2 can be assigned without a
5-code because 5-codes are not assigned in lead V1

Fig. 7-77

4-1-1 + 5-2

T-wave inversion (negative positive)

FIGURE 7.77. Remember in coding 4- and 5-codes the J-point depression and T-wave amplitude are
measured independently. For example, a 4-1 code does not automatically have with
it a 5-1 code. Each is measured and coded separately. The ST depression measures
2 mm to qualify for a 4-1-1 code and the negative T-wave measured 1 mm but
< 5 mm and qualies for a 5-2 code
93
 4-1-1 Fig. 7-78

5-1

FIGURE 7.78. A 4-1-1- and 5-1 code coexist

Fig. 7-79

4-1-2
5-2

FIGURE 7.79. Remember, with a diphasic T-wave, the negative phase is measured vertically from
the most level part of the following T-P interval

94
 4-1-1 Fig. 7-80

5-2

FIGURE 7.80. It is a more dramatic example than is usually seen, but it would be a mistake to
measure the T-wave 5 mm negative instead of 4 mm. The correct code is 5-2

Fig. 7-81

end of ST segment end of ST segment

positive T wave negative - positive T wave

FIGURE 7.81. Another problem in distinguishing negative T-wave along with different shaped ST
segments is shown. The rst beat shows an upward sloping ST segment regarded as
ending at the peak of the T-wave. Therefore, no negative T-wave is coded. The next
beat has a negative T-wave code because although the ST segment is upward sloping,
the T-wave eventually slopes downward. The end of the ST segment is recorded as
the beginning of the T-wave, a negativepositive, diphasic T
95
 Fig. 7-82

4-4 4-4
5-4 5-3

FIGURE 7.82. Illustration of possible codes for the patterns shown in the previous gure

Fig. 7-83

4-4
5-3

FIGURE 7.83. It is possible to have a negative T-wave with an upward sloping ST segment

96
 Fig. 7-84
5-2 5-2

5-3 5-3 Fig. 7-85

T amplitude < 0.5 mm.

5-4 Fig. 7-86

FIGURES 7.847.86. 5-codes do not require a 4-code. A 5-code can be present in a given lead
and be recorded without a 4-code in contrast to 4-codes, which are generally
accompanied by 5-codes. In Fig. 7.847.86, the T-waves are codable but the
ST segments are not
97
8
Atrioventricular (A-V) Conduction Defects (6-Codes)
The 6-codes classify defects in conduction between the atria and ventricles and are
concerned with the relation of the P-wave to the QRS complex. Usually atrial activity (the
P-wave) causes the ventricles to re (QRS complex). There is usually a xed and regular
P-R interval.

P-R interval Fig. 8-1

Fig. 8-2
normal P-R interval: a=b=c=d=e

a b c d e

FIGURES 8.1 and 8.2. The time between the onset of the P-wave and the onset of QRS is called
the P-R interval, whether or not the QRS complex commences with an R- or
a Q-wave

98
 Fig. 8-3

FIGURE 8.3. From the onset of the P-wave to the onset of the Q wave is also the P-R interval

Fig. 8-4
P wave positive

find the point on top of the

baseline where

the rise begins to form

the P wave

Fig. 8-5
P wave negative

A is the beginning

of the P wave - measure

P-R from point B

FIGURES 8.48.7. When conduction from the atria to the ventricles is accelerated, delayed, or
blocked, the length of the P-R interval and the relationship of the P-wave to
the QRS complex changes. If inspection reveals a short or long P-R interval, it
must be measured precisely. To do so, the rst step is to determine the onset of
the P-wave at the top margin of the baseline for positive P-wave (see Fig. 8.4)
and from the bottom of the baseline when the P-wave is negative (Fig. 8.5). This
point is the origin of the P-R interval (Fig. 8.48.7)
99
 Fig. 8-6

beginning of the P wave

Fig. 8-7

beginning of the P wave

100
 Fig. 8-8

onset of the QRS and

end of P-R interval

Fig. 8-9

onset of the P wave

onset of the QRS

FIGURES 8.8 and 8.9. Next, determine the end of the P-R interval where the QRS complex begins.
The beginning of QRS depends on its pattern. In the case of an initial R, the
beginning of QRS is measured from the top edge of the baseline (see Fig. 8.8).
When the QRS pattern is a QS-wave, onset of QRS is also measured at the
upper edge of the baseline (see Fig. 8.9)

101
 Fig. 8-10

onset of the P wave onset of the QRS

Fig. 8-11
the P-R interval is the

distance between the

beginning of the P and

the beginning of the

QRS

Fig. 8-12
P-R interval

102
 Fig. 8-13

P-R interval

Fig. 8-14

P-R interval

FIGURES 8.108.14. When the QRS begins with a Q, its onset is measured from the upper edge of
the baseline (see Fig. 8.10). So, the P-R interval is the distance between the
onset of the P-wave and the onset of the QRS complex (see Figs. 8.118.14)

103
 Fig. 8-15

0.10 sec. < P-R interval < 0.12 sec.

Fig. 8-16

0.12 sec. < P-R interval < 0.13 sec.

FIGURES 8.15 and 8.16. The following are examples of P-R interval measurements. In Fig. 8.15,
the P-R interval is >0.10 second but <0.12 second. In Fig. 8.16 the P-R
interval is >0.12 but <0.13 second

104
 6-1 Fig. 8-17

R1 R2 R3

P1 P2 P3 P4

P
T

FIGURE 8.17. Code 6-1 is for complete (or third-degree) A-V block. In this nding, the atrial and
ventricular rates are regular, but independent and therefore, the P-R interval varies
greatly from beat to beat. The code requires the atrial rate be faster than the ventricu-
lar rate and the ventricular rate less than 60 beats per min. The QRS complex is often
wide, which can be in all beats or intermittent. A P-wave may distort the shape of the
T-wave as seen in the same gure

Fig. 8-18
6-2-1
Mobitz type II
constant P-R interval
constant P-P interval
dropped QRS and T

PR

P1 P2 P3 P4

FIGURE 8.18. All 6-2 codes for second degree A-V blocks have QRS complexes preceded regularly
by a P-wave. Code 6-2-1 is for an A-V conduction defect called Mobitz type II. This
defect is characterized by a P-wave followed by unexpected absence of a ventricular
complex (QRS and T). The P-waves are normal and the P-R intervals constant and
regular. The R-R interval including the dropped beat should be within 10% of the
normal R-R interval x 2
105
 Fig. 8-19
6-2-2
2:1 A-V block

R1 R2 R3 R4

P1 P2 P3 P4 P5 P6 P7

6-2-2 Fig. 8-20

3:1 A-V block

P1 P2 P3 P4 P5 P6 P7 P8 P9

FIGURES 8.19 and 8.20. Code 6-2-2 is also a form of second degree or partial A-V block. The atrial
and ventricular rates are both regular with the atrial rate a multiple of
the ventricular rate (10%). The P-R and R-R intervals are quite regular.
The block may be 2 Ps for 1 QRS, 3:1, 4:1, etc

106
 6-2-3 Fig. 8-21
Wenckebach

FIGURE 8.21. Code 6-2-3 is another second degree A-V conduction defect known as Wenckebachs
phenomenon. It is characterized by a dropped ventricular complex, but unlike the 6-2
code (Mobitz type II), the P-R intervals progressively lengthen, and nally, a P-wave
is not followed by a QRS-T

6-3 Fig. 8-22

in any leads I, II, III, aVL or AVF

P-R interval 0.22 sec.
in the majority of beats

Fig. 8-23
6-3

FIGURES 8.22 and 8.23. Code 6-3 is a prolonged P-R interval or rst-degree block. The P-R interval
must be 0.22 second in the majority of beats in any of the following leads:
I, II, III, aVL, or aVF. There are no absent P-wave or QRS complexes
107
 6-4-1
Fig. 8-24
WPW pattern

normal P waves
short P-R interval < 0.12 sec
co-existing
in same prolonged QRS 0.12 sec.
beat
R peak duration 0.06 sec.
must be present in the majority of beats in any of leads I, II, aVL, V4, V5, V6

FIGURE 8.24. Code 6-4-1 is a unique pattern of short P-R and prolonged or slurred QRS, called
the Wolff-Parkinson-White Pattern (WPW).The WPW is characterized by normal
P-waves, short P-R interval <0.12 second, prolonged QRS duration 0.12 second
with a slurred upstroke to the QRS complex, and R peak duration 0.06s. All these
conditions must coexist in the same beat and also be present in the majority of beats
in any of the following leads: I, II, aVL, V4, V5, or V6

6-4-2 Fig. 8-25

intermittent WPW
leads I, II, aVL, V4-V6

WPW pattern 50% or less of the beats in any one

of the appropriate leads

FIGURE 8.25. Code 6-4-2 is for intermittent WPW-waves when the above criteria occur in 50% of
the beats in any one of the appropriate leads: I, II, aVL, V4, V5, or V6

108
FIGURES 8.26 and 8.27. Code 6-5 is a short P-R interval <0.12 second. This condition must be
present in every beat in any two of the following leads: I, II, III, aVL, or
aVF. This is one if the few codes that is an exception to the majority rule

109
 6-6
Fig. 8-28

normal P wave
P-R interval 0.12 sec.
bizarre QRS which is 0.12 sec.
occurs in 50% of the beats in any lead

FIGURE 8.28. Code 6-6 is for intermittent aberrant ventricular conduction. When present, it usually
occurs with only one or two beats in the ECG. Criteria for this code are a wide QRS
complex, which is 0.12 second and a P-R interval 0.12 s. This combination must
only occur in 50% of the beats in any lead (if >50% of beats, a 7-4 code is present,
see Chap. 9). Most of the beats are sinus rhythm and there are no ventricular prema-
ture beats (8.1-2) on the record

6-8 Fig. 8-29

Artificial pacemaker

FIGURE 8.29. Code 6-8 is for the presence of an articial pacemaker. Criteria are sharp vertical
spikes having a duration too short to measure and occurring at absolutely regular
intervals preceding each QRS complex
110
9
Intraventricular Conduction Defects (7-Codes)
With ventricular conduction defects, the pathway of excitation through the heart is unusual
and inefcient. The electrical wavefront follows a different, often slower path through one
bundle at a time and through the ventricular muscle. The result of this slower, roundabout
path is seen as a wider than normal QRS complex coded as 7-codes.

Fig. 9-1

QRS duration

onset offset

Fig. 9-2

QRS duration

FIGURES 9.19.6. The rst step is to detect a wide QRS, the duration of which is then measured
precisely. A QRS duration of 0.12 second or more is wide. The shape of the
QRS complex determines the measurement procedure. For example, an R com-
plex is measured at the lower margin of the baseline (see Figs. 9.1 and 9.2). If the
QRS is a QS-wave, the duration is measured at the upper margin (see Fig. 9.3).
If the complex begins with a Q-wave and ends with an R-wave (QR), the dura-
tion measurement starts at the upper edge of the baseline and ends at the lower,
measured horizontally not diagonally (see Fig. 9.49.6)
111
QRS duration onset offset Fig. 9-3

QRS duration Fig. 9-4

QRS duration Fig. 9-5

QRS duration Fig. 9-6

112
 Fig. 9-7
QRS duration

Fig. 9-8
QRS duration

Fig. 9-9
QRS duration

FIGURES 9.79.12. Other possible QRS duration measurements are shown

113
 Fig. 9-10
QRS duration

Fig. 9-11
QRS duration

Fig. 9-12
QRS duration

114
 Fig. 9-13

when there are two possible

J points take the earlier J point
for measuring QRS duration

Fig. 9-14

when there are two possible

J points take the earlier
J point for measuring QRS
duration

FIGURES 9.13 and 9.14. If there are two J-points, the earliest is used for measuring QRS duration
(in contrast to ST segment measurement)

115
 Fig. 9-15

QRS duration

FIGURE 9.15. QRS complexes without sharp offsets are represented. In such cases, a tangent is
drawn along the inner margin of the terminal QRS deection. The point where the
tracing departs from the tangent is the end of the QRS

Fig. 9-16

QRS duration = 0.09 sec.

FIGURE 9.16. An example of a QRS duration of exactly 0.09 second

116
 Fig. 9-17

0.11 sec. < QRS duration < 0.12 sec.

Fig. 9-18

QRS duration = 0.09 sec.

FIGURES 9.179.20. More examples of QRS duration measurement

117
 Fig. 9-19

0.09 sec. < QRS duration < 0.10 sec.

Fig. 9-20

QRS duration = 0.08 sec.

118
 R peak duration = 0.04 sec. Fig. 9-21

R peak duration = 0.06 sec. Fig. 9-22

0.05 sec. < R peak duration <0.06 sec. Fig. 9-23

FIGURES 9.219.23. An additional measurement needed for certain 7-codes is the R peak duration,
from the onset of QRS to the peak of R. All these Figures show examples of
this measurement
119
 R peak duration Fig. 9-24

FIGURE 9.24. An R-wave with two peaks. In such cases, the second peak is used for the R peak
measurement

R peak duration Fig. 9-25

S-wave
depth

FIGURE 9.25. RSR (R-S-R prime) pattern. In this case, the R peak duration is measured from the
beginning of QRS to the peak of the R-wave, not the R. For an R to be present, the
S-wave must be 0.25 mm below the lower edge of the baseline

S peak duration Fig. 9-26

FIGURE 9.26. In the event that the QRS complex is a QS, there is, of course, no R peak duration. In
this case, measure QS peak duration

120
 Fig. 9-27

7-1-1
in any of leads I, II, III, aVL or aVF
QRS duration 0.12 sec. in
the majority of beats

PLUS

Fig. 9-28

7-1-1 (2nd requirement)

in any of leads I, II, aVL, V5 or V6
R peak duration 0.06 sec.
in the majority of beats

FIGURES 9.27 and 9.28. Code 7-1-1 is for complete left bundle branch block and requires two
conditions. First, the QRS duration must be 0.12 second in the majority
of beats in any one of the following lead: I, II, III, aVL, or aVF. Second,
the R peak duration must be 0.06 second in the majority of beats in any
one the following leads: I, II, aVL, V5, or V6

121
 7-1-2 Fig. 9-29
intermittent left bundle branch block

FIGURE 9.29. Code 7-1-2 is for intermittent left bundle branch block. The criteria are the same
as 7-1-1. However, in this case they occur in some beats in conjunction with some
normally conducted beats. The normal beats are 50% of total beats

Fig. 9-30
7-2-1
in any of leads I, II, III, aVL or aVF
QRS duration 0.12 sec. in
the majority of beats

PLUS

FIGURES 9.309.33. Code 7-2-1 is for complete right bundle branch block, requiring two condi-
tions. The rst is a QRS duration 0.12 second in the majority of beats in any
one of leads I, II, III, aVL, or aVF (see Fig. 9.30). In addition, any one of the fol-
lowing conditions must also be met: (a) there must rst be an RSR pattern in
V1 or V2 with R amplitude greater than R and second an S wave 0.25 mm in
the majority of beats of the lead (see Fig. 9.31), or (b) the QRS must be mainly
positive with an R peak duration 0.06 second in the majority of beats in V1
or V2 (see Fig. 9.32), or (c) the duration of the S-wave must be greater than
R-wave duration in all beats in either lead I or lead II (see Fig. 9.33)
122
 Fig. 9-31
7-2-1 (2nd requirement)

R' > R in V1 or V2

OR

7-2-1 (2nd requirement) Fig. 9-32

lead V1 or V2

QRS mainly upright

(R>Q or S)

R peak duration 0.06 sec.

in the majority of

beats

OR

7-2-1 (2nd requirement) Fig. 9-33

lead I or II

S duration > R duration

in all beats

S duration

R duration

123
 7-2-2 Fig. 9-34

FIGURE 9.34. Code 7-2-2 is for intermittent right bundle branch block. The criteria are the same as
7-2-1 code except that some normally conducted QRS complexes are present. The
normal beats are 50% total beats

Fig. 9-35
7-3
R'>R in lead V1 or V2
QRS duration < 0.12 sec. in the
majority of beats in each of
leads I, II, III, aVL, aVF
S wave 0.25 mm. below the PR
baseline in the majority of beats
initial R 0.25 mm. in the majority
of beats
terminal R amplitude 1 mm. in
the majority of beats

FIGURE 9.35. Code 7-3 is for an incomplete right bundle branch block. Two criteria must exist to
code a 7-3. First, the QRS duration must <0.12 second in a majority of beats in all of
the following leads: I, II, III, aVL, and aVF. Also, there must be an R-R pattern with
the R amplitude greater than R in the majority of beats in either lead V1 or V2
124
 Fig. 9-36

initial R 0.25 mm.

terminal R1 mm.

FIGURE 9.36. An initial R amplitude must be 0.25 mm and the terminal R or R amplitude
must be 1.0 mm. The terminal R amplitude is measured from the upper margin of
the previous P-R interval at the onset of QRS vertically to the R peak

Fig. 9-37
7-3
initial R
a 0.25 mm.
b reaches peak 0.02 sec,
terminal R
c 1 mm.
b
c
a

FIGURE 9.37. It is important to remember the criteria for initial R-waves when coding a 7-3. The R
amplitude must reach a peak of 0.25 mm within 0.02 second of its onset

125
 RSR' pattern notched R Fig. 9-38
S 0.25 mm. S < 0.25 mm.

FIGURE 9.38. To have an R-R pattern, there must be an S-wave present of at least 0.25 mm amplitude. If
the S-wave is <0.25 mm, then the pattern is considered a notched R rather than an R-R

7-4 Fig. 9-39

in any of leads I, II, III, aVL or aVF

QRS duration 0.12 sec. in

the majority of beats

FIGURE 9.39. Code 7-4 is for intraventricular block. The only criterion for that is the QRS duration
must at least be 0.12 second in the majority of beats in any one of the following
leads: I, II, III, aVL, or aVF
7-5 Fig. 9-40
R R' in the majority of beats

in lead V1 or V2

both R amplitudes 0.25 mm. above

the previous PR baseline in the

majority of beats

S wave amplitude 0.25 mm. below

the PR baseline

FIGURE 9.40. Code 7-5 is another R-R pattern in which the R must be less or equal in amplitude
to the R in the majority of beats in either lead V1 or V2
126
 7-5 Fig. 9-41
R > R'

S 0.25 mm.

FIGURE 9.41. Remember as with 7-3 code, for a 7-5 code, the S amplitude must be 0.25 mm.
Also, both R and Ramplitudes are measured from the upper margin of the preceding
P-R baseline
7-5 Fig. 9-42
R' < R

R and R' 0.25 mm.

FIGURE 9.42. In all cases except for 7-5 codes, terminal R amplitude must be 1.0 mm (see Chap. 4).
For 7-5 code, initial and terminal R amplitude need only be 0.25 mm

terminal R (definition for 7-5 only) Fig. 9-43

a 0.25 mm

b 0.04 sec.

c 0.25 mm.
b

a c

FIGURE 9.43. Sometimes, it is difcult to determine whether a terminal R-wave is present. It must
meet the following conditions: R amplitude 0.25 mm, dropping at least 0.25 mm
within 0.04 second of its peak, in the majority of beats of the lead in question
127
 Fig. 9-44

7-6
leads I, aVL and V5 or V6
0.10 sec. QRS duration
< 0.12 sec.

FIGURE 9.44. Code 7-6 is for incomplete left bundle branch block. For this code, QRS must be
0.10 second and <0.12 second in the majority of beats in each of the following
leads: I, aVL, and V5 or V6. A 7-6 is not codable if there is a 1-code (Q- or QS-code)
anywhere on the record

Fig. 9-45
7-7
QRS < 0.12 sec. in majority (a) a
of beats in each of leads
I, II, III, aVL or aVF
Q wave 0.25 mm. amplitude (b)
and < 0.03 sec. duration in c
lead I (c)
QRS axis < -45

FIGURE 9.45. Code 7-7 is for left anterior fascicular block that has three criteria. First, the QRS
duration must be <0.12 second in the majority of beats in each one of the following
leads: I, II, III, aVL, or aVF. In addition, there must be at least one Q-wave in lead
I or aVL 0.25 mm amplitude and <0.03 second in duration, in the absence of any
codable initial R-wave. This code and 7-8 are the only exceptions for a Q-wave to
be 1 mm amplitude to be codable. Finally, there must be an axis measurement more
negative than 45, i.e., 46, 47, etc
128
 Fig. 9-46
7-8
QRS duration 0.12 sec.
in the majority of beats
in any of leads I, II, III, aVL or aVF
Q 0.25 mm. amplitude and
< 0.03 sec. duration in
lead I

PLUS

Fig. 9-47
QRS axis < -45

-3

PLUS 1 of the following

-7.5

FIGURES 9.469.50. Code 7-8 is a combination of 7-2-1 and 7-7, sometimes called bifascicular
block. All criteria for 7-2-1 and for 7-7 must be present. First, the QRS duration
must be 0.12 second in the majority of beats in any one of the following leads:
I, II, III, aVL, or aVF. There must also be a Q-wave in lead I that is 0.25
mm in amplitude and <0.03 second in duration (see Fig. 9.46). In addition,
the axis measurement must be to the left of 45 (see Fig. 9.47). One of the
following conditions must also be met: R greater than R amplitude in the
majority of beats in lead V1 or V2 (see Fig. 9.48); or QRS mainly positive with
an R peak duration 0.06 second in the majority of beats in V1 or V2 (see Fig.
9.49); or S-wave duration greater than R-wave duration in every beat in either
lead I or II (see Fig. 9.50).
129
 R'
Fig. 9-48
lead V1 or V2

R' > R

OR

Fig. 9-49
lead V1 or V2

QRS mainly upright (R>Q or S)

R peak duration 0.06 sec. in

the majority of beats

OR

Fig. 9-50
lead I or II

S duration > R duration

in all beats

130
Code 7-9 is for the Brugada pattern1,2 and is a combination of persistent ST elevation in
the right precordial leads in at least two of leads V1 to V3. Code 7-3 may or may not be
present in addition.

Fig. 9-51
7-9-1

V1

V2

V3

FIGURE 9.51. Type 1 Brugada pattern1,2 7-9-1 has a convex (coved) ST segment elevation 2 mm plus
a negative T-wave with little or no isolelectric (baseline) separation in at least 2 leads
of V1-V3

7-9-2 Fig. 9-52

V1

V2

V3

FIGURE 9.52. Type 2 Brugada pattern 7-9-2 has ST segment elevation 2 mm plus a positive or
diphasic T-wave that results in a saddle-back shape in at least 2 leads of V1-V3
131
FIGURE 9.53. Type 3 Brugada pattern 7-9-3 has an ST segment elevation 1 mm plus a saddle-
back conguration in at least 2 leads of V1-V3

FIGURE 9.54. Code 7-10 for fragmented QRS shows different morphologies of a fragmented
QRS on a 12-lead ECG3-6

132
Rules for coding of Fragmented QRS: (1) notching is a sudden change in slope (down or
up) of any part of the Q-, R-, or S-wave such that the angle formed is <90; (2) notching
can occur anywhere from beginning to end of QRS; (3) no restrictions on height or dura-
tion of the notch; (4) small initial R or small Q-waves at the start of the QRS complex are
not considered notching; (5) the notch may not cross the baseline; (6) notching must be
present in a majority of beats in a lead in order to be counted for that lead or ECG; (7) no
notching is found in a lead with 60 cycle interference. This could be noted with technical
problem code 9-8-2.

References

1. Brugada P, Brugada J, Right bundle branch block, persistent ST segment elevation and sudden
cardiac death: a distinct clinical and electrocardiographic syndrome. A multicenter report, J Am
Coll Cardiol 1992; 20:1391-1396.
2. Benito B, Brugada R, Brugada J, Brugada P. Brugada syndrome Prog Cardiovasc Dis. 2008;
Jul-Aug;51(1):1-22. Review.
3. Das MK, Zipes DP. Fragmented QRS: a predictor of mortality and sudden cardiac death. Heart
Rhythm 2009;6(3 Suppl):S8-14.
4. Morita H, Kusano KF, Miura D, Nagase S, Nakamura K, Morita ST, Ohe T, Zipes DP, Wu J.
Fragmented QRS as a marker of conduction abnormality and a predictor of prognosis of Brugada
syndrome. Circulation 2008;118(17):1697-1704.
5. Peters S, Trmmel M, Koehler B. QRS fragmentation in standard ECG as a diagnostic marker of
arrhythmogenic right ventricular dysplasia-cardiomyopathy. Heart Rhythm 2008;(5):1417-1421
6. Das MK, Khan B, Jacob S, Kumar A, Mahenthiran J. Signicance of a fragmented QRS com-
plex versus a Q wave in patient with coronary artery disease. Circulation 2006;113:2495-2501.

133
10
Arrhythmias, 8-Codes
Disordered heart rhythm is represented by 8-codes. Abnormal foci for generating P-waves or QRS-
T complexes can occur from diseased segments of the atria or ventricles. The function of the sinus
node and the conduction through the heart can also be affected by disease processes in the heart
muscle, disordered metabolism, or drug effects. In general, the arrhythmias are signaled by the
presence of unusual beats, irregular R-R intervals, or very fast or very slow heart rates.

Fig. 10-1
8-1-1

22 mm. 22 mm. 15 mm.

FIGURES 10.110.6. Code 8-1-1 is for a supraventricular premature beat (SVPB). SVPB must be
premature by at least 10%, that is, if the normal R to R interval is 22 mm, the
beat in question needs to be at least 2.2 mm (10%) premature or 19.8 mm (see
Fig. 10.1). In addition, the P-wave and/or P-R interval of the abnormal beat
must be absent or different from the normal P and/or P-R interval in the tracing.
In addition to being premature, an SVPB can occur in the following ways: (a)
a P-wave of different form occurring without a QRS complex (see Fig. 10.2),
(b) an absent P-wave occurring with a normal-looking QRS complex (see Fig.
10.3), (c) a different P-wave occurring with a prolonged QRS complex (see Fig.
10.4), (d) a different P-wave occurring with a normal-looking QRS complex
(see Fig. 10.5), or (e) an absent P-wave occurring with a bizarre QRS complex
that is <0.12 second duration (see Fig. 10.6). In summary, to code an 8-1-1, the
beat must be 10% premature and have an absent or different P-wave
134
8-1-1 Fig. 10-2

20 mm. 14 mm. 24 mm.

Fig. 10-3
8-1-1

Fig. 10-4
8-1-1

22 mm. 22 mm. 19 mm. 26 mm.

135
 Fig. 10-5
8-1-1

22 mm. 19 mm. 24 mm.

Fig. 10-6
8-1-1

22 mm. 19.8 mm. 24 mm.

136
FIGURES 10.7 and 10.8. Code 8-1-2 is for a ventricular premature beat (VPB). Unlike an SVPB,
the degree of prematurity of a VPB is not specied for coding. However,
its QRS must be 0.12 second duration (see Fig. 10.7). The duration of
the QRS complex is measured in identical fashion to that for 7-codes. It
must have a different shape or direction from the normal QRS complex
(see Fig. 10.8), and the P-wave must be absent

137
 Fig. 10-9
8-1-2
3 VPB

Fig. 10-10
8-1-2
2 SVPB

FIGURES 10.910.12. In the event that there are two or more SVPBs or VPBs in one lead which
look alike, that is, they have the same shape and direction, they must be coded
all VPB or all SVPB. Measure all the beats to determine how wide they are.
Greater than 50% of the ectopic beats must have a QRS 0.12 second dura-
tion in order to count them all as VPB, and normal and abnormal (of different
shape) P-waves must be absent preceding the VPB (see Fig. 10.9). If only
50% or less of the ectopic beats are 0.12 second, then for all ectopic beats of
the same shape and direction, the same lead must be coded as SVPB (see Fig.
10.10). If more than one ectopic beat occurs in one lead but they are not the
same shape and direction and some beats meet the criteria for a VPB and oth-
ers meet the criteria for a SVPB (but not fusion beats, see below), both VPB
and SVPB may be counted and 8-1-3 is coded (see Fig. 10.11 and 10.12)

138
139
 Fig. 10-13
fusion beat
normal P
bizarre QRS here the duration 0.12 sec.

FIGURES 10.1310.19. A special form of VPB, a fusion beat, may occur soon after a normal P-
wave and can be of different shape and duration from other VPBs, because
the QRS complex is caused by a wave of electrical activity spreading
simultaneously from the atria and the ventricles. This beat may be greater
than or less than 0.12 second duration (see Figs. 10.13 and 10.14). Fusion
beats are counted as VPBs and coded 8-1-2. To identify and code the fusion
beat as a VPB, there must be at least one other clear VPB somewhere else
on the ECG. It may be in any lead. The shape of the fusion beat is usually
intermediate between a normal complex and the other VPB on the record.
If two VPBs occur in an ECG with one fusion beat, then three VPBs are
counted (see Fig. 10.15). However, if an apparent fusion beat is present
with no other VPB present, then no VPB fusion beat is counted and no 8-1-
2 code made (see Fig. 10.16).
However, if this situation occurs (i.e. no clearcut VPB or SVPB present
with the abnormal beat) a 6-code may be appropriate. One possibility is a
6-4-2, which has a normal P-wave with a P-R interval <0.12 second, a QRS
0.12 second and an R peak duration 0.06s. All these conditions must co-
exist in the same beat in any one of the following leads: I, II, aVL, V4, V5,
or V6 and be present in 50% of the beats on the ECG (see Fig. 10.17). An-
other possible 6-code to consider is a 6-6, intermittent aberrant conduction.
A 6-6 has a normal P-wave with a P-R interval > 0.12 second duration and a
wide QRS complex 0.12 second duration. This is coded when most of the
complexes on the ECG are normal sinus beats (see Fig. 10.18). If any VPBs
occurred elsewhere in the record, 8-1-2 would be coded and the abnormal
beat in Fig. 10.18 would be counted as a VPB and there would be no 6-code.
Similarly, in Fig. 10.19, where a VPB appeared elsewhere on the ECG, the
abnormal beat would be counted as a VPB and 8-1-2 would be coded and
there would be no 6-4-2 code

140
 Fig. 10-14
fusion beat
normal P
bizarre QRS here the duration < 0.12 sec.
requires a VPB to be present elsewhere in the ECG

Fig. 10-15
8-1-2
3 VPB

aVL

141
 Fig. 10-16
no VPB
no 8 code

Fig. 10-17
6-4-2
Intermittent WPW

normal P wave
coexisting in the same beat in the
PR < 0.12 sec.
MINORITY of beats in any of leads
QRS 0.12 sec.
I, II aVL, V4-V6
R peak duration 0.06 sec.

142
 Fig. 10-18
6-6
intermittent aberrant conduction
normal P wave
PR 0.12 sec.
QRS 0.12 sec.

Fig. 10-19
8-1-2
2 VPB
I

V6

143
 Fig. 10-20
7-1-1
8-1-2
1 VPB

Fig. 10-21
7-1-1
8-1-1
1 SVPB

FIGURES 10.20 and 10.21. In the presence of a bundle branch block where all beats are 0.12
second, the duration of the ectopic QRS complex is not a determining
factor in classifying it as a VPB or SVPB. In these cases, classication
will be determined by the shape and direction of the ectopic complex.
If the abnormal beat is of a different shape or direction than the usual
beats in the lead, then it will be classied as a VPB provided it is not
preceded by an abnormal P-wave (see Fig. 10.20). If the ectopic com-
plex is of the same shape and direction as the other beats in the lead,
then it will be classied as an SVPB, provided that it is at least 10%
premature (see Fig. 10.21)
144
FIGURE 10.22. Whenever two or more ectopic beats of similar duration occur in one lead and look
different, that is, have a different shape or direction, they are considered mutliform
SVPBs or VPBs (see Fig. 10.22)

FIGURE 10.23. Code 8-1-3 is for the presence of any SVPB (8-1-1) in combination with any VPB
(8-1-2). These two types of ectopic beats do not need to occur in the same lead

145
 Fig. 10-24
8-1-4

Fig. 10-25
8-1-1
1 SVPB

FIGURES 10.24 and 10.25. Code 8-1-4 is for wandering atrial pacemaker. Criteria for this code are
varying P-wave shapes and varying R-R intervals, but without ectopic
beats. Varying P-R intervals may be present but not invariably. The QRS
complexes are normal with one P-wave for each QRS (see Fig. 10.24). The
difference between an 8-1-1 and an 8-1-4 is sometimes slight, but look for
these characteristics for an 8-1-1: absent or unusual P-wave associated with
a premature beat and a regular, normal R-R interval, except for the prema-
ture beats (see Fig. 10.25). In contrast, an 8-1-4 has a generally irregular
R-R interval with no SVPB and varying P-wave form (see Fig. 10.24)

Fig. 10-26
8-1-5

FIGURE 10.26. Code 8-1-5 is a combination of 8-1-4 and 8-1-2

146
 Fig. 10-27
8-2-1
ventricular fibrillation

Fig. 10-28
8-2-1
ventricular asystole

FIGURES 10.27 and 10.28. Code 8-2-1 is for ventricular brillation or ventricular asystole. Ven-
tricular brillation shows irregular undulations of the baseline without
clear P-, QRS-, and T-wave complexes (see Fig. 10.27). In the most
severe case, ventricular brillation will turn into ventricular asystole.
Ventricular asystole is simply a straight line ECG trace with no atrial
or ventricular activity (see Fig. 10.28). This signies that the heart has
stopped functioning

147
 Fig. 10-29
8-2-2
Persistent ventricular rhythm
QRS0.12 sec.
absence of preceding P wave

FIGURES 10.29. Code 8-2-2 is for persistent ventricular rhythm. The characteristics for this code
are wide QRS complexes 0.12 second (usually occurring regularly) and absence
of preceding P-wave. This pattern must persist in all beats of the ECG

Fig. 10-30
8-2-3
Intermittent ventricular tachycardia
3 or more consecutive ventricular premature beats occurring at a rate 100 bpm

FIGURES 10.30. Code 8-2-3 is for intermittent ventricular tachycardia, requiring three or more con-
secutive VPBs at a rate of 100 beats min (bpm)
148
 Fig. 10-31

coupling
interval
=13mm.

8-2-4 Fig. 10-32

ventricular parasystole
14mm. 18mm.

unifocal VPB
coupling interval (shortest to longest) varying by 0.12 sec. (3mm.)

FIGURES 10.31 and 10.32. Code 8-2-4 is for ventricular parasystole, requiring at least two unifocal
VPBs in the ECG in which the shortest to the longest coupling intervals
must vary by 0.12 second. The coupling interval is the distance from
the onset of the normally conducted QRS complex to the onset of the
VPB complex (see Figs. 10.31 and 10.32). If any multiform VPBs
occur, code as 8-1-2 and not 8-2-4

149
 Fig. 10-33
8-3-1
atrial fibrillation

Fig. 10-34
8-3-3
intermittent atrial fibrillation
3 or more consecutive sinus beats in presence of atrial fibrillation

FIGURES 10.33 and 10.34. Code 8-3-1 is for atrial brillation. The criteria for this code are absent
P-waves, irregular undulations in the baseline, normal QRS duration
but sometimes variable shaped QRS complexes, and a totally irregular
ventricular rate. This pattern shall persist in all ECG leads but the cri-
teria are often most identiable in lead V1 (see Fig. 10.33). If three or
more clear-cut consecutive sinus beats occur in any lead in the presence
of atrial brillation, code an 8-3-3, which is intermittent atrial brilla-
tion (see Fig. 10.34)
150
FIGURES 10.35 and 10.36. Code 8-3-2 is for atrial utter. The criteria for this code are a regularly
sharp undulating baseline (resembling a sawtooth edge), normally shaped
QRS complexes, and a regular or irregular ventricular rate varying from
a 2:1 to 8:1 block. The number of sawtoothed F waves between each
QRS complex is frequently constant. The number of F waves is usually
between 2 and 8 per QRS complex. This pattern persists throughout the
ECG (see Fig. 10.35). If there are three or more clear-cut, consecutive
sinus beats in any lead in the presence of atrial utter, code an 8-3-4 for
intermittent atrial utter (see Fig. 10.36)
151
 Fig. 10-37
8-4-1
persistent supraventricular
rhythm
QRS duration < 0.12 sec.
absent or abnormal P
regular rhythm II

III

aVF

Fig. 10-38
8-4-1
absent P waves
regular ventricular rhythm

FIGURES 10.37 and 10.38. Code 8-4-1 is for persistent supraventricular rhythm. The criteria for this
code are normal QRS complexes (<0.12 second duration). A regular
ventricular rhythm, absent or unusual P-waves, and a P-R interval <0.12
second. The P-waves are often negative in leads II, III, and aVF. Re-
member the P-R interval must be <0.12 second in the majority of beats
associated with the abnormal P-waves (see Figs. 10.37 and 10.38)

152
FIGURE 10.39. Code 8-4-2 is for intermittent supraventricular tachycardia. To code this, there must
be three or more consecutive SVPBs occurring at a rate 100 beats min

FIGURES 10.40 10.42. Code 8-5-1 is for sinoatrial arrest. The characteristic of this code is the
unexpected absence, together, of P-, QRS-, and T-waves. There is a regular
ventricular rhythm and the abnormal R-R interval, including one dropped
beat, should be of a duration twice the normal R-R interval 10% (see Fig.
10.40). If two beats are dropped, then the abnormal R-R interval should be
10% of the 3 times the normal R-R interval, etc. (see Fig. 10.41). Usually
the resulting pause is shorter than a xed multiple of the normal R-R in-
tervals. This code is very similar to a 6-2-1 (Mobitz type II), the difference
being that in the case of an 8-5-1 code the P-, QRS-, and T-wave will all be
dropped, while in the case of 6-2-1, the P-wave will occur on time, but the
QRS- and T-waves are dropped (see Fig. 10.42)

153
 Fig. 10-41
8-5-1
sino-atrial arrest

P P P

Fig. 10-42
6-2-1
Mobitz type II

P P P P

154
FIGURES 10.43. Code 8-5-2 is for sinoatrial block. This condition is caused by periodic failure of a
sinus impulse to reach atrial tissue. It is a situation analogous to partial A-V block
but occurring at the level of the sinus node. Criteria require both the following
ndings (a) Characteristics of an 8-5-1 code. An unanticipated failure of P, QRS,
and T to occur at its expected time produces a pause (long R-R interval) which is
an exact multiple of the P-P cycle length, but occasionally may be 3 or 4 times
greater. If one beat is dropped, the R-R interval should be of a duration twice the
normal R-R interval 10%; if two beats are dropped then the abnormal R-R inter-
val should be of duration 3 times the normal R-R interval 10%; etc. (b) PLUS
progressive shortening of P-P intervals preceding dropped P-, QRS-, and T-wave
with the resulting pause less than a multiple of P-P interval, and P-P interval fol-
lowing the pause greater than P-P interval preceding it

FIGURES 10.44. Code 8-6-1 is for A-V dissociation with a ventricular pacemaker without capture
beats. These disturbances are characterized by two independent rhythms (pace-
makers) one sinus or atrial and the other junctional or ventricular. The P-P in-
terval is longer than the R-R interval; the ventricular rate is regular and there is no
xed relation between P-waves and QRS complexes. The P-wave occurs at vari-
able intervals from each QRS complex and appears to move closer to or farther
from the QRS in each cycle. For 8-6 codes, the ventricular rate must be 60 beats
min. If <60 beats min, code 6-1. The QRS duration is 0.12 second
155
 Fig. 10-45
8-6-2
A-V dissociation with ventricular pacemaker with capture
QRS0.12 sec.

