High Resolution Electrocardiography
Abstract
Over the past decade, significant advances were
made in
the research, diagnosis, and treatment of cardiovascular diseases. Such
progress was in every sphere of cardiology that includes non-invasive,
minimally invasive, and invasive technologies. Interpretive
electrocardiography, cardiac pacemakers, cardiac stents, and
angioplasty are some areas where the progress has been significant.
Non-invasive methods of diagnosis
of cardiac disorders involve digital recording of cardiac signals at
the body surface (chest) and subsequent computerized analysis. Such
methods
and instruments provide a vital first step to the diagnosis of the
heart
without involving surgical procedures. One such non-invasive field is
High
Resolution Electrocardiography (HRECG). A high-resolution
electrocardiogram
detects very low amplitude signals in the ventricles called 'Late
Potentials'
in patients with abnormal heart conditions. A standard
electrocardiogram
cannot detect these signals. The presence of late potentials is widely
accepted to have prognostic significance in patients after Acute
Myocardial Infarction (AMI)1,2,3.
High Resolution Electrocardiography enhances the
diagnostic capabilities of ECGs. This article describes the principles
involved
in HRECG and the techniques that are employed to derive such superior
diagnostic capabilities. The use of these techniques may lead to more
discoveries in the causes of cardiac disorders and improved drug
discoveries
to combat such conditions.
1. Introduction
The
Electrocardiogram (ECG) is a graphical representation of the electrical
potentials generated by the heart. Such electrical signals are recorded
via ECG leads placed on the body surface. Based on the resolution of
the digital recording of analog ECG signals, the instruments and
techniques may be categorized into two types: 1) Low-resolution (or
standard) ECG, and 2) High-Resolution Electrocardiogram (HRECG). A
standard 12-Lead ECG is a typical example of a widely used
low-resolution instrument that records 9 seconds of cardiac data. A
Signal-Averaged Electrocardiogram (SAECG) is a typical example of a
High-Resolution Electrocardiogram. The latter signifies recent
innovations in advanced ECG analysis and diagnosis. These instruments
record ventricular ECG signals of very low magnitudes called
'Ventricular Late Potentials' (VLP) by averaging a number of signals.
The presence of VLPs is indicative of increased risk for subsequent
occurrence of arrhythmic events, mainly sustained ventricular
tachycardia1,2,3.
2. Cardiac Late Potentials
Cardiac late potentials are low amplitude signals
that occur in the ventricles. Also called Ventricular Late Potentials
(VLPs), these signals are caused by slow or delayed conduction of the
cardiac activation sequence. Under certain abnormal conditions, there
may be small regions of the ventricles within a diseased or ischemic
region that generate such delayed conduction. This results in
depolarization
signals that prolong past the refractory period of surrounding tissues
and re-excite the ventricles4. This re-excitement is known as 'reentry'. Reentry is
believed to be a key factor that causes VLPs.
2.1 Late Potentials and the Electrocardiogram
The
electrocardiogram represents the depolarization and repolarization of
the major chambers of the heart. Under normal conditions, the
ventricular regions to be activated coincide with the end of the QRS
complex. However, in conditions of reentry, VLPs appear a few
microvolts extending well beyond the terminal portion of the QRS in the
ECG.
Due
to their very low magnitudes, late potentials are not visible in a
standard ECG. Moreover, factors such as increased distance of the body
surface electrodes from the heart, and inherent noise in patients make
identification of VLPs beyond the resolution limits of a standard ECG.
As a result, high-resolution recording techniques and computerized ECG
processing are necessary for detection of late potentials. Such ECG
signal processing includes techniques to improve the signal-to-noise
ratio (SNR). One widely used technique to improve the SNR of ECG
signals for the detection of late potentials is signal averaging.
2.2 Signal Averaging
Signal Averaging is a signal processing technique that is applied to
signals that are periodic. Every signal inherently has two components:
1) signal, and 2) noise. By adding many signals, the overall noise
component of the signal sum will decrease while the desired signal
component remains unchanged5. Since ECG
signals are periodic
in nature, inherent noise in the signals can be minimized thus
enhancing the signal component in the process. This signal enhancement
will improve the resolution of the overall signal especially the
detection of late potentials. Figure 1.0 shows an ECG signal and a
SAECG signal of the same patient. The noise components are
significantly reduced in the SAECG signal
3. HRECG Instrument
Figure 3.0 illustrates the block diagram of a typical HRECG instrument.
A HRECG instrument may be divided into two major parts: 1) signal
acquisition front-end, and 2) computerized SAECG processing. The
components of
each part are described in greater detail in the following sections.
Figure 1(a). A ECG signal with noise components
Figure 1(b). A SAECG signal
Figure 2. HRECG block diagram
3.1 Components
As
illustrated in Figure 2.0, an HRECG instrument consists of 4 key
components: 1) Amplifiers, 2) Bandpass filters, 3) Analog/Digital
converter, and 4) SAECG Processor. The SAECG Processor may in turn be
functionally divided into the following components: a) Signal Averager,
b) Bidirectional Bandpass Filter, c) Filtered Vector Magnitude, and d)
SAECG Quantifier. In addition, the instrument includes 7 ECG leads.
These leads are bipolar, orthogonal electrodes comprising X+, X-, Y+,
Y-, Z+, Z-, and ground placed in a particular fashion on the body
surface6. These electrodes are usually
referred to as XYZ leads.