FIGURES 10.45. Code 8-6-2 is for A-V dissociation with a ventricular pacemaker and capture beats.
This disturbance is similar to 8-6-1 except that there are momentary irregularities
in the ventricular rate. This appears as an earlier than expected QRS complex,
which is preceded by a P-R interval > 0.12 second. The QRS complex of this
earlier beat is narrower unless there is a conduction defect, in which case the QRS
of captured lead will not be narrower than the regular QRS complexes, which
are 0.12 second

Fig. 10-46
8-6-3
A-V dissociation with atrial pacemaker and no capture beats
QRS 0.12 sec.

FIGURES 10.46. Code 8-6-3 is for A-V dissociation with an atrial pacemaker and no capture
beats. This situation is similar to 8-6-1, but all the QRS complexes are <0.12
second in duration

156
FIGURES 10.47. Code 8-6-4 is for A-V dissociation with an atrial pacemaker and capture beats. This
situation is similar to an 8-6-1 except all QRS complexes including early QRS are
<0.12 second in duration

FIGURES 10.48. When heart rate is not measured or recorded exactly (see Chap. 11), very slow or
very fast sinus rhythm is coded as an arrhythmia. Code 8-7 is for sinus tachycardia.
The R-R intervals occur 15 mm apart to give a heart rate 100 beats min. Normal
P-waves, P-R intervals, and QRS complexes are present. Measure at least 3 R-R
intervals in lead I and average. If there are not 3 R-R intervals in lead I, measure
and average the R-R intervals in leads I and V6
157
 Fig. 10-49
8-8
sinus bradycardia
31mm.
I

V6 32mm. 30mm.

FIGURES 10.49. Code 8-8 is for sinus bradycardia. Here, the R-R intervals average 30 mm apart to
give a heart rate of 50 beats min. Normal P-R intervals and QRS complexes are
present. As with Code 8-7 leads I and V6 may be needed to provide at least 3 R-R
intervals for measurement

Code 8-9 is for other arrhythmias not coded above.

In reporting 8-2-2 (persistent ventricular rhythms) and 8-4-1 (persistent supraventricular
rhythm), the HR should also be stated. Selected heart rates, for example, 60 beats min or
140 beats min may be arbitrarily chosen to report persistent ventricular or supraventricular
tachycardia (as distinct from 8-2-3 and 8-4-2 codes of intermittent ventricular or supraven-
tricular tachycardia).

158
11
Miscellaneous Codes (9-Codes)
These 9-codes are reserved for miscellaneous ECG items 9-1 through 9-5 and 9-6 is for
p-wave abnormality associated with left atrial hypertrophy, and error code 9-8. Code
9-7 is left open for any additional ECG ndings of interest to individual investigators.

9-1 Fig. 11-1

lead I

calibration = 10 mm. QRS peak to peak amplitude

<5mm. in ALL beats in each
of leads I, II, III

lead II
OR

*check calibration before coding

lead III

FIGURES 11.111.6. Code 9-1 is for low QRS amplitude. Criteria are a QRS peak-to-peak ampli-
tude of < 5 mm in every beat in each of leads I, II, and III (see Fig. 11.1) or if
the QRS peak-to-peak amplitude is < 10 mm in every beat in each of leads V1-V6
(see Fig. 11.2). Note how QRS peak-to-peak amplitude is measured. When
the beat has both positive and negative waves, the baseline is included in the
measurement (see Fig. 11.3). When the QRS complex has only a negative
wave (QS), the amplitude is measured from the lower margin of the baseline
at the beginning of the QRS complex to the QS nadir (see Fig. 11.4). When the
QRS complex has only a positive wave (R), the amplitude is measured from
the upper margin of the baseline at the beginning of the QRS complex to the
R peak (see Fig. 11.5). Be certain to check the calibration before coding a 9-1
code (see Fig. 11.6)

See following pages for Figures 11-2 through 11-6.

159
 Fig. 11-2
9-1

QRS peak to peak

amplitude < 10mm.
V1-V6 in ALL beats in each
of leads V1-V6

calibration
= 10 mm.

*check calibration before coding

FIGURE 11.2.

Fig. 11-3

FIGURE 11.3.
160
 Fig. 11-4

FIGURE 11.4.

Fig. 11-5

FIGURE 11.5.

Fig. 11-6

calibration=10mm.

FIGURE 11.6.
161
 Fig. 11-7
9-2
in any of leads I, II, III, aVL, V5 or V6
ST elevation 1 mm.
ST horizontal or upward
sloping

Fig. 11-8
9-2
lead V1-V4
ST elevation 2 mm.

FIGURES 11.7 and 11.8. Code 9-2 is for ST elevation. For this the ST segment must be elevated
1 mm in the majority of beats in any of leads I, II, III, aVL, V5, or V6.
The ST segment may be upward sloping (see Fig. 11.7), horizontal, or
downward sloping. In leads V1-V4 the segment must be elevated 2 mm
in the majority of beats in a lead (see Fig. 11.8)

162
FIGURE 11.9. A 9-2 code very often has a downward sloping ST segment. The gradient for coding
is a slope of 0.5 mm in 0.08 sec

FIGURES 11.1011.17. The method for nding the J-point when it is elevated and not readily appar-
ent is similar to that used for J-point depression. Draw a tangent along the
attest part of the ST segment and extend it toward the QRS complex. J is
the point along the curve where the tangent departs from the ECG trace. A
2 mm at segment is not essential for construction of the tangent (see Figs.
11.1011.13). When an upwardly convex elevated ST segment is present,
measure on the horizontal from the highest point of the curved ST segment
to the upper margin of the preceding P-R baseline (see Fig. 11.1411.17)
163
 Fig. 11-11
9-2

FIGURE 11.11.

Fig. 11-12
9-2

FIGURE 11.12.

Fig. 11-13
9-2

FIGURE 11.13.
164
 Fig. 11-14
9-2
J

FIGURE 11.14.

J
Fig. 11-15
9-2

FIGURE 11.15.

Fig. 11-16
9-2

FIGURE 11.16.
165
 Fig. 11-17
9-2

FIGURE 11.17.

P wave amplitude Fig. 11-18

in any of leads II, III or aVF

Fig. 11-19
9-3

P wave
amplitude 2.5 mm.
in majority of beats
in any of leads II, III or aVF

FIGURES 11.18 and 11.19. Code 9-3 is for high P-wave amplitude, 2.5 mm in the majority of
beats in any of leads II, III, or aVF. The P amplitude is measured from
the upper margin of the baseline of the segment preceding the P-wave
to the P peak
166
 Fig. 11-20
usual chest lead QRS axis transition

V1 V4

V2 V5

V3 V6

FIGURE 11.20.

Fig. 11-21
V1
9-4-1
all beats mainly positive
in any lead V1-V3

V2

V3

FIGURE 11.21.
FIGURES 11.2011.23. Codes 9-4-1 and 9-4-2 are for the QRS transition zone in the leads. Most
often the transition from mainly negative QRS complexes to mainly posi-
tive occurs between V3 and V4. Usually QRS complexes in V1-V3 are mainly
negative and in V4-V6 are mainly positive (see Fig. 11.20). Where the QRS
is not negative in V1-V3 and not positive in V4-V6, either 9-4-1 or 9-4-2 will
be coded. The transition to mainly positive QRS complexes occurring in V1,
V2, or V3 is a 9-4-1 code (see Fig. 11.21). All beats in any one of those leads
must be mainly positive. A positive wave of 5 mm and a negative wave of 6
mm gives a sum of 1 mm (see Fig. 11.22). Note also that only the highest
positive wave and the deepest negative waves are measured for 9-4 codes
(see Fig. 11.23)
167
 Fig. 11-22

QRS mainly negative

-6

FIGURE 11.22.

Fig. 11-23

QRS mainly positive

-3

FIGURE 11.23.

168
 9-4-2 Fig. 11-24
V4
all beats mainly negative
in any lead V4-V6
and all leads negative in V1-V3

V1

V5

V2

V3 V6

FIGURE 11.24. Code 9-4-2 is for transition of a mainly negative QRS to a mainly positive QRS at
V4 or to the left of V4 (i.e., V5 or V6)

Fig. 11-25
9-5
in any of leads I, II, III aVL, aVF, V1-V6
T wave > 12 mm in the majority of beats

FIGURE 11.25. Code 9-5 is for a T-wave amplitude > 12 mm in the majority of beats in any of leads
I, II, III, aVL, aVF, V1-V6. The T-wave is measured from the T-P baseline to the
peak of the T-wave

169
FIGURE 11.26. Code 9-6 is for notched and widened P wave (duration 0.12 second) in frontal
plane (usually lead II), and/or a deep negative component to the P wave in lead V1
(i.e., P-terminal force) duration 0.04 second and depth 1 mm

FIGURE 11.27. Code 9-7 is for early repolarization (ER),1-13 which is an electrocardiographic (ECG)
pattern characterized by prominent J point and ST-segment elevation with upward
concavity ending in a positive high amplitude T wave in most ECG leads especially
the midchest leads thought for decades to be benign but now possibly related to
the Brugada syndrome (see Chap. 9).
Code 9-7-1 is for denite ER: STJ elevation 1 mm in the majority of beats, plus
T wave amplitude 5 mm, prominent J point, upward concavity of the ST, plus a
distinct notch or slur on the down-stroke of the R wave in any of V3-V6; OR STJ
elevation 2 mm in the majority of beats and T wave amplitude 5 mm, promi-
nent J point, and upward concavity of the ST segment in any of V3-V6 without the
distinct notch on the downstroke of the R-wave.

170
FIGURE 11.28. Code 9-7-2 is for probable ER: STJ elevation 1 mm in the majority of beats,
prominent J point, and upward concavity of the ST segment in any of V3-V6 and T
wave amplitude 8 mm in any of the chest leads
ER may be more prominent in limb leads in about 15% of people and these may be
coded separately or use code 9-2 or 9-5 to screen for possible ER in the lead groups
with limb leads (see Chapter 13).
Incompatible codes, codes that should suppress the ER code: 6-1, 6-4-1. 6-8, 7-1-1,
7-2-1, 7-4, 8-2-1, 8-2-2.
Lead Reversals
Misplacement of electrodes cause a subgroup of poor quality ECGs. These are commonly
called lead reversals and can often be recognized by the particular patterns produced. Some
lead reversals are correctable by visual inspection and coding can correct the error (correctable
lead reversals). Other lead reversals produce tracings that cannot be coded (uncorrectable lead
reversals). In electronic analysis of digital ECGs, lead reversals are often misdiagnosed.
Correct electrode cable connection for limb leads should exactly be placed as in Fig. 14.4
(Chap. 14), and the chest lead placements should follow the order V1, V2, V3, V4, V5, and
V6 as in Fig. 14.9.
Minnesota Code 9-8-X codes are used for coding lead reversals and poor quality/technical
problems that interfere with coding. If a future division into subtypes is required, the accom-
panying Novacode (see Appendix B) subtypes (0.3.x.x.) can be used. These are illustrated
in the following gures:

(a) 9-8-1 Uncorrectable lead reversal

(Novacode 0.3.1.1; 0.3.1.2; 0.3.1.3) Fig. 11.2911.31

(b) 9-8-2 Poor quality/technical problems which interfere with coding Fig. 11.32

(c) 9-8-3 Correctable lead reversal

i. Correctable limb lead connection error
(Novacode 0.3.2.1; 0.3.2.2; 0.3.2.3; 0.3.2.4; 0.3.2.5) Fig. 11.3311.37
ii. Correctable chest lead connection errors in V1-V3
(Novacode 0.3.3.1; 0.3.3.2; 0.3.3.3; 0.3.3.4; 0.3.3.5) Fig. 11.3811.42
iii. Correctable chest lead connection errors in V4-V6
(Novacode 0.3.4.1; 0.3.4.2; 0.3.4.3; 0.3.4.4; 0.3.4.5) Fig. 11.4311.47
iv. Correctable other chest lead connection errors
(Novacode 0.3.5.1; 0.3.5.2; 0.3.5.3; 0.3.5.4; 0.3.5.9) Fig. 11.4811.51

(d) 9-8-4 Technical problems that do not interfere with coding Fig. 11.52
171
 TABLE 11.1. Electrode cable connection errors in the limb leads
Minnesota Novacode Electrode cable Resulting
Code misplacements pattern

9-8-0 0.3.0 Normal ECG RA LA

RL LL
9-8-1 0.3.1.1 RL LA LL RA RL LL RL
RA LA
9-8-1 0.3.1.2 RL RA RL LA
RA LL
9-8-1 0.3.1.3 RL LA LL RL RA RL
LL LA
9-8-3 0.3.2.1 RA LA LA RA
RL LL
9-8-3 0.3.2.2 RA LL LL LA
RL RA
9-8-3 0.3.2.3 LA LL RA LL
RL LA
9-8-3 0.3.2.4 RA LA LL RA LL RA
RL LA
9-8-3 0.3.2.5 RA LL LA RA LA LL
RL RA
RA electrode cable on right arm; LA electrode cable on left arm; RL electrode cable on right leg; LL electrode cable on left leg

9-8-1 Fig. 11-29

(RL LA LL RA RL)
Lead I -- Flat
Lead II, III, aVF -- QRS are mainly negative;
Lead aVR, aVL -- QRS are mainly upright

Correct Leads

Lead reversal

Uncorrectable limb lead reversal code: Minnesota code 9-8-1 (Novacode 0.3.1.1)

FIGURE 11.29.

(a) 9-8-1 Uncorrectable Lead Reversal (Novacode 0.3.1.1; 0.3.1.2; 0.3.1.3) Fig. 11.2911.31.
* See Table 11.1 for reversal lead arrangments
172
 9-8-1 Fig. 11-30
(RA RL)
Lead II Flat
Lead I, aVL QRS are mainly negative;
Lead III, aVR, aVF QRS are mainly upright
P Negative in Lead I and aVL;
P Positive in Lead aVR.

Correct Leads

Lead reversal

Uncorrectable limb lead reversal code: Minnesota code 9-8-1 (Novacode 0.3.1.2)

FIGURE 11.30.

9-8-1 Fig. 11-31

(RL LA LL RL)
Lead III Flat
Lead I, II, aVL, aVF -- QRS are mainly upright

Correct Leads

Lead reversal

Uncorrectable limb lead reversal code: Minnesota code 9-8-1 (Novacode 0.3.1.3)

FIGURE 11.31.
173
 Fig. 11-32

9-8-2
example
possible 3-1
top of beat cut off

V5

(b) 9-8-2 Poor quality/technical problems that interfere with coding Fig. 11.32.

FIGURE 11.33.

9-8-3 Fig. 11-33

(RA LA)
Lead I Mirror image down;* Lead aVR Lead aVL;
Lead II Lead III; Lead aVL Lead aVR;
Lead III Lead II; Lead aVF Unchanged.

Correct Leads

Lead reversal

Correctable limb lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.2.1)

(c) 9-8-3 Correctable lead reversal Fig. 11.3311.51

i) Correctable limb lead connection error Fig. 11.3311.37
(Novacode 0.3.2.1; 0.3.2.2; 0.3.2.3; 0.3.2.4; 0.3.2.5)
* Mirror image down = reversed polarity
174
 9-8-3 Fig. 11-34
(RA LL)
Lead I Lead III mirror image down; Lead aVR Lead aVF;
Lead II Lead II mirror image Lead aVL Unchanged;
Lead III Lead I mirror image Lead aVF Lead aVR.

Correct Leads

Lead reversal

Correctable limb lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.2.2)

FIGURE 11.34.

9-8-3 Fig. 11-35

(LA LL)
Lead I Lead II; Lead aVR Unchanged;
Lead II Lead I; Lead aVL Lead aVF;
Lead III Lead III mirror image down; Lead aVF Lead aVL.

Correct Leads

Lead reversal

Correctable limb lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.2.3)

FIGURE 11.35.
175
9-8-3 Fig. 11-36
(RA LA LL)
Lead I Lead II mirror image down; Lead aVR Lead aVF;
Lead II Lead III mirror image down; Lead aVL Lead aVR;
Lead III Lead I; Lead aVF Lead aVL.

Correct Leads

Lead reversal

Correctable limb lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.2.4)

FIGURE 11.36.

Fig. 11-37
9-8-3
(RA LL LA)
Lead I Lead III; Lead aVR Lead aVL;
Lead II Lead I mirror image down; Lead aVL Lead aVF;
Lead III Lead II mirror image down; Lead aVF Lead aVR.

Correct Leads

Lead reversal

Correctable limb lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.2.5)

FIGURE 11.37.
176
 9-8-3 Fig. 11-38
(V1 / V2)

Correct Leads

Lead reversal

Correctable chest lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.3.1)

FIGURE 11.38.

(ii) Correctable chest lead connection errors in V1-V3 Figs. 11.3811.42

(Novacode 0.3.3.1; 0.3.3.2; 0.3.3.3; 0.3.3.4; 0.3.3.5)

9-8-3 Fig. 11-39

(V1 / V3)

Correct Leads

Lead reversal

Correctable chest lead reversal code: Minnesota code 9-8-3 (Novacode 0.3.3.2)

FIGURE 11.39.
177
FIGURE 11.40.

FIGURE 11.41.
178
FIGURE 11.46.

FIGURE 11.47.
181
FIGURE 11.50.

FIGURE 11.51.

183
References
1. Gussak I, Antzelevitch C. Early repolarization syndrome: clinical characteristics and possible
cellular and ionic mechanisms. J Electrocardiol. 2000;33:299-309.
2. Klatsky A, Oehm R, Cooper R, et al. The Early Repolarization normal variant electrocar-
diogram: correlates and consequences. Am J Med. 2003;115:171-177.
3. Vacanti LJ. Thoracic pain and early repolarization syndrome at the cardiologic emergency unit.
Arq Bras Cardiol. 1996;67:335-338.
4. Balady GJ, Cadigan JB, Ryan TJ. Electrocardiogram of the athlete: an analysis of 289 profes-
sional football players. Am J Cardiol. 1984;53:1339-1343.
5. Akhmedov NA. Early ventricular repolarization syndrome and heart function in the inhabitants
of Asia, Africa, and Latin America. Kardiologiia. 1986;26:63-65.
6. Mehta MC, Jain AC. Early repolarization on scalar electrocardiogram. Am J Med Sci. 1995;309:305-
311.
7. Marriott HJL: Practical Electrocardiography. 7th ed. Baltimore, MD; Williams & Wilkins,
1988.
8. Riera A, Uchida A, Schapachnik E, et al. Early repolarization variant: Epidemiological aspects,
mechanism, and differential diagnosis. Cardiol J. 2008 (article in press, www.cardiologyjour-
nal.org).
9. Boineau J. The early repolarization variantnormal or a marker of heart disease in certain
subjects. J. Electrocardiol. 2007;40: 3.e11-3.e16.
10. Boineau JP. The early repolarization variantan electrocardiographic enigma with both QRS
and J-STT anomalies. J Electrocardiol. 2007;40(1):3.e1-10.
11. Lux RL. Early repolarization variant: interesting electrocardiographic anomaly or marker of
arrhythmogenic risk? J Electrocardiol. 2007;40(1):4-5.
12. Dowdy L, Wagner GS, Birnbaum Y, et al. Early repolarization: friend or foe? Am J Med. 2003;115:237-
240.
13. Wasserburger RH, Alt WJ. The normal RS-T segment elevation variant. Am J Cardiol. 1961;8:184-
192.

186
12
Exact Measurements
In addition to codes 19, other measurements may be required to characterize the ECG in
population studies. Coding of the actual heart rate (apart from 8-7 and 8-8) may provide
important predictive information. Continuous heart rate and frontal plane axis measurements
also allow the computer to choose different cutoff points for dening certain arrhythmias.
When exact amplitude measurements are made, less severe forms of cardiac hypertrophy
can be followed than by sole use of 3-codes.
Heart rate and frontal plane QRS axis are measured with the use of charts that convert
intervals to rates and amplitudes to axes. Other ECG items may be measured including P-R
interval, Q-T interval, etc.

FIGURE 12.1. The heart rate (HR) is calculated for each ECG as beats/min in lead I. Measure
the distance in millimeters between the rst four complete sinus beats in lead I, that
is, three complete R-R intervals. The measurement should be made with a plastic
ruler to the nearest 0.5 mm

On the HR chart (Table 12.1), the distance in column A is transposed to HR in column D.

This chart is only for measurement of HR in ECGs recorded at 25 mm/second paper speed

187
TABLE 12.1. Heart rate per minute from R-R intervals in millimeters.
5-RR 3-RR 1-RR HR 5-RR 3-RR 1-RR HR 5-RR 3-RR 1-RR HR

A B C D A B C D A B C D
25 15 5 300 95 57 19 79 165 99 33 45
27.5 16.5 5.5 273 97.5 58.5 19.5 77 167.5 100.5 33.5 45
30 18 6 250 100 60 20 75 170 102 34 44
32.5 19.5 6.5 231 102.5 61.5 20.5 73 172.5 103.5 34.5 43
35 21 7 214 105 63 21 71 175 105 35 43
37.5 22.5 7.5 200 107.5 64.5 21.5 70 177.5 106.5 35.5 42
40 24 8 188 110 66 22 68 180 108 36 42
42.5 25.5 8.5 176 112.5 67.5 22.5 67 182.5 109.5 36.5 41
45 27 9 167 115 69 23 65 185 111 37 41
47.5 28.5 9.5 158 117.5 70.5 23.5 64 187.5 112.5 37.5 40
50 30 10 150 120 72 24 63 190 114 38 39
52.5 31.5 10.5 143 122.5 73.5 24.5 61 192.5 115.5 38.5 39
55 33 11 136 125 75 25 60 195 117 39 38
57.5 34.5 11.5 130 127.5 76.5 25.5 59 197.5 118.5 39.5 38
60 36 12 125 130 78 26 58 200 120 40 38
62.5 37.5 12.5 120 132.5 79.5 26.5 57 202.5 121.5 40.5 37
65 39 13 115 135 81 27 56 205 123 41 37
67.5 40.5 13.5 111 137.5 82.5 27.5 55 207.5 124.5 41.5 36
70 42 14 107 140 84 28 54 210 126 42 36
72.5 43.5 14.5 103 142.5 85.5 28.5 53 212.5 127.5 42.5 35
75 45 15 100 145 87 29 52 215 129 43 35
77.5 46.5 15.5 97 147.5 88.5 29.5 51 217.5 130.5 43.5 34
80 48 16 94 150 90 30 50 220 132 44 34
82.5 49.5 16.5 91 152.5 91.5 30.5 49 222.5 133.5 44.5 34
85 51 17 88 155 93 31 48 225 135 45 33
87.5 52.5 17.5 86 157.5 94.5 31.5 48 227.5 136.5 45.5 33
90 54 18 83 160 96 32 47 230 138 46 33
92.5 55.5 18.5 81 162.5 97.5 32.5 46 232.5 139.5 46.5 32
Using a Metric Ruler to measure 5 R to R intervals (5RRa), or 3 R to R intervals (3RRb), or 1 R to R interval (1RRc) in
millimeters (mm). The Table lists the corresponding heart rate (HR)
a
HR = 7,500 mm/5RR;
b
HR = 4,500 mm/3RR;
c
HR = 1,500 mm/1RR

188
 Fig. 12-2

42 mm.
I

21 mm.

V6

63 mm. = a heart rate of 71 beats per minute

Fig. 12-3

I
20mm.

V6 40mm.

60 mm = a heart rate of 75 beats per minute

FIGURES 12.2 and 12.3. In the event that there are less than three complete normal R-R intervals in
lead I, use a combination of three intervals from leads I and V6
189
 Fig. 12-4
45 mm.

45mm. = a heart rate of 100 beats per minute

V6 48 mm.

48 mm. = a heart rate of 94 beats per minute

48 mm. + 45 mm. = 93 mm.

93 mm. 2 = 46.5 mm.
46.5 mm.=a heart rate of 97 beats per minute

FIGURE 12.4. Measure three R-R intervals in lead I and three R-R intervals in lead V6 and average
the two HR estimates from the chart. Look up that number in column B of the HR chart
to determine the HR

190
Frontal Plane QRS Axis

Fig. 12-5

I
6

-3

III

-5

I +3
III -2
QRS axis -11

FIGURE 12.5. The rst step in determining QRS axis is to measure the next to last complete normal
beat in lead I. Measure the largest positive and negative wave of the QRS complex
to the nearest 0.25 mm and add them together algebraically. Do the same for the next
to last complete normal beat in lead III

Next, look up these values on the QRS axis charts shown in Table 12.2.1 In the example
shown in Fig. 12.5, lead I positive, lead III negative, look across the top of the chart along
the lead axis row and nd the number 3. Then look down the lead III axis column and nd
2. Read across and down to nd the corresponding axis value, in the case, 11.

1
From Friedman HH: Diagnostic Electrocardiography and Vectrocardiography. New York, McGraw-Hill Co.,
1970, pp 470-473. (Reprinted with permission)
191
 TABLE 12.2. QRS axis charts.
Lead I Positive (+)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 20.0
0.0 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30
0.5 90 60 49 44 41 39 38 37 36 35 35 34 33 33 33 32 32 32 32 32 32 31
1.0 90 71 60 53 49 46 44 42 41 40 39 38 37 36 35 35 34 34 34 33 33 32
1.5 90 76 67 60 55 52 49 47 45 44 43 41 39 38 38 37 36 36 36 35 35 33
2.0 90 79 71 65 60 56 53 51 49 47 46 44 42 41 40 39 38 38 37 37 36 35
2.5 90 81 74 68 64 60 57 54 52 51 49 47 45 43 42 41 40 39 39 38 38 36
3.0 90 82 76 71 67 63 60 57 55 53 52 49 47 45 44 43 42 41 40 39 39 37
Positive (+)

3.5 90 83 78 73 69 66 63 60 58 56 54 51 49 47 46 44 43 42 42 41 40 38
4.0 90 84 79 75 71 68 65 62 60 58 56 53 51 49 47 46 45 44 43 42 42 39
4.5 90 85 80 76 73 69 67 64 62 60 58 55 53 51 49 48 47 45 44 43 43 40
5.0 90 85 81 77 74 71 68 66 64 62 60 57 55 52 51 49 48 47 46 45 44 41
6.0 90 86 82 79 76 73 71 69 67 65 63 60 57 55 53 52 50 49 48 47 46 43
Lead III

7.0 90 87 83 81 78 75 73 71 69 67 65 63 60 58 56 54 53 51 50 49 48 44
8.0 90 87 84 82 79 77 75 73 71 69 68 65 62 60 58 56 55 53 52 51 50 46
9.0 90 87 85 82 80 78 76 74 73 71 69 67 64 62 60 58 57 55 54 53 52 48
10.0 90 88 85 83 81 79 77 76 74 72 71 68 66 64 62 60 58 57 56 54 53 49
11.0 90 88 86 84 82 80 78 77 75 73 72 70 67 65 63 62 60 59 57 56 55 50
12.0 90 88 86 84 82 81 79 78 76 75 73 71 69 67 65 63 61 60 59 57 56 52
13.0 90 88 86 84 83 81 80 78 77 76 74 72 70 68 66 64 63 61 60 59 58 53
14.0 90 88 87 85 83 82 80 79 78 77 75 73 71 69 67 66 64 63 61 60 59 55
15.0 90 88 87 85 84 82 81 80 78 77 76 74 72 70 68 67 65 64 62 61 60 55
20.0 90 89 88 87 85 84 83 82 81 80 79 77 76 74 70 71 70 68 67 65 65 60

Lead I Positive (+)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 20.0
0.0 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30
0.5 -90 -30 0 11 16 19 21 22 23 24 25 26 26 27 27 27 28 28 28 28 28 29
1.0 -90 -60 -30 -11 0 7 11 14 16 18 19 21 22 23 24 25 25 26 26 26 27 27
1.5 -90 -71 -49 -30 -16 -7 0 5 7 11 13 16 18 20 21 22 23 23 24 24 25 26
2.0 -90 -76 -60 -44 -30 -19 -11 -5 0 4 7 11 14 16 18 19 20 21 22 22 23 25
2.5 -90 -79 -67 -53 -41 -30 -21 -14 -8 -4 0 6 9 12 14 16 17 19 20 20 21 23
3.0 -90 -81 -71 -60 -49 -39 -30 -22 -16 -11 -7 0 5 8 11 13 15 16 17 18 19 22
Negative (-)

3.5 -90 -82 -74 -65 -55 -46 -38 -30 -23 -18 -13 -6 0 4 7 10 12 14 15 16 17 21
4.0 -90 -83 -76 -68 -60 -52 -44 -37 -30 -24 -19 -11 -5 0 4 7 9 11 13 14 15 19
4.5 -90 -84 -78 -71 -64 -56 -49 -42 -36 -30 -25 -16 -9 -4 0 3 6 8 10 12 13 18
5.0 -90 -85 -79 -73 -67 -60 -53 -47 -41 -35 -30 -21 -14 -8 -4 0 3 6 8 9 11 16
6.0 -90 -86 -81 -76 -71 -66 -60 -54 -49 -44 -39 -30 -22 -16 -11 -7 -3 0 3 5 7 13
Lead III

7.0 -90 -86 -82 -78 -74 -69 -65 -60 -55 -51 -46 -38 -30 -23 -18 -13 -9 -6 -3 0 2 10
8.0 -90 -87 -83 -80 -76 -72 -68 -64 -60 -56 -52 -44 -37 -30 -24 -19 -15 -11 -8 -5 -2 7
9.0 -90 -87 -84 -81 -78 -74 -71 -67 -64 -60 -56 -49 -42 -36 -30 -25 -20 -16 -13 -9 -7 3
10.0 -90 -87 -85 -82 -79 -76 -73 -70 -67 -63 -60 -53 -47 -41 -35 -30 -25 -21 -17 -14 -11 0
11.0 -90 -88 -85 -83 -80 -77 -75 -72 -69 -66 -63 -57 -51 -45 -40 -35 -30 -26 -22 -18 -15 -3
12.0 -90 -88 -86 -83 -81 -79 -76 -74 -71 -68 -66 -60 -54 -49 -44 -39 -34 -30 -26 -22 -19 -7
13.0 -90 -88 -86 -84 -82 -80 -77 -75 -73 -70 -68 -63 -57 -52 -47 -43 -38 -34 -30 -26 -23 -10
14.0 -90 -88 -86 -84 -82 -80 -78 -76 -74 -72 -69 -65 -60 -55 -51 -46 -42 -38 -34 -30 -27 -13
15.0 -90 -88 -87 -85 -83 -81 -79 -77 -75 -73 -71 -67 -62 -58 -53 -49 -45 -41 -37 -33 -30 -16
20.0 -90 -89 -87 -86 -85 -83 -82 -81 -79 -78 -76 -73 -70 -67 -63 -60 -57 -53 -50 -47 -44 -30

192
 TABLE 12.2. continued

Lead I Negative (-)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 20.0
0.0 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150
0.5 -90 -120 -131 -136 -139 -141 -142 -143 -144 -145 -145 -146 -147 -147 -147 -148 -148 -148 -148 -148 -148 -149
1.0 -90 -109 -120 -127 -131 -134 -136 -138 -139 -140 -141 -142 -143 -144 -145 -145 -146 -146 -146 -147 -147 -148
1.5 -90 -104 -113 -120 -125 -128 -131 -133 -135 -136 -137 -139 -141 -142 -142 -143 -144 -144 -144 -145 -145 -147
2.0 -90 -101 -109 -115 -120 -124 -127 -129 -131 -133 -134 -136 -138 -139 -140 -141 -142 -142 -143 -143 -144 -145
2.5 -90 -99 -106 -112 -116 -120 -123 -136 -128 -129 -131 -133 -135 -137 -138 -139 -140 -141 -141 -142 -142 -144
3.0 -90 -98 -104 -109 -113 -117 -120 -123 -125 -127 -128 -131 -133 -135 -136 -137 -138 -139 -140 -141 -141 -143
Negative (-)

3.5 -90 -97 -102 -107 -111 -114 -117 -120 -122 -124 -124 -129 -131 -133 -134 -136 -137 -138 -138 -139 -140 -142
4.0 -90 -96 -101 -105 -109 -112 -115 -118 -120 -122 -124 -127 -129 -131 -133 -134 -135 -136 -137 -138 -138 -141
4.5 -90 -95 -100 -104 -107 -111 -113 -116 -118 -120 -122 -125 -127 -129 131 -132 -133 -135 -136 -137 -137 -140
5.0 -90 -95 -99 -103 -106 -109 -112 -114 -116 -118 -120 -123 -125 -128 -129 -131 -132 -133 -134 -135 -136 -139
6.0 -90 -94 -98 -101 -104 -107 -109 -111 -113 -115 -117 -120 -123 -125 -127 -128 -130 -131 -132 -133 -134 -137
Lead III

7.0 -90 -93 -97 -99 -102 -105 -107 -109 -111 -113 -115 -117 -120 -122 -124 -126 -127 -129 -130 -131 -132 -136
8.0 -90 -93 -96 -98 -101 -103 -105 -107 -109 -111 -112 -115 -118 -120 -122 -124 -125 -127 -128 -129 -130 -134
9.0 -90 -93 -95 -98 -100 -102 -105 -106 -107 -109 -111 -113 -116 -118 -120 -122 -123 -125 -126 -127 -128 -132
10.0 -90 -92 -95 -97 -99 -101 -103 -104 -106 -108 -109 -112 -114 -116 -118 -120 -122 -123 -124 -126 -127 -131
11.0 -90 -92 -94 -96 -98 -100 -102 -103 -105 -107 -108 -110 -113 -115 -117 -118 -120 -121 -123 -124 -125 -130
12.0 -90 -92 -94 -96 -98 -99 -101 -102 -104 -105 -107 -109 -111 -113 -115 -117 -119 -120 -121 -123 -124 -128
13.0 -90 -92 -94 -96 -97 -99 -100 -102 -103 -104 -106 -108 -110 -112 -114 -116 -117 -119 -120 -121 -122 -127
14.0 -90 -92 -93 -95 -97 -98 -100 -101 -102 -103 -105 -107 -109 -111 -113 -114 -116 -117 -119 -120 -121 -125
15.0 -90 -92 -93 -95 -96 -98 -99 -100 -102 -103 -104 -106 -108 -110 -112 -113 -115 -116 -118 -119 -120 -125
20.0 -90 -91 -92 -93 -95 -96 -97 -98 -99 -100 -101 -103 -104 -106 -108 -109 -110 -112 -113 -115 -115 -120

Lead I Negative (-)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 20.0
0.0 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150 -150
0.5 90 150 180 -169 -164 -161 -159 -158 -157 -156 -155 -154 -154 -153 -153 -153 -152 -152 -152 -152 -152 -151
1.0 90 120 150 169 180 -173 -169 -166 -164 -162 -161 -159 -158 -157 -156 -155 -155 -154 -154 -154 -153 -153
1.5 90 109 131 150 164 173 180 -175 -172 -169 -167 -164 -162 -160 -159 -158 -157 -157 -156 -156 -155 -154
2.0 90 104 120 136 150 161 169 175 180 -176 -173 -169 -166 -164 -162 -161 -160 -159 -158 -158 -157 -155
2.5 90 101 113 127 139 150 159 166 172 176 180 -174 -171 -168 -166 -164 -163 -161 -160 -160 -159 -157
3.0 90 99 109 120 131 141 150 158 164 169 173 180 -175 -172 -169 167 -165 -164 -163 -162 -161 -158
Positive (+)

3.5 90 98 106 115 125 134 142 150 157 162 167 174 180 -176 -173 -170 -168 -166 -165 -164 -163 -159
4.0 90 97 104 112 120 128 136 143 150 156 161 169 175 180 -176 -173 -171 -169 -167 -166 -165 -161
4.5 90 96 102 109 116 124 131 138 144 150 155 164 171 176 180 -177 -174 -172 -170 -168 -167 -162
5.0 90 95 101 107 113 120 127 133 139 145 150 159 166 172 176 180 -177 -174 -172 -171 -169 -164
6.0 90 94 99 104 109 114 120 126 131 136 141 150 158 164 169 173 177 180 -177 -175 -173 -167
Lead III

7.0 90 94 98 102 106 111 115 120 125 129 134 142 150 157 162 167 171 174 177 180 -178 -170
8.0 90 93 97 100 104 108 112 116 120 124 128 136 143 150 156 161 165 169 172 175 178 -173
9.0 90 93 96 99 102 106 109 113 116 120 124 131 138 144 150 155 160 164 167 171 173 -177
10.0 90 93 95 98 101 104 107 110 113 117 120 127 133 139 145 150 155 159 163 166 169 180
11.0 90 92 95 97 100 103 105 108 111 114 117 123 129 135 140 145 150 154 158 162 165 177
12.0 90 92 94 97 99 101 104 106 109 112 114 120 126 131 136 141 146 150 154 158 161 173
13.0 90 92 94 96 98 100 103 105 107 110 112 117 123 128 133 137 142 146 150 154 157 170
14.0 90 92 94 96 98 100 102 104 106 108 111 115 120 125 129 134 138 142 146 150 153 167
15.0 90 92 93 95 97 99 101 103 105 107 109 113 118 122 127 131 135 139 143 147 150 164
20.0 90 91 93 94 95 97 98 99 101 102 104 107 110 113 117 120 123 127 130 133 136 150

193
 Fig. 12-6
QRS axis

7.5

-3.75

2.25

-6.75

I 3.75 round up to 4
III -4.5 remains -4.5
QRS axis -36

QRS axis Fig. 12-7

I
6

-0.5

III

4.5

-4.5

I 5.5 round up to 6
III 0 remains 0
QRS axis 30

FIGURES 12.6 and 12.7. Between 0 and 5 mm, the axis chart is divided in increments of 0.5 mm.
If the sum of the measurements of the positive and negative waves is
not exactly a whole number or on the 0.5 mm mark, round to the nearest
whole or 0.5 mark above. For example, 1.25 mm is rounded off to 1.5
and 1.75 is rounded off to 2 mm. From 5 to 15 mm, the chart is divided
into increments of 1 mm. Always round off the sum of the positive and
negative waves in leads I and III to an even number. As noted earlier, the
separate positive and negative waves in leads I and III are measured to the
nearest 0.25 mm. Therefore, a sum of positive and negative waves of 7.5
mm is rounded off to 8 mm, 7.75 mm is rounded off to 8 mm, 7.25 mm is
rounded off to 7 mm, 6.75 mm is rounded off to 7 mm, etc
194
FIGURE 12.8. For values above 15 mm, divide both values from lead I and lead III in half and read
those values from the chart

195
Amplitude Measurements

maximum in leads I, II, III Fig. 12-09

11

7.5

maximum R is 11 mm.

maximum R wave in leads V4, V5, V6

Fig. 12-10

10

maximum R is 10 mm.

FIGURES 12.9 and 12.10. R-wave amplitude is measured to the nearest whole millimeter in the
next to last complete normal beat in the appropriate leads. For example,
the maximum R-wave in lead I, II, and III is the highest R-wave in the
next to last beat of lead I or II or III (see Fig. 12.9). A similar measure-
ment is made in V4, V5, and V6 (see Fig. 12.10)
196
 maximum S wave in leads I, II, III Fig. 12-11

10

maximum S is 10 mm.

maximum S wave in leads V1, V2, V3 Fig. 12-12

12

maximum S is 12 mm.

FIGURES 12.11 and 12.12. The maximum S-wave or QS is measured in the same manner as the
maximum R-wave

197
 Fig. 12-13
T wave measurement in V5

2
positive T wave

-3 mm

negative T wave

FIGURE 12.13. The maximum T-wave measurement is taken only in lead V5 on the next to last beat.
It is measured to the nearest whole millimeter. The measurement may be positive
or negative

Fig. 12-14
T wave measurement

-2 mm

FIGURE 12.14. If the T-wave in V5 is diphasic and a negativepositive type, measure only the nega-
tive part of the T-wave

198
 Fig. 12-15
T wave measurement

V5

-2mm
-2mm

Fig. 12-16
T wave measurement

V5

1.5 mm

-0.5 mm

T wave is 1 mm

FIGURES 12.15 and 12.16 If the T-wave in V5 is diphasic and a positivenegative type, with the
negative part 1 mm, measure only the negative part (see Fig. 12.15).
If the negative part is <1 mm, take the algebraic sum of the positive and
negative parts (see Fig. 12.16)

199
Q-X, Q-T Intervals

FIGURE 12.17. The Q-X and Q-T intervals are measured with the loupe in hundredths of a second
on the next to last complete normal beat in the lead to be measured. The Q-X is the
distance along a horizontal tangent from the upper edge of the baseline from the onset
of the QRS complex to the point where the beginning of the T-wave intersects this
line. This measurement starts at the beginning of the QRS complex. Whenever the ST
segment is elevated, or there is any 4-code present, or when the T-wave is at or zero,
Q-X cannot be measured in the next to last beat in lead I, any other beat in lead I may
be used. Choose the beat in lead I most clear, normal and free from noise

Heart rate affects the QT interval, which is of longer duration with slower heart rates. The
commonest clinical correction for this heart rate variability of the QT interval is Bazetts
QTc. Unfortunately this is a awed correction that is quite misleading at high and low heart
rates. For a full discussion of rate corrections, see ref. 2. An easy formulation without the
problems of QTc is QTI or QT index = (QT interval/656)(HR + 100); similarly JTI = (JT
interval/518)(HR + 100). Change in the QT interval is a measure used by national drug
agencies and pharmaceutical companies to test the safety of new drugs. Relative increase
in the rate corrected QT indicates an arrhythmogenic potential, so exact measurement is
important. The repeatability of measurement should be high (see Table 17.3, 17.4 and 17.6
in Chap. 17).

200
 Fig. 12-18

measure QT from
the beginning of
the QRS to the end
of the T wave

FIGURE 12.18. The Q-T interval is the distance from the onset of the QRS complex to the end the
T-wave. Measure from the onset of the QRS to the offset of the T-wave. If the Q-T
cannot be measured on the next to last complete beat, any other beat may be used.
Choose the most clear normal beat. For tracings from digital recording, a median or
average beat will be available for measurement. Measure the maximum QT interval
in the ECGusually in V2 or V3. If quality is poor in these leads, then move to lead
aVR

201
 Fig. 12-19
The end of T-wave without clear offset

FIGURE 12.19. If the end of the T-wave is without clear offset, the Q-T interval is the distance from
the beginning of QRS wave to the baseline crossing point of the tangent of the end
of the T-wave and the baseline (representing the isoelectric line). In such cases, a
tangent is drawn along the inner margin of the down slope of the T-wave

T-wave followed by a U-wave Fig. 12-20

FIGURE 12.20. When the T-wave is followed by a U-wave. The end point of T-wave can be decided
by the same method as in Fig. 12.19. The Q-T interval is the distance from the be-
ginning of QRS wave to the crossing point of the tangent of end of the T-wave and
the baseline
References

1. Friedman HH: Diagnostic Electrocardiography and Vectrocardiography. New York,

McGraw-Hill 1970, pp 470-473. Reprinted with permission
2. Rantaraju P and Rantaraju F: Ivestigative Electrocardiography in Epidemilogical Studies and
Clinical Trials. Newyork: Springer; 2007.
202
13
Coding the Whole ECG
The complete ECG is scanned with the code kept as a reference. First, Q-QS waves are detected
and coded. Then 2-codes or frontal plane axis is recorded, and so on in order for 39-codes.
Practice soon leads to detection of all codable ndings, which are then coded in order.
Absence of codes will soon be recognized. Single coding rates of 3050 ECGs/hour can
be expected for normal records. Coding rates for records in cardiac patients are likely to be
much slower, from 9 to 15 per hour.
In some schemes, particularly for patient populations, coding is made separately for each
of three major lead groups: anterior lead group (leads V1-V5), anterolateral lead group (leads
I, aVL, V6), and posterior (inferior) lead group (leads II, III, aVF). Lead aVR is not used
for coding purposes except to identify switched arm leads or for counting premature beats.
Within each lead group, the major code found for that group is recorded. If a 1-1 pattern and
a 1-2 pattern are present, only the 1-1 pattern is recorded. However, a 1-1 code in one lead
group and a 1-2 code in another lead group may be coded for certain applications.

Coding Hierarchy
Certain codes suppress others. The presence of some ECG ndings makes other codes
meaningless. For example, bundle branch block codes (7-1, 7-2) are almost always accom-
panied by negative T-waves, so 5-codes give no added meaning. Other codes indicate that
ner coding is unnecessary, for example, 6-8, with a cardiac pacemaker. The suppression
codes listed in the Appendix are checked before settling on the nal codes.

ECG CODING FORM

Fig. 13-1
07315

Shipment Lot Seq.

Participant ID Number No. Number
Current Year of Date of Recording Date of Coding
Clinic Followup
ID Label
Month Day Year Month Day Year

Rest Supine
Heart PR QRS QT QRS T R Height S Depth Quality Lead Clear
Rate interval duration interval Axis Axis Lead aVL Lead V3 Reversal
(/min) (ms) (ms) (ms) (uV) (uV) (1, 3, 5) (0, 1) 1.0

Q and QS S-T Segment T Wave ST Segment Miscellaneous

A-V Ventricular Arrhythmias Ectopic
patterns Depression Items Elevation R Items
Conduction Conduction (8X) Codes
(1X) (4X) (5X) (9.2) 9X
Defect Defect
1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 3X (6X) (7X) 1 2 3 4 5 6 7 8 9 SVPB VPB 6
1 3 4X 5

203
Data Recording
Codes for the ECG are recorded on a standard form designed for a particular study. Fig.
13.1 shows such a form. Continuous measurements and codes specic for, or unnecessary
to, a particular program may be added or subtracted in a similar format.

In Fig. 13.2 is an ECG having a number of codable ndings. The correct codes for this
ECG are shown in Fig. 13.3.
Fig. 13.4 illustrates an ECG with both a suppression code and a code that it suppresses,
and Fig. 13.5 shows the correct coding for this ECG. So that even though a 4-1-1 pattern is
present in lead II, it is not coded because it is suppressed by code 7-1-1. An additional check
is made by use of an editing program in the computer to correct errors after data are entered.

Fig. 13-2
lead I
1-1-1
4-1-1
5-1

ECG CODING FORM Fig. 13-3

07315

Shipment Lot Seq.

Participant ID Number No. Number
Current Year of Date of Recording Date of Coding
Clinic Followup 012A 01 001
ID Label
Month Day Year Month Day Year

A 123456 7ABC A 1

Rest Supine
Heart PR QRS QT QRS T R Height S Depth Quality Lead Clear
Rate interval duration interval Axis Axis Lead aVL Lead V3 Reversal
(/min) (ms) (ms) (ms) (uV) (uV) (1, 3, 5) (0, 1) 1.0

Q and QS S-T Segment T Wave ST Segment Miscellaneous

Items A-V Ventricular Arrhythmias Ectopic
patterns Depression Elevation R Items
(5X) Conduction Conduction (8X) Codes
(1X) (4X) (9.2) 9X
Defect Defect
1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 3X (6X) (7X) 1 2 3 4 5 6 7 8 9 SVPB VPB 6
1 3 4X 5

11 11 1

204
 Fig. 13-4

lead II
7-1-1
4-1-1
5-1

4-1-1 and 5-1 are not coded. They are suppressed by the 7-1-1

ECG CODING FORM Fig. 13-5

07315

Shipment Lot Seq.

Participant ID Number No. Number
Current Year of Date of Recording Date of Coding
Clinic Followup 012A 01 001
ID Label
Month Day Year Month Day Year

B 123456 7ABC B 1

Rest Supine
Heart PR QRS QT QRS T R Height S Depth Quality Lead Clear
Rate interval duration interval Axis Axis Lead aVL Lead V3 Reversal
(/min) (ms) (ms) (ms) (uV) (uV) (1, 3, 5) (0, 1) 1.0

Q and QS S-T Segment T Wave ST Segment Miscellaneous

Items A-V Ventricular Arrhythmias Ectopic
patterns Depression Elevation R Items
(5X) Conduction Conduction (8X) Codes
(1X) (4X) (9.2) 9X
Defect Defect
1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 1LV6 23F V1-5 3X (6X) (7X) 1 2 3 4 5 6 7 8 9 SVPB VPB 1 4X 6
3 5

11

205
14
ECG Data Acquisition Procedures and Maintenance of Recording Quality
Including Technician Training
Electrocardiographic recording in clinical trials and population studies requires standard-
ization of ECG equipment, electrodes, and the ECG data acquisition protocols. This, com-
bined with triplicate independent ECG coding, ensures accurate, repeatable ECG wave-
form classication. Technicians are centrally trained in ECG acquisition procedures and
their performance certied before recording study patient data. A standard recording pro-
tocol is used for 12-lead rest ECGs. Although single or multichannel ECG machines may
be specied for a study, the recording procedures for 12-lead rest ECGs differ mainly in
simultaneous recording of all 12 leads and automatic lead group sequencing.

Preparation of Study Participant

Participant Position

Prior to electrode placement, there are some steps and precautions to be followed. For
example, participants should be relaxed and comfortable in a supine or semi recumbent
position. The examination table/bed should be adequate to comfortably accommodate the
participant. Examination tables that are too narrow are more likely to cause limb movement
or tenseness. A drape for exposed upper torso will be needed. An additional covering may
be needed to prevent the participant from becoming chilled. Coldness can also introduce
tremor in the participant which causes artifact in the recording of the isoelectric line (base-
line). Make sure ankles and wrists are accessible for electrode application. ECG electrode
placement should always be performed with the technician standing to the participants left
side. Supplies needed for ECG acquisition should be assembled and arranged efciently.

Fasting State

Many cardiovascular disease natural history studies and clinical trials are currently or did
in the past record the ECG in the fasting state in order to avoid the effects of recent food
ingestion on the ECG in a non standard fashion. Any study that does not require fasting
state ECG recording introduces random error into ECG recording of autonomic and isch-
emic ECG parameters and degrades the precision of ECG comparisons of serially recorded
electronic or visual data from one study visit to another.
There is a moderate-sized and consistent literature on the effect of food ingestion on the
ECG. Findings include an effect of alterations in the size and shape of T waves (including
T wave attening and negativity) and increase in the heart rate;13 heart rate variability,4,5
QTc,6 and A-V conduction7 are all altered and ST segment depression and more abnormal
Minnesota Code ndings occur.7,8 All such alterations could cause systematic error (eg.
if ECGs are scheduled at a xed time after a clinic snack), or random error if there is no
206
prescribed pre-fasting time before recording. New subclinical ECG predictors will lose
power and serial change for ECG outcome data will be subject to more random error. Our
own experience in epidemiologic studies where a large amount of participant data is to
be collected makes us fully aware of the logistic problems that may arise for participant
scheduling after an overnight fast. And so the study Steering Committees may want to
compromise by requesting that scheduled ECGs be collected fasting whenever possible. If
it is impossible, then a data form from the study Coordinating Center should be completed
to indicate time from last food ingestion by the participant. Just as there are good reasons
for collecting other study data in the fasting state such as blood pressure and blood lipids,
the ECG should also be obtained in the fasting state.

Single Channel Electrocariographs

Machine Standardization

Single channel recorders are standardized by requiring a at writing stylus and 2.5-inch
wide heat sensitive ECG paper; all equipment must meet or exceed the latest internation-
ally recognized certifying body (e.g., National heart Associations) ECG technical speci-
cations. Electrocardiographic recording equipment from a single manufacturer, for a
particular study, is encouraged to reduce machine variability. Multichannel recorders are
recommended for simultaneous 12-lead recording to provide accurate measurement for the
onset and offset of waveforms.

Electrodes and Electrode Placement

Electrodes continue to improve. Current electrodes are tabs, prejelled with conductive
media for both limb leads and chest electrodes.

Electrode Position Measurement and Marking

Mark each electrode site with a good quality felt tip pen, taking special care to locate
precordial sites because they constitute the most important nonbiologic source of ECG
variability. The procedure is given below.
Electrode V2. (1) Locate the angle between sternum and second left rib with the index
and middle ngers of, the right hand. (2) Count down to the fourth intercostal space below
it. (3) Locate V2 in the fourth intercostal space at the left sternal border. Mark V2 location
with a dot.
Electrode V1. Locate electrode V1 in the fourth intercostal space at the right sternal
border. This should be at the same level as V2 and immediately to the right of the sternum.
Mark V1 location with a dot.
Anterior 5th interspace marker (E Point). Identify the fth intercostal space below V2
in the manner previously described. Follow this space to the midsternal line and mark this
point. This is the E point.
Electrodes V3-V6. (1) Locate the V6 electrode position at the same level of the E point
in the midaxillary line. Mark this location with a dot. This identies the horizontal level
for V4-V6 electrodes. (2) Using a metric tape, measure the horizontal distance in centimeters
from the E point to V6 to the nearest 0.5 cm. The midpoint distance is the V4 electrode
207
location. Mark this location. (If a study involves longitudinal ECG recordings the E to V6
distance may be recorded each time in order to assess placement repeatability). (3) Using
a exible ruler, measure the distance between V4 and V6. The midpoint is the location of
the V5 electrode. Place a dot at this site. (4) In a similar manner measure the distance
between V2 and V4. The midpoint is the location of the V3 electrode. This site should also
be marked.
Limb leads. (1) Locate electrode LL on the left leg. (2) Locate electrode RL on the right
leg. (3) Locate electrode LA on the left forearm (inside). (4) Locate electrode RA on the
right forearm (inside).

Application of Electrodes

As the prejelled electrodes are placed at each site they should rmly adhere to the skin. It
is important that the electrode jelly is not smeared over a large area in order to avoid ECG
waveform distortion resulting from paste overlap in adjacent electrode sites. The limb lead
electrodes are placed in the appropriate locations and the patient lead wires are attached to
the corresponding electrodes, using care not to entangle or pull on any of the lead wires.

ECG Recording

Record a full minute of ECG, data which consists of at least 2.5 seconds of each of
leads I, II, III, aVR, aVL, aVF, and V1-V6. Approximately 0.04 second of 1 mm cali-
bration pulses are recorded prior to any ECG data, followed by the standard 12 leads.
Tracings are recorded at a paper speed of 25 mm/second. Leads recorded at half standard
amplitude are preceded by a half standard calibration pulse and the notation 1/2 STD.

Mounting Electrocardiograms

Although many modern electrocardiographs produce ECGs that do not require mounting,
many archival records are still resurrected (and many commercially available data are still
collected on single channel electrocardiographs) for answers to new questions and proper
presentation of these single strip ECGs makes the coding presentation much more efcient.
The alternative is to use scanned records and apply digital measures on the computer. At
the time of publication, there are no satisfactory programs available to accomplish this
task. Each lead should be clearly labeled as Lead I, Lead II, Lead V1, etc. The original
unmounted continuous ECG is mounted at in the following manner:
ECG quality checks. Every tracing should be checked before mounting to determine that
(1) a standardization strip is included, (2) each lead is approximately 5 inches long, (3) the
ECG contains the complete patient identifying data (clinic ID, subject ID, visit code, date),
(4) all 12 leads are recorded and labeled in the proper order, and (5) the tracings have been
taken at the speed of 25 mm/second. Prior to any cutting or mounting of the ECGs, exactly
60s of data are measured starting with lead I and including as many of the 12 leads as neces-
sary to make 1 min. One minute can be determined by placing the ECG along a measured
150cm tape (=1 min at 25 mm/second) and marking this point on the record. Within this
1-min interval code all ectopic beat information, thus referencing arrhythmic events to a
standardized time.
208
 Cutting procedure. If ECG leads are to be cut and some data discarded, the following
procedure is used.
(1) The standardization strip is cut to the size of 3 0.5 inches. Then each of the 12 leads
are cut to approximately 3.5 1.75 inches with a Littmann mounter. (2) The length of each
tracing taken by the clinic should be approximately 5 inches but is trimmed to 3.5 inches
when being prepared for mounting. In general, select the portion to be mounted from as
near the center of the original tracing as possible.
ECG mounting procedure with cut strips. The standardization strip and the cutout trac-
ings of the 12 leads are mounted on a single sheet size 8.5 11 inches (see Fig. 14.1). First,
mount the standardization strip on the upper right edge of the mount sheet. If the standard-
ization strip is shorter than 3 inches or missing, cover the glued portion with long lasting
transparent tape. Next, mount the 12 individual tracings in order; limb leads on the left and
precordial leads to the right. Mounting sequence is indicated by the numbering sequence.

209
 FIGURE 14.1.
ECG mounting continuous strips. The ECG data is a continuous strip about ve feet in length. Indi-
vidual leads are separated from the each other without data loss and attached to two sheets of pressure
sensitive paper. Leads I, II, and III are placed on the left side and leads aVR, aVL, and aVF to the right
(see Fig. 14.2); the calibration and the patients ID label are attached at the bottom. On the other sheet,
V1-V3 is attached on the left and V4-V6 on the right side (see Fig. 14.3). When an ECG lead strip ex-
tends past the edge of the mounting sheet, it is folded to the back. The sheets are placed in clear plastic
loose-leaf protectors so that the limb leads are toward the front and precordial leads at the back.
210
 FIGURE 14.2.

FIGURE 14.3.

Rechecking mounted ECG. After complete ECG mounting, check for errors. One of the most
common is to mount a lead upside down. Identify such a tracing by the relation of P- and T-waves
to QRS complex. The ECG supervisor reviews records after mounting to identify any one of the
following problems: (1) clipped voltages, (2) misplaced lead mounts, (3) leads which are mounted
upside down, (4) mis-identied leads, and (5) duplicated leads labeled to suggest that they represent
different leads or missing tracings. File only those records that are properly mounted and labeled.
211
Twelve-Lead ECG Using Multichannel Electrocardiographs

Standard rest supine 12-lead electrocardiograms recorded on multichannel machines re-

quire the same procedures described for single channel machines, except calibrations and
lead groups record automatically and the technician places all six precordial leads at one
time. Mounting this ECG data is accomplished by inserting the entire record into a prela-
beled plastic insert cover which identies individual leads or preferably recording the
electronic signal to a disk or transmitting the signal directly over analogue phone lines to
a central ECG reading facility.

Electrode Position Measuring and Marking

1. With the participants consent, excess hair is removed from each site using a razor or
electric shaver. This is often not tolerated and then more vigorous rubbing in is required.
2. At each electrode site the skin should be rubbed vigorously with an alcohol swab until the
skin is red. The limb lead electrodes are placed in the appropriate locations (Fig. 14.4) and
the patient lead wires attached to the corresponding electrodes, using care not to entangle
or pull any of the lead wires. If the skin preparation has removed the felt pen marking at
any of the electrode sites, these should be accurately reestablished by carefully repeating
the procedure described in Electrode Position Measurement and Marking.

Electrode Placement

The adhesive portion of a new (disposable) electrode is exposed, and the electrode care-
fully positioned and rmly stuck at each electrode site. A light circular pressure over the
adhesive material will insure good electrode-skin contact.

Limb Limb
Electrode Electrode

RA LA

Limb Limb
Electrode Electrode

RL LL
FIGURE 14.4.
212
 Reference Point E for Locating V4, V5, and V6: From the location of V2, palpate with the
middle nger of your right hand the intercostal space and follow it laterally outside the
sternal border and at a slight angle down. Feel the fth rib between your index and middle
ngers and feel the fth intercostal space with your index nger.
At the level of the fth intercostal space, mark a (+) at the midsternal line below your
X mark for V1-V2 level. This (+) is the reference level E for V4, V5 and V6 (Fig. 14.6). In
overweight persons and in women with tender breast tissue, it is often difcult to locate the
fth intercostal space. In such a case, mark the + for E 1 in (3 cm) below your reference
level X for V1 and V2 (in smaller adults, 1 inch. 2.5cm) is enough).

Approximate Location of V6: Move the left elbow laterally without moving it anteriorly
or posteriorly, while observing the anterior and posterior axillary folds. The left elbow
must be supported properly. Follow a line exactly in the vertical midplane of the thorax
(mid-axillary line) down where the line meets the horizontal plane of Point E. Using your
marker, make a vertical one inch long line there as an approximate location of V6.

Exact Location of V6: Exact location of V6 is determined by using the HEARTSQUARE.9

Place the HEARTSQUARE with the wider arm (E arm) horizontally at level E (Fig. 14.7).
Slide the V6 arm of the HEARTSQUARE toward the midaxillary line until the arrow points
to the mark at the midaxillary line. Mark the exact location of V6 at the level of the arrow
on the V6 arm.

E Point

22
21 20
19 18
17 16
15 14
24
8

13 12 16 18 20 22
7

10 12 23 25
6

17 19 21
5

9 11 13 15
4

11 10
3
2
1

9 8
7 6
5 4
3 2
1 0

Exact V6 location

FIGURE 14.7.

Exact Location Of V4: Keeping the HEARTSQUARE in the horizontal position with the
arrow on the V6 arm pointing toward the V6 position, observe the reading at E point
(Fig. 14.8). Use this E reading on the centimeter scale on the V6 arm, and follow this same
E reading along the 45 lines toward the torso to locate the exact position of V4. Once you
have located V6 and V4, secure the V6 arm with your thumb to prevent it from sliding.
Note the V6 reading which is the distance from the arrow on the V6 arm to where this arm
intersects the E arm at right angles. You may then remove the HEARTSQUARE. Write
214
down the E and V6 measurements for later entry to the patient information data section
of the electrocardiograph or data form. Enter the E measurement in the HEIGHT eld of
your MAC1200 patient setup (by pressing the pat info key in the ECG machine) and the V6
measurement in the WEIGHT eld (do not enter the Height and Weight of the participant).
E and V6 measurements should be entered as three digits. For example, Fig. 14.8 shows
that the E and V6 measurements should be entered as 165 and 110 for the readings of 16.5
cm and 11.0 cm, respectively.
22

21 20

19 18

17 16

15 14
HEARTSQUARE
8

13 12
7
6
5

10 12 14 16 8 20 22 24
4
3

11 10 9 11 13 15 17 19 21 23 25
2
1

8 9
V6
measurement
6 7
V4 E
4 5

2 3

0 1
E measurement = 16.5
V6 measurment = 11.0
V6 V6

FIGURE 14.8.

Locations of V3 and V5: Mark V3 exactly halfway between V2 and V4. Mark V5 exactly
halfway between V4 and V6 (Fig. 14.9)

V1 V2
V3 V4
V5 V6

FIGURE 14.9.
215
Attaching the electrodes

After you have located electrode positions, rubbed them with alcohol swabs and gauze
pads, and marked their position with a water-soluble marker, you may apply the electrodes.
Do not place electrodes directly over bone. Attach lead wires in the same, correct order
every time to establish a routine and to eliminate lead swaps. Attaching a lead wire to a wrong
location can result in abnormal ECG patterns that cannot be corrected. Position the lead
harness on the patients abdomen. Grasp each lead of the lead harness attachment. Follow
lead wire to the electrode attachment end. Attach wire to electrode, making sure that the
attachment is not in contact with electrode adhesive. Make sure lead wires have some slack
and are hanging loosely. Secure the lead wire to the skin by applying paper tape 1-inch
below the clip, if the ECG shows baseline noise despite careful preparation.

Fault Detection Procedures

The quality of ECG data should be previewed prior to the actual recording. If problems with
noise or drift are encountered, electrodes should be replaced. The following is a guide for de-
termining which electrodes may be at fault. The underlined electrodes are the predominant
determinants of the lead and, therefore, are the usual electrodes affecting a given lead. After ad-
justment and/or replacement of suspect electrodes, all leads should be tested again for quality.

Lead affected Possible faulty electrode

I RL, RA, LA
II RL, RA, LL
III RL, LA, LL
aVR RL, RA, LL, LA
aVL RL, LL, RA, LA
aVF RL, LL, RA,,LA
V1 RL, LL, RA, LA, V1
V2 RL, LL, RA, LA, V2
V3 RL, LL, RA, LA, V3
V4 RL, LL, RA, LA, V4
V5 RL, LL, RA, LA, V5
V6 RL, LL, RA, LA, V6

Self-Evaluation of Technical Performance

A reasonable estimate of the ECG data quality can be obtained by measuring the amplitude
of noise drift on the ECG recording. Based on requirements of the Minnesota Code, accept-
able levels of noise and baseline drift have been established by ve grades of quality shown in
Table 14.1 (self-evaluation of performance). The grades given in this table take into account
visual measurement precision and the ability of trained coders to achieve reliable results.
Baseline noise is measured as peak-to-peak noise uctuations (sensitivity 10 mm/mV).
The beat-to-beat drift level is determined by searching for the pair of successive QRS com-
plexes having the largest amplitude differences (vertical distance) between successive P-R
segments. Overall drift is measured from maximum and minimum baseline levels for a
216
specic lead (as determined by the PR and TP segments) and is the vertical distance between
them. Fig. 14.10 illustrates these measurements. In general, technicians should achieve a
grade 2 or better in the preview data before making a study ECG recording. Examples of
several technical problems encountered in the ECG recording and suggested remedial
actions are illustrated in Figs. 14.1114.13.
TABLE 14.1. Self-evaluation of technical quality performance grade.
Drift
Noise (number of small Overall (number of Beat-to-Beat (number of
Quality grade Level paper divisions) small paperdivisions) small paper divisions)
1 1 7 1.75
2 2.25 8 2.5
3 3.5 9 3
4 4.5 10 3.5
5 >4.5 >10 >3.5

noise
6mm

beat to beat overall

19mm 25mm

FIGURE 14.10.

Baseline noise. Muscle tremor causes irregular oscillations (deections) of low amplitude
and varying rapidity, superimposed upon the ECG waveform (see Fig. 14.11). Muscle tremor
may be due to the involuntary muscle activity in a participant who is tense, apprehensive,
or uncomfortable. A brief explanation of recording procedure may be all that is required to
eliminate this problem. The participant should be asked if the room temperature is too cool,
and is covered with a light blanket if needed. During exercise testing, muscle potential arti-
facts are more difcult to avoid, but minimizing arm movement and isometric holding of
ergometric equipment reduces the problem. Occasionally an intermittent shorting of patient
lead wire(s) may be the cause; checking the continuity of lead wires will detect this.

FIGURE 14.11.
217
 Baseline uctuations. Short- and long-term drift is typically caused by faulty skin-
electrode interface. During exercise testing, this pattern of baseline deviations is com-
monly associated with excess perspiration. (However, baseline wandering or drift only in
the precordial leads [V1-V6] might be due to the subjects respiratory movements.) Fig.
14.12 illustrates beat-to-beat and overall baseline drift. Technical performance limits for
these are indicated in Table 14.1.

FIGURE 14.12.

Careful adherence to the skin preparation protocol usually eliminates this problem. Sus-
pect electrodes should be checked again for high impedance or excess offset voltage. Vis-
ible perspiration on the chest and in the axilla should be removed with a gauze pad. Poor
skin electrode adherence is corrected by a light circular pressure around the adhesive area
of the electrode.

Regular baseline oscillation. Sixty-Hertz interference is characterized by regular oscil-

lations occurring at the rate of 60 times/second (see Fig. 14.13). Electrical equipment of
any kind may be the source of AC interference on an electrocardiogram in all leads or only
particular ones. AC interference appearing only in two standard limb leads (i.e., in two of
leads I, II, and III) casts suspicion on the extremity electrode which is common to them.

1. Lead I is the potential difference between LA and RA.

2. Lead II is the potential difference between LL and RA.
3. Lead III is the potential difference between LL and LA.

Therefore, if only leads II and III show AC interference, the left leg, being the common
member, must be at fault. It must, therefore, be checked with regard to quality of skin
preparation and electrode contact, secure attachment of the LL cable tip to the electrode,
possible contact to left leg with any metal part of bed or other equipment (or proximity to
a wall with hidden wiring), or a partially broken cable.

FIGURE 14.13.
218
Continued Quality Assurance for Electrocardiographs

Routine systematic checks on an electrocardiographs performance during a study is

needed to minimize gradual change in ECG waveform characteristics resulting from the
aging of electronic components. We recommend bimonthly evaluation of at least paper
speed and linearity. A rugged, reliable ECG calibrator with several amplitude pulse
options and a series of pulses with constant interpulse differences provide a method to
document linearity and paper speed accuracy. Use either qualitative assessment of ST slope
(frequency response) on simulated ECG waveforms or measurement of the decay constant
(amplitude decrease of a 10 mm calibration pulse within 3 s). Procedures for documenting
linearity, frequency response, and paper speed accuracy are described below.
Linearity test. Three calibration pulse levels, 5, 10, and 25 are input to the electrocar-
diograph using a stable ECG calibrator. Every output calibration pulse amplitude must be
within 5% of the input pulse amplitude, and overshoot or undershoot must be within 10%
of the allowable limit.
Measurement of calibration pulse. The amplitude of the pulse step is measured from
the top of the baseline to the top of the pulse. Fig. 14.14 illustrates this measurement with
acceptable deviations.

Linearity Check

0.5mm 0.5mm

FIGURE 14.14.

Input-pulse Amplitude (mm) Output Response (mm)

5 5 0.25
10 10 0.5
25 25 1.25

Overshoot and undershoot are measured 0.5 mm to the right of the rst vertical line of the
step-pulse (Fig. 14.15). For overshoot, measure from the top margin of the baseline to the
point of overshoot. For undershoot, measure from the lower margin of the baseline to the
nish of the undershoot. Fig. 14.15 illustrates these measurements and the acceptable toler-
ances are shown below.
219
 Undershoot and Overshoot Checks

Overshoot (OS) Undershoot (US)

end of OS bottom
OS
*
of BL
top of OS
US *
end of
US

0.5mm 0.5mm

FIGURE 14.15.

Step-pulse Amplitude (mm) Undershoot or Overshoot Allowed (mm)

5 0.5
10 1.0
25 2.5

Importance of linearity. Systematic amplitude distortion of ECG waveforms produce

major consequences in epidemiologic studies. Prevalence of LVH (3-codes), Q-codes
(1-codes), ST depressions (4-codes), and T inversions (5-codes) may be over or under
reported when ECG calibration changes during a study.

220
 Low Frequency Response

3 sec.
(75mm)

top of BL
top of BL 3mm

starting level of measurement

FIGURE 14.16.

Low frequency response test. The low frequency response is determined by measuring
the amplitude of a continuous 10 mm DC square wave pulse. When the residual pulse am-
plitude is 3 mm at or after 3 second of pulse duration, the frequency response is satisfac-
tory (Fig. 14.16). However, when the residual pulse amplitude is <3 mm at the 3-second
point, the time when the amplitude is = 3 mm is determined and the distance to the start of
the pulse measured. Again, this measurement is from the top of the baseline to the top
of the baseline. The 3 mm amplitude time point must occur within 6575 mm of the pulse
start.
Importance of frequency response. The frequency response determines the accuracy with
which the ST segment is reproduced (downsloping, at, or upsloping). A low frequency
response cutoff above 0.05 Hz changes the prevalence of 4-code ndings.

Paper speed accuracy. Both short- and long-term paper speed accuracy need to be as-
sessed. A 10-second series of calibration pulses with known width and frequency is input
to the electrocardiograph.
Short-term response. This test measures transient changes in paper speed as with an
intermittent sticking of the chart paper drive. Measure Five different calibration pulses
within the 10 second. Individual pulse widths must be within 5% of the known pulse width.
Measure from the lower margin of the top baseline of one pulse to the corresponding point
on the adjacent pulse (see Fig. 14.17).
Long-term response This test measures overall accuracy of paper speed in order to detect
persistent alteration of chart drive. In this test a 10-s interval of calibration pulses should
contain 60 pulses. Deviation within 2% is allowed. The pulses must be individually count-
ed and the 60th calibration pulse must be within 5 mm from the desired ending point (see
Fig. 14.18).

221
 short term response

calibration
top
of BL

bottom
of BL

4mmx<4.5 mm

FIGURE 14.17.

long term response

10 seconds [250 mm]

1 2 3 4 5 6 7 8 9
pulse
10 sec = 60 calibration pulses

FIGURE 14.18.

Importance of accurate paper speed. The impact of inaccurate paper speed is to distort
QRS width measurements. This results primarily in the change in frequency of 1-code items
and 7-code items (Q-wave duration and QRS duration).
Whenever calibration measurements exceed acceptable levels, clinic technical staff are
notied about and advised on the specic ECG machine performance decit. They, in turn,
contact local service representatives for their machine and have the specied decit cor-
rected. An additional calibration strip is required to document corrective action.
222
ECG Quality Grading for Electronic ECGs

The ECG quality grade denitions are listed below. They are determined electronically,
except for lead reversals (see Chap. 11) that are detected by a combination of electronic
measurement and visual editing. All lead reversals are classied as grade 5.
Generally the more electrical artifact introduced by poor acquisition techniques, the worse
the grade.
Grade1 is perfect, grades 24 have increasing amounts of artifacts and are indicators of a
technicians lack of attention to detail in recording an ECGbut are all fully retrievable for
study data. Whereas grade 5 ECGS are almost always lost to study data analysis. These are
denitions of the quality grades:
Median ST-T signal is used to measure the rms value of the difference between the origi-
nal and smoothed (= ltered) signal for the ST-T segment from J + 10 ms to the end of T in
each lead. These values are then used to assign quality grades according to the maximum
noise in any one of the 12 leads.

Training of Clinic Site ECG Technicians Helps to Maintain High Quality ECG Recording

In multicenter studies, centralized training together with production of a standardized

ECG training manual with details of standardized recording techniques (see above) will
aid the process. Centralized training should be provided for clinic coordinators and ECG
technicians.
An intensive training session in handling and programming ECG acquisition units forms
an integral part of the centralized training session performed by the Central ECG Reading
Center (CERC).
A certication process should also be instituted that requires successful recording and
transmission (for electronic ECGs) of ve successive, adequate quality ECGs to the CERC.
Personnel turnover is anticipated and necessitates special consideration for training of new
ECG technicians. Usually, new technicians will be trained by their clinic coordinator or by
a previously certied ECG technician, and they will go through the standard certication
process before being authorized to record ECGs for the study.
Centralized training would also include the handling of paper ECG ow with detailed
instructions for completing transmittal and ndings forms.
Production of an ECG Section for the study Manual of Operations also aids in maintaining
high quality recording. This task can be completed as soon as relevant procedural details
of scheduling have been nalized and would include all of the directions above specifying
study specic protocol needs and the electrocardiographs to be employed

Quality Control of ECG Data Collection and Processing Procedures

Quality Control (QC) of ECG data acquisition and processing needs to be instituted
from the very beginning of data collection and applied unabated throughout the study. QC
reports, whether produced by the CERC or the study coordinating center (CC), should be
documents not just used for review and led away but as instruments to allow study man-
agement to be proactive. Reports of maintenance of continued high quality should be used
to bolster morale an important activity in a multiyear study. Declining quality, by individ-
uals, clinics or the study as a whole, in the case of the ECG, should trigger the necessity for
223
local or central retraining, or fact-nding clinic site visits by the CC or the CERC. Secular
drift is important to check because it can produce the appearance of biological change that
is false and due only to procedural or technological change. By constant review, retrain-
ing, machinery renewal, and procedural correction, false secular drift can be avoided. ECG
recording QC requires attention in four areas: ECG recording technician, visual records
coding, electronic ECG processing, and electrocardiographs.

Quality Control at Field Centers

All ECG recording technicians should be required to be certied and recertied at annual
or semiannual intervals.

Quality Trend Monitoring

Each ECG technician should have a personal study ID (that will not change for the dura-
tion of a specic study) entered into the ECG acquisition unit and the transmittal/ndings
form at the time that each new ECG recording begins. This ID should be transmitted or
mailed along with the ECG and becomes part of that record. Quality grades assigned to
each ECG are used to compile continuous quality trend analysis data for each clinic center
ECG technician to spot emerging problems, particularly with the change of ECG personnel
over the duration of the study. The overall goal for each ECG technician is to keep the frac-
tion of poor quality recordings (grade 5) below 5% throughout the study. Quality grade 5
records are, in general, still usable for classication, but they pose difculties, particularly
with classication of borderline abnormalities (see above and Table 14.1).
A special eld needs to be kept for lead reversal problems (see Chap. 11) which, for the
standard resting ECG, are more common than is generally recognized.
Regular reinforcement of ECG recording technician performance can be achieved by
regular central training of all study ECG recording technicians, which should be done
after the recruitment period of a study, if the study Steering Committee decides to repeat
the baseline ECG, or if there is a broad decrement in performance judged by a study QC
Committee review of regularly submitted CERC QC reports.
Additionally, ECG recording technicians in each clinic can have their procedural perfor-
mance checked on an annual or semiannual basis and a formal report submitted to the CC
for distribution and a summary for the study QC Committee.

Minimizing Biologic Variability

Rest ECG
ECG biological variability at rest can be reduced by standard patient instructions. The
ECG should be recorded no sooner than two hours after eating, smoking, or exercise. Fre-
quently, an overnight fast may be required in the context of other measurements. It is
important that the same requirements are followed for all study subjects in a clinical trial
or survey. On the same occasion that the ECG is recorded other tests may occur. The ECG
should always be recorded before the administration of a glucose load, at least 30 min after
blood drawing, and before stress testing. Repeat ECG records for study participants should
be performed under similar conditions. Others have summarized the factors affecting vari-
ability in the recording.10 Testing of combined variability, from technician, lead placement,
electrocardiograph and biologic can be made by repeating recording on an appropriate
group of test subjects.11-12
224
References

1. Widerlov E, Jostell KG, Claesson L, Odlind B, Keisu M, Freychuss U. Inuence of food intake
on electrocardiograms of healthy male volunteers. Eur J Clin Pharmacol. 1999; 55:619-624.
2. Anderson M. Fasting electrocardiogram. Acta Med Scand. 1970;187:385-390.
3. Dear HD, Buncher CR, Sawayama T. Changes in electrocardiogram and serum potassium
values following glucose ingestion. Arch Intern Med. 1969 124:25-28.
4. Lu CL, Zou X, Orr WC, Chen JD. Postprandial changes of sympathovagal balance measured
by heart rate variability. Dig Dis Sci. 1999, 44:857-861.
5. Paolisso G, Manzella D, Ferrara N, et al. Glucose ingestion affects cardiac ANS in healthy
subjects with different amounts of body fat. Am J Physiol. 1997; 273:E471-478.
6. Nagy D, DeMeersman R, Gallagher D, et al. QTc interval (cardiac repolarization): lengthen-
ing after meals. Obes Res. 1997, 5:531-537.
7. Hoon TJ, McCollam PL, Beckman KJ, Hariman RJ, Bauman JL. Impact of food on the pharma
cokinetics and electrocardiographic effects of sustained release verapamil in normal subjects.
Am J Cardiol. 1992;70:1072-1076.
8. Riley CP, Oberman A, Shefeld LT. Electrocardiographic effects of glucose ingestion. Arch
Intern Med. 1972;130:703-707.
9. Rautaharju PM, Park L, Rautaharju FS, Crow R. A standardized procedure for locating ECG
chest electrode positions: consideration of the effect of breast tissue on ECG amplitudes in
women. J Electrocardiol. 1998;31:17-29.
10. Schijvenaars BJA, Van Herpen G, Kors JA. Intraindividual variability in electrocardiograms.
J Electrocardiol. 2008;41:190-196.
11. Schroeder EB, Whitsel EA, Evans GW, Prineas RJ, Chambless LE, Heiss G. Repeatability of
heart rate variability measures. J Electrocardiol. 2004;37:163-172.
12. Vaidean GD, Schroeder EB, Whitsel EA, et al. Short-term repeatability of electrocardiographic
spatial T-wave axis and Qt interval. J Electrocardiol. 2005;38:139-147.

225
15

Criteria for Signicant Electrocardiographic Change

The Minnesota Code was originally developed for determining prevalence information in
populations. Later, in an attempt to document incident events, criteria were developed to
document signicant ECG changes that were based on apparent changes in the Minne-
sota Code of sequential ECGs. However, because the Minnesota Code is categorical and
ECG parameters are continuous, this approach yielded high false positive misclassica-
tions as judged by clinical assessment. We realized that a new and quantitative approach
was required to overcome the deciencies of categorical Minnesota Code change. We de-
veloped a compilation of procedures and rules for serial comparison that took into account
the limitations of the previous Minnesota Code change criteria by requiring side-by-side,
or direct comparison of reference and follow-up ECGs, rather than relying solely on direct
categorical change in Minnesota Codes.1 This method simulates the clinicians approach
for comparing ECGs, but includes a standardized method for documenting signicant ECG
pattern change.2 Just as the original Minnesota Code uses specied measurement rules
to reduce coding variability and a systematic classication code based on these measure-
ments, serial comparison uses a set of measurement rules and a systematic classication
code to document categories of ECG pattern changes.
Briey, the procedure for direct serial ECG comparison uses 4 steps. First, Minnesota
prevalence Codes are assigned to all ECGs using Minnesota Code criteria and measure-
ment rules. Second, specic Minnesota Code change(s) between an initial (reference) and
follow-up (event) ECG) is/are identied by a computer algorithm and used to select ECG
pairs for direct comparison using criteria in Table 15-1. Third, each prevalence code change
is evaluated for signicance by direct evaluation of the corresponding average reference
and follow-up ECG waveform (Q, ST, T-wave, QRS duration or R-wave amplitude) using
criteria in Table 15-2. To be regarded as a signicant ECG change, these criteria are ap-
plied to the follow-up electrocardiogram in the same lead or lead group. For example, if
the reference ECG has a 5-3 T-wave inversion code, the follow-up ECG must have a 5-2
or more severe T-wave inversion code, and the average amplitude of the follow-up ECG T-
wave must be 100% greater than the reference ECG T-wave in the same lead group. When
the Minnesota Code change that triggered the comparison meets the criteria it is considered
to be a true and signicant change (this could be either an increase or decrease). If it does
not meet the criteria, it is labeled no change, or, when signicant noise or baseline drift
interferes with serial comparison it is labeled technical, Fourth, the serial change infor-
mation from each Minnesota Code change (increase, decrease, no change or technical) is
used in a computer algorithm to dene the serial change categories shown in Table 15-3
(evolving Q-wave, evolving ST elevation, evolving ST-T, evolving BBB, and evolving
LVH). In summary, the Minnesota Codes requiring evaluation for ECG pattern change,
should be compared using the following guidelines:
226
 For increasing codes, the follow-up Minnesota code determines the lead group and
sometimes the lead to examine side-by-side comparison.
For decreasing codes, the reference ECG code determines which lead group or lead to
examine at follow-up for side-by-side comparison.
Side-by-side ECG comparison requires measurement of amplitudes/durations for the
ECG pattern under consideration. Measurement rules adhere generally to the Minnesota
Code measurement procedures but with specic exceptions that will be addressed in
detail as specic codes changes are addressed.
When a Minnesota Code change is evaluated by serial comparison, the result must be one of
four outcomes: signicant increase, signicant decrease, no change or technical problem.
There are seven serial ECG change categories for Q-wave evolution with or without
ST-T wave evolution. There are ve ST elevation serial change categories, 14 serial ECG
change categories for ST-T wave evolution without Q-wave evolution. There are three
criteria for right, left and non-specic bundle branch block evolution (E-BBB) and a six
change categories for ECG LVH (E-LVH). In addition, there are signicant serial change
limits for continuous measures of LVH (Sokolow-Lyon, Cornell Voltage, Cornell Voltage
Product, Sum of 12 leads QRS voltage, and Sum of 12 leads QRS product). Severity for
Q-wave and ST-T wave evolution is hierarchically ranked; evolving Q is more severe than
evolving ST or T-wave evolution (the latter two are equivalent). Within the Q or ST-T
categories separately, a lower number supersedes a higher number. The above procedure
may be determined entirely by computer analysis or by visually determined ndings and
manually entered into a computer algorithm

TABLE 15.1. Minnesota Code Changes That Prompts Direct ECG Wave Form Comparison.
Minnesota Code Minnesota Code
ECG Item Criteria Number on Earlier ECG on Later ECG
Q-QS Patterns 1 None of 1-1 to 1-3 1-1 or 1-2
(major) 2 Any 1-3 1-1

Q-QS Patterns 3 None of 1-1 to 1-3 1-3

(minor) 4 Any 1-3 1-2
5 1-2 1-1

T-waves 6 None of 5-1 to 5-3 5-1 or 5-2

7 5-3 5-2 or 5-1
8 5-2 5-2a or 5-1
9 5-1 5-1a

ST depression 10 None of 4-1 to 4-2 4-1 to 4-2

11 4-2 4-1
12 4-1 4-1a

ST elevation 13 No 9-2 9-2

a
For ST codes a signicant increase or decrease in amplitude may occur within code 4-1, for T wave codes a signicant
increase or decrease in amplitude may occur within code 5-2 or 5-1
Minnesota Code change of any ECG item may occur within any of the following lead groupings: anterolateral I, aVL,
and V6; inferior II, III, and aVF; anterior V1V5

227
TABLE 15.2. Criteria for Determining Signicant ECG Pattern Change.
Event ECG
Minnesota Code
Q-Code Comparison Rule for Determining Signicant ECG Pattern Change
1-1-1 Requires 50% increase in event Q/R ratio or 1 mm initial R-wave amplitude decrease
in event ECG compared with corresponding lead(s) of reference ECG. If no Q-wave in
reference ECG, event record must have codable Q-wave.

1-1-2 Requires 50% increase in event ECG Q/R ratio and 1.5 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave. New appearance of
QS complex in leads to the left of V1 when V1 does not show change will also be judged
as positive evidence of new infarction.

1-1-3 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave.

1-1-4 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with lead III of reference ECG, plus a new Q-wave in
aVF. If no Q-wave in reference ECG, event record must have codable Q-wave.

1-1-5 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave.

1-1-6 Requires 1 mm decrease in event ECG initial R-wave amplitude compared with
corresponding lead(s) of reference ECG.

1-1-7 Requires 1 mm decrease in event ECG initial R-wave compared with corresponding
lead(s) of reference ECG. If no QS-wave in reference ECG, event record must have
codable QS-wave.

1-2-1 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event records must have codable Q-wave.

1-2-2 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave.

1-2-3 Requires 1 mm decrease in event ECG initial R-wave amplitude compared with
corresponding lead(s) of reference ECG. If no QS-wave in reference ECG, event record
must have codable QS-wave.

1-2-4 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with lead III of reference ECG, plus new Q-wave in
aVF. If no Q-wave in reference ECG, event record must have codable Q-wave.

1-2-5 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event compared with corresponding lead of reference ECG. If no Q-wave in
reference ECG, event record must have codable Q-wave.

228
TABLE 15.2. (continued)
Event ECG
Minnesota Code
Q-Code Comparison Rule for Determining Signicant ECG Pattern Change
1-2-7a Requires 1 mm decrease in event ECG initial R-wave amplitude compared with
corresponding lead(s) of reference ECG. If no QS-wave in reference ECG, event record
must have codable QS-wave.

1-3-1 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave.

1-3-2 Requires 1 mm decrease in event ECG initial R-wave amplitude compared with
corresponding lead(s) of reference ECG. If no QS-wave in reference ECG, event record
must have codable QS-wave.

1-3-3 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have codable Q-wave.

1-3-4 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with lead III of reference ECG, plus a new Q-wave in
aVF. If no Q-wave in reference ECG, event record must have a codable Q-wave.

1-3-5 Requires 50% increase in event ECG Q/R ratio or 1 mm initial R-wave amplitude
decrease in event ECG compared with corresponding lead(s) of reference ECG. If no
Q-wave in reference ECG, event record must have a codable Q-wave.

1-3-6 Requires 1 mm decrease in event ECG initial R-wave amplitude compared with
corresponding lead(s) of reference ECG. If no QS-wave in reference ECG, event record
must have a codable QS-wave.

1-3-8a Requires 1 mm decrease in event ECG initial R-wave amplitude in the lead to the left
compared with corresponding lead(s) of reference ECG.

ST Elevation
Code

9-2 Requires reference record have no 9-2 code, or there is 100% change in event ECG ST
segment elevation compared with corresponding lead(s) of reference ECG.

ST Depression
Code

4-1 Requires reference record have no 4-1 or 4-2, or there is 100% change in event ECG
ST segment depression.

OR

4-2 Requires reference record have no 4-1 or 4-2 codes, or there is 100% change in event
ECG ST segment depression.

229
TABLE 15.2. (continued)
T-wave
Inversion Code
5-1 Requires reference record have no T-wave inversion 5-1, 5-2 codes, or there is 100%
change (as in 9-2 codes) in event T-wave inversion.
OR
5-2 Requires reference record have no T-wave inversion 5-1, 5-2 codes, or there is 100%
change (as in 9-2 codes) in event T-wave inversion.
Bundle Branch
Block Code
7-1-1, 7-2-1 Requires new 7-1-1, 7-2-1 or 7-4 code in event ECG with a QRS-duration increased by
0.02 sec in event ECG compared with reference ECG.
LVH
Code
3-1, 3-3 Requires event ECG to have a Minnesota 3-code change specied in evolving ECG-LVH.
a
Note: Code 1-2-6 in past Minnesota Coding has now been excluded from serial comparison and code 1-3-8 is a new code
(see chapt. 4) that in the past was coded as 1-2-8

TABLE 15.3. ECG Criteria for Signicant Serial ECG Pattern Change
Evolving Q Wave
Evolving Q1. No Q-code in reference ECG followed by a record with a Diagnostic Q-code (Minn. code
1-1-1 through 1-2-7) OR an Equivocal Q-code (any 1-3-x) in reference ECG followed by
a record with any code 1-1-x Q-code.

Evolving Q2. An Equivocal Q-code (any 1-3-x code) and no major ST-segment depression (4-0, 4-4,
4-3) in reference ECG followed by a record with a Diagnostic Q-code (Minn. code 1-2-1
through 1-2-7) PLUS a major ST-segment depression (Minn. code 4-1-x or 4-2).

Evolving Q3. An Equivocal Q-code (any 1-3-x) and no major T-wave inversion (5-4, 5-3 or 5-0) in
reference ECG followed by a record with a Diagnostic Q-code (Minn. code 1-2-1 through
1-2-7) PLUS a major T-wave inversion (Minn. code 5-1 or 5-2).

Evolving Q4. An Equivocal Q-code (any 1-3-x) and no ST-elevation in reference ECG followed by a
record with a Diagnostic Q-code (Minn. code 1-2-1 through 1-2-7) PLUS an ST-segment
elevation (Minn. code 9-2).

Evolving Q5. No Q-code and no 4-1-x or 4-2 in reference ECG followed by a record with an Equivocal
Q-code (any 1-3-x) PLUS 4-1-X or 4-2.

Evolving Q6. No Q-code and no 5-1 or 5-2 in reference ECG followed by a record with an Equivocal
Q-code (any 1-3-x) PLUS a 5-1 or 5-2.

Evolving Q7. No Q-code and no 9-2 in reference ECG followed by a record with and Equivocal Q- code
(any 1-3-x) PLUS a 9-2.

(Evolving Q1-Q7 must be conrmed by serial comparison rules)

230
TABLE 15.3. (continued)
Evolving ST-Elevation
Evolving STE-1: Code 9-0 in reference ECG followed by a record with 9-2 in at least 2 leads and 100%
increase in ST elevation in both leads.

Evolving STE-2: Code 9-2 in reference ECG followed by a record with 9-2 in at least 2 leads and 100%
increase in ST elevation in both leads.

Evolving STE-3: Code 9-2 and no code 5-1 or 5-2 in reference ECG, followed by a record with code 5-1 or
5-2 and 100% increase in T wave inversion in at least 2 leads.

Evolving STE-4: Reversal of evolving STE-1 - within the hospital ECG only.

Evolving STE-5: Reversal of evolving STE-2 - within the hospital ECG only.

Evolving ST-Depression / T Wave Inversiona

Evolving ST-T1: Either 4-0 (no 4-code), 4-4 or 4-3 in reference ECG followed by a record with 4-2 or 4-1-2
or 4-1-1 and 100% increase in ST segment depression

Evolving ST-T2: Either 4-2 or 4-1-2 in reference ECG, followed by a record with 4-1-1 and 100%
increase in ST segment depression.

Evolving ST-T3: Either 5-0, 5-4 or 5-3 in reference ECG, followed by a record with 5-2 or 5-1 and 100%
increase in T-wave inversion.

Evolving ST-T4: Code 5-2 in reference ECG, followed by a record with 5-1 and 100% in T-wave inversion.

Evolving ST-T5: Code 4-1-1 in reference ECG, followed by a record with 4-1-1 and 100% increase in ST
depression.

Evolving ST-T6: Code 5-1 in reference ECG, followed by a record with 5-1 and 100% increase in T- wave
inversion.

Evolving ST-T7: Code 5-2 in reference ECG, followed by a record with 5-2 and 100% increase in T- wave
inversion.

Evolving ST-T1R through ST-T7R: Reverse of ST-T1 to ST-T7. Requires 100% decrease in ST depression
or T-wave inversion of follow-up record compared to reference.

Evolving Bundle Branch Block:

E-BBB1: No code 7-1 in the reference ECG followed by an ECG with 7-1-1 in event ECG and QRS
duration increased by 0.02 second.

E-BBB2: No code 7-2 in the reference ECG followed by an ECG with 7-2-1 in event ECG and QRS
duration increased by 0.02 second.

E-BBB3: No code 7-4 in the reference ECG followed by an ECG with 7-4 in event ECG and QRS
duration increased by 0.02 second.
a
Code changes must occur in the same lead groups. For example no Q-code in V-leads in reference must be followed by
new Q-code in the V-leads. However, 4-code, 5-code and 9-2 code changes do not have to be in the same lead groups as
Q-waves, but must occur in the same lead group as the reference 4, 5, or 9-2. For example no 4-code in lead l, aVL and
V6 must be followed by a 4-1-x or 4-2 in the same lead group. All changes are conrmed by side-by-side ECG comparison and
require 100% change in the event compared to the reference ECG.

231
Evolving ECG-LVH (Minnesota Code 3-1 and 3-3 or ECG Measures for LVH)

The rst step in 3-code evaluation is to determine in which lead the most severe 3-code
occurred. Compare this lead with the corresponding lead in the other ECG. Code 3-1 is
considered more severe than Code 3-3. If both ECGs have the same 3-code, the follow-
up record determines which lead to use to compare with the reference ECG. If the 3-code
occurred in different leads i.e. leads V5 and aVL, use the following hierarchy to determine
which lead to compare: V5 /V6 (whichever R-amplitude is higher)>I>II>III> aVL.
When making 3-code comparisons, only the second to last complete normal beat of the
lead is used. If there are only two beats in a lead, the last beat is used. Moreover, there
are various ECG criteria for LVH by voltage pattern alone or QRS duration and voltage
product. The following criteria are yet to be validated by prognostic follow-up studies.

TABLE 15.4. Evolving ECG-Minnesota Code 3 for LVH.

Code change Leads Change criteria
3-0 3-1 I, II, III 36%
3-0 3-1 aVL >60%
3-0 3-1 V5/V6 >30%
3-0 3-3 I >36%
3-0 3-3 V5/V6 >25%
3-1 3-1 I, II, III 36%
3-1 3-1 aVL 60%
3-1 3-1 V5/V6 30%
3-3 3-3 I 36%
3-3 3-3 V5/V6 +25%

E-LVH 1: Minnesota Code 3-0 in reference ECG followed by an ECG with a 3-1in the
follow-up ECG, conrmed as a signicant increase.

E-LVH 2: Minnesota Code 3-0 in reference ECG followed by an ECG with a 3-3 in the
follow-up ECG, conrmed as a signicant increase.

E-LVH 3: Minnesota Code 3-1 in reference ECG, followed by an ECG with a 3-0 in the
event ECG, conrmed by a signicant decrease.

E-LVH 4: Minnesota Code 3-3 in reference ECG, followed by an ECG with a 3-0 in the
event ECG, conrmed by a signicant decrease.

E-LVH 5: Minnesota Code 3-1 in the reference ECG, followed by an ECG with a 3-1 in
the event ECG, conrmed by a signicant increase or a signicant decrease.

E-LVH 6 Minnesota Code 3-3 in the reference ECG, followed by an ECG with a 3-3 in
the event ECG, conrmed by a signicant increase or decrease.

(LVH must by conrmed by serial change criteria for E-LVH)

232
TABLE 15.5. Evolving ECG Measures for LVH.
Denition Formula Cut-Point for LVH Change Criteria
Sokolow-Lyon SV1 + RV5 or RV6 3500 V >900 V
Cornell Voltage RaVL + SV3 2800 V (men) >400 V
2200 V (women) >400 V
Cornell Product (RaVL + SV3) QRS duration 243.6 V.s >41 V. s.
Sum of 12 leads 12-lead QRS sum (except lead aVR) 17900 V >2319 V
12 leads Product 12-lead QRS sum QRS duration 1747.2 V.s >355.6 V. s.
Units: V, microvolts for Q.R.S of amplitudes
V .s. microvolts.seconds for the product of ECG voltage and QRS duration.

Serial Change for Acute Myocardial Infarction

A subset of evolving Q and evolving ST-T were labeled as Evolving Diagnostic ECG
or Positive ECG in the AHA Scientic Statement for Case Denitions for Acute Cor-
onary Heart Disease in Epidemiology and Clinical Research Studies.1 Evolving Q1-
Q4 are labeled evolving diagnostic for the AHA epidemiologic case denition of Acute
Myocardial Infarction (AMI). Any of these document a denite AMI. Evolving Q5-Q7,
evolving STE 1-5, or EBBB1 document positive ECG.

These serial change criteria have been used in many epidemiologic studies and clinical
trials,1 and have been validated in the Multiple Risk Factor Intervention Trial and the Min-
nesota Heart Study.2,3 There are seven criteria for Q-wave MI and nine for ST-T-wave
evolution. Evolving Q-waves have adjusted relative risks about 4.0, and evolving ST-T
waves have an adjusted relative risk from 1.52.0. These risk levels persist for both silent
and symptomatic ischemic cardiac events.

Categories of Signicant ECG Waveform Change Determined by Minnesota Code

Serial Comparison

Evolving Diagnostic Q-code pattern with or without ST-T wave changes (ED)
Evolving ST-T wave without evolving Q-code pattern (EV)
Evolving Bundle Branch Block (E-BBB)
Evolving Left Ventricular Hypertrophy (E-LVH)

When classifying Minnesota Code changes into the above categories, it is important to
remember the following:
The reference ECG is the reference record for non-hospital visits; the reference ECG is
the earliest hospital record for in-hospital acute events.
An equivocal Minnesota Q-code is any 1-3-x code.
A diagnostic Minnesota Q-code is any 1-1-x or any 1-2-x.
All ED patterns are conrmed as signicant increase by serial comparison.
All EV patterns are conrmed as signicant increases or decreases by serial comparison.
All E-BBB are conrmed as signicant increases by serial comparisons.
All E-LVH are conrmed as signicant increases or decreases by serial comparison criteria.

233
 REFERENCE ECG EVENT ECG

FIGURE 15.1. Minnesota Code 1-1-1. Reference ECG shows initial R-waves V1-V3. Event ECG
has 1-1-1 code in V3. Signicant pattern change IS conrmed because 1mm R-
wave amplitude decrease occurs between the ECGs in V3 (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.2. Minnesota Code 1-1-1. Reference ECG shows initial R-waves V2. Event ECG
shows no initial R-wave in lead V2 (QR complex) and a 1-1-1 code. Signicant ECG
pattern change IS NOT conrmed because < 1 mm R-wave amplitude decrease
occurs between the reference and event ECG in V2 (EVQ0)

234
 REFERENCE ECG EVENT ECG

FIGURE 15.3. Minnesota Code 1-1-1. Reference ECG shows a small noncodable Q-waves in lead
II. Event ECG shows a 1-1-1 in lead II. Signicant ECG pattern change IS conrmed
in lead II because there is 1mm R-wave amplitude decrease between reference
ECG and event ECG in lead II (EVQ1)

FIGURE 15.4. Minnesota Code 1-1-2. Reference ECG shows initial R-waves V1V3. Event ECG
shows tiny initial R-wave in lead V1 but not in the majority of beats), and QS com-
plex in V2V3. Lead V1 meets 1-1-2 code criteria. Signicant pattern change IS con-
rmed because 1 mm R-wave amplitude decrease occurs between the reference
and event ECG in lead V1 (EVQ1)

235
 REFERENCE ECG EVENT ECG

FIGURE 15.5. Minnesota Code 1-1-2. Reference ECG shows QS complex in V1. Event ECG
has QR pattern in V1, meeting 1-1-2 code criteria. Signicant ECG pattern change
IS NOT conrmed because < 50% increase in Q/R occurs between the reference
and Event ECGs in lead V1. (The Q/R of reference ECG is innite because of QS
complex.) (EVQ0)

REFERENCE ECG EVENT ECG

FIGURE 15.6. Minnesota Code 1-1-2. Reference ECG shows tiny initial R-wave in V1. Event
ECG has QR complex in V1, meeting 1-1-2 code criteria. Signicant ECG pattern
change IS NOT conrmed because <1.0 mm R-wave amplitude decrease occurs
between the reference and event ECGs (EVQ0)

236
 REFERENCE ECG EVENT ECG

FIGURE 15.7. Minnesota Code 1-1-4. Reference ECG shows initial R-waves in the inferior leads.
Event ECG shows 1-1-4 in lead III and a new Q-wave in aVF. Signicant ECG pattern
change IS conrmed because 1 mm R-wave amplitude decrease occurs between
the reference and event ECGs in lead III accompanied by new Q-wave in aVF (Q-wave
in aVF must be 1 mm in depth but not necessarily 0.02 s). (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.8. Minnesota Code 1-1-4. Reference ECG shows tiny initial R-wave in lead III and no
Q-wave in aVF. Event ECG shows QR in lead III with a Q-wave in aVF, making a
1-1-4 code. Signicant ECG pattern change IS NOT conrmed because < 1 mm R-wave
amplitude decrease occurs between the reference and event ECGs in lead III (EVQ0)
237
 REFERENCE ECG EVENT ECG

FIGURE 15.9. Minnesota Code 1-1-6. Reference ECG shows initial R-waves V2-V3. The event
ECG shows an initial R-wave in lead V2 and QS complex in V3. Signicant ECG
pattern change IS conrmed because 1 mm R-wave amplitude decrease occurs
between the reference and event ECGs in V3 (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.10. Minnesota Code 1-1-6. Reference ECG shows initial R-waves V1-V3. Event ECG
shows tiny initial R-wave in lead V1 and QS complex in V2 making a 1-1-6 code.
Signicant ECG pattern change IS conrmed because 1 mm R-wave amplitude
decrease occurs between the reference and event ECGs in V2 (EVQ1)

238
 REFERENCE ECG EVENT ECG

FIGURE 15.11. Minnesota Code 1-1-6. Reference ECG shows small initial R-waves in V1V3.
Event ECG shows initial R-wave in V1 and QS pattern in V2. Signicant ECG
pattern change IS NOT conrmed because < 1 mm R-wave amplitude decrease
occurs between the ECGs in V2 (EVQ0)

FIGURE 15.12. Minnesota Code 1-2-1. Reference ECG shows narrow Q-waves in lead II. Event
ECG has broader Q-wave in lead II, making 1-2-1 code. Signicant ECG pattern
change IS conrmed because 50% increase in Q/R occurs between the reference
and event ECGs in lead II (EVQ1)

239
FIGURE 15.13. Minnesota Code 1-2-1. Reference ECG shows small noncodable Q-waves in V6.
Event ECG shows a broader, deeper Q-wave in V6 making a 1-2-1 code. Signi-
cant ECG pattern change IS conrmed because 1 mm R-wave decrease occurs
between reference and event ECGs in lead V 6 (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.14. Minnesota Code 1-2-3. Reference ECG shows initial R-wave in lead II. Event ECG
shows QS pattern in lead II (note terminal R-waves are < 1 mm in amplitude and
therefore complex is a QS). Signicant ECG pattern change IS conrmed because
1 mm initial R-wave amplitude decrease occurs between the reference and event
ECGs in lead II (EVQ1)

240
 REFERENCE ECG EVENT ECG

FIGURE 15.15. Minnesota Code 1-2-3. Reference ECG shows a tiny initial R-wave in lead II and is,
therefore, an RS complex. Event ECG shows QS pattern in lead II, making a 1-2-3
code. Signicant pattern change IS NOT conrmed because < 1 mm R-wave ampli-
tude decrease occurs between the reference and event ECGs in lead II (EVQ0)

REFERENCE ECG EVENT ECG

FIGURE 15.16. Minnesota Code 1-2-4. Reference ECG shows small transitional complex in lead
III. Event ECG shows deep Q-wave in lead III making 1-2-4 code and a new
codable Q-wave in aVF. Signicant ECG pattern change IS conrmed because 1 mm
initial R-wave amplitude decrease occurs between the reference and event ECGs
in lead III and there is a new Q-wave in the event ECG in aVF (EVQ1)

241
REFERENCE ECG EVENT ECG

FIGURE 15.17. Minnesota Code 1-2-7. Reference ECG shows initial R-waves V1-V3. Event ECG
shows QS pattern V1-V3, making a 1-2-7 code. Signicant ECG pattern change IS
conrmed because 1 mm R-wave amplitude decrease occurs between the refer-
ence and event ECGs in lead V3. (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.18. Minnesota Code 1-2-7. Reference ECG shows small initial R-waves in V2 and V3.
Event ECG shows QS pattern V1V3, making a 1-2-7 code. Signicant ECG pattern
change IS NOT conrmed because < 1 mm R-wave amplitude decrease occurs
between the reference and event ECGs in V3. (Note majority of initial R-waves in V3
at reference are < 1 mm.) (EVQ0)

242
 REFERENCE ECG

EVENT ECG

FIGURE 15.19. Minnesota Code 1-2-7, 9-2 and 5-2. Reference ECG shows a 5-3 T-wave inversion
code in lead V5V6 and no ST elevation code (9-2). Event ECG shows QS com-
plexes in V1V4 making a 1-1-7 code with a 9-2 ST elevation code in V2 and a 5-2
T-wave inversion code in lead V4V5. Signicant ECG pattern change IS conrmed
because 1 mm initial R-wave amplitude decrease occurs between the reference
and event ECGs in V4, and because 100% increase in ST elevation amplitude
occurs in V2 and 100% increase in T-wave negative occurs in V5 between refer-
ence and event ECGs (EVQ1)

243
 REFERENCE ECG EVENT ECG

FIGURE 15.20. Minnesota Code 1-2-3 and Stable Minnesota Code 5-2. Reference ECG shows
initial R-wave in lead II and no 5-2 in lead II. Event ECG shows 1-2-3 code and 5-2
T-wave in lead II. Signicant pattern change IS conrmed for code 1-2-3 because
1 mm R-wave amplitude decrease occurs between the reference and event ECGs
in lead II. Signicant pattern change IS conrmed for T-wave negativity because
100% increase in T-wave negativity occurs between the ECGs in lead II. This
illustrates that a signicant pattern change can occur within the same lead for both
Q-code and T-wave code. It may also occur in a different lead group from Q-code
change (EVQ1)

REFERENCE ECG EVENT ECG

FIGURE 15.21. Minnesota Code 1-2-7. No Q-code in reference ECG to a 1-2-7 code in event ECG.
Signicant pattern change IS conrmed because 1 mm R-wave decrease occurs
between reference and event ECGs in lead V3 (EVQ1)

244
 REFERENCE ECG EVENT ECG

FIGURE 15.22. Minnesota Code 1-2-7. No Q-code in reference ECG to a 1-2-7 in event ECG.
Signicant pattern change is not conrmed because < 1 mm R-wave amplitude
decrease occurs between reference and event ECGs (EVQ0)

REFERENCE ECG EVENT ECG

FIGURE 15.23. Minnesota Code 1-1-4. Code 1-3-4 in reference ECG to a 1-1-4 in event ECG.
Signicant pattern change IS conrmed because 50% Q/R ratio increase occurs
between the reference and event ECGs (EVQ1)

245
 REFERENCE ECG EVENT ECG

FIGURE 15.24. Minnesota Code 1-1-4 Code 1-3-4 in reference ECG to a 1-1-4 in event ECG. Sig-
nicant pattern change is not conrmed because < 50% Q/R ratio occurs between
reference and event ECGs. (EVQ0)

246
FIGURE 15.25. Minnesota Code 1-2-7 and 4-2-1. MC 1-3-2 and no 4 code in reference ECG and the
code 1-2-7 and code 4-1-2 in event ECG. Signicant pattern change IS conrmed
because 1 mm R-wave decrease occurs in lead V3 and 100% ST depression in
lead V5 occurs between the reference and event ECGs (EV-Q2)

247
 REFERENCE ECG EVENT ECG

FIGURE 15.26. Minnesota Code 1-2-4 and 5-2. Reference ECG shows 1-3-4 code and no 5-code.
Event ECG shows a 1-2-4 in lead III PLUS 5-2 code in lead I. Signicant pattern
change IS conrmed because 50% Q/R ratio increase in lead III and 100% T-wave
inversion in lead I occurs between reference and event ECGs. Note, this illustrates
that Q-code and T-wave code evolution may occur in the same or different lead
groups (EV-Q3)

248
 REFERENCE ECG EVENT ECG

FIGURE 15.27. Minnesota Code 1-2-7 and 9-2. MC 1-3-2 and no 9-2 code in reference ECG and code
1-2-7 and code 9-2 in event ECG. Signicant pattern change IS conrmed because
1 mm decrease in V3 R-wave amplitude and 100 % increase in V4 and V5 ST
elevation occurs between reference and event ECGs (EV-Q4)

249
 REFERENCE ECG EVENT ECG

FIGURE 15.28. Minnesota Code 1-3-2 and 4-1-2. No Q-code and no 4-code in reference ECG
followed by a record with 1-3-2 PLUS 4-1-2. Q-code evolution and T-wave evo-
lution may be in the same or different lead groups. Signicant pattern change
IS conrmed because 1 mm R-wave amplitude decrease in V2 and 100% ST
depression in leads V4-V6 occurs between reference and event ECGs (EV-Q5)

250
FIGURE 15.29. Minnesota Code 1-3-2 and 5-2. MC 1-0 and no 5 code in reference ECG and the
code 1-3-2 and code 5-2 in event ECG. Signicant pattern change IS conrmed
because 1 mm R-wave amplitude decrease in V2 and 100% increase in T-wave
negativity in V3 occurs between reference and event ECGs (EV-Q6)

251
 REFERENCE ECG EVENT ECG

FIGURE 15.30. Minnesota Code 1-3-2 and 9-2. No Q-code and no 9-2 in reference ECG followed
by a record with a Code 1-3-2 Plus Code 9-2, Signicant pattern change IS conrmed
by serial comparison rules for Q-code evolution and 100% increase in STE
(EV-Q7)

252
 REFERENCE ECG EVENT ECG

FIGURE 15.31. Minnesota Code 1-3-2. No Q-code in reference ECG in V1-V3. Event ECG shows
code 1-3-2 in V2. Signicant pattern change IS conrmed because 1 mm R-
wave decrease occurs in V2 between reference and event ECGs. Minor Q-code
evolution but no EV-Q pattern. (EV-Q0)

REFERENCE ECG EVENT ECG

FIGURE 15.32. Minnesota Code 1-3-8. No Q-code in V1-V3 in reference ECG. Event ECG shows
R-wave decrease between V2-V3 meeting code 1-3-8. Signicant pattern change
IS conrmed because R-wave amplitude decrease 1 mm in V3 occurs between
reference and event ECGs. Minor Q-code evolution but no EV-Q pattern (EV-Q0)

253
 REFERENCE ECG EVENT ECG

FIGURE 15.33. Minnesota Code 4-1-2. No 4-code and 5-3 T-wave inversion code in reference ECG
in V6. Event ECG shows a 4-1-2 ST depression code and 5-2 T-wave inversion
code in V6. Signicant pattern change IS conrmed because 100% increase in
J point amplitude negativity and T-wave inversion occurs between reference and
event ECGs in V6. (EV-ST-T1)

FIGURE 15.34. Minnesota Code 4-0. 4-2 ST depression code and 5-2 T-wave inversin code in V4 of
reference ECG. Event ECG shows no 4-code and no 5-code in V4. Signicant
pattern change IS conrmed because 100% increase occurs for these codes
between reference and event ECGs (EVST-T1R)

254
 REFERENCE ECG EVENT ECG

FIGURE 15.35. Minnesota Code 5-1 No 5-code (T-wave inversion code) in reference ECG. Event
ECG shows a 5-1 T-wave inversion code in V4. Signicant pattern changes IS
conrmed because 100% increase in T-wave negative amplitude occurs between
reference and event ECGs (EVST-T3)

REFERENCE ECG EVENT ECG

FIGURE 15.36. Minnesota Code 5-0 A Reverse Code. Code 5-2 T-wave negative in reference
ECG V5-V6. Event ECG shows no 5-code in V5-V6. Signicant pattern change IS
conrmed because 100% increase in T-wave amplitude occurs between reference
and event ECGs (EVST-T3R)

255
 REFERENCE ECG EVENT ECG

FIGURE 15.37. Minnesota Code 9-2. No 9-2 STE in reference ECG. Event ECG shows code 9-2
in V2-V3. Signicant pattern change IS conrmed because 100% increase in STE
occurs between reference and event ECGs in V2V3 (EV-STE-1)

REFERENCE ECG EVENT ECG

FIGURE 15.38. Minnesota Code 7-1-1. No 7-1-1 code in reference ECG and a 7-1-1 code (LBBB)
in event ECG. Signicant pattern change IS conrmed because 7-1-1 code occurred
in event ECG and QRS duration was 0.02 second longer than reference ECG
(EBBB1)

256
 REFERENCE ECG

EVENT ECG

FIGURE 15.39. Minnesota Code 7-1-1. Reference ECG shows no 7-code. Event record shows a
7-1-1 code (LBBB). Signicant pattern change IS conrmed because 0.02 second
QRS duration increase occurs between the reference and event ECGs (EBBB1)

257
 REFERENCE ECG

EVENT ECG

FIGURE 15.40. Minnesota Code 7-2-1. No 7-2-1 code in reference ECG and a 7-2-1 code (RBBB)
in event ECG. Signicant pattern change IS conrmed because 7-2-1 code occurs in
event ECG and QRS duration was 0.02 second longer than reference ECG (EBBB2)

258
 REFERENCE ECG EVENT ECG

FIGURE 15.41. Minnesota Code 7-2-1. MC 7-3 code in reference ECG and a 7-2-1 code (RBBB)
in event ECG with the event ECG QRS 0.02 second longer than reference ECG
(EBBB2)

259
 REFERENCE ECG

EVENT ECG

FIGURE 15.42. Minnesota Code 7-4. No 7-4 code in reference ECG and a 7-4 code in event ECG
with the event ECG QRS 0.02 second longer than reference ECG (EBBB3)

260
 REFERENCE ECG EVENT ECG

FIGURE 15.43. Minnesota Code 3-1. No 3-1 code or 3-3 code in reference ECG and a 3-1 code in
event ECG. Signicant pattern change IS conrmed because 30% R-wave ampli-
tude increase occurs between reference and event ECGs. (E-LVH1)

REFERENCE ECG EVENT ECG

FIGURE 15.44. Minnesota Code 3-1. No 3-1 or 3-3 in reference ECG and a 3-3 in event ECG.
Signicant pattern change IS conrmed because 25% R-wave amplitude increase
occurs between reference and event ECGs (E-LVH2)

261
References

1. Luepker RV, Apple FS, Christenson RH, et al. Case Denitions for Acute Coronary Heart
Disease in Epidemiology and Clinical Research Studies: A Statement From the AHA Council
on Epidemiology and Prevention; AHA Statistics Committee; World Heart Federation Council
on Epidemiology and Prevention; the European Society of Cardiology Working Group on
Epidemiology and Prevention; Centers for Disease Control and Prevention; and the National
Heart, Lung, and Blood Institute. Circulation. 2003;108:2543-2549.
2. Crow RS, Prineas RJ, Jacobs DR Jr, Blackburn H. A New Epidemiologic Classication System
for Interim Myocardial Infarction from Serial Electrocardiographic Changes. Am J Cardiol.
1989;4:454-461.
3. Crow RS, Prineas RJ, Hannan PJ, Grandits G, Blackburn H. Prognostic Associations of Minnesota
Code Serial Electrocardiographic Change Classication with Coronary Heart Disease Mortality
in the Multiple Risk Factor Intervention Trial. Am J Cardiol. 1997;80:138-144.

262
16
ECG Indices that Add to Independent Prognostication for
Cardiovascular Disease Outcomes
There are a number of ECG indices derived from continuous measurements of duration,
voltage, and intervals that add to the prognostic information that can be derived from
the electrocardiogram. These are easily derived from analysis of electronically recorded
ECG signals but some, such as heart rate variability can be derived by measurement of
hard copy ECGs.

QRS/T Angle and Spatial T Axis

The ECG spatial T axis and the angles between the QRS and T spatial vector axes have
been demonstrated to be strong independent predictors of incident coronary heart disease
and total mortality.1-4 These variables can be calculated by the parameters from resting,
standard twelve-lead electrocardiogram. These indices do require programmed analysis of
electronic ECG signals.

Spatial QRS/T angle is the angle between the mean QRS vector and T vector. Mean spatial
QRS and T vectors are calculated from quasi-orthogonal X, Y, and Z leads reconstructed
from standard ECG leads using a matrix transformation method.1,2 However, the orthogo-
nal X, Y, and Z leads presently are not routinely recorded. As the initial step, a current
method commonly used for QRS/T angle determination requires the generation of the X,
Y, and Z leads by using an algorithm obtained from a matrix transformation of the 12-lead
ECG. Then the mean X, Y, and Z values in the QRS and T windows are calculated to ob-
tain the mean QRS and T vectors, and then the spatial QRS/T angle is calculated as the
scalar product between these two vectors. This procedure is very cumbersome; therefore, a
simple procedure by using the net QRS and T amplitudes in three standard leads for estima-
tion of the spatial QRS/T angle has been developed recently, which can explain 79% of the
variance of QRS/T angle from X, Y, and Z leads.3

Moreover, the frontal plane QRS/T angle, easily obtained as the difference between the
frontal plane axes of QRS and T, also has been identied as a strong independent predictor
for CHD and total mortality, and it is a suitable clinical substitute for spatial QRS/T angle
for risk prediction5 (see Chap. 5 for derivation of these axes from ECG paper tracings if
necessary, where the axes are not provided with each ECG printout).

263
The following denitions are used for spatial QRS/T angle, frontal QRS/T angle, and
spatial T axis.

(A) The spatial QRS/T angle from X, Y, and Z leads by a matrix transformation
methods QRS/T matrix

Step 1. Using a matrix transformation algorithm from the digital les of the 12-lead ECG
signals to generate the X, Y, and Z leads. First, calculate the mean X, Y, and Z lead
amplitudes for QRS and T waves in their respective time windows. And the subscripts
x, y, and z refer to the X, Y, and Z components of the mean QRS and T vectors;

Step 2. Calculate the spatial magnitude between the QRS and the T vectors.
QRSxyz = (QRSx2 +QRS y2 +QRSz2)0.5
Txyz = (T x2 + T y2 + T z2)0.5

Step 3. The algorithm for QRS/T angle (matrix)

QRS/T angle = ACOS[(QRSx Tx) + (QRSy Ty) + (QRSz Tz)]/(QRSxyz Txyz )
where ACOS is the inverse cosine;
If QRS/T angle is obtained in radians, as in Excel, they are converted to degrees
by multiplying the inverse cosine by the factor 57.3.

(B) The spatial QRS/T angle by using the net QRS and T amplitudes in 3 standard
leads QRS/T simple

Step 1. Using the net QRS and T amplitudes in 3 standard leads from the 12-lead ECG to
generate the X, Y, and Z components.
QRSnet = R amplitude abs (S or QS, whichever is larger)
Tnet = (signed T + signed Tprime) amplitude of the T wave
where abs refers to the absolute value (S and QS amplitudes may be reported as
signed or unsigned values.)

Step 2. Calculate the spatial magnitude between the QRS and the T vectors (magnitude).
QRSsm = [(QRSnetV6)2 + (QRSnetaVF)2 +(QRSnetV2)2]0.5
Tsm = [(TnetV5)2 + (TnetaVF)2+ (TnetV2) 2]0.5
(Subscript sm refers to spatial magnitude.)

Step 3. The algorithm for QRS/T angle (simple)

QRS/T angle = ACOS[(QRSnetV6 TnetV5) + (QRSnetaVF TnetaVF) +
(QRSnetV2 TnetV2)] /(QRSsm Tsm)
where ACOS is the inverse cosine;
If QRS/T angle is obtained in radians, as in Excel, they are converted to degrees
by multiplying the inverse cosine by the factor 57.3.
264
(C) The frontal plane QRS/T angle dened as the absolute value of the difference
between the frontal plane QRS axis and T axis in the route ECG report directly
QRS/T frontal

Frontal plane QRS/T angle = abs (QRS axis T axis).

where abs is the absolute value.
using (360 angle) for an angle > 180 to adjust to the minimal angle.

(D) Spatial T-wave axis based on areas of the wave components of the QRS complex
and T wave

Step 1. Using a matrix transformation algorithm from the digital les of the 12-lead ECG
signals to generate the X, Y, and Z leads. First calculate the mean X, Y, and Z lead
amplitudes for QRS and T waves in their respective time windows. The spatial T-axis
is an estimate of the deviation from the normal reference direction (1/3=0.5774.
X = 1/3, Y = 1/3, and Z = 1/3, where x, y, and z are the unit vector components in
X, Y, and Z directions). Thus, this reference direction is 45 anteriorly, with a 45 angle
from the +Y axis in the frontal and sagittal plane projections.

Step 2. Calculate the spatial magnitude of the T vector.

Txyz = (T x2 + T y2 + T z2)0.5

Step 3. The algorithm Spatial T-wave axis

Spatial T axis = ACOS[(0.5774 * Tx) + (0.5774 * Ty) + (0.5774 * Tz)]/Txyz
where ACOS is the inverse cosine.
If spatial T axis is obtained in radians, as in Excel, they are converted to degrees
by multiplying the inverse cosine by the factor 57.3.

FIGURE 16.1. Spatial QRS/T angle (matrix) = 125; spatial QRS/T angle (simple) = 127; frontal QRS/T
angle (frontal) = 124; spatial T-wave axis = 63

265
FIGURE 16.2. Spatial QRS/T angle (matrix) = 141; spatial QRS/T angle (simple) = 138; frontal QRS/T
angle (frontal) = 114; spatial T-wave axis = 48

Heart Rate Variability

Heart rate variability (HRV) analysis is designed for quantitative assessment of the
autonomic nervous system (ANS). Reduced HRV in ultrashort standard ECG tracings
indicating time domain but not frequency has been associated in numerous studies with
mortality risk.6-13

HRV analysis is based on measuring variability in heart rate, specically, variability in intervals
between R waves RR intervals. Originally, HRV was assessed manually from the calculation
of the mean R-R interval, and its standard deviation was measured on short-term (e.g. 5 minutes)
ECG. The smaller the standard deviation in R-R intervals, the lower is the HRV. To date, HRV
also can be assessed from an ultrashort ECG strips (one or multiple 10 second ECGs).

There are many different types of arithmetic manipulations of R-R intervals, which
have been used to represent HRV. The basic parameters include the standard deviations
(SD) of all normal mean R-R intervals (SDNN) and square root of the mean of
the difference of successive R-R intervals (RMSSD). For short-term ECG recordings
(<20 min), only short-term components of HRV can be calculated, and for brief
recordings (as in 10 second), only particular time-domain measures of HRV can be
calculated. It is, however, possible to derive indices from differences between normal
interbeat time intervals (NN), RMSSD (see below) for an estimate of short-term
components of HRV, and SDNN (see below) for an estimate of overall HRV.

There will be n beats per record. Let each beat be designated N (0), for normal beat, or A
(1), for abnormal or ectopic beat. The time from the onset of recording to the onset of each
beat is designated by the QRSTIM in ms from the rst beat to the last beat (approximately
10,000 ms later).

It is the interval between N or NN from which HRV will be calculated. So we rst remove
all NA, AN and AA intervals and also the rst NN interval following an NA, AN or AA
interval. Then, from all remaining NN intervals estimate 2 HRV indices: (a) SDNN and
(b) RMSSD.
266
 Also, remove (ignore) the N1 N2 (i.e., interval between the rst beat and the second
beat because it is unknown if this could be an AN interval) of each of the 3 ECG
recordings.
If < 50% of all NN, AN, AA, or AN intervals are NN, DO NOT ESTIMATE HRV,
i.e., code to -.

HRV indices from ultrashort records

(A) SDNN is the standard deviation of eligible NN intervals.

Let, x = mean of eligible NN intervals and
n = the number of eligible NN intervals
and NNj = the jth interval
and Nj = the jth beat
Then, each eligible NNj is obtained by subtracting QRSTIM for Nj from QRSTIM
for Nj+1.

Then, SDNN =
(x NN ) j
2

ms
n

(B) RMSSD is the square root of the mean value of the squares of differences between all
eligible successive NN intervals. Let, ni = the number of eligible intervals per ECG.

Then, RMSSD =
(NN j+1 NN j )2
ms
(n i 1)

QRS_n qrstyp01 qrstyp02 qrstyp03 qrstyp04 qrstyp05 qrstyp06 qrstyp07 qrstyp08 qrstyp09 qrstyp10 qrstyp11
11 0 0 0 0 0 0 0 0 0 0 0

qrstim01 qrstim02 qrstim03 qrstim04 qrstim05 qrstim06 qrstim07 qrstim08 qrstim09 qrstim10 qrstim11
108 1046 1952 2820 3730 4672 5562 6550 7476 8380 9304

RRmean rrb_01 rrb_02 rrb_03 rrb_04 rrb_05 rrb_06 rrb_07 rrb_08 rrb_09 rrb_10
917.57 906 868 910 942 890 988 926 904 924

SumMean_dif SDNN meand02 meand03 meand04 meand05 meand06 meand07 meand08 meand09 meand10
9262 (n=9) 32.08 134 2456 57 598 759 4962 71 184 42

SumRR_dif RMSDD rrdif_01 rrdif_02 rrdif_03 rrdif_04 rrdif_05 rrdif_06 rrdif_07 rrdif_08 rrdif_09
21268 (n=8) 51.56 1444 1764 1024 2704 9604 3844 484 400

FIGURE 16.3. SDNN = 32.08; RMSDD = 51.56; percent of accepted normal R-R interval = 100%
(9/9)

267
 QRS_n qrstyp01 qrstyp02 qrstyp03 qrstyp04 qrstyp05 qrstyp06 qrstyp07 qrstyp08 qrstyp09 qrstyp10 qrstyp11 qrstyp12 qrstyp13

13 0 0 0 0 0 0 1 0 0 0 0 1 0

qrstim01 qrstim02 qrstim03 qrstim04 qrstim05 qrstim06 qrstim07 qrstim08 qrstim09 qrstim10 qrstim11 qrstim12 qrstim13

398 1148 1898 2650 3402 4154 4618 5654 6424 7178 7930 8408 9444

RRmean rrb_01 rrb_02 rrb_03 rrb_04 rrb_05 rrb_06 rrb_07 rrb_08 rrb_09 rrb_10 rrb_11 rrb_12

752 750 752 752 752 754 752

SumMean_dif SDNN meand02 meand03 meand04 meand05 meand06 meand07 meand08 meand09 meand10 meand11 meand12

8 (n=6) 1.15 4 0 0 0 4 0

SumRR_dif RMSDD rrdif_01 rrdif_02 rrdif_03 rrdif_04 rrdif_05 rrdif_06 rrdif_07 rrdif_08 rrdif_09 rrdif_10 rrdif_11

8 (n=4) 1.41 4 0 0 4

FIGURE 16.4. SDNN = 1.15; RMSDD = 1.41; percent of accepted normal R-R interval = 55%
(6/11)

QRS_na The number of total beats in this sample

qrstyp01-qrstyp13a QRS type: 0 = normal beat; 1 (or 2) = abnormal beat; .
= missing
qrstim01-qrstim13a Time for R-R interval in 10 second ECG strip which
begins from 0 ms to 10,000 ms
rrb_01-rrb_12b R-R interval between two beats
meand01-meand12b The square of the difference between a normal R-R
intervals and mean RR interval
rrbdif_01-rrbdif_11c The square of difference between two normal R-R
intervals;
total R-R interval QRS_n - 2
accepted normal R-R interval rrb_n
RRmean Sum of RR intervals/number of RR intervals
SumMean_dif Sum of (RR intervalmean of RR intervals)2
SumRR_dif Sum of (difference between two normal RR intervals)2
rMSSD (sum of RR_dif/number of RR_dif)0.5
SDNN (sum of mean_dif/number of mean_dif)0.5
a
Will be same number as QRS_n The number of total beats in this 10-second ECG sample
b
or the number as (QRS_n 1)
c
or the number as (QRS_n 2)

268
References

1. Rautaharju PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic Abnormalities that

predict coronary heart disease events and mortality in postmenopausal women: The Womens
Health Initiative. Circulation. 2006;113:473-480.
2. Rautaharju PM, Prineas RJ, Wood J, Zhang ZM, Crow R, Heiss G. Electrocardiographic
predictors of new-onset heart failure in men and in women free of coronary heart disease (from
the Atherosclerosis in Communities [ARIC] Study). Am J Cardiol. 2007; 100:1437-1441.
3. Rautaharju PM, Prineas RJ, Zhang ZM. A simple procedure for estimation of the spatial QRS/T
angle from the standard 12-lead ECG. J Electrocardiol. 2007;40:300-304.
4. Rautaharju PM, Nelson JC, Kronmal RA, Usefulness of T-axis deviation as an independent risk
indicator for incident cardiac events in older men and women free from coronary heart disease
(the Cardiovascular Health Study). Am J Cardiol. 2001;88:118-123.
5. Zhang ZM, Prineas RJ, Case D, Soliman EZ, Rautaharju PM. for the ARIC Research Group.
Comparison of the prognostic signicance of the electrocardiographic QRS/T angles in
predicting incident coronary heart disease and total mortality (from the Atherosclerosis Risk in
Communities Study). Am J Cardiol. 2007;100:844-849.
6. Dekker JM, Schouten EG, Klootwijk P, Pool J, Swenne CA, Kromhout D. Heart rate variability
from short electrocardiographic recordings predicts mortality from all causes in middle-aged
and elderly men: the Zutphen Study. Am J Epidemiol. 1997;145:899908.
7. Crow RS, Folsom AR, Hannan PJ, Liao D, Swenne CA, Schouten EG. Low heart rate
variability in a 2-minute rhythm strip predicts risk of coronary heart disease and mortality
from several causes: the ARIC Study. Atherosclerosis Risk In Communities study. Circulation.
2000;102(11):1239-1244.
8. Schroeder EB, Liao D, Chambless LE, Prineas RJ, Evans GW, Heiss G. Hypertension, blood
pressure, and heart rate variability. The Atherosclerosis Risk in Communities (ARIC) Study.
Hypertension. 2003;42:1106-1111.
9. Schroeder EB, Chambless LE, Liao D, et al. Diabetes, glucose, insulin, and heart rate variability:
the Atherosclerosis Risk in Communities (ARIC) Study. Diabetes Care. 2005;28(3):668-674.
10. Carnethon MR, Prineas RJ, Temprosa M, Zhang ZM, Uwaifo G, Molitch ME; Diabetes Prevention
Program Research Group. The association among autonomic nervous system function, incident
diabetes, and intervention arm in the Diabetes Prevention Program. Diabetes Care. 2006;29:
914-919.
11. Mozaffarian D, Prineas RJ, Stein PK, Siscovick DS. Dietary sh and omega-3 fatty acid
consumption and heart rate variability in U.S. adults. Circulation. 2008;117: 1130-1137.
12. Ohira T, Diez Roux AV, Prineas RJ, Kizilbash MA, Carnethon MR, Folsom AR. Associations
of psychosocial factors with heart rate and its short-term variability: multi-ethnic study of
atherosclerosis. Psychosom Med. 2008;70(2):141-146.
13. Whitsel EA, Quibrera PM, Christ SL, et al. Heart rate variability, ambient particulate matter air
pollution, and glucose homeostasis: the environmental epidemiology of arrhythmogenesis in
the womens health initiative. Am J Epidemiol. 2009;169(6):693-703.

269
17
Quality Control of Visual and Electronic Coding
Maintaining accuracy of recording and coding in a central ECG laboratory is essential to
a clinical trial or health survey nationally and internationally. Random error may occur
by haphazard application of the rules. When this occurs, true differences between groups
are obscured. Systematic error is potentially more serious and can lead to false apparent
differences between centers and over time. This is particularly serious when incidence of
specic ECG markers of disease, such as indicators of ischemic heart disease, left ven-
tricular hypertrophy, or subclinical predictors of disease, is monitored by measurement of
sequential ECGs.

Attention should be paid to four levels of quality control (QC): careful and comprehensive
training, repeated testing of day-to-day comparisons of independent codes of each ECG by
different coders, internal circulation of records to check for total laboratory contribution to
random and systematic errors over time, and repeated exchange of test records with other
centers or a monitoring center. QC also encompasses the need for maintaining the quality of
the ECGs recorded at clinics in epidemiologic studies and clinical trials (see Chap. 14).

Visual Coding

Initial training and testing for coders: Careful training is necessary for standard ECG
coding procedures, and reduces coding errors and variations in coding. For those newly
introduced to the code, there is a learning curve, over which time quality will improve to
an optimum after coding about 1,000 records.

Internal QC: When coding is done by nonphysician coders, accuracy is increased by

repeated, blinded coding of each ECG. For example, if each ECG is coded separately by
three coders, each coder records ndings without the knowledge of the results of the other
two. The third coder, who is always an experienced or senior coder, notes differences
in coding results among the three. These differences are discussed and a consensus is
reached. If a consensus cannot be reached, the differences are adjudicated by another
trained supervisor or electrocardiographer. Records of rst, second, and third codings can be
examined weekly. Any coder whose results consistently differ from the nal coding results
then should have remedial instruction. Such a need may occasionally arise even with senior
coders, given the large number of guidelines that must be followed. If the coding is done
by a trained electrocardiographer, then regular intra-coder repeatability should be all that is
necessary to maintain visual reading quality with a regular review of a random selection of
records by another trained electrocardiographer or ECG Reading Center supervisor.

Intra-coder QC: Recirculation of ECGs coded previously allows comparison of whole

ECG Reading Center performance, and evaluates coders stability and repeatability. It
may be that a particular coding rule is forgotten and interpreted differently over time, or
270
that new supervisory staff is engaged so that systematic error may enter at some point. The
data manager in a central laboratory can select 25% coded ECGs randomly from the study
reporting ID list monthly, assign a new designated QC ID, have a data entry technician
log into the study database, and produce a coding sheet for the current study. The newly
identied records are then included in the regular work load of the original coders to re-
read. The records to be recirculated should be examples of ECGs currently being measured,
but enriched with abnormal ECGs. The ECGs should be marked in such a way that the
coders do not recognize the records as repeat ECGs. This may require alteration of the date
or label on the ECG. Preparation of such repeat samples should obviously not be made by
anyone whose judgment could affect the nal coding of the ECG.
Inter-coder QC: For a central ECG Reading Center, it is necessary to build an extensive,
validated ECG testing library with a variety of wave form patterns for comprehensive
quality control. The library samples can be selected randomly from the central laboratorys
electronic ECG database, previously carefully coded, and also veried by senior
electrocardiographers in the Center. These validated Gold Standard sets can then be
used to document inter-coder variability of ECG measurement and coding. For studies that
extend over years, or for an ECG Reading Center engaged in coding for multiple studies,
there is also a need to recirculate test records not previously coded for a year or more.
These batches can be included in the regular work load at quarterly intervals.
Comparison between different ECG centers or studies: Comparison from ECG Reading
Centers from different geographical areas is enhanced by the exchange of test ECGs
enriched with study-specic abnormalities. Such exchange of test records is particularly
important for centers having a small work load or that code ECGs intermittently.

Analysis of Repeatability Tests

Coded data from test sets are entered to a QC database in the central laboratory. The analysis
includes a comparison of intra-coder or inter-coder and descriptive statistics: correlation
coefcients, standard deviation of measurement, Kappa coefcients to evaluate coding
consistency, and related graphs (easily plotted with Excel or similar graphics software).
Any systematic trends and signicant deviations from the standard will trigger appropriate
corrective action and discussion with and retraining of the coding personnel.

Qualitative Variables
There are a number of ways suitable for the reporting of intra-coder or inter-coders repeatability
of the coding items by the Minnesota Code. For example, repeatability of Q/QS codes.

First Report
1:0 1:1 1:2 1:3 Total
1:0 A t2
1:1 b
Second
1:2 c
Report
1:3 d
Total t1 T

271
The results may then be summarized as:

(1) The frequency of disagreement as to the presence of any codable Q/QS item, as
a proportion of the number of records on which any codable Q-wave was reported at
least once i.e.,
(t1 - a) + (t2 a)
100
(T - a)

(2) The frequency of disagreement on exact coding of individual items, as a proportion of

the number of records on which any codable item was reported at least once i.e.,
(T - a - b - c - d)
100
(T - a)

Measurement of agreement on exact coding of individual items can also be tested by

Cohens Kappa Coefcient,1,2 which is a statistical measure of intra-rater/inter-rater reliability.
PROC FREQ by SAS3 computes tests and measures of agreement for square tables (i.e., for
tables where the number of rows equals the number of columns). PROC FREQ computes
the kappa coefcients (simple and weighted), their asymptotic standard errors, and
their condence limits when the AGREE option is specied in the TABLES statement.
Generally, an acceptable Kappa coefcient should be larger than 0.60. A Kappa greater
than 0.75 represents excellent agreement beyond chance, a Kappa below 0.40 represents
poor agreement (Poor: < 0.40; Acceptable: > 0.40 and < 0.60; Good: > 0.60 and < 0.75;
Excellent: > 0.75).

Below is an example of agreement for Minnesota code 1 between two coders.

First Report
1:0 1:1 1:2 1:3 Total
1:0 39 0 0 1 40
1:1 0 2 0 0 2
Second
1:2 0 0 3 0 3
Report
1:3 0 0 0 8 8
Total 39 2 3 9 53

Kappa Statistics

Statistic Value ASE 95% Condence Limits

Simple Kappa 0.9545 0.0452 0.8658 1.0000
Weighted Kappa 0.9428 0.0564 0.8322 1.0000

Quantitative Variables

Disagreements between replicate determinations often have skewed distributions i.e.,

most pairs of readings may show relatively small disagreements, but a few may vary widely.

272
Ordinary statistical summaries are based on variance estimates, which are very sensitive
to a few outlying observations. It is therefore as well to examine the actual distribution
of disagreements on at least a sample of readings and calculate variance, etc., only if the
distribution is normal or approximately so.

Summarizing the repeatability of a series of repeated measurements such as heart rate, P-R
interval, QRS duration, and JT and QT intervals by SAS3 Procedures of MEANS, CORR,
and TTEST should be a suitable way to produce a mean of all rst readings, mean of all
second readings, correlation coefcients, and standard deviation of differences between
pairs of readings.

The following example is a plotted graph for heart rate measurement by Excel graphics
software which rst selects paired heart rate measurements to test, then chooses a Chart
type XY (scatter) from Chart Wizard in Excel then draws a central line for an exact match
between two measurements, and adds two other lines either ve up or ve down to give
the accepted range for this ECG parameter.

Quality Control Testing -- Visual ECG Reading for Heart Rate

110
105
Heart Rate (/min) - Second Reading

100
95
90
85
80
75
70
65
60
55
50
45
45 50 55 60 65 70 75 80 85 90 95 100 105 110
Heart Rate (/min) - Standard (First Reading)

The Minimum Accepted Standard for Repeatability Tests

The difference between coders and the standard for heart rate should be less than 5 beats/
min, the difference for P-R interval and QRS duration should be less than 10 ms, the
difference for JT and QT intervals should be less than 20 ms, and the difference for QTI
and JTI should be less than 5%.

Repeatability of MI and continuous measurement coding from our Reading Center, EPI-
CARE, in the recent decade are shown in Tables 1 to 6.

273
(A) Internal Quality Control Analysis between the ECG Coder and Standard (Quarterly)

TABLE 17.1. The Agreement for Myocardial Infarction by Minnesota Code

(MC-MI see appendix A).
Final Adjudicated Code (Standard)
MC-MI = 0 MC-MI = 1 Total
Initial Visual Code
MC-MI = 0 34 1 35
MC-MI = 1 0 18 18
Total 34 19 53

Simple Kappa Coefcient

Kappa 0.9585
ASE 0.0411
95% Lower Conf Limit 0.8782
95% Upper Conf Limit 1.0000

TABLE 17.2. The Agreement for Myocardial Infarction by Novacode

(NC-MI see appendix B).
Final Adjudicated Code (Standard)
NC-MI = 0 NC-MI = 1 Total
Initial Visual Code
NC-MI = 0 36 2 38
NC-MI = 1 1 14 15
Total 37 16 53

Simple Kappa Coefcient

Kappa 0.8633
ASE 0.0765
95% Lower Conf Limit 0.7134
95% Upper Conf Limit 1.0000

TABLE 17.3. Correlation Analysis Between

the Coder and Standard.
Pearson Correlation Coefcients
Heart Rate 0.9981
P-R Interval 0.9829
QRS Duration 0.8092
Q-T Interval 0.9638
QTI 0.9064
QRS Axis 0.9803

274
 TABLE 17.4. T-Test for the Difference Between the Coder and Standard.
Variables Mean SD P Value
Heart Rate 0.26 0.68 0.0069
PR interval 0.19 5.71 0.8110
QRS Duration 0.19 7.00 0.8452
QT interval 1.43 9.07 0.2550
QTI 0.52 2.37 0.1159
QRS Axis 0.06 7.9 0.9589

(B) Intra-coder Quality Control Analysis between of an ECG coder (From 2000 2008)

TABLE 17.5. Kappa for Myocardial Infarction

by Minnesota Code & Novacode (see Appendies A and B).
Years MC-MI NC-MI
2000 0.9077 0.8100
2001 1.0000 0.8383
2002 0.9529 0.8628
2003 0.9113 0.8129
2004 0.9548 0.9459
2005 0.8643 0.9170
2006 0.8586 0.9462
2007 0.9548 0.8525
2008 0.9207 0.8591

TABLE 17.6. The Correlation for ECG Measurements with year 2008
Year Heart rate PR interval QRS duration QT interval

2001 0.9961 0.9965 0.8839 0.9087

2002 0.9977 0.9706 0.9361 0.9168
2003 0.9986 0.9733 0.9164 0.8787
2004 0.9907 0.9784 0.9454 0.9322
2005 0.9962 0.9562 0.9326 0.9145
2006 0.9979 0.9662 0.9379 0.9273
2007 0.9980 0.9681 0.9380 0.9272

Quality Control of Electronically Processed ECGs

The variability of current, directly electronically transmitted ECG source data will be 0%
due to the digital nature of the stored and transmitted data. This was not always so with
digital data transmitted on cassettes with wave averaging analysis.4 With current systems,
the median (most representative) P-QRS-T complex produced is used by EPICARE to
classify ECG ndings according to the Minnesota Code and Novacode algorithms. Interval
measurements by the program are ideal for the assessment of time trends. The measurements
are very robust, with the exception of the rare occurrences of missed detection of low
amplitude P waves and misplacement of the T wave at the end of the U wave when T-U

275
fusion takes place. Every ECG is checked for these possible wave detection errors and
an interactive computer graphics terminal with special software is used to correct these
errors. It can be categorically stated that when the global onsets and offsets of ECG waves
are properly detected, wave amplitude measurements used to assign Minnesota Codes are
invariable, done with a precision far superior to that possible with visual inspection (even
with fourfold magnication).

As with the use of any computer-ECG program, built-in safeguards have to be in place to
protect against software changes that may produce secular time trends in ECG measurements
due to possible software upgrades during a prolonged period of an observational study.
This task can be achieved by using a test library of say, 200+ ECGs enriched with abnormal
ECG patterns (MIs, conduction defects etc.). The ECGs from this library should be
processed using the Minnesota Code classication algorithms in the beginning of the study
and annually. Contingency tables demonstrating the invariability of classication can then
be produced for each Minnesota Code classication category.

Trend Analysis

The digital raw measurements for PR QT interval and QRS duration should be maintained
by a study Coordinating Center to check for unsuspected technological, recording procedural
changes, or editing changes that might occur during the course of a long recruitment
period as in some population based epidemiologic studies or clinic-based clinical trials.
Monitoring of QRS, QT and PR intervals will reect seasonal variation due to heart rate
(HR) sensitive intervals, such as QT. With HR correction of QT, trend analysis based on
electronic measurements (obtained from the electronically transmitted ECGs from the
eld centers or clinics) will provide data on aging changes due to biological alteration
of these intervals with time. More importantly, any sudden unexplained deviation from
a steady state in these parameters would signal procedural or software alteration (at local
electrocardiographs or central analytic programs) that can be investigated and corrected by
the study directorate. Such monitoring may uncover problems not otherwise apparent from
other QC monitoring. For the same reasons, QRS and T axis should be monitored for trend
analysis reecting frontal plane voltage changes.

References

1. Streiner DL, Norman GR. Health Measurement Scales. 4th ed. Oxford: Oxford University
Press; 1994.
2. Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall; 1991.
3. SAS Version 9.1: The FREQ Procedure. Cary, North Carolina. SAS Institute, Inc.,
4. Rautaharju PM, Broste SK, Prineas RJ, et al. Quality control procedures for the resting
electrocardiogram in the Multiple Risk Factor Intervention Trial. Controlled Clin Trials.
1986;7:46S-65S.

276
Appendix A
MINNESOTA CODE 2009

Q and QS Patterns

Do not code in the presence of WPW code 6-4-1, or articial pacemaker code 6-8 or code
6-1, 8-2-1. 8-2-2, or 8-4-1 with a heart rate 140. To qualify as a Q-wave, the deection
should be at least 0.1 mV (1 mm in amplitude).

Anterolateral Site (leads I, aVL, V6)

1-1-1 Q/R amplitude ratio 1/3, plus Q duration 0.03 s in lead I or V6.
1-1-2 Q duration 0.04 s* in lead I or V6.
1-1-3 Q duration 0.04 s plus R amplitude 3 mm in lead aVL.
1-2-1 Q/R amplitude ratio 1/3, plus Q duration 0.02 s and < 0.03 s in lead I or V6.
1-2-2 Q duration 0.03 s and < 0.04 s in lead I or V6.
1-2-3 QS pattern in lead I. Do not code in the presence of 7-1-1.
1-3-1 Q/R amplitude ratio 1/5 and < 1/3, plus Q duration 0.02 s and < 0.03 s in
lead I or V6.
1-3-3 Q duration 0.03 s and < 0.04 s, plus R amplitude 3 mm in lead aVL.
1-3-81 Initial R amplitude decreasing to 2 mm or less in every beat (and absence of codes
3-2, 7-1-1, 7-2-1, or 7-3) between V5 and V6. (All beats in lead V5 must have an
initial R > 2 mm.)

Posterior (inferior) site (leads II, III, aVF)

1-1-1 Q/R amplitude ratio 1/3, plus Q duration 0.03 s in lead II.
1-1-2 Q duration 0.04 s in lead II.
1-1-4 Q duration 0.05 s in lead III, plus a Q-wave amplitude 1.0 mm in the majority
of beats in lead aVF.
1-1-5 Q duration 0.05 s in lead aVF.
1-2-1 Q/R amplitude ratio 1/3, plus Q duration 0.02 s and < 0.03 s in lead II.
1-2-2 Q duration 0.03 s and < 0.04 s in lead II.
1-2-3 QS pattern in lead II. Do not code in the presence of 7-1-1.
1-2-4 Q duration 0.04 s and < 0.05 s in lead III, plus a Q-wave 1.0 mm amplitude in
the majority of beats in aVF.
1-2-5 Q duration 0.04 s and < 0.05 s in lead aVF.
1-3-1 Q/R amplitude ratio 1/5 and < 1/3, plus Q duration 0.02 s and < 0.03 s in lead II.
1-3-4 Q duration 0.03 s and < 0.04 s in lead III, plus a Q-wave 1.0 mm amplitude in
the majority of beats in lead aVF.
1-3-5 Q duration 0.03 s and < 0.04 s in lead aVF.
1-3-6 QS pattern in each of leads III and aVF. (Do not code in the presence of 7-1-1.)
1-3-72 QS pattern in lead aVF only. (Do not code in the presence 7-1-1)
* s = second
277
Anterior Site (leads V1, V2, V3, V4, V5)
1-1-1 Q/R amplitude ratio 1/3 plus Q duration 0.03 s in any of leads V2, V3, V4, V5.
1-1-2 Q duration 0.04 s in any of leads V1, V2, V3, V4, V5.
1-1-6 QS pattern when initial R-wave is present in adjacent lead to the right on the chest,
in any of leads V2, V3, V4, V5, V6.
1-1-7 QS pattern in all of leads V1-V4 or V1-V5.
1-2-1 Q/R amplitude ratio 1/3 plus Q duration 0.02 s and < 0.03 s, in any of leads V2,
V3, V4, V5.
1-2-2 Q duration 0.03 s and < 0.04 s in any of leads V2, V3, V4, V5.
1-2-7 QS pattern in all of leads V1, V2, and V3. (Do not code in the presence of 7-1-1.)
1-3-1 Q/R amplitude ratio 1/5 and < 1/3, plus Q duration 0.02 s and < 0.03 s in any of
leads V2, V3, V4, V5.
1-3-2 QS pattern in lead V1 and V2. (Do not code in the presence of 3-1 or 7-1-1.)
1-3-81 Initial R amplitude decreasing to 2.0 mm or less in every beat (and absence of codes
3-2, 7-1-1, 7-2-1, or 7-3) between any of leads V2 and V3, V3, and V4, or V4 and V5.
(All beats in the lead immediately to the right on the chest must have an initial R >
2 mm.)

QRS Axis Deviation

Do not code in presence of low-voltage QRS code 9-1, WPW 6-4-1, articial pacemaker
code 6-8, ventricular conduction defects 7-1-1, 7-2-1, 7-4, or 7-8.
2-1 Left. QRS axis from 30 through 90 in leads I, II and III. (The algebraic sum of
major positive and major negative QRS waves must be zero or positive in I, negative
in III, and zero or negative in II.)
2-2 Right. QRS axis from +120 through 150 in leads I, II, and III. (The algebraic
sum of major positive and major negative QRS waves must be negative in I, and
zero or positive in III, and in I must be one-half or more of that in III.)
2-3 Right (optional code when 2-2 is not present). QRS axis from +90 through +119
in leads I, II, and III. (The algebraic sum of major positive and major negative QRS
waves must be zero or negative in I and positive in II and III.)
2-4 Extreme axis deviation (usually S1, S2, S3 pattern). QRS axis from 90 through
149 in leads I, II and III. (The algebraic sum of major positive and major negative
QRS waves must be negative in each of leads I, II, and III.)
2-5 Indeterminate axis. QRS axis approximately 90 from the frontal plane. (The algebraic
sum of major positive and major negative QRS waves is zero in each of leads I, II
and III, or the information from these three leads is incongruous.)

High Amplitude R Waves

Do not code in the presence of codes 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.
3-1 Left: R amplitude > 26 mm in either V5 or V6, or R amplitude > 20.0 mm in any
of leads I, II, III, aVF, or R amplitude > 12.0 mm in lead aVL measured only on
second to last complete normal beat.

278
3-2 Right: R amplitude 5.0 mm and R amplitude S amplitude in the majority of
beats in lead V1, when S amplitude is > R amplitude somewhere to the left on the
chest of V1 (codes 7-3 and 3-2, if criteria for both are present).
3-3 Left (optional code when 3-1 is not present): R amplitude > 15.0 mm but 20.0 mm
in lead I, or R amplitude in V5 or V6, plus S amplitude in V1 > 35.0 mm.
3-4 Criteria for 3-1 and 3-2 both present.

ST Junction (J) and Segment Depression

Do not code in the presence of codes 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8. When 4-1, 4-2,
or 4-3 is coded, then a 5-code most often must also be assigned except in lead V1.

Anterolateral Site (leads I, aVL, V6)

4-1-1 STJ depression 2.0 mm and ST segment horizontal or downward sloping in any
of leads I, aVL, or V6.
4-1-2 STJ depression 1.0 mm but < 2.0 mm, and ST segment horizontal or downward
sloping in any of leads I, aVL, or V6.
4-2 STJ depression 0.5 mm but < 1.0 mm and ST segment horizontal or downward
sloping in any of leads I, aVL, or V6.
4-3 No STJ depression as much as 0.5 mm, but ST segment downward sloping and segment
or T-wave nadir 0.5 mm below P-R baseline, in any of leads I, aVL, or V6.
4-4 STJ depression 1.0 mm and ST segment upward sloping or U-shaped, in any of
leads I, aVL, or V6.

Posterior (inferior) Site (leads II, III, aVF)

4-1-1 STJ depression 2.0 mm and ST segment horizontal or downward sloping in lead
II or aVF.
4-1-2 STJ depression 1.0 mm but < 2.0 mm and ST segment horizontal or downward
sloping in lead II or aVF.
4-2 STJ depression 0.5 mm but < 1.0 mm and ST segment horizontal or downward
sloping in lead II or aVF.
4-3 No STJ depression as much as 0.5 mm, but ST segment downward sloping and
segment or T-wave nadir 0.5 mm below P-R baseline in lead II.
4-4 STJ depression 1.0 mm and ST segment upward sloping, or U-shaped, in lead II.

Anterior Site (leads V1, V2, V3, V4, V5)

4-1-1 STJ depression 2.0 mm and ST segment horizontal or downward sloping in any
of leads V1, V2, V3, V4, V5.
4-1-2 STJ depression 1.0 mm but < 2.0 mm and ST segment horizontal or downward
sloping in any of leads V1, V2, V3, V4, V5.
4-2 STJ depression 0.5 mm but < 1.0 mm and ST segment horizontal or downward
sloping in any of leads V1, V2, V3, V4, V5.
4-3 No STJ depression as much as 0.5 mm, but ST segment downward sloping and segment
or T-wave nadir 0.5 mm below P-R baseline in any of leads V2, V3, V4, V5.
4-4 STJ depression 1.0 mm and ST segment upward sloping or U-shaped in any of
leads V1, V2, V3, V4, V5.

279
T-Wave Items

Do not code in the presence of codes 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.

Anterolateral Site (leads I, aVL, V6)

5-1 T amplitude negative 5.0 mm or more in either of leads I, V6, or in lead aVL when
R amplitude is 5.0 mm.
5-2 T amplitude negative or diphasic (positivenegative or negativepositive type) with
negative phase at least 1.0 mm but not as deep as 5.0 mm in lead I or V6, or in lead
aVL when R amplitude is 5.0 mm.
5-3 T amplitude zero (at), or negative, or diphasic (negativepositive type only) with
less than 1.0 mm negative phase in lead I or V6, or in lead aVL when R amplitude
is 5.0 mm.
5-4 T amplitude positive and T/R amplitude ratio < 1/20 in any of leads I, aVL, V6; R
wave amplitude must be 10.0 mm.

Posterior (Inferior) Site (leads II, III, aVF)

5-1 T amplitude negative 5.0 mm or more in lead II, or in lead aVF when QRS is
mainly upright.
5-2 T amplitude negative or diphasic with negative phase (negativepositive or positive
negative type) at least 1.0 mm but not as deep as 5.0 mm in lead II, or in lead aVF
when QRS is mainly upright.
5-3 T amplitude zero (at), or negative, or diphasic (negative-positive type only) with
less than 1.0 mm negative phase in lead II; not coded in lead aVF.
5-4 T amplitude positive and T/R amplitude ratio < 1/20 in lead II; R wave amplitude
must be 10.0 mm.

Anterior Site (leads V2, V3, V4, V5)

5-1 T amplitude negative 5.0 mm or more in any of leads V2, V3, V4, V5.
5-2 T amplitude negative, or diphasic (negativepositive or positivenegative type)
with negative phase at least 1.0 mm but not as deep as 5.0 mm, in any of leads V2,
V3, V4, V5.
5-3 T amplitude zero (at), or negative, or diphasic (negativepositive type only) with
less than 1.0 mm negative phase, in any of leads V3, V4, V5.
5-4 T amplitude positive and T/R amplitude ratio < 1/20 in any of leads V3, V4, V5; R
wave amplitude must be 10.0 mm.

A-V Conduction Defect

6-1 Complete (third degree) A-V block (permanent or intermittent) in any lead. Atrial
and ventricular complexes independent, and atrial rate faster than ventricular rate,
with ventricular rate < 60.
6-2-1 Mobitz Type II (occurrence of P-wave on time with dropped QRS and T).
6-2-2 Partial (second degree) A-V block in any lead (2:1 or 3:1 block).
6-2-3 Wenckebachs Phenomenon (P-R interval increasing from beat to beat until QRS
and T dropped).

280
6-3 P-R (P-Q) interval 0.22 s in the majority of beats in any of leads I, II, III, aVL, aVF.
6-4-1 Wolff-Parkinson-White Pattern (WPW), persistent. Sinus P-wave. P-R interval
< 0.12 s, plus QRS duration 0.12 s, plus R peak duration 0.06 s, coexisting in
the same beat and present in the majority of beats in any of leads I, II, aVL, V4, V5, V6.
(6-4-1 suppresses 1-2-3, 1-2-7, 1-3-2, 1-3-6, 1-3-8, all 3, 4, 5, 7, 9-2, 9-4, 9-5 codes.)
6-4-2 WPW Pattern, intermittent. WPW pattern in 50% of beats in appropriate leads.
6-5 Short P-R interval. P-R interval < 0.12 s in all beats of any two of leads I, II, III,
aVL, aVF.
6-6 Intermittent aberrant atrioventricular conduction. P-R > 0.12 s (except in presence
of 6-5 or heart rate greater than 100), and wide QRS complex > 0.12 s, and normal
P-wave when most beats are sinus rhythm. (Do not code in the presence of 6-4-2.)
6-8 Electronic pacemaker.

Ventricular Conduction Defect

7-1-1 Complete left bundle branch block (LBBB). (Do not code in presence of 6-1, 6-4-1,
6-8, 8-2-1 or 8-2-2.) QRS duration 0.12 s in a majority of beats (of the same
QRS pattern) in any of leads I, II, III, aVL, aVF, plus R peak duration 0.06 s in a
majority of beats (of the same QRS pattern) in any of leads I, II, aVL, V5, V6.
(7-1-1 suppresses 1-2-3, 1-2-7, 1-3-2, 1-3-6, 1-3-7, 1-3-8, all 2, 3, 4, 5, 9-2, 9-4, 9-5
codes. If any other codable Q-wave coexists with the LBBB pattern, code the Q and
diminish the 7-1-1 code to a 7-4 code.)
7-1-2 Intermittent left bundle branch block. Same as 7-1-1 but with presence of normally
conducted QRS complexes of different shape than the LBBB pattern.
7-2-1 Complete right bundle branch block (RBBB). (Do not code in the presence of 6-1,
6-4-1, 6-8, 8-2-1 or 8-2-2.) QRS duration 0.12 s in a majority of beats (of the same
QRS pattern) in any of leads I, II, III, aVL, aVF, plus: R > R in V1; or QRS mainly
upright, plus R peak duration 0.06 s in V1 or V2; or S duration > R duration in all
beats in lead I or II. (Suppresses 1-3-8, all 2-, 3-, 4-and 5-codes, 9-2, 9-4, 9-5.)
7-2-2 Intermittent right bundle branch block. Same as 7-2-1 but with presence of normally
conducted QRS complexes of different shape than the RBBB pattern.
7-3 Incomplete right bundle branch block. QRS duration < 0.12 s in each of leads I, II,
III, aVL, aVF, and R > R in either of leads V1, V2 (Code as 3-2 in addition if those
criteria are met. 7-3 suppresses code 1-3-8.)
7-4 Intraventricular block. QRS duration 0.12 s in a majority of beats in any of leads
I, II, III, aVL, aVF. (7-4 suppresses all 2, 3, 4, 5, 9-2, 9-4, 9-5 codes.)
7-5 R-R pattern in either of leads V1, V2 with R amplitude R.
7-6 Incomplete left bundle branch block. (Do not code in the presence of any codable
Q- or QS-wave.) QRS duration 0.10 s and < 0.12 s in the majority of beats of each
of leads I, aVL, and V5 or V6.
7-7 Left anterior hemiblock (LAH). QRS duration < 0.12 s in the majority of beats in
leads I, II,III, aVL, aVF, plus Q-wave amplitude 0.25 mm and < 0.03 s duration
in lead I or aVL, plus left axis deviation of 45 or more negative. (In presence of
7-2, code 7-8 if axis is < 45 and the Q-wave in lead I meets the above criteria.)
7-8 Combination of 7-7 and 7-2.

281
7-9-12 Type 1 Brugada pattern convex (coved) ST segment elevation 2 mm plus T-wave
negative with little or no isolelectric (baseline) separation in at least 2 leads of V1-V3.
2
7-9-2 Type 2 Brugada pattern ST segment elevation 2 mm plus T-wave positive or
diphasic that results in a saddle-back shape in at least 2 leads of V1-V3.
2
7-9-3 Type 3 Brugada pattern. 7-2-1 plus ST segment elevation 1 mm plus a saddle-back
conguration in at least 2 leads of V1V3.
2
7-10 Fragmented QRS.

Arrhythmias

8-1-1 Presence of any atrial or junctional premature beats.

8-1-2 Presence of any ventricular premature beats.
8-1-3 Presence of both atrial and/or junctional premature beats and ventricular premature
beats.
8-1-4 Wandering atrial pacemaker.
8-1-5 Presence of 8-1-2 and 8-1-4.
8-2-1 Ventricular brillation or ventricular asystole.
8-2-2 Persistent ventricular (idioventricular) rhythm.
8-2-3 Intermittent ventricular tachycardia. Three or more consutive ventricular premature
beats occurring at a rate 100. This includes more persistent ventricular tachycardia.
8-2-4 Ventricular parasystole (should not be coded in presence of 8-3-1).
8-3-1 Atrial brillation (persistent).
8-3-2 Atrial utter (persistent).
8-3-3 Intermittent atrial brillation (code if 3 or more clear-cut, consutive sinus beats are
present in any lead).
8-3-4 Intermittent atrial utter (code if 3 or more clear-cut, consutive sinus beats are
present in any lead).
8-4-1 Supraventricular rhythm persistent. QRS duration < 0.12 s; and absent P-waves or
presence of abnormal P-waves (inverted or at in II, III and aVF); and regular rhythm.
8-4-2 Supraventricular tachycardia intermittent. Three consutive atrial or junctional pre-
mature beats occurring at a rate 100.
8-5-1 Sinoatrial arrest. Unexpected absence of P, QRS and T, plus a R-R interval at a
xed multiple of the normal interval, 10%.
8-5-2 Sinoatrial block. Unexpected absence of P, QRS and T, preceded by progressive
shortening of P-P intervals. (R-R interval at a xed multiple of the normal interval,
10%.)
8-6-1 A-V dissociation with ventricular pacemaker (without capture). Requires: P-P and
R-R occur at variable rates with ventricular rate as fast as or faster than the atrial
rate, plus variable P-R intervals, plus no capture beats.
8-6-2 A-V dissociation with ventricular pacemaker (with capture).
8-6-3 A-V dissociation with atrial pacemaker (without capture).
8-6-4 A-V dissociation with atrial pacemaker (with capture).
8-7 Sinus tachycardia (100/min).
8-8 Sinus bradycardia (50/min).
8-9 Other arthythmias. Heart rate may be recorded as a continuous variable.

282
ST Segment Elevation

Do not code in the presence of codes 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.

Anterolateral Site (leads I, aVL, V6)

9-2 ST segment elevation 1.0 mm in any of leads I, aVL, V6.

Posterior (Inferior) Site (leads II, III, aVF)

9-2 ST segment elevation 1.0 mm in any of leads II, III, aVF.

Anterior site (Leads V1, V2, V3, V4, V5)

9-2 ST segment elevation 1.0 mm in lead V5 or ST segment elevation 2.0 mm in
any of leads V1, V2, V3, V4.

Miscellaneous Items

9-1 Low QRS amplitude. QRS peak-to-peak amplitude < 5 mm in all beats in each of
leads I, II, III, or < 10 mm in all beats in each of leads V1, V2, V3, V4, V5, V6. (Check
calibration before coding.)
9-3 P-wave amplitude 2.5 mm in any of leads II, III, aVF, in a majority of beats.
9-4-1 QRS transition zone at V3 or to the right of V3 on the chest. (Do not code in the
presence of 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.)
9-4-2 QRS transition zone at V4 or to the left of V4 on the chest. (Do not code in the
presence of 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.)
9-5 T-wave amplitude > 12 mm in any of leads I, II, III, aVL, aVF, V1, V2, V3, V4, V5,
V6. (Do not code in the presence of 6-4-1, 6-8, 7-1-1, 7-2-1, 7-4, or 7-8.)
9-62 Notched and widened P wave (duration 0.12 s.) in frontal plane (usually lead II),
and/or deep negative component to the P wave in lead V1 duration 0.04 s. and
depth 1 mm.
9-7-12 Definite Early Repolarization.STJ elevation 1mm in the majority of beats, T wave
amplitude 5 mm, prominent J point, upward concavity of the ST segment, and a
distinct notch or slur on the down-stroke of the R wave in any of V3 V6, OR STJ
elevation 2 mm in the majority of beats and T wave amplitude 5 mm, prominent
J point and upward concavity of the ST segment in any of V3 V6.
2
9-7-2 Probable Early Repolarization. STJ elevation 1 mm in the majority of beats,
prominent J point, and upward concavity of the ST segment in any of V3 V6 and
T wave amplitude 8 mm in any of the leads V3 V6.
9-8-12 Uncorrectable lead reversal.
9-8-23 Poor Quality/Technical problems which interfere with coding.
9-8-32 Correctable lead reversal
i. Correctable limb lead connection error
ii. Correctable chest lead connection error in V1-V3
iii. Correctable chest lead connection error in V4-V6
iv. Correctable other chest lead connection error
3
9-8-4 Technical problems that do not interfere with coding.

283
Incompatible Codes

The codes in the left column suppress codes in the right column.
Code Suppresses this code(s)
All Q-, QS-codes 7-6
Q 0.03 in lead I 7-7
3-1 1-3-2
3-2 1-3-8, 7-3
6-1 All other codes except 8-2
6-4-1 All other codes
6-8 All other codes
7-1-1 1-2-3, 1-2-7, 1-3-2, 1-3-6, 1-3-7, 1-3-8, all 2-, 3-, 4-, and
5-codes, 7-7, 7-8, 7-9, 7-10, 9-2, 9-4, 9-5, 9-7-1, 9-7-2
7-2-1 1-3-8, all 2-, 3-, 4-, and 5-codes, 9-2, 9-4, 9-5, 9-7-1, 9-7-2
7-3 1-3-8
7-4 All 2-, 3-, 4-, and 5-codes, 9-2, 9-4, 9-5
7-8 1-3-8, all 2-, 3-, 4-, and 5-codes, 9-2, 9-4, 9-5, 9-7-1, 9-7-2
8-1-2 8-2-4
8-1-4 8-1-1, 9-3
8-2-1 All other codes
8-2-2 All other codes
8-2-3 8-1-2
8-3-1 8-1-1, 8-1-2
8-3-2 6-2-2, 8-1-1, 8-1-2
8-3-3 8-1-1, 8-1-2
8-3-4 6-2-2
8-4-1 6-5
8-4-1 + heart rate 140 All other codes except 7-4 or 6-2
Heart rate > 100 6-5
8-4-2 8-1-1
9-1 All 2-codes
1
1-3-8 was previously 1-2-8
2
New code from rst edition
3
9-8-2 in the rst edition was 9-8-1, and 9-8-4 was 9-8-2 in the rst edition.

ECG Criteria for Signicant Serial ECG Change

See Chap. 15 for serial comparison rules

Evolving Q-wave
Q1. No Q-code in reference ECG followed by a record with a diagnostic Q-code (MC
1-1-1 through 1-2-7) OR an Equivocal Q-code (any MC 1-3-x) in reference ECG
followed by record with any code 1-1-x Q-code.
Q2. An Equivocal Q-code (any MC 1-3-x code) and no major ST-segment depression
(MC 4-0, 4-4, 4-3) in reference ECG followed by a record with a diagnostic Q-
code (MC 1-2-1 1-2-7) Plus a major ST-segment depression (MC 4-1-x or 4-2).
284
Q3. An Equivocal Q-code (any MC 1-3-x) and no major T-wave inversion (MC 5-4,
5-3 or 5-0) in reference ECG followed by a record with a diagnostic Q-code
(MC1-2-1 through 1-2-7) Plus a major T-wave inversion (MC 5-1 or 5-2).
Q4. An Equivocal Q-code (any MC 1-3-x) and no ST-elevation in reference ECG
followed by a record with a diagnostic Q-code (MC1-2-1 through 1-2-7) Plus
an ST-segment elevation (MC 9-2).
Q5. No Q-code and no MC 4-1-x or 4-2 in reference ECG followed by a record with
an Equivocal Q- code (any MC 1-3-x) Plus MC 4-1-x or 4-2.
Q6. No Q-code and no MC 5-1 or 5-2 in reference ECG followed by a record with
an Equivocal Q- code (any MC 1-3-x) Plus a MC 5-1 or 5-2.
Q7. No Q-code and no MC 9-2 in reference ECG followed by a record with an
Equivocal Q-code (any MC 1-3-x) Plus a MC 9-2.

Evolving ST-Elevation
STE-1 MC 9-0 in reference ECG followed by a record with MC 9-2 in at least 2 leads
and >100% increase ST elevation in both leads.
STE-2 MC 9-2 in reference ECG followed by a record with MC 9-2 in at least 2 leads
and >100% increase in ST elevation in both leads.
STE-3 MC 9-2 and no MC 5-1 or 5-2 in reference ECG followed by a record appear-
ance of MC 5-1 or 5-2 with 100% increase in T wave inversion in at least 2
leads.
STE-4 Reversal of evolving STE-1 (within the hospital ECG only).
STE-5 Reversal of evolving STE-2 (within the hospital ECG only).

Evolving ST-Depression / T Wave Inversion

ST-T1 Either MC 4-0 (no 4-code), 4-4 or 4-3 in reference ECG followed by a record with
MC 4-2 or 4-1-2 or 4-1-1 and > 100% increase in ST segment depression.
ST-T2 Either MC 4-2 or 4-1-2 in reference ECG followed by a record with MC 4-1-1
and >100% increase in ST segment depression.
ST-T3 Either MC 5-0, 5-4 or 5-3 in reference ECG followed by a record with MC 5-2
or 5-1 and > 100% increase in T-wave inversion.
ST-T4 MC 5-2 in reference ECG followed by a record with MC 5-1 and >100% in
T-wave inversion.
ST-T5 MC 4-1-1 in reference ECG followed by a record with MC 4-1-1 and > 100%
increase in ST depression.
ST-T6 MC 5-1 in reference ECG followed by a record with MC 5-1 and >100%
increase in T-wave inversion.
ST-T7 MC 5-2 in reference ECG followed by a record with MC 5-2 and >100%
increase in T-wave inversion.
ST-T1R Reverse of ST-T14
ST-T2R Reverse of ST-T24
ST-T3R Reverse of ST-T34
ST-T4R Reverse of ST-T44
ST-T5R Reverse of ST-T54

285
ST-T6R Reverse of ST-T64
ST-T7R Reverse of ST-T74
4
Requires >100% decrease in ST depression or T-wave inversion of follow-up record compared to
reference ECG, and code changes must occur in the same lead groups.

Evolving Bundle Branch Block

E-BBB1 No MC 7-1 in the reference ECG followed by an ECG with MC 7-1-1 in fol-
low-up ECG and QRS duration increased by > 0.02 s.
E-BBB2 No MC 7-2 in the reference ECG followed by an ECG with MC 7-2-1 in fol-
low-up ECG and QRS duration increased by > 0.02 s.
E-BBB3 No MC 7-4 in the reference ECG followed by an ECG with MC 7-4 in follow-
up ECG and QRS duration increased by > 0.02 s.

Evolving ECG LVH 5

E-LVH 1 MC 3-0 in reference ECG followed by an ECG with a MC 3-1in the follow-up
ECG, conrmed as a signicant increase.
E-LVH 2 MC 3-0 in reference ECG followed by an ECG with a MC 3-3 in the follow-up
ECG, conrmed as a signicant increase.
E-LVH 3 MC 3-1 in reference ECG followed by an ECG with a MC 3-0 in the follow-up
ECG, conrmed by a signicant decrease.
E-LVH 4 MC 3-3 in reference ECG followed by an ECG with a MC 3-0 in the follow-up
ECG, conrmed by a signicant decrease.
E-LVH 5 MC 3-1 in the reference ECG followed by an ECG with a MC 3-1 in the fol-
low-up ECG, conrmed by a signicant increase or a signicant decrease.
E-LVH 6 MC 3-3 in the reference ECG followed by an ECG with a MC3-3 in the follow-
up ECG, conrmed by a signicant increase or signicant decrease.

5
see page 232 for signicant 3-code change

286
Appendix B
The Novacode Criteria for Classication of ECG Abnormalities
and Their Clinically Signicant Progression and Regression
The Novacode classication system is an extension of the Minnesota Code. It was
developed initially in the late 1980s and further rened in 1998 and is still evolving.1,2 The
Novacode ECG classication system provides a comprehensive hierarchical set of criteria
for prevalent ECG abnormalities and for clinically signicant serial ECG changes.

The Structure of the Novacode -- Coding Categories for Prevalent ECG Abnormalities
Baseline ECGs are commonly used to categorize a study population into groups based
on major and minor abnormalities. The sequential coding categories of prevalent ECG
abnormalities are summarized in Table B.1 (with corresponding Minnesota Codes). The
rst category of the coding system is applied to indicate conditions that suppress coding
of some or all ECG abnormalities (code 0). The other categories are rhythms (code
1), atrioventricular conduction (code 2), complete bundle branch blocks (code 3), prolonged
repolarization (code 4), myocardial infarction (MI) and ischemia (code 5), left ventricular
hypertrophy (LVH) (code 6), left atrial enlargement (code 7), right ventricular hypertrophy
(code 8), right atrial enlargement (code 9), and fascicular blocks (code 10).

Detailed denitions for various ECG criteria used in the code for prevalent ECG
abnormalities are listed below, which also contains denitions of the variables used to
denote ECG measurements and waveform labels, as well as a table that provided a simple
scheme for visual assessment of ECG record quality.

Measurement
The rules for basic ECG measurement and calculation as heart rate, durations and amplitudes
for P-Q-R-S-T, intervals for P-R, Q-T, and QRS axis are same as the rules of the Minnesota
Code in this manual, except the cut point for the codable Q wave (75 V for Novacode and
100 V for Minnesota Code).

TABLE B.1. Classication Code for Prevalent ECG Abnormalities (with Corresponding Minnesota
Codes).
Denition and Description Novacode Minnesota Code
Baseline ECG Suppression Codes Novacode 0
ECG not available NC-0.1 MC-9.8.2
Inadequate quality or missing leads NC-0.2
Inadequate quality NC-0.2.1 MC-9.8.2
Missing leads NC-0.2.2 MC-9.8.2
Lead connection interchange NC-0.3
Uncorrectable lead connection interchanges (ECG uncodable) NC-0.3.1.x MC-9.8.1

287
TABLE B.1. (continued)
Denition and Description Novacode Minnesota Code
Correctable limb lead connection error (ECG codable) NC-0.3.2.x MC-9.8.3
Correctable chest lead in V1-V3 connection error (ECG codable) NC-0.3.3.x MC-9.8.3
Correctable chest lead in V4-V6 connection error (ECG codable) NC-0.3.4.x MC-9.8.3
Correctable other chest lead connection error (ECG codable) NC-0.3.5.x MC-9.8.3
QRS duration 120 ms NC-0.4 MC-7.1, MC-7.2, MC-7.4,
MC-7.8, MC-6.4
Atrial brillation or atrial utter NC-0.5 MC-8.3.1, MC-8.3.2, MC-8.3.3
Electronic pacemaker NC-0.6 MC-6.8
Other suppression codes NC-0.7 MC-6.2, MC-6.1, MC-8.6
Uncertain P wave detection NC-0.8

Rhythm Codes Novacode 1

Basic sinus rhythm (SR) NC-1.0
Normal sinus rhythm (NSR), rate from 5194 cpm NC-1.0.1
Sinus Bradycardia (SB) NC-1.0.2 MC-8.8
Sinus Tachycardia (ST) NC-1.0.3 MC-8.7
Supplementary codes to sinus rhythm
With ectopic supraventricular complexes (ESVC) NC-1.0.S.1 MC-8.1.1
With aberrant supraventricular complexes (ASVC) NC-1.0.S.2 MC-8.1.1
With ectopic ventricular complexes (EVC) NC-1.0.S.3 MC-8.1.2
With pause (possible sinal arrest or sinoatrial block) NC-1.0.S.4 MC-8.5.1
Wandering atrial pacemaker (WAP) NC-1.1 MC-8.1.4
Junctional rhythm (JR) NC-1.2
Junctional rhythm, rate from 45-64 cpm NC-1.2.1 MC-8.4.1
Junctional bradycardia (JB) NC-1.2.2 MC-8.4.1
Accelerated junctional rhythm (AJR) NC-1.2.3 MC-8.4.1
Ectopic atrial rhythm (EAR) NC-1.3
Ectopic atrial rhythm, rate from 5090 cpm NC-1.3.1 MC-8.4.1
Ectopic atrial bradycardia (EAB) NC-1.3.2 MC-8.4.1
Ectopic atrial tachycardia (EAT) NC-1.3.3 MC-8.4.1
Supraventricular tachycardia (SVT) NC-1.4
Supraventricular tachycardia, rate < 130 cpm NC-1.4.1 MC-8.4.2
Supraventricular tachycardia, rate 130 cpm NC-1.4.2 MC-8.4.2
Atrial utter or atrial brillation (AFLF) NC-1.5
Atrial utter type 1 (AFL1) NC-1.5.1 MC-8.3.2
Atrial utter type 2 (AFL2) NC-1.5.2 MC-8.3.2
Atrial brillation (AF) NC-1.5.3 MC-8.3.1
Electronic pacemaker (PM) NC-1.6
Ventricular pacemaker (VPM) or combination pacemaker (CPM) NC-1.6.1 MC-6.8
Atrial pacemaker only (APM) NC-1.6.2 MC-6.8
Ventricular Tachycardia (VT) NC-1.7 MC-8.2.3
Other abnormal rhythm codes NC-1.8
Other Atrial Rhythms NC-1.8.1 MC-8.9
Other Ventricular Rhythms NC-1.8.2 MC-8.2.x
Indeterminate rhythm classication NC-1.9
Inderminate Atrial Rhythms NC-1.9.1 MC-8.9
Indeterminate ventricular rhythm NC-1.9.2 MC-8.2.x

AV Conduction Abnormalities Novacode 2

First-degree AV block (AVB1) NC-2.1 MC-6.3
Second-degree AV block (AVB2) NC-2.2
Second-degree AV block type Wenckebach or Mobitz 1 (AVB2W) NC-2.2.1 MC-6.2.3
Second-degree singular AV block or type Mobitz 2 (AVB2S) NC-2.2.2 MC-6.2.2
Second-degree multiple AV block (AVB2M) NC-2.2.3 MC-6.2.1
High-grade AV dissociation (AVD) NC 2.3
Third-degree (complete) AV block (AVB3) NC-2.3.1 MC-6.1
AV dissociation with capture (AVDC) NC-2.3.2 MC-8.6.x
Ventricular preexcitation pattern (WPW) NC-2.4 MC-6.4
(continued)

288
TABLE B.1. (continued)
Denition and Description Novacode Minnesota Code
Prolonged Ventricular Excitation Novacode 3
Left bundle branch block (LBBB) NC-3.1
LBBB without ECG evidence of myocardial infarction (MI) NC-3.1.0 MC-7.1
LBBB with possible MI NC-3.1.1 MC-7.4 plus MC-1.x
Right bundle branch block (RBBB) NC-3.2
RBBB without ECG evidence of MI NC-3.2.0 MC-7.2, MC-7.8
RBBB with possible MI NC-3.2.1 MC-7.2, 7.8 plus MC-1.x
Indeterminate ventricular conduction delay (IVCD) NC-3.3
IVCD without ECG evidence of MI NC-3.3.0 MC-7.4
IVCD with possible MI NC-3.3.1 MC-7.4 plus MC-1.x
Borderline prolonged ventricular excitation NC-3.4
Borderline delay of right ventricular excitation NC-3.4.1a MC-7.3
Borderline delay of left ventricular excitation NC-3.4.2 MC-7.6
Prolonged Ventricular Repolarization Novacode 4a
Prolonged ventricular repolarization NC-4.1
Marginal prolongation of ventricular repolarization NC-4.1.1
Signicant prolongation of ventricular repolarization NC-4.1.2

ECG Categories Associated With Myocardial

Infarction / Ischemia Novacode 5a MC-1, MC-4, MC-5, MC-9.2
Q wave MI
Q wave MI; major Q waves with or without ST-T abnormalities NC-5.1 MC-1.1.x
Q wave MI; moderate Q waves with ST-T abnormalities NC-5.2 MC-1.2.x plus MC-4.1, 4.2,
MC-5.1, 5.2
Possible Q wave MI; moderate Q waves without ST-T abnormalities NC-5.3 MC-1.2.x
Possible Q wave MI; minor Q waves with ST-T abnormalities NC-5.4 MC-1.3.x plus MC-4.1, 4.2,
MC-5.1, 5.2
Isolated ischemic abnormalities
ST abnormalities without Q waves NC-5.5 MC-4.1, 4.2
T wave abnormalities without Q waves NC-5.6 MC-5.1, 5.2
Isolated minor Q and ST- T abnormalities
Minor Q waves without ST - T abnormalities NC-5.7 MC-1.3.x
Minor ST-T abnormalities NC-5.8 MC-4.3, 4.4, MC-5.3, 5.4

Left Ventricular Hypertrophy Novacode 6*

Left ventricular hypertrophy (LVH) NC-6.1
Left ventricular hypertrophy without ST-T NC-6.1.0 MC-3.1 (if meet the Cornell Votage Criteria)
Left ventricular hypertrophy with ST-T NC-6.1.1 MC-3.1 plus MC-5.1, 5.2,
MC-4.1, 4.2
(if meet the Cornell Voltage Criteria)

Left Atrial Enlargement Novacode 7

Left atrial enlargement (LAE) NC-7.1 MC-9.6

Right Ventricular Hypertrophy Novacode 8

Right ventricular hypertrophy (RVH) NC-8.1 MC-3.2

Right Atrial Enlargement Novacode 9

Right Atrial Enlargement (RAE) NC-9.1 MC-9.3

Fascicular Blocks Novacode 10

Left anterior fascicular block (LAFB) NC-10.1 MC-7.7
Left posterior fascicular block (LPFB) NC-10.2 MC-7.6
a
The criteria are different between Novacode and Minnesota Code
NC-3.1 with QRS duration 125 ms (MC-7.1 with QRS duration 120 ms)
NC-3.4.1 with QRS duration between 110 and 119 ms (MC-7.3 no QRS duration limit requirement)
NC-3.4.2 with QRS duration between 110 and 119 ms (MC-7.6 with QRS duration between 100 and 119 ms)
NC-4 (No MC code for prolonged ventricular repolarization)
NC-5 (Novacode for MI/Ischemiasee Tables B. 25 below. MC combines the codes of MC 1, 4, 5, and 92 for MI/Ischemia see
Chap. 4, 7, and 11)
NC-6 (NC-LVH by Cornell Voltage Criteria; MC-LVH by Sokolow-Lyon Voltage Criteria)

289
Dictionary of Variables and Novacode Denitions of ECG Wavesa

Amplitudes are in microvolts and durations in milliseconds (V after variable name implies
the amplitude and ms the duration of the wave; if not clear from the context, subscript a is
used to indicate amplitude and d to indicate duration). The amplitudes of P, PP (P prime), J,
ST, T, and T prime are positive or negative (signed); the amplitudes of other waves (QRS)
are absolute values.

Denitions of ECG Waveform Variables

HR Heart (ventricular) rate, complexes per min (CPM)
J j point (QRS offset); used to dene time point reference for ST-segment measure-
ments and time point for amplitude measurement for ST-segment elevation
JT QTd QRSd
JTI JT prolongation index (%) = (JT/518)*(Heart Rate + 100)
P Initial P wave exceeding 25 V (absolute value)
P2 Secondary P (prime) deection (with opposite polarity of P) exceeding 25 V
(absolute value)
PR P-R interval
Q Initial deection within QRS complex 75 V and 20 ms followed by a
positive deection (R) 100 V
QRS QRS interval
QRSn Net QRS deection (maximum positive absolute value of maximum negative
QRS deection)
QRSp Peak-to-peak amplitude (V) of QRS (maximum positive + absolute value of
maximum negative amplitude)
QS Initial deection within QRS 75 V, with no positive deection 100 V
QT QT interval
QTI QT prolongation index (%) = (QT/656)*(Heart Rate + 100)
R First positive deection within QRS (in Q score, used to ag the presence of Ri)
Ri Initial R within QRS with 25 V and no Q wave preceding (Qa 50 V)
RPT R peak time (ms), from QRS onset to R maximum
R2 Secondary R (prime) wave, a positive deection 100 V within QRS complex
following an S wave
R/Q Ra/Qa ratio
R/S Ra/Sa ratio
S2 Secondary S (prime) wave, a negative deection 100 V within ORS complex
following an RP wave
STD ST depression amplitude at J + 60 ms
STE ST elevation amplitude at J point if > 0
T First deection within T wave, positive or negative
Tn Net T deflection (maximum positive absolute value of maximum negative
T deection)
TP T prime deection, positive or negative, with opposite polarity of T
a
Dictionary reproduced from reference 2 with permission of ELSEVIER publishing company.

290
Tp Peak-to-peak amplitude (V) of T wave (maximum positive + absolute
value of maximum negative amplitude)
TQ Interval from end of T to beginning of QRS complex
X Flag used to denote in Q score assignment initial R with amplitude of
100 199 V
Z Flag for QRS waveform pattern with no codable Q or QS wave and initial
R 200 V (no ag E or X)

General Denitions Related to Rhythm Codes

Aberrant QRS complex (generally ectosinal premature) with blocked conduction

in some branch of the conduction system
Coalescent QRS complex overlapping ST-T of the preceding QRS-T complex
(R-on-T)
cpm cycles per minute, referring to ventricular rate, atrial rate, or the rate of
utter waves
Ectosinal (ES) Ectopic supraventricular excitation originating from outside the sinus node
Junctional Atrial and/or ventricular complex or rhythm originating from atrioven-
tricular (AV) junction
Nodal Atrioventricular nodal or atrioventricular junctional complex or rhythm
Sinal Originating at SA node
Sinal P P wave with normal P axis (PII 0 V and P aVR 0 V)
Sinal QRS QRS complex coupled with a sinal P wave

Codes for Prevalent ECG Abnormalities

0. Baseline EGG Suppression Codes
Note: Character - or . on the coding or report form indicates codes suppressed.
0.1 ECG not available
0.2 Inadequate quality or missing leads
0.2.1 Inadequate quality
Criteria
C1: Quality grade 5
C2: ECG hard copy contrast inadequate
Code 0.2 = C1 or C2
Note: C1 = poor quality and is present if either there is a baseline drift > 4
mm (choose three successive P-QRS-T complexes and measure peak-
to-peak baseline drift in millimeters in the worst lead); OR > 2 mm
perturbation of the isoelectric line (measurement in millimeters from
the highest peak-to-peak random [muscle] or 60-Hz noise in the worst
lead).
0.2.2 Missing leads
0.3 Lead connection interchange
0.3.1 RA/RL interchange (ECG uncodable)
Criteria
C1: PII < 50 V, QRSplI < 50 V and TpII < 50 V
C2: I = -III (P, ORS, and T are mirror images of each other)
Code 0.3 = C1 and C2
291
 0.3.2 Other lead connection interchanges (correction made and ECG coded)
0.3.3 Correctable chest lead in V1-V3 connection error (ECG codable)
0.3.4 Correctable chest lead in V4-V6 connection error (ECG codable)
0.3.5 Correctable other chest lead connection error (ECG codable)
0.4 Any conditions with ORS duration 120 ms (codes 2.4, 3.1, 3.2, 4.3, and 1.9)
Note: Code 0.4 suppresses ST-T scores.
0.5 Atrial brillation or utter
Criteria
C1: Rhythm codes 1.5.1 or 1.5.2 (atrial utter type 1 or type 2) or 1.5.3 (atrial bril-
lation) present
Code 0.5 = C1
Note: Code 0.5 suppresses code 2, code 7, and code 9. Atrial utter suppresses ST-T
scores and the corresponding prevalent and incident ischemic codes.
0.6 Ventricular or dual-chamber electronic pacemaker
Note 1: Except demand pacemaker with more than two adequate-quality nonpaced
complexes is available for coding from all lead groups.
Note 2: Code 0.6 suppresses all codes (except code 1.6.2).
0.7 Other suppression codes
Note: Includes atrial electronic pacemaker, which suppresses code 3, code 7, and
code 9.
0.8 Uncertain P wave detection
0.0 No suppression codes

1. Rhythm Codes
1.0 Basic sinus rhythm (SR)
Criteria
C1: PII 0 V and PaVR 0 V
C2: P amplitude variation < 100 V
C3: Presence of three or more P-QRS-T complexes meeting criteria C1 and C2
Code 1.0 = C1 and C2 and C3
Note 1: C1 implies P axis 120 and -30.
Note 2: Usually sinus rhythm is associated with xed coupling between the normal
P waves and the following QRS complexes, with P-R interval variation <
20% from the median PR. However, in Mobitz type 1 rst degree AV block
(code 2.2.1), the PR interval varies.
1.0.1 Normal sinus rhythm (NSR)
Criteria
C1: Basic rhythm sinus (code 1.0)
C2: Ventricular rate 51-94 cpm
Code 1.0.1 = C1 and C2
1.0.2 Sinus bradycardia (SB)
Criteria
C1: Basic rhythm sinus (code 1.0)
C2: Ventricular rate 50 cpm
Code 1.0.2 = C1 and C2

292
 1.0.3 Sinus tachycardia (ST)
Criteria
C1: Basic rhythm sinus (code 1.0)
C2: Ventricular rate 95 cpm
Code 1.0.3 = C1 and C2
1.1 Wandering atrial pacemaker (WAP)
Criteria
C1: Presence of more than one P wave trains with three or more P waves in each and
with P amplitudes changing by 100 V or more
Code 1.1 = C1
Note: Sinus rhythm with segments of transient ectopic atrial or junctional rhythm
(dened below) is coded as WAP
1.2 Junctional rhythm (JR)
Criteria
C1: PaVR > 0 V and PI 0 V
C2: PR 120 ms and PR variation < 20 ms
C3: Retrograde P waves
C4: No P waves identiable and no atrial utter or brillation waves
C5: R-Rmax - R-R min 40 ms
Code 1.2 = (C1 and C2) or ([C3 or C4] and C5)
Note 1: Criterion C1 implies a P axis from -60 to -90.
Note 2: C1 with PR > 120 ms (possible coronary sinus rhythm or JR with delayed
antegrade conduction) is coded under code 1.3 (EAR).
Note 3: For C5, exclude possible competing ectopic or sinus complexes.
Note 4: Atrial brillation with junctional rhythm is coded under code 1.5.3 (atrial
brillation).
1.2.1 Junctional rhythm, rate 45-64 cpm
Criteria
C1: Code 1.2
C2: Ventricular rate 45-64 cpm
Code 1.2.1 = C1 and C2
1.2.2 Junctional bradycardia (JBR)
Criteria
C1: Code 1.2
C2: Ventricular rate < 45 cpm
Code 1.2.2 = C1 and C2
1.2.3 Accelerated junctional rhythm (AJR)
Criteria
C1: Code 1.2
C2: Ventricular rate from 65 cpm and < 89 cpm
Code 1.2.3 = C1 and C2
Note: Code as narrow QRS tachycardia (code 1.4) if rate 90 cpm.
1.3 Ectopic atrial rhythm (EAR)
Criteria
C1: PII < 0 V

293
 C2: PaVR > 0 V
C3: No code 1.2 (JR)
Code 1.3 = (C1 or C2) and C3
Note: Criterion C1 or C2 implies a P axis < -30 or > 120 .
1.3.1 Ectopic atrial rhythm, rate 50-90 cpm
Criteria
C1: Code 1.3
C2: Ventricular rate 50-89 cpm
Code 1.3.1 = C1 and C2
1.3.2 Ectopic atrial bradycardia (EABR)
Criteria
C1: Code 1.3
C2: Ventricular rate < 50 cpm
Code 1.3.2 = C1 and C2
1.3.3 Ectopic atrial tachycardia (EAT)
Criteria
C1: Code 1.3
C2: Ventricular rate 90 cpm
Code 1.3.3 = C1 and C2
1.4 Supraventricular (SVT) or narrow QRS tachycardia
Criteria
C1: Six or more successive supraventricular ectopic complexes, with QRS < 120 ms
C2: Ventricular rate 95 cpm
Code 1.4 = C1 and C2
Note: If fewer than six ectopic complexes, include them in supraventricular ectopic
complex count.
1.4.1 Supraventricular tachycardia, rate < 130 cpm
Criteria
C1: Code 1.4
C2: Ventricular rate during episode < 130 cpm
Code 1.4.1 = C1 and C2
1.4.2 Supraventricular tachycardia, rate 130 cpm
Criteria
C1: Code 1.4
C2: Ventricular rate during episode 130 cpm
Code 1.4.2 = C1 and C2
1.5 Atrial utter or brillation (AFLF)
Screening criteria
C1: No P waves present
C2: Flutter (F) waves 100 V peak to peak with repetitive, regular morphology,
present in lead II, aVF or V1
C3: Fibrillation (f) waves or F waves with irregular cycle intervals or amplitudes,
in II, III, or aVF
C4: R-R intervals irregular (fewer than three R-R within 40-ms class interval)
Code 1.5 = C1 and (C2 or C3 or C4 or C5)

294
1.5.1 Atrial utter type 1 (AFL1)
Criteria
C1: Five or more R-R intervals, each with F wave amplitudes 100 V peak
to peak
C2: F 333 cpm (F cycle interval 180 ms) (i.e., 4.5 mm at 25 mm/second)
C3: At least partial regularity of R-R intervals, with three or more R-R inter-
vals within 40 ms of each other (i.e., within two adjacent class intervals
of 20 ms); if ventricular rate 100 cpm, four or fewer R-R intervals are
required to be within 40 ms of each other
Code 1.5.1 = C1 and C2 and C3
1.5.1.1 Classic atrial utter type 1 (AFL1C)
Criteria
C1: Code 1.5.1
C2: F waves predominant in lead II or aVF, sawtooth pattern, with the
initial leading edge notch of the F wave negative with respect to
ECG baseline (onset of QRS)
Code 1.5.1.1 = C1 and C2
1.5.1.2 Variant atrial utter type 1 (AFL1V)
Criteria
C1: Code 1.5.1
C2: F waves predominant in lead II or aVF sawtooth pattern, with the
initial leading edge notch of the F wave positive with respect to
ECG baseline
Code 1.5.1.2 = C1 and C2
Note 1: Possible atrial utter with 1:1 AV conduction is coded under
code 1.4 (supraventricular tachycardia).
Note 2: Supplementary codes for dominant and variable AV conduc-
tion with atrial utter are dened under codes 1.5.1 S 9 and
1.5.1 S 11.
1.5.2 Atrial utter type 2 (AFL2)
Criteria
C1: F waves 100 V peak to peak sustained for 5 or more R-R intervals
C2: F > 333 cpm and < 430 cpm (F cycle interval 141179 ms, i.e., 3.6-4.4
mm at 25 mm/second
Code 1.5.2 = C1 and C2
Note: In type 2 atrial utter at F > 350 cpm, F wave morphology tends to
become irregular in amplitude and in cycle length, and ventricular rate
often does not meet the partial regularity criteria characteristics of atrial
utter type 1.
1.5.3 Atrial brillation (AF)
Criteria
C1: Code 1.5
C2: Criteria for codes 1.5.1 and 1.5.2 not met
Code 1.5.3 = C1 and C2
Note: Supplementary code for atrial brillation with probable junctional
rhythm is dened under code 1.5.3 S 12 - 1.5.3 S 14.
295
 1.5.4 Atrial brillation/utter with possible dominant AV conduction
Criteria
C1: Fibrillation waves or four or less R-R intervals, each with F waves 100
V peak to peak
C2: Possible dominant AV conduction present, with R-R within two adjacent
class intervals of 20 ms (i.e., within 40 ms from each other). If mean R-R
600 ms, four or more R-R intervals within 40 ms required to dene pos-
sible dominant conduction.
Code 1.5.4 = C1 and C2
1.6 Electronic pacemaker (PM)
1.6.1 Ventricular pacemaker (VPM) or combination (dual chamber) pacemaker
(CPM)
Criteria
C1: Coupled pacemaker spikes with spikespike interval 80 ms
C2: Single pacemaker spikes
C3: QRS 120 ms
Code 1.6.1 = (C1 and C3) or (C2 and C3)
1.6.2 Atrial electronic pacemaker (APM)
Criteria
C1: Single pacemaker spikes >80 ms before QRS complex, preceding P
waves
C2: QRS < 120 ms
Code 1.6.2 = C1 and C2
1.7 Ventricular tachycardia (VT)
Criteria
C1: Three or more successive ventricular ectopic complexes with rate 130 cpm
Code 1.7 = C1
Note: Other ventricular ectopic complexes counted separately (under continuous
measurements and counts).
1.8 Other abnormal rhythm classication
1.8.1 Other atrial rhythms
1.8.2 Other ventricular rhythms
1.9 Indeterminate rhythm classication
1.9.1 Indeterminate atrial rhythm
1.9.2 Indeterminate ventricular rhythm (IVR)

2. Atrioventricular Conduction Abnormalities

2.0 AV conduction normal
Criteria
C1: PR interval 120-219 ms
C2: PR intervals within 40-ms class interval (excluding ectopic complexes)
Code 2.0 = C1 and C2
2.1 First-degree AV block (AVB1)
Criteria
C1: PR 220 ms

296
 C2: P-R intervals within 40-ms class interval (excluding ectopic complexes)
Code 2.1 = C1 and C2
2.2 Second-degree AV block (AVB2)
Criteria
C1: Occurrence of one or more blocked P waves within the R-R interval of conducted
P waves
Code 2.2 = C1
2.2.1 Second-degree AV block type Wenckebach or Mobitz 1 (AVB2W)
Criteria
C1: Repetitive cycles of progressive prolongation of PR followed by a blocked
P wave
Code 2.2.1 = C1
2.2.2 Second-degree singular AV block or type Mobitz 2 (AVB2S)
Criteria
C1: Singular blocked P at variable or xed (1:2, 1:3, etc.) block ratio
Code 2.2.2 = C1
Note 1: Singular blocked P wave equal to no more than one blocked P wave
within any R-R interval. Blocked P is in a regular P wave train
(rather than an early premature P within the refractory period of AV
node).
Note 2: Block ratio for the second degree singular AV block is the ratio of
blocked P waves to conducted P waves. Block ratio = 1:x, where x
is the number of conducted P waves following the blocked P wave.
Second-degree singular AV block with 1:1 block ratio is often con-
fusingly called 2:1 AV block, and with a 1:2 block ratio, it is called
3:1 AV block.
Note 3: Second-degree singular AV block with 1:1 block ratio may be a Wenck-
ebach block that is revealed only if the block ratio changes to 1:2 or
1:3. A true Mobitz 2 second-degree AV block is commonly associated
with a complete bundle branch block or with bifascicular block.
2.2.3 Second-degree multiple AV block (AVB2M)
Criteria
C1: Two or more blocked P waves within R-R interval of conducted atrial
complexes
Code 2.2.3 = C1
2.3 High-grade AV dissociation (AVD)
Criteria
C1: P wave train and QRS wave train independent (with no consistent relationship)
for the majority of the complexes
C2: R-R intervals of the independent QRS complexes regular within 100 ms, with
QRS duration varying less than 10 ms
Code 2.3 = C1 and C2
Note 1: Second-degree AV blocks can also be considered to present a form of AV
dissociation.
Note 2: AV dissociation with atrial utter is coded under Code 1.5.1 S 10.

297
 2.3.1 Third-degree (complete) AV block (AVB3)
Criteria
C1: Criteria for code 2.3 persist throughout the record
Code 2.3.1 = C1
Note 1: Ventricular rate is usually slower than atrial rate.
Note 2: If several types of AV blocks coexist in the same record, code the
highest block.
2.3.2 AV dissociation with captured ventricular or atrial complexes (AVDC)
Criteria
C1: Code 2.3
C2: Occurrence of a P wave followed by a QRS complex with preceding
R-R interval 100 ms shorter than R-R intervals meeting the regularity
criteria
C3: Occurrence of a retrograde P wave with preceding P-P interval 100 ms
shorter than those in the regular P wave train
Code 2.3.2 = C1 and (C2 or C3)
2.4 Ventricular preexcitation pattern Wolff-Parkinson-White syndrome (WPW)
Criteria
C1: PR < 120 ms
C2: P axis from 1 to 90
C3: QRS complex 120 ms
C4: Delta wave present
Code 2.4 = C1 and C2 and C3 and C4
Subclassication of Preexcitation Patterns
Criteria
C1: Delta wave positive in lead I
C2: Delta wave positive in lead V1
C3: Delta wave negative in lead V1
C4: Delta wave isoelectric in lead V1
C5: Q or QS waves in leads I and V6
C6: Q or QS waves in leads II and aVF
C7: QRS mainly positive in lead III
C8: QRS mainly negative in lead III
C9: QRS mainly positive in lead V1
C10: QRS mainly negative in lead V1
C11: Delta wave positive in leads II and aVF and positive or isoelectric in lead III
C12: Delta wave negative in leads II, III, and aVF
C13: Delta wave positive in leads V2, V3, V4, V5 and V6
C14: Delta wave negative or isoelectric in lead V6
C15: Ri < 100 V and S 100 V in lead V1
C16: Ri 100 V and S 100 V in lead V2
Code 2.4.1 Anterior right ventricular preexcitation = C1 and C4 and C8 and C10
Code 2.4.2 Posterior right ventricular preexcitation = C3 and C6 and C10
Code 2.4.3 Posterior left ventricular preexcitation =C1 and C2 and C6 and C8 and
C9 and C14

298
 Code 2.4.4 Lateral left ventricular preexcitation = C2 and C5 and C7 and C14
Code 2.4.5 Anterior paraseptal preexcitation = C1 and C4 and C11
Code 2.4.6 Posterior paraseptal preexcitation = C1 and C12 and C13 and C15 and C16
2.5 Other AV conduction abnormalities (concealed conduction, AV dissociation other
than third-degree AV block, etc.)

3. Prolonged Ventricular Excitation

3.0 QRS duration normal
Criteria
C1: QRS < 110 ms
Code 3.0 = C1
3.1 Left bundle branch block (LBBB)
Criteria
C1: QRS 125 ms
C2: WPW absent (no code 2.4)
C3: R peak time or R2 peak time 60 ms in leads V5 or V6 or I or aVL
Code 3.1 = C1 and C2 and C3
3.1.0 LBBB without ECG evidence of myocardial infarction (MI)
Criteria
C1: Code 3.1
C2: Q score < 25
Code 3.1.0 = C1 and C2
3.1.1 LBBB with possible MI
Criteria
C1: Code 3.1
C2: Q score 25
Code 3.1.1 = C1 and C2
3.2 Right bundle branch block (RBBB)
Criteria
C1: QRS 120 ms
C2: WPW absent (no code 2.4)
C3: R peak time or R2 peak time 60 ms in leads V1 or V2
C4: S duration R duration in I or V6
Code 3.2 = (C1 and C2 and C3) or (C1 and C2 and C4)
3.2.0 RBBB without ECG evidence of MI
Criteria
C1: Code 3.2
C2: Q score < 25
Code 3.2.0 = C1 and C2
3.2.1 RBBB with possible MI
Criteria
C1: Code 3.2
C2: Q score 25
Code 3.2.1 = C1 and C2

299
 3.3 Indeterminate ventricular conduction delay (IVCD)
Criteria
C1: QRS 120 ms
C2: WPW absent (no code 2.4)
C3: No code 3.1 or 3.2
Code 3.3 = C1 and C2 and C3
Note: Code 3.3 includes LBBB pattern with QRS 120124 ms.
3.3.0 IVCD without ECG evidence of MI
Criteria
C1: Code 3.3
C2: Q score < 25
Code 3.3.0 = C1 and C2
3.3.1 IVCD with possible MI
Criteria
C1: Code 3.3
C2: Q score 25
Code 3.3.1 = C1 and C2
3.4 Borderline prolonged ventricular excitation
Criteria
C1: QRS 110119 ms
Code 3.4 = C1
3.4.1 Borderline delay of right ventricular excitation
Criteria
C1: Code 3.4
C2: R2 in V1
Code 3.4.1 = C1 and C2
3.4.2 Borderline delay of left ventricular excitation
Criteria
C1: Code 3.4
C2: No code 3.4.1
Code 3.4.2 = C1 and C2

4. Prolonged Ventricular Repolarization

4.0 No prolonged ventricular repolarization
Criteria
C1: QTI < 112%
Code 4.0 = C1
4.1 Prolonged ventricular repolarization
Criteria
C1: QTI 112%
Code 4.1 = C1
4.1.1 Marginal prolongation of ventricular repolarization
Criteria
C1: QTI 112% to 116%
Code 4.1.1 = C1

300
 4.1.2* Denite prolongation of ventricular repolarization
Criteria
C1: QTI > 116%
Code 4.1.2 = C1
Note 1: QTI (%) = (QT/656) (HR + 100). QT is in ms. At heart rate 60 cpm,
QTI 112% corresponds to a QT of 460 ms.
Note 2: It is essential to use JTI rather than QTI for coding prolonged repolar-
ization if QRS 120 ms. JT prolongation index JTI (%) = (JT/518)
(HR + 100) where JT = QT QRS.
Note 3: It should be recognized that QTI includes not only the period of ven-
tricular repolarization but also ventricular excitation. The inclusion
of separate terms for ventricular repolarization (JT) and excitation
(QRS) may be warranted.

5. ECG Categories Associated with Prevalent Myocardial Infarction/Ischemia (MI

Likelihood)
5.0 No signicant Q waves and no signicant STT abnormalities
Criteria
C1: No codes 5.1 through 5.8
Code 5.0 = C1
High Likelihood of Q Wave MI
5.1 Q wave MI with major Q waves
Criteria
C1: Q score 35 in any lead group
Code 5.1 = C1
5.2 Q wave MI with moderate Q waves and with ST-T abnormalities
Criteria
C1: Q score 25 in any lead group
C2: ST-segment depression (STD) or T wave negativity (TN) score 20 or higher in
any lead group
Code 5.2 = C1 and C2
Moderate Likelihood of MI
5.3 Possible Q wave MI with moderate Q waves and without ST-T abnormalities
Criteria
C1: Q score 25 in any lead
C2: STD and TN score < 20 in all lead groups
Code 5.3 = C1 and C2
5.4 Possible Q wave MI with minor Q waves and with ST-T abnormalities
Criteria
C1: Q score 15 in any lead
C2: STD or TN score 20 in any lead group
Code 5.4 = C1 and C2
Isolated Ischemic Abnormalities
5.5 ST abnormalities without Q waves
Criteria
C1: STD score 20 in any lead group
301
 C2: Q score < 15 in all leads
Code 5.5 = C1 and C2
5.6 T wave abnormalities without Q waves
Criteria
C1: TN score 20 in any lead group
C2: Q score < 15 in all leads
Code 5.6 = C1 and C2
Minor Q Wave or ST- T Abnormalities
5.7 Minor Q waves without ST-T abnormalities
Criteria
C1: Q score 15 in any lead
C2: STD and TN score < 20 in all lead groups
Code 5.7 = C1 and C2
5.8 Minor ST-T abnormalities
Criteria
C1: STD or TN score 10 in any lead group
Code 5.8 = C1
Note 1: Code 0.1 (ECG not available) and codes 0.2.1 and 0.2.3 (inadequate quality
or missing leads, including RA/RL reversal) interfere with morphologic
codes 0.3 (ventricular pacemaker), 0.4 (complete bundle branch block and
Wolff-Parkinson-White pattern), 0.6 (electronic pacemaker) and suppress
all code 5 items.
Note 2: Code 6.0 (no left ventricular hypertrophy) is recommended as an additional
condition for codes 5.5 and 5.6.

6. Left Ventricular Hypertrophy

6.0 No ventricular hypertrophy
Criteria
C1: Code 6.1 not present
Code 6.0 = C1
6.1 Left ventricular hypertrophy (LVH)
Criteria
C1: RaVL + SV3 2,800 V in men
C2: RaVL + SV3 2,200 V in women
Code 6.1 = C1 or C2
6.1.0 LVH without ST-T abnormalities
Criteria
C1: Code 6.1
C2: STD or TN score < 20
Code 6.1.0 = C1 and C2
6.1.1 LVH with ST-T abnormalities
Criteria
C1: Code 6.1
C2: STD or TN score 20
Code 6.1.1 = C1 and C2

302
7. Left Atrial Enlargement
7.0 No left atrial enlargement
Criteria
C1: Code 7.1 not present
Code 7.0 = C1
7.1 Left atrial enlargement
Criteria
C1: PII 120 ms
C2: P2V1 -100 V, or (PV1 -100 V if P2V1 = 0)
C3: (P2aV1 x P2dV1) < -4000 Vms
Code 7.1 = C1 or C2 or C3
Note: Use C3 alone, if available, for higher specicity.

8. Right Ventricular Hypertrophy

8.0 No right ventricular hypertrophy
Criteria
C1: Code 8.1 not present and QRSd < 120 ms
Code 8.0 = C1
8.1 Right ventricular hypertrophy (RVH)
Criteria
C 1: QRSnaVR 0 V and QRSnaVL 0 V
C2: (R/S) I 1 and (R/S) II 1 and (R/S) III 1
C3: (R/S) V5 1
C4: (R/S) V6 1
Code 8.1 = (C1 or C2) and (C3 or C4)
Note: Criterion 1 (net QRS amplitude in aVR 0 V and in aVL 0 V) implies
QRS axis from 120 to 240 (right axis deviation).

9. Right Atrial Enlargement

9.0 No right atrial enlargement
Criteria
C1: Code 9.1 not present
Code 9.0 = C1
9.1 Right atrial enlargement
Criteria
C1: PII > 250 V
Code 9.1 = C1

10. Fascicular Blocks

10.0 No fascicular block
Criteria
C1: QRS axis > 45
C2: QRS axis 90
Code 10.0 = C1 and C2

303
 10.1 Left anterior fascicular block (LAFB)
Criteria
C1: QRS < 120 ms
C2: QRSII < 0 V and QRSIII < 0 V
C3: RiS patterns in lead II, with initial Ra < 200 V
C4: Q 25-100 V in lead aVL
C5: R 200 V and R peak time 40 ms in lead aVL
Code 10.1 = C1 and C2 and C3 and C4 and C5
Note: Criterion 2 (net QRS amplitude < 0 V in leads II and III) implies QRS axis
from -31 to -119.
10.2 Left posterior fascicular block (LPFB)
Criteria
C1: QRS < 120 ms
C2: QRSI < 0 V and QRSaVF > 0 V
C3: Code 8.0 (no RVH)
C4: Q from 25 V to 99 V and R 100 V in III and aVF
C5: Q < 40 ms in III and aVF
Code 10.2 = C1 and C2 and C3 and C4 and C5
Note: Criterion C2 implies QRS axis from 91 to 179 .

11. Other Clinically Signicant Abnormalities

Note: Code 11 is reserved for other abnormalities not included in codes 1 through
10 and not specied in this context

Denitions of Supplementary Codes (S)

Note: The rst two digits (preceding S) identify the associated main rhythm code
(i.e. 1.0 S = supplementary condition to sinus rhythm, code 1.0).

Supplementary Codes to Sinus Rhythm

1.0 S 1 With ectopic (ectosinal) supraventricular complexes (ESVC)
Criteria
C1: Code 1.0.1 or code 1.0.2 (NSR or SB)
C2: QRS morphology matches sinal QRS complexes (QRS duration within
20 ms and QRS wave labeling same as for sinal QRS)
C3: P amplitude differs by more than 100 V from sinal P waves, or retro-
grade P or no P wave discernible
C4: PR 40 ms shorter than PR of sinal QRS complexes
C5: R-R of the early complex 200 ms shorter than the preceding R-R and
240 ms shorter than the R-R following the ectopic complex
Code 1.0 S 1 = (C1 and C2) and (C3 or C4 or C5)
Note: C5 reects the minimum expected combined effect of the ectopic
complex on the prematurely (R-R preceding) and SA node suppression
(R-R following).
1.0 S 1.1 With atrial bigeminy (ABG)
Criteria
C1: Code 1.0 S.1
304
 C2: Ectopic complex follows every sinal QRS complex
Code 1.0 S 1.1 = C1 and C2
1.0 S 1.2 With atrial trigeminy (ATG)
Criteria
C1: Code 1.0 S.1
C2: Ectopic complex after every pair of sinal QRS complexes
Code 1.0 S 1.2 = C1 and C2
1.0 S 2 With aberrant supraventricular complexes (ASVC)
Criteria
C1: Code 1.0.1 or code 1.0.2 (NSR or SB)
C2: QRS 20 ms longer than normally conducted sinal QRS
C3: P wave precedes wide QRS complex
C4: RSR2 in V1, with R2 > R
C5: QRSd 140 ms
Code 1.0 S 2 = C1 and C2 and [C3 or (C4 and C5)] or (C5 and (C6 or C7))]
1.0 S 3 With ectopic ventricular complexes (EVC)
Criteria
C1: QRS 120 ms or QRS 20 ms longer than normally conducted QRS
complexes.
C2: Criteria for code 1.0 S.2 not met
Code 1.0 S 3 = C1 and C2
Note 1: Code 1.0 S.2 includes interpolated ectopic ventricular complexes
Note 2: QRS fusion complexes (preceding sinal P occurring with normal timing
in the P wave train) are counted as ectopic ventricular complexes.
1.0 S 3.1 With a doublet of ectopic ventricular complexes (DEVC)
Criteria
C1: Code 1.0 S.3
C2: Two EVCs in succession within one R-R interval of sinal
QRS complexes
Code 1.0 S 3.1 = C1 and C2
Note: A triplet of EVCs is coded under code 1.9 (VT).
1.0 S 3.2 With coalescent ventricular ectopic complexes (CEVC)
Criteria
C1: Ectopic ventricular complex overlaps the ST-T of the preceding
QRS-T complex
Code 1.0 S 3.2 = C1
1.0 S 3.3 With polymorphic ectopic ventricular complexes (PEVC)
Criteria
C1: QRSa (net QRS amplitudes) of EVCs differ by 50%
C2: QRS durations of EVCs differ by 20 ms
Code 1.0 S 3.3 = C1 and C2
1.0 S 3.4 With ventricular bigeminy (VBG)
Criteria
C1: Code 1.0 S.3
C2: Ectopic complex follows every sinal QRS complex
Code 1.0 S 3.4 = C1 and C2
305
 1.0 S 3.5 With ventricular trigeminy (VTG)
Criteria
C1: Code 1.0 S.3
C2: Ectopic complex after every pair of sinal QRS complexes
Code 1.0 S 3.5 = C1 and C2
1.0 S 4 With pause (possible sinoatrial arrest or block)
Criteria
C1: Code 1.0
C2: P-P interval containing pause prolonged 90% compared with median
P-P of sinal P waves
C3: No P wave in the prolonged P-P interval until the next PQRS complex
C4: Preceding PQRS complex not an ectopic supraventricular or ventricular
complex
Code 1.0 S 4 = C1 and C2 and C3 and C4
1.0 S 5 With reduced heart rate variability (HRV)
Criteria
C1: Code 1.0.1 (normal sinus rhythm)
C2: R-R interval variation range of normally conducted QRS complexes <
40 ms, excluding complexes following ectopic (ventricular or supraven-
tricular) complexes (interpolated or with compensatory pause)
Code 1.0 S 5 = C1 and C2
Note: With computer measurements of all R-R intervals of normally conducted
QRS complexes (N-N intervals) available, code 1.0 S.5 = SDNN 5 ms,
where SDNN is the standard deviation of N-N intervals.
1.0 S 6 With increased heart rate variability
Criteria
C1: Code 1.0.1 (normal sinus rhythm)
C2: Largest successive difference of R-R intervals of normally conduct-
ed QRS complexes > 100 ms excluding complexes following ectopic
(ventricular or supraventricular) complexes (interpolated or with com-
pensatory pause)
Code 1.0 S 6 = C1 and C2
Note: With computer measurements of all R-R intervals of normally con-
ducted QRS complexes (N-N intervals) available, 1.0 S.6 = SDNN
30 ms, where SDNN is the standard deviation of N-N intervals.

Supplementary Codes to Ectopic Atrial Rhythm

1.3 S 7 Probably left atrial ectopic focus
Criteria
C1: Code 1.3
C2: PI < 0
C3: PV1 > 0 and PPV1 0
Code 1.3 S 7 = C1 and C2 and C3
Note: C3 implies that PV1 is positive of bid.

306
 1.3 S 8 Probably right atrial ectopic focus
Criteria
C1: Code 1.3
C2: No code 1.3 S 7
Code 1.3 S 8 = C1 and C2

Supplementary Codes to Atrial Flutter

1.5.1 S 9 With dominant AV conduction ratio 1 to x (specify x)
Criteria
C1: Conduction ratio (x) is constant for the majority of R-R intervals
Code 1.5.1 S 9 = C1
Note 1: In AV conduction ratio, x = (1 + N), where N is the number
of nonconducted F waves following a conducted F wave (x
is also the ratio of the dominant regular R-R interval to the F
cycle length).
1.5.1 S 10 With AV dissociation
Criteria
C1: The majority of R-R intervals are constant within 40 ms
C2: FR intervals are irregular (except for possible captured ventricular
complexes)
C3: Dominant R-R is not a multiple of the F cycle length within 40 ms
Code 1.5.1 S 10 = C1 and C2 and C3
1.5.1 S 11 With variable AV conduction
Criteria
C1: No code 1.5.1 S 10
C2: Conduction ratio varies
Code 1.5.1 S 11 = C1 and C2

Supplementary Codes to Atrial Fibrillation

1.5.3 S 12 With AV dissociation and junctional rhythm
Criteria
C1: Code 1.5.3
C2: R-R interval variation range < 40 ms
Code 1.5.3 S 12 = C1 and C2
1.5.3 S 13 With slow ventricular response
Criteria
C1: Code 1.5.3
C2: Ventricular rate < 50 cpm
Code 1.5.3 S 13 = C1 and C2
1.5.3 S 14 With rapid ventricular response
Criteria
C1: Code 1.5.3
C2: Ventricular rate > 95
Code 1.5.3 S 14 = C1 and C2

307
Supplementary Codes to AV Dissociation
2.3 S 15 With narrow QRS complex
Criteria
C1: Code 2.3
C2: QRS < 120 ms (in the majority group)
Code 2.3 S 15 = C1 and C2
Note: Code 2.3 S 15 indicates that the likely block site is the AV node
or His bundle.
2.3 S 16 With wide QRS complex
Criteria
C1: Code 2.3
C2: QRS 120 ms (in the majority group)
Code 2.3 S 16 = C1 and C2
2.3 S 17 With wide QRS complex and slow ventricular rate
Criteria
C1: Code 2.3 S 16
C2: Ventricular rate 45 cpm
Code 2.3 S 17 = C1 and C2

The Novacode Criteria for Classication of Myocardial Infarction

and Ischemic Abnormalities
Waveform Pattern Labels for Coding of Q Wave Abnormalities

The Novacode MI coding system is based on the use of a hierarchy of waveform pattern la-
bels (ags) to identify Q wave abnormalities (Table B.2). The coder rst determines for
each ECG lead if ags QS, EQS, E, X, or Z are present, corresponding to patterns in rows
Ha, Hb, I, J, and K of Table B. 2 (Flag QS in row Ha = QS wave; Flag EQS in row Hb= if
Flag E, or X, or Z in V1-V5, and QS wave in next lead V2-V6; Flag E in row I = initial
R < 100 V; Flag X in row J = initial R < 200 V; Flag Z in row K = initial R 200 V,
respectively [see Fig. B. 1, B. 2]). Identication of these ags is essential for reliable serial
ECG coding. If present for a given lead, the appropriate ag is entered on the coding entry
form; otherwise the coder checks for the presence of Q waves. Q wave durations exceeding
the limits 20, 30, 40, and 50 ms carry ags 20, 30, 40, and 50, respectively. An additional
symbol 4 is entered with these Q wave duration ags if the R/Q ratio is less than 4 for
example, 30/4 for a Q duration of 3039 ms with an R/Q ratio less than 4 (R/Q ratios need
not be considered for leads aVL, aVF, III, V1, and V2). These ags in the reference and
follow-up ECG coding forms determine the prevalent MI codes and MI incidence codes
by assigning lead-specic Q scores, listed in Table B. 2. The advantage of this approach is
that the ECG coder does not need to memorize any coding criteria or even the Q scores as-
signed to different abnormal patterns as long as a proper ag is entered for each lead on the
coding data entry form. A look-up table can then be used, or a simple algorithm can assign
the appropriate scores and classication codes (Fig. B.3 and B.4).

308
 In case electronic data entry is not used for ECG coding forms, the scores assigned for
different waveform ags are entered for each ECG lead on the coding form. These scores
are relatively easy to memorize. It is noted that the Q scores increase uniformly in steps of
10 points for Q wave duration increments of 10 ms, with ve extra points added if the R/Q
ratio is less than 4 (except for leads aVL, aVF, III, V1 and V2). The weights for the Q scores
are otherwise identical for all leads except that they are 5 points lower for leads aVF and
III, and 10 points lower for lead aVL. The Q score limits <15, 15, 25, and 35 identify
grade 0, grade 1, grade 2, and grade 3 Q waves, respectively.

Lead Groups for Coding of Myocardial Infarctions and Ischemic Abnormalities

The lead groups for localizing abnormalities are formed by considering the directional
components and relative sensitivity of the lead vectors in detecting cardiac source events
from various regions of the left ventricle. Five lead groups are identied (Table B. 2),
reecting abnormalities primarily in the lateral, inferior, posterolateral, anteroseptal, and
anterior regions. It is recognized that the leads in all these lead groups do not represent
selectively the activity in any single region. For instance, the anterior chest leads V3-V6
do not reect only the activity in the anterior left ventricle because their lead vectors also
have a strong component or sensitivity for the apical and lateral regions. Leads V1 and V2
are actually shared by two lead groups, anteroseptal, and posterolateral. The Q scores for
the posterolateral lead group (listed in the footnote of Table B.2) were derived from the
weights used for the variables in the Selvester score for MI size.3 The inverted lead -aVR
is included with the lateral lead group. The extra lead in this lead group is expected to
enhance MI detection sensitivity, particularly in coding some of the MI incidence codes,
which require Q wave evolution in two leads. The use of Q wave-equivalent information
from inverted leads poses no problems in computer coding. In visual coding, the reversal of
their polarity on the ECGs reproduced would facilitate their use because the initial R waves
in them would appear as more familiar Q waves. This holds true also for the initial R waves
in leads V1 and V2 when they are used for coding MIs for the posterolateral lead group.

Coding of Repolarization Abnormalities

ST-segment depression, T wave negativity, and ST-segment elevation exceeding specied

amplitude thresholds determine the presence of waveform abnormality ags for each lead
group (Table B.3). The T wave negativity is not coded for leads V1 and V2; it is coded for
lead aVL only if the R wave is 500 V or higher and for lead aVF only if the QRS complex
is mainly positive (R/S ratio > 1). The ECG coder identies for each lead group the ECG
lead exceeding the highest threshold, and the score of this lead denes the score assigned
for its lead group. With computer coding, each of the 12 leads is scored separately for
ST-segment elevation, ST depression, and T wave patterns, and the lead with the largest
score determines the overall score in each lead group. The ST-segment elevation score
is used primarily for identication of an evolving ST-T abnormality from acute-phase
hospitalization ECGs and is offered as an optional item for coding of regularly scheduled
follow-up ECGs.

309
Classication Criteria for Prevalent Myocardial Infarction and Categorization for
Risk Stratication

The Novacode category 5 contains a hierarchical classication scheme categorizing MI/

ischemia into high, moderate, marginal, or low likelihood (Table B.4). The categories with
high and moderate likelihood of MI are considered as major ECG abnormalities and the
marginal MI likelihood categories as minor abnormalities. Novacode 5 categories 5.1 5.4
contain criteria for old Q wave infarction, category 5.5 represents profound myocardial
ischemia, possibly associated with a non-Q wave infarction, and category 5.6 isolates ma-
jor ST-T abnormalities without signicant Q waves. The remaining two categories (5.7 and
5.8) are for borderline Q waves and ST-T abnormalities. The primary focus of Novacode 5
is on the manifestations of coronary heart disease, and for this reason LVH shown by ECG
is excluded for classication of codes 5.5 and 5.6.
Novacode 5 abnormalities listed in Table B.4 are also used together with additional
information on the history of MI to categorize study participants into high-, moderate-,
marginal-, and low-risk subgroups for evaluation of the risk of coronary heart disease
mortality.

Classication of Incident Myocardial Infarctions

For classication of incident MI from regularly scheduled follow-up ECGs, the Q score
and ST-T score algorithms are identical to those used for evaluating prevalent MIs. The
classication of an incident MI is based on the change of the Q score exceeding specied
limits and on whether the ST-T has evolved signicantly from the reference ECG (Table
B.5). Incident MI codes I 5.1 to I 5.4 are ranked according to the likelihood of Q wave MIs.
Evolving ischemic repolarization abnormalities are categorized as profound ST-T evolu-
tion (code I 5.5), evolving ST-T with nonevolving Q waves (code I 5.6.1), or isolated ST-T
evolution (code I 5.6.2).
In case of moderate Q wave evolution (grade 1 change, or Q score increase of 15 points),
a two-lead involvement is required for the higher-order codes. A borderline Q wave evolu-
tion in a single lead without ST-T evolution is included as the lowest category in the inci-
dence code 5 hierarchy (code I 5.7) because chest lead placement errors or changes in QRS
patterns due to secondary directional changes (due to respiration, obesity, minor changes
with time, poor record quality, etc.) may produce a false ECG event in this category.
An evolving ST-T pattern for regularly scheduled follow-up ECGs is dened by an
increase of 20 or more points in ST depression or T wave negativity score, without consi-
dering ST elevation. For acute-phase, unscheduled hospitalization ECGs, ST-T evolution
is dened as an ST elevation score increase in any event ECG of 40 or more points or a
change (increase or decrease) in ST elevation score of 20 or more points, with the highest
increase in the ST depression or T wave negativity score being 20 or more points in the
same or in some other ECG recorded later on in the acute phase (Note 2 in Table B.5). Note
3 in Table B.5 denes criteria for incident acute MI in the presence of left bundle branch
block, adapted from the 1996 report of the GUSTO-1 trial.4 Both prevalent and incident
Novacode major and minor ECG abnormalities are independent signicant predictors of
future cardiovascular disease events and mortality.5

310
 Novacode Flag

QS E X Z

R
R
R

FIGURE B.1. Novacode ags for ECG waveform pattern: Flag QS for QS wave; Flag E for
initial R 2599 V; Flag X for initial R 100199 V; Flag Z for initial R 200 V,
respectively

Novacode Flag EQS

V1 V2 V3 V4 V5

or or or or

V2 V3 V4 V5 V6

FIGURE B.2. Flag EQS = if Flag E, or X, or Z in V1-V5, and QS wave in next lead V2V6

311
 Group L Group I Group S Group A
Novacode 5
aVL I -aVR II aVF III V1 V2 V3 V4 V5 V6
ECG QRS Flag Z Z 20/4 40/4 50 50 Z Z Z 30 40/4 40/4
ST Elevation __
0 0 0 0 0 0
(maximum)
ST Depression __
25 0 0 0 0 0
(maximum)
T Negative __ __ __ __
50 0 0 100
(maximum)
Max. Q Score 35 Max. ST-T Score 20 Novacode 5 5.1
FIGURE B.3. The ECG with codable QRS and ST-T patterns, Q waves in the leads II, aVF, III, and
V4-V6, ST-T abnormality in lead I, aVL and V5V6, producing a myocardial infarction
code Novacode 5.1

Group L Group I Group S Group A

Novacode 5
aVL I -aVR II aVF III V1 V2 V3 V4 V5 V6
ECG QRS Flag Z Z Z Z QS QS QS QS QS Z Z Z
ST Elevation __
0 50 0 0 0 0
(maximum)
ST Depression __
50 0 0 0 0 0
(maximum)
T Negative __ __ __ __
100 0 0 0
(maximum)
Max. Q Score 30 Max. ST-T Score 20 Novacode 5 5.2
FIGURE B.4. The ECG with codable QRS and ST-T patterns, QS waves in the lead V1-V3 and lead
aVF and III, ST-T abnormality in lead I and aVL, producing a myocardial infarction
code Novacode 5.2
312
TABLE B.2. Hierarchy and Denitions of Waveform Patterns and Corresponding Novacode
Q Scores for ECG Leads of Four lead Groups.a
L (Lateral) I (Inferior) S (Septal) A (Anterior)
(L-I-API) (API-I-S) (S-A) (A-API-I-L)
Waveform Pattern FLAG aVL I -aVR II aVF III V1 V2 V3 V4 V5 V6
A Q 50 ms 50 30 40 40 40 35 20 20 40 40 40 40 40
B R/Q <4 and Q 40ms 40/4 20 35 35 35 25 15 20 30 35 35 35 35
C Q 40ms 40 20 30 30 30 25 15 20 30 30 30 30 30
D R/Q <4 & Q 30ms 30/4 10 25 25 25 15 5 20 20 25 25 25 25
E Q 30ms 30 10 20 20 20 15 5 20 20 20 20 20 20
F R/Q <4 & Q 20ms 20/4 0 15 15 15 5 0 10 15 15 15 15 15
G Q 20ms 20 0 10 10 10 5 0 10 10 10 10 10 10
Ha Flag for QS QS 0 30 30 30 15 5 10 20 30 40 30 20
Hb Flag for EQS EQS 25 35 45 35 25
I Flag for Ri 25-99 V E 0 0 0 0 0 0 0 0 0 0 0 0
J Flag for Ri 100-199 V X 0 0 0 0 0 0 0 0 0 0 0 0
K Flag for Ri 200 V Z 0 0 0 0 0 0 0 0 0 0 0 0
1. Score limits 35, 25, and 15 indicate grade 3, grade 2, and grade 1 abnormal Q waves for prevalent MI classication.
Grade 2 and grade 1 evolving Q waves are dened by Q score increases of 25 and 15 points, respectively. The leads
are allocated to various lead groups on the basis of their sensitivity to different myocardial regions. L, lateral, I inferior;
S, septal; API, apical; A, anterior (referring to regionality of leads). Ri initial R wave, RP R prime, R/Q and R/S refer
to amplitude ratios. LBBB left bundle branch block, IVCD indeterminate ventricular conduction defect
2. Using max Q score and max_STD_Tneg score in 12 leads to categorize Novacode 5, criteria for determining change
scores for incident MIs are listed as Note 1 in Table B.4
3. EQS = If Flag Z or X or E in V1 - V5, and QS in next lead V2 - V6. Q score for EQS = [QS score + 5 point]
in V2 - V6.
If Q score < 5 in aVF and Q score 5 in III modies Q score in lead III, reducing it by 5 points
If R amplitude in lead aVL < 300 uV, then Q Score aVL = 0
For inverted aVR, Q = Ri and R/Q ratio = S/Ri ratio.
Codes 3.1 and 3.3 (LBBB and IVCD) modify QS score to 0.
4. Q score for posterolateraI (PL) MI: For lead V1, score = 40 for Ri 60 ms; score = 30 for Ri 50 ms and Ri/S 1.5;
and score = 20 for Ri 30 ms and Ri/S 1.5. For lead V2, score = 30 for Ri 60 ms; score = 20 for Ri 50 ms (Ri
2000 uV and Ri/S 2); and score = 10 for Ri 40 ms (Ri 2000 mV and Ri/S 2)
a
Tables B.2B.5 reproduced from reference 2 with permission of ELSEVIER Publishing Company.

TABLE B.3. ST Depression, T wave Negative, and ST Elevation Pattern labels and Scores.
Amplitude Thresholds ST Depression Score T wave Negative Scorea ST Elevation Score ST Elevation Score
(uV) (All lead groups) (Groups L-I-A) (Groups L-I-A) (Group S)
500 50 50 50 50
400 50 50 50 40
300 50 40 40 30
200 40 30 30 20
100 30 20 20 10
50 20 10 10 0
25 10 10b 0 0
a
T wave score is not determined for leads V1, V2 and III; for lead aVF only if R/S > 1; and for lead aVL only if R 500 mV.
b
Or at T wave, with positive phase < 25 uV. L lateral; I inferior; A anterior lead group; S septal lead group.

313
TABLE B.4. Novacode Hierarchy and Criteria for Myocardial Infarction/Ischemia (Code 5)
Stratied According to Likelihood of Q Wave Infarction and Ischemic Injury and Risk of
Coronary Heart Disease Mortality by the History of Heart Attack.
Likelihood of MI/Ischemic Injury
and Risk of CHD Mortality

History of No history of
Code Category Criteria Heart Attack Heart Attack
5.1 Q wave MI; major Q waves Q score 35 in any lead High High

5.2 Q wave MI; moderate Q waves with Q score 25 in any lead and STD or TN High Moderate
ST-T abnormalities score 20 in any lead group

5.3 Possible Q wave MI; moderate Q Q score 25 in any lead and STD and TN Moderate Marginal
waves without ST-T abnormalities score < 20 in all lead groups

5.4 Possible Q wave MI; minor Q waves Q score 15 in any lead and STD or TN Moderate Marginal
with ST-T abnormalities score 20 in any lead group

5.5 Isolated ST abnormalities STD score 20 in any lead group and Q Moderate Marginal
score < 15 in all leads and code 6.0 (no LVH)

5.6 Isolated T wave abnormalities TN score 20 in any lead group and Q Moderate Marginal
score < 15 in all leads and code 6.0 (no LVH)

5.7 Minor Q waves Q score 15 in any lead and STD and TN Moderate Marginal
score < 20 in all lead groups

5.8 Minor ST-T abnormalities STD or TN score 10 in any lead group Marginal Low

5.0 No signicant Q waves or ST-T Q score < 15 and STD and TN scores < 10 Low Low
MI myocardial infarction, CHD coronary heart disease, STD ST-segment depression, TN T wave negativity, LVH left ventricular
hypertrophy

TABLE B.5. Novacode Hierarchy for Classication of Incident Myocardial Infarction (MI) and
Ischemia
Category Criteria
Evolving Q wave MI
I 5.1 Major Q wave evolution Q score increase 35 in one lead and 15 in an additional lead
I 5.2 Moderate Q wave evolution with evolving ST-T (Q score increase 25 in one lead or 15 in two leads)
and STD or TN score increase 20
Possible evolving Q wave MI
I 5.3 Moderate Q wave evolution with nonevolving ST-T Q score increase 25 in one lead or 15 in two leads
and STD and TN score increase < 20
I 5.4 Borderline Q wave evolution with evolving ST-T Q score increase 15 in a single lead
once STD or TN score increase 20
Ischemic ST-T evolution
I 5.5 Profound ST-T evolution without evolving Q waves STD or TN score increase 30 and Q score increase < 15
I 5.6.1 Evolving ST-T with nonevolving Q waves STD or TN score increase 20 and Q score 15 in ECG 2
and Q score increase < 15
I 5.6.2 Isolated ST-T evolution STD or TN score increase 20 and Q score < 15 in ECG 2
Borderline Q wave change
I 5.7 Borderline Q wave evolution with nonevolving ST-T Q score increase 15 in a single lead
and STD and TN score increase < 20
No signicant Q or ST-T evolution
I 5.0 None of the above No codes I 5.1I 5.7
(continued)

314
TABLE B.5. (continued)
Note 1
Determining Q score change (follow-up ECG score - reference ECG score, listed in Table B.2) according adjusted scores, if modied.
(For lead III, use unmodied reference ECG score.)
Comparing Q score by lead and ST-T score by groups between reference ECG and Event ECG.
Using maximum Q score change and maximum STD/Tneg score change to determine Novacode I_5.
Comparing Q score with the Flags of QS, E, X, or Z:
If Flag QS in lead V2 in Reference ECG and QS in lead V3 in Event ECG, then Q Score reduces 10 points for lead V3 in Event [-10];
If Flag QS in lead V3 in Reference ECG and QS in lead V4 in Event ECG, then Q Score reduces 20 points for lead V4 in Event [-20];
If Flag E in Reference ECG and QS or Q in Event ECG, then Q Score = 0 in all leads in Event;
If Flag X in Reference ECG and QS or Q in Event ECG, then Q Score = (QS Score - 10 point) for Lead V3-V5 in Event ECG;
If Flag Z in Reference ECG and Flag E in Event ECG, then Q Score = 20 points for Lead V3-V5 in Event ECG;
Note 2
In case of code I 5.0, prevalent MI classication codes 5.1-5.8 in the follow-up or event ECG are categorized as nonevolving (NE)
abnormalities (eg, code I 5.0 with 5.4 as code NE 5.4, nonevolving isolated ST-T abnormalities)
Note 3
I 5.5 criteria for acute-phase hospitalization ECGs are modied as follows:
I 5.5 Profound ST-T evolution (non-Q wave MI) criteria:
C1. STE score increase 30
C2. STE score decrease 30
C3. STE score increase 20
C4. STE score decrease 20
C5. TN score increase 20
Code I 5.5 = C1 or C2 or (C3 and C5) or (C4 and C5)
Note 4
In the presence of incident left bundle branch block (code 3.1), the criteria for acute MI are modied as follows:
I 5.5 Profound ST-T evolution criteria:
C1. Code I 3.1
C2. STE score 20 in lead group L
C3. STD score 30 in lead group S
Code I 5.5 = C1 and (C2 or C3)
STD, ST-segment depression; STE, ST-segment elevation; TN, T wave negativity

315
Serial change (i.e., incident) myocardial infarction (MI) and ischemia by Minnesota Code
(Chap. 15) and Novacode differ in their pattern of occurrence and result in a different
distribution of MI cases by the Minnesota Code than by the Novacode. This is illustrated
in Table B.6 examining the change from baseline to rst follow-up for nearly 100, 000
persons, where it can be seen that a considerable proportion of Novacode MI are coded as
serial STT worsening by the Minnesota Code rather than as serial MI. In either case, serial
change of new MI or STT worsening carry a signicant independent adverse prognosis for
future cardiovascular disease.5

TABLE B.6. Percentage frequency a of incident myocardial infarction and ischemia by Minnesota
Code (MC) by incident Novacode (NC) myocardial infarction (MI) and ischemia among 99,738
men and women aged 1865 years from ve combined study populations of well and diseased
people in up to 3 years of follow-up.
Incident Incident NC MI
c
MC MI 5.0 5.1 5.2 5.3 5.4 5.5 5.6.1 5.6.2 5.7

Q0 83.75 0.01 0.01 0.20 0.03 0.03 0.07 0.39 3.26

Q1 0.02 0.06 0.07 0.22 0.02 0.00 0.00b 0.00 0.11

Q2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Q3 0.00 0.00 0.00 0.00 0.00b 0.00 0.00b 0.00 0.00

Q4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Q5 0.00b 0.00b 0.01 0.00 0.02 0.00b 0.00 0.00b 0.01

Q6 0.00b 0.00b 0.00b 0.00 0.02 0.00b 0.00 0.00b 0.01

Q7 0.00 0.00 0.00b 0.00 0.00 0.00 0.00 0.00 0.00b

ST1 0.46 0.00b 0.02 0.01 0.06 0.27 0.10 0.62 0.03

ST2 0.00 0.00 0.00 0.00 0.00 0.00b 0.00 0.00b 0.00

ST3 0.94 0.00b 0.02 0.01 0.08 0.28 0.16 0.75 0.05

ST4 0.01 0.00 0.00b 0.00b 0.00b 0.00b 0.00b 0.01 0.00b

ST5 0.45 0.00b 0.00b 0.00b 0.00 0.00b 0.01 0.02 0.02

ST6 0.08 0.00 0.00 0.00 0.00 0.00b 0.00 0.01 0.00b

ST7 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00

ST8 1.31 0.00b 0.00b 0.00b 0.00b 0.01 0.03 0.06 0.07

Total 87.03 0.08 0.14 0.43 0.24 0.60 0.37 1.86 3.57
a
5.68% of records removed because QRS 120 ms
b
cells with < 5 cases and hence 0.00%
c
See Chap. 15
Total will slightly vary from column additions because of small cells include in the total

316
Criteria for Other (than Incident MI) Incident or Progressing EGG Abnormalities

Note 1: Event ECG refers to an ECG from a regularly scheduled follow-up examination
or an acute event hospitalization ECG used for serial comparison.
Note 2: Incident abnormalities marked with an asterisk (*) denote clinically signicant
changes or unfavor able responses to therapeutic or other preventive interven-
tions. Incident abnormalities not marked with an asterisk are categorized as ab-
normal change, not clinically signicant.
Note 3: Criteria for regressing abnormalities: Regressing abnormalities or favorable
changes are denoted by regression (R) codes and follow denitions identical to
those for incident (I) codes dened below except that the change occurs in the
opposite (favorable) direction.
Note 4: Clinically signicant favorable change to basic sinus rhythm (regression code
Criteria
C1: Any other rhythm code except 1.0.1, 1.0.2, 1.0.3 or 1.1 in reference ECG
C2: Code 1.0 in event ECG
Code R 1.0 = C1 and C2

I0 Serial Change Uncodable (Totally or Partially) Due to Suppression Code

Note: Character - or . on the coding or report form identies the suppressed codes.

I 0.0 No suppression codes present

I 0.1 Reference or event ECG not available. (This code suppresses all incidence codes.)
I 0.2 Inadequate quality or missing leads
I 0.3 Lead connection interchange in event ECG
I 0.3.1 RA/RL connection interchange. (This code suppresses all I codes.)
Note: Code 0.3.1 in reference ECG is coded as I 0.1 for event ECG.
I 0.3.2 Other lead connection interchanges (correction made and ECG coded)
I 0.4 QRS duration 120 ms in event ECG
Note: Code I 0.4 suppresses ST-T scores and codes I 5.5, I 5.6, and I 5.8. Code I 0.4
with code 3.2 in event ECG suppresses code I 8.
I 0.5 Atrial brillation or utter in event ECG
Note: Code I 0.5 suppresses codes I 2, I 7, and I 9.
I 0.6 Electronic pacemaker
Note: Code I 0.6 with code 1.6.1 in event ECG suppresses all incident codes and,
with code 1.6.1, incidence codes I 1.1-1.8, I 2, I 7, and I 9.
I 0.7 Other suppression codes
I 0.8 Uncertain P wave detection in event ECG

I 1 Incident Arrhythmias
I 1.0 No codable incident arrhythmias
Criteria
C 1: Code 1.0.1 (sinus rhythm) in event ECG
Code I 1.0 = C1
I 1.0.2 Incident sinus bradycardia (SB)

317
 Criteria
C1: Code 1.0.2 in event ECG
C2: Code 1.0.1 or 1.0.3 in reference ECG
C3: Heart rate decrease > 20 cpm from reference ECG
Code I 1.0.2 = C1 and C2 and C3
I 1.0.3 Incident sinus tachycardia (ST)
Criteria
C1: Code 1.0.3 in event ECG
C2: Code 1.0.1 or 1.0.2 in reference ECG
C3: Heart rate increase > 20 cpm from reference ECG
Code I 1.0.3 = C1 and C2 and C3
I 1.1 Incident wandering atrial pacemaker (WAP)
Criteria
C1: Code 1.1 in event ECG
C2: Code 1.0 in reference ECG
Code I l.1 = C1 and C2
I 1.2 Incident junctional rhythm (JR)
Criteria
C1: Code 1.2 in event ECG
C2: Code 1.0 in reference ECG
Code I 1.2 = C1 and C2
I 1.3 Incident ectopic atrial rhythm (EAR)
Criteria
C1: Code 1.3 in event ECG
C2: Code 1.0 in reference ECG
Code I 1.3 = C1 and C2
I 1.4 Incident supraventricular tachycardia (SVT)
Criteria
C1: Code 1.4 in event ECG
C2: Code 1.0.0 or 1.0.1 in reference ECG
Code I 1.4 = C1 and C2
I 1.4.1 Incident SVT, rate < 130 cpm
Criteria
C1: Code 1.4.1 in event ECG
C2: Code 1.0.1 or 1.0.2 in reference ECG
Code I 1.4.1 = C1 and C2
I 1.4.2 Incident SVT, rate 130 cpm
Criteria
C1: Code 1.4.2 in event ECG
C2: Heart rate increase 30 cpm from the reference ECG
Code I 1.4.2 = C1 and C2
I 1.5 Incident atrial brillation/utter (AFLF)
Criteria
C1: Code 1.5 in event ECG
C2: No code 1.5 in reference ECG
Code I 1.5 = C1 and C2
I 1.5.1 Incident atrial utter type 1 (AFL1)
Criteria
C1: Code 1.5.1 in event ECG
C2: No code 1.5 in reference ECG
Code I 1.5.1 = C1 and C2
I 1.5.1.1 Incident atrial utter type 1 classic (AFL1C)

318
 Criteria
C1: Code 1.5.1.1 in event ECG
C2: No code 1.5.1 in reference ECG
Code I 1.5.1.1 = C1 and C2
I 1.5.1.2 Incident atrial utter type 1 variant (AFL1V)
Criteria
C1: Code 1.5.1.2 in event ECG
C2: No code 1.5.1 in reference ECG
Code I 1.5.1.2 = C1 and C2
I 1.5.2 Incident atrial utter type 2 (AFL2)
Criteria
C1: Code 1.5.2 in event ECG
C2: No code 1.5 in reference ECG
Code I 1.5.2 = CI and C2
I 1.5.3 Incident atrial brillation (AF)
Criteria
C1: Code 1.5 in event ECG
C2: No code 1.5.1 in event ECG
C3: No code 1.5.2 in event ECG
Code I 1.5.3 = C1 and C2 and C3
I 1.6 Incident electronic pacemaker (PM)
Criteria
C1: Code 1.6 in event ECG
C2: No code 1.6 in reference ECG
Code I 1.6 = C1 and C2
I 1.6.1 Incident ventricular pacemaker (VPM) or combination pacemaker (CPM)
Criteria
C1: Code 1.6.1 in event ECG
C2: No code 1.6 in reference ECG
Code I 1.6.1 = C1 and C2
I 1.6.2 Incident atrial pacemaker (APM)
Criteria
C1: Code 1.6.2 in event ECG
C2: No code 1.6 in reference ECG
Code I 1.6.2 = C1 and C2
I 1.7 Incident ventricular tachycardia (VT)
Criteria
C1: Code 1.7 in event ECG
C2: No code 1.7 in reference ECG
Code I 1.7 = C1 and C2
I 1.8 Other incident abnormal rhythms
Criteria
C1: Code 1.8 in event ECG
C2: No code 1.8 in reference ECG
Code I 1.8 = C1 and C2
I 1.9 Indeterminate incident rhythm (IAR)
Criteria
C1: Code 1.9 in event ECG
C2: No code 1.9 in reference ECG
Code I 1.9 = C1 and C2
Note: Supplementary conditions listed under prevalent arrhythmia codes (S 1 through S 16)
can also be coded for incident arrhythmic codes.
I 2 Incident AV Conduction Abnormalities

319
I 2.0 No codable incident AV conduction abnormalities
Criteria
C1: Code 2.0 in event ECG
Code I 2.0 = C1
I 2.1 Incident rst-degree AV block (AVB1)
Criteria
C1: Code 2.1 in event ECG
C2: Code 2.0 in reference ECG
C3: PR increase 40 ms from reference ECG
Code I 2.1 = C1 and C2 and C3
I 2.2 Incident second-degree AV block (AVB2)
Criteria
C1: Code 2.2 in event ECG
C2: Code 2.0 in reference ECG
Code I 2.2 = C1 and C2
I 2.2.1 Incident second-degree AV block type Wenckebach or Mobitz 1 (AVB2W)
Criteria
C1: Code 2.2.1 in event ECG
C2: Code 2.0 or 2.1.1 in reference ECG
Code I 2.2.1 = C1 and C2
I 2.2.2 Incident second-degree singular AV block or type Mobitz 2 (AVB2S)
Criteria
C1: Code 2.2.2 in event ECG
C2: Code 2.0 or 2.1.1 in reference ECG
Code I 2.2.2 = C1 and C2
I 2.2.3 Incident second-degree multiple AV block (AVB2M)
Criteria
C1: Code 2.2.3 in event ECG
C2: Code 2.0 or 2.1.1 in reference ECG
Code I 2.2.3 = C1 and C2
I 2.3 Incident third-degree or complete AV block (AVB3)
Criteria
C1: Code 2.3 in event ECG
C2: Code 2.0 or 2.1 or 2.2 in reference ECG
Code I 2.3 = C1 and C2
I 2.4 Incident ventricular preexcitation (WPW)
Criteria
C1: Code 2.4 in event ECG
C2: Code 2.0 in reference ECG
C3: QRS duration increase 20 ms from reference ECG
C4: PR duration decrease 20 ms from reference ECG
Code I 2.4 = C1 and C2 and C3 and C4
I 2.5 New ventricular preexcitation pattern, change not signicant
Criteria
C1: Code 2.4 in event ECG
C2: Code 2.0 in reference ECG
C3: QRS duration increase < 20 ms from reference ECG
C4: PR duration decrease < 20 ms from reference ECG
Code I 2.5 = C1 and C2 and (C3 or C4)
I 3 Incident Prolonged Ventricular Excitation

320
I 3.0 No incident prolonged ventricular excitation
Criteria
C1: Code 3.0 in event ECG (QRS < 115 ms)
Code I 3.0 = C1
I 3.1 Incident left bundle branch block (LBBB)
Criteria
C1: Code 3.1 in event ECG
C2: Code 3.0 in reference ECG
C3: QRS duration increase 20 ms from reference ECG
Code I 3.1 = C1 and C2 and C3
I 3.1.0 Incident LBBB without ECG evidence of incident MI
Criteria
C1: Code I 3.1 in event ECG
C2: Q score increase 15 from reference ECG
Code I 3.1.0 = C1 and C2
I 3.1.1 Incident LBBB with possible incident MI
Criteria
C1: Code I 3.1 in event ECG
C2: Q score 20
C3: Q score increase 15 or more
Code I 3.1.1 = C1 and C2 and C3
Note: Criteria for acute MI in the presence of LBBB are listed in Table B.5, Note 4.
I 3.2 Incident right bundle branch block (RBBB)
Criteria
C1: Code 3.2 in event ECG
C2: Code 3.0 in reference ECG
C3: QRS duration increase 20 ms
Code I 3.2 = C1 and C2 and C3
I 3.2.0 Incident RBBB without ECG evidence of incident MI
Criteria
C1: Code I 3.2 in event ECG
C2: Q score increase less than 15 from reference ECG
Code I 3.2.0 = C1 and C2
I 3.2.1 RBBB with possible incident MI
Criteria
C1: Code I 3.2 in event ECG
C2: Q score 20 in event ECG
C3: Q score increase 15 or more from reference ECG
Code I 3.2.1 = C1 and C2 and C3
I 3.3 Incident indeterminate ventricular conduction delay (IVCD)
Criteria
C1: Code 3.3 in event ECG
C2: Code 3.0 in reference ECG
C3: QRS duration increase 20 ms
Code I 3.3 = C1 and C2 and C3
I 3.3.0 Incident IVCD without ECG evidence of incident MI
Criteria
C1: Code I 3.3 in event ECG
C2: Q score increase less than 15 from reference ECG
Code I 3.3.0 = C1 and C2
I 3.3.1* Incident IVCD with possible incident MI

321
 Criteria
C1: Code I 3.3 in event ECG
C2: Q score 20 in event ECG
C3: Q score increase 15 or more from reference ECG
Code I 3.3.1 = C1 and C2 and C3
I 3.4 New borderline prolonged ventricular exaltation
Criteria
C1: Code 3.4 in event ECG
C2: Code 3.0 in reference ECG
C3: QRS duration increase 20 ms from reference ECG
Code I 3.4 = C1 and C2 and C3
I 3.5 New bundle branch block, QRS duration increase < 20 ms
Criteria
C1: Code 3.1 or 3.2 or 3.3 in event ECG
C2: QRS duration increase < 20 ms from reference ECG
Code I 3.5 = C1 and C2
I 3.5.0 New bundle branch block with QRS duration increase < 20 ms and with no ECG
evidence of incident MI
Criteria
C1: Code I 3.5 in reference ECG
C2: Q score increase less than 15 from the reference ECG
Code I 3.5.0 = C1 and C2
I 3.5.1 New bundle branch block with QRS duration increase < 20 ms and possible
incident MI
Criteria
C1: Code I 3.5 in reference ECG
C2: Q score increase 15 or more from the reference ECG
Code I 3.5.1 = C1 and C2

I 4 Incident Prolonged Ventricular Repolarization

I 4.0 No incident prolonged ventricular repolarization

C1: QTI < 112 in event ECG
C2: QTI 112 in event ECG and QTI < I12 in reference ECG and QTI increase <10%
Code I 4.0 = C1 or C2
I 4.1 Incident prolonged ventricular repolarization
Criteria
C1: QTI 112% in event ECG and QTI 112% in reference ECG
C2: QTI increase 10% from reference ECG
Code I 4.1 = C1 and C2

I 5 Incident ECG Abnormalities Related to Myocardial Infarction and Ischemia

See Table B.5 for denitions.

I 6 Incident Left Ventricular Hypertrophy (LVH)

I 6.0 No incident LVH

Criteria
C1: Code 6.0 in event ECG
Code I 6.0 = C1
I 6.1 Incident LVH

322
 Criteria
C1: Code 6.0 in reference ECG
C2: Code 6.1 in event ECG
C3: (RaVL + SV3) increase > 400 V from reference ECG
Code I 6.1 = C1 and C2 and C3
I 6.2 New LVH in ECG 2, change not signicant
Criteria
C1: Code 6.1 in event ECG
C2: (RaVL + SV3) increase 400 V from reference ECG
Code I 6.2 = C1 and C2
Note: For risk analysis, the following optional criteria are suggested:
I 6.3 LVH progression
Criteria
C1: (RaVL + SV3) increase > 400 V from reference ECG
C2: ST-segment depression or T wave negativity score increase 20 points
Code I 6.3 = C1 and C2
I 6.4 LVH regression
Criteria
C1: (RaVL + SV3) decrease > 400 V from reference ECG
Code I 6.4 = C1
Note: Incident and regression codes for other abnormalities are not dened in this version
of the Novacode due to unavailability of criteria or limits for short-term biologic and
technical variation.

***Update from the 1998 article (2)

I 7 Incident Left Atrial Enlargement (LAE)

I 7.0 No incident LAE

Criteria
C1: Code 7.0 in event ECG
Code I 7.0 = C1
I 7.1 Incident LAE
Criteria
C1: Code 7.1 in event ECG
C2: Code 7.0 in reference ECG
Code I 7.1 = C1 and C2

I 8 Incident Right Ventricular Hypertrophy (RVH)

I 8.0 No incident RVH

Criteria
C1: Code 8.0 in event ECG
Code I 8.0 = C1
I 8.1 Incident RVH
Criteria
C1: Code 8.1 in event ECG
C2: Code 8.0 in reference ECG
Code I 8.1 = C1 and C2
I 9 Incident Right Atrial Enlargement (RAE)

323
I 9.0 No incident RAE
Criteria
C1: Code 9.0 in event ECG
Code I 9.0 = C1
I 9.1 Incident RAE
Criteria
C1: Code 9.1 in event ECG
C2: Code 9.0 in reference ECG
Code I 9.1 = C1 and C2

I 10 Incident Fascicular Block

I 10.0 No incident Fascicular Block

Criteria
C1: Code 10.0 in event ECG
Code I 10.0 = C1
I 10.1 Incident Left Anterior Fascicular Block (LAFB)
Criteria
C1: Code 10.1 in event ECG
C2: Code 10.0 in reference ECG
Code I 10.1 = C1 and C2
I 10.2 Incident Left Posterior Fascicular Block (LPFB)
Criteria
C1: Code 10.2 in event ECG
C2: Code 10.0 in reference ECG
Code I 10.2 = C1 and C2

References

1. Rautaharju PM, Calhoun HP, Chaitman BR. Novacode serial ECG classication system for
clinical trials and epidemiological studies. J Electrocardiol 1992;24(suppl):179-187.
2. Rautaharju PM, Park LP, Chaitman BR, Rautaharju F, Zhang ZM. The Novacode criteria for
classication of ECG abnormalities and their clinically signicant progression and regression. J
Electrocardiol 1998;31(3):157-187.
3. Selvester R. Wagner G, Hindman N. The Selvester QRS scoring system for estimating
myocardial infarct size. The development and application of the system, Arch Intern Med.
1985;145:1877-1881.
4. Sgarbossa EB, Pinski SL, Barbagelata A, et al. The GUSTO-1 (Global Utilization of
Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries) Investigators.
Electrocardiographic diagnosis of evolving acute myocardial infarction in the presence of left
bundle-branch block. N Engl J Med. 1996; 334:481-487.
5. Denes P, Larson JC, Lloyd-Jones DM, Prineas RJ, Greenland P. Major and minor ECG
abnormalities in asymptomatic women and risk of cardiovascular events and mortality. JAMA.
2007;298 (9):978-985.

324
Appendix C
Major and Minor ECG Abnormalities for Population Comparisons
with Minnesota Code and Novacode Equivalents
Major Abnormalities
Abnormality Minnesota Codes Novacodes
Major Q wave abnormalities MC 1-1, 1-2 NC 5.1, 5.2, 5.3
(Old prevalent MI)

Minor Q wave abnormalities plus MC I-3 plus NC 5.4

ST-T abnormalities (Possible old MI) MC 4-1 or 4-2, or 5-1 or 5-2

Major Isolated ST-T abnormalities MC 4-1 or 4-2 or 5-1 or 5-2 NC 5.5, 5.6

Complete or intermittent LBBB MC 7-1 NC 3.1

Complete or intermittent RBBB MC 7-2 NC 3.2
Nonspecic intraventricular block MC 7-4 NC 3.3
RBBB with left anterior hemiblock MC 7.8 NC 3.2 plus NC 10.1
Brugada pattern MC 7-9

Left ventricular hypertrophy plus MC 3-1 plus NC 6.1.1

ST-T abnormalities MC 4-1 or 4-2 or 5-1 or 5-2

Major QT prolongation QTI 116% NC 4.1.2

Atrial Fibrillation or Flutter MC 8-3 NC 1.5

(Continuous or intermittent)

Major AV conduction abnormalities

Third-degree AV block (AVB3) MC 6-1 NC 2.3
Second-degree AV block (AVB2) MC 6-2 NC 2.2
Ventricular preexcitation pattern (WPW) MC 6-4 NC 2.4
Articial pacemaker MC 6-8 NC 1.6

Other major arrhythmias

Ventricular brillation or MC 8-2 NC 1.7, 1.8.2, 1.9.2
Ventricular asystole

Supraventricular tachycardia (SVT) MC 8-4-2 or NC 1.4 or

MC 8-4-1with HR>140 NC 1.3.3, 1.2.3 with HR>140

325
Minor Abnormalities
Abnormality Minnesota Codes Novacodes
Minor Isolated Q/QS waves MC 1-3 NC 5.7
Minor ST/T abnormalities MC 4-3, 4-4, 5-3, 5-4 NC 5.8
High R waves (left ventricular) MC 3-1, 3-3, 3-4 NC 6.1.0
High R waves (right ventricular) MC 3-2 NC 8.1
ST segment elevation MC 9-2 (Event ECG I-5.5)
Incomplete RBBB MC 7-3 NC 3.4.1
Incomplete LBBB MC 7-6, 7-7 NC 3.4.2, 10.1, 10.2
Minor QT prolongation QTI 112% NC 4.1.1
Short PR interval MC 6-5
Long PR interval MC 6-3 NC 2.1
Left axis deviation MC 2-1
Right axis deviation MC 2-2
Premature beats (supraventricular) MC 8-1-1 NC 1.0.A.1, 1.0.A.2
Premature beats (ventricular) MC 8-1-2 NC 1.0.A.3
Premature beats (combined) MC 8-1-3, 8-1-5 NC 1.0.A.1, 1.0.A.3,
(with NC 1.1)
Wandering atrial pacemaker MC 8-1-4 NC 1.1
Sinus tachycardia MC 8-7 NC 1.0.3
Sinus bradycardia MC 8-8 NC 1.0.2
Supraventricular rhythm persistent MC 8-4-1 NC 1.2, 1.3
Low voltage QRS MC 9-1
High amplitude P wave MC 9-3 NC 9.1
Left atrial enlargement (LAE) MC 9-6 NC 7.1
Fragmented QRS MC 7-10
(Early Repolarization)a MC 9-7a
a
If Early Repolarization is shown to be independently predictive of future mortality

326
Index
1-Codes, 16, 3448, 128, 220, 222
2-Codes, 4954
3-Codes, 5559, 187, 220, 232
4-Codes, 60, 63, 7580
5-Codes, 60, 6497, 203, 220, 248,
254, 255
6-Codes, 98, 105110, 140143
7-Codes, 111, 119132
8-Codes, 134158
9-Codes, 159185
A Heart rate, 187190
Aberrant ventricular conduction, 110
Acute myocardial infarction, 233257, 308316
Arrhythmias, 134158
Atrial utter, 151, 184, 307
Atrial brillation, 150, 184, 307
A-V dissociation, 155157
B
Beats to be measured, 13
Bifascicular block, 129
Biologic variability, 224
Bipolar limb leads, 6
Brugada pattern, 131, 132
C Lead reversals, 171185
Calibration deection, 13, 159161
Chest leads, 8
Coding forms, 203205
Criteria for serial change, 226262
D Left Atrial hypertrophy, 159, 170, 383
Depolarization, 3
Digital Records, 15
E Major ECG abnormalities by
E point, 208, 214
Early repolarization, 170, 171
ECG leads, 6
ECG recording, 207225
Electrode positions, 207, 212215
Electronic (articial) pacemaker, 110
Electronic records, see Digital
Evolving ST segment elevation, 231
Evolving ST depression, 231
Evolving T wave inversion, 227
Evolving Q waves, 227230
F
Fasting, 206
Forms, see Coding form
Fragmented QRS, 132133
Frontal plane QRS/T angle, 265
Fusion beats, 138, 140, 141
H
Heart block complete (third degree), 105
Heart block (second degree), 105107
Heart block (rst degree), 107
Heart rate variability, 266268
High R-wave (MC 3), 5559
History, 1
I
Intraventricular block (non-specic), 126
Incompatible codes, 284
J
J Point, 6079, 115, 170, 171
J Point depression, 63, 67, 7279, 93
J Point elevation, 67, 162165
L
Left anterior fascicular block, 128
Left axis deviation, 50
Left bundle branch block (LBBB), 121,
122, 128
Left ventricular hypertrophy, 5559, 232, 302
Low QRS amplitude, 159161
M
Minnesota code, 325
Major ECG abnormalities by Novacode, 325
Mathematical symbols, 15
Measurement differences, 15
Measuring devices, 10
Measuring loupe, 12
Minor ECG abnormalities by
Minnesota code, 326
Minor ECG abnormalities by Novacode, 326
Mobitz type I heart block, 107
Mobitz type II heart block, 105, 107, 153, 154
327
N Recording paper grid, 10
Novacode, 287324 Recording form, 203
Novacode criteria, 287324 Repolarization, 5
Novacode criteria for myocardial infarction, Right axis deviation, 51, 52
308316 Right bundle branch block (RBBB),
Novacode criteria for serial change myocardial 122124, 259
infarction, 308315
Novacode/Minnesota code equivalents, 287289, S
325326 S wave amplitude, 120, 126, 197
Serial ECG change, 226261
P Short PR interval, 109
P wave amplitude, 166 Single channel electrocardiographs, 207
P wave duration, 170 Sino-atrial arrest, 153154
P wave offset (QRS onset), 98, 101 Sino-atrial block, 155
P wave onset, 98, 101 Sinus bradycardia, 158
Persistent ventricular rhythm, 148 Sinus tachycardia, 157
Plastic ruler, 13 Spatial QRS/T angle, 264, 265
Position for ECG recording, 207208 Spatial T axis, 265
PR interval, 98105 ST segment codes by site, 279
Premature beats supraventricular, 134140, 144146 ST segment depression, 60, 63
Premature beats ventricular, 137146, 149 ST segment elevation, 162165, 170
P-terminal force, 170 ST segment slope, 7174
Summary of Minnesota codes, 277286
Q Supraventricular rhythm, 152, 153
Q wave amplitude, 19 Suppression codes, see Incompatible codes
Q wave codes by site, 277, 278
Q wave onset, 22, 99 T
Q wave duration, 15, 21, 22, 222, (309, 310?) T wave amplitude, 8688, 90, 93, 169171,
QRS axis, 4954, 191195 198199, 255
QRS duration, 111118, 222 T wave axis, 5, 54
QRS onset, 18, 26, 50, 74, 98, 101103, 119, 125, T wave codes by site, 280
200, 201 T wave offset, 201, 202
QRS offset, 111118 T wave negative, 8284
QRS/T matrix, 264 T wave diphasic, 81, 82, 86, 89, 90, 95, 131,
QRS transition zone, 167169 198, 199
QS waves, 1620, 2326, 28, 3234, 45, 47, 48 Terminal R wave, 29, 30
QT interval, 200202 The electricity part of the ECG, 25
Quality of ECG recording, 216218
Quality control of ECG data, 223225 U
Quality control of visual coding, 270275 Unipolar limb leads, 7
Quality control of digital coding, 275
QRS/T simple, 264 V
QRS/T frontal, 265 Ventricular asystole, 147
Ventricular brillation, 147
R Ventricular parasystole, 149
Rate corrected QT interval, 200 Ventricular tachycardia, 149
R peak duration, 119122, 129, 130, 140 Visual records, 15, 270
R wave voltage, 5559
R wave amplitude, 2429, 125, 196, 227 W
R wave initial amplitude, 125 W pattern QRS, 31, 32
R wave terminal amplitude, 125, 127 What is the Electrocardiogram or ECG?, 1
Ratio Q/R, 3335, 41, 45 Wolff-Parkinson-White (WPW), 108, 142

328