3.1.1 Amplifiers
ECG
signals are acquired using precision instrumentation amplifiers. This
electronic circuitry is a critical part in the signal conditioning
front-end, as noise is a serious hindrance to high-resolution signal
acquisition. These devices combine the positive and negative components
of a signal (X+, X-) to yield a differential signal with reduced noise.
3.1.2 Bandpass Filter
Most
of the energy in ECG signals are concentrated in the bandwidth between
1Hz - 80Hz. This analog base-bandwidth is typical to standard ECGs.
However, for HRECG signal acquisition, a broad bandwidth of 0.05Hz -
300Hz is
considered in order to acquire as many signal frequencies as possible.
As a result, the signals are sampled at very high frequencies. Typical
minimum sampling frequencies are >= 1000Hz (1KHz).
3.1.3 Analog / Digital Converter
An
analog-to-digital converter (ADC) allows the digitization of an analog
signal. After conversion, analog signals that are continuous in nature
are represented as a digital stream of data. This digital data is
stored in memory as binary bits. ADCs in HRECG instruments should have
a minimum of 12-bit resolution i.e. every data point is 12-bits of
digital data. Once converted, the signal representation is transformed
from the analog domain to the digital domain.
3.1.4 SAECG Processor
The
SAECG processing functionality is digitally implemented either through
hardware components or through software via computerized processing. In
hardware, a general purpose Digital Signal Processor (DSP) can perform
all the 4 key SAECG functional blocks. On the other hand, each SAECG
block can be implemented in software. Each of these functions is
described in greater detail below.
3.1.4.1 Signal Averager
Signal averaging is the crux of an HRECG instrument. As explained in
Section 2.2, a signal averager computes the average value of each data
point in a cardiac signal after averaging many such signals.
Consequently, the resultant average results in minimized noise, and
allows for high resolution processing of the signal average.
3.1.4.2 Bidirectional Bandpass Filter
Signal averaging may introduce ST segments shifts. Filters are used on
a signal average to eliminate any bias especially in the terminal QRS
and ST segment regions. Typically a recommended bandpass filter is a
bi-directional Butterworth implementation7.
This implementation is a 2nd order low-pass filter and a 4th order
highpass filter with cutoff frequencies of 250Hz and 400Hz
respectively. These filters inherently cause ringing in the signal.
Consequently, artifacts may be introduced in the regions of interest,
which may be mistaken for late potentials. A bi-directional
implementation of the Butterworth filter shifts this ringing into
regions of the QRS without distorting the terminal QRS endpoints8.
3.1.4.3 Filtered Vector Magnitude
The filtered vector magnitude is computed as (X2 + Y2 + Z2)1/2. The
results of this computation are the basis for quantifying the SAECG
parameters for late potential identification.
3.1.4.4 SAECG Quantifier
The
SAECG quantifier determines the HRECG parameters that identify late
potentials. The criteria for late potential identification are
described in Section 4.0. The SAECG parameters are based on these
criteria.
4. HRECG Parameters: Criteria to Identify Late Potentials
The
task force committee of the European Society of Cardiology, the
American Heart Association and the American College of Cardiology9 suggested a representative criteria for the
identification of late potentials. This criteria is defined in Table 1.0
|
HRECG
Parameter
|
Description
|
Criteria
|
|
QRSD
|
Filtered QRS duration
|
> 114 msec
|
|
RMS40
|
Root mean square voltage of the last 40
msec of the QRS complex
|
< 20 µV
|
|
LAS40
|
The duration of the low amplitude signals
that are < 40µV of
the terminal QRS complex (LAS40)
|
> 38 msec
|
Table 1. Criteria for Late Potential Identification
4.1 A High Resolution Electrocardiogram
A
high-resolution electrocardiogram is aimed at identifying late
potentials. It graphically represents the filtered Vector Magnitude of
the XYZ leads. Figure 3 illustrates a high-resolution electrocardiogram
of a patient with positive late potentials.
4.2 Standard ECG vs HRECG
As
described in the sections above, a HRECG differs from a standard ECG in
several ways. Table 2.0 summarizes some of the key differences between
the two.
Figure 3 HRECG of a patient with positive late
potentials
|
Standard ECG
|
HRECG
|
|
Low digital signal resolution
|
High digital signal resolution
|
|
ECG signals are bandlimited to low
frequencies, typically 0-80/100Hz
|
ECG signals have a wide analog bandwidth,
typically 0.05Hz-300Hz
|
|
ECG signals are sampled at low rates,
typically 300 Hz
|
ECG signals are sampled at very high
rates, typically >= 1KHz
|
|
The analog-to-digital converters have a
low resolution, typically 8 bits
|
The ADCs have a high resolution, typically
12-bits
|
|
Cannot detect late potentials
|
Can detect late potentials after digitally
processing the signals
|
|
The focus is on interpreting all portions
of a ECG signal
|
The focus is on interpreting the terminal
portion of the QRS complex and regions beyond
|
|
The PR interval, QRS duration, ST segment
elevation / depression are some key analysis parameters
|
QRS duration, RMS40, and LAS40 are the key
analysis parameters
|
Table 2. Some differences between Standard ECG and HRECG
5. Clinical Significance
The ability to detect cardiac late potentials in a High Resolution
ECG has significant clinical implications. What was originally
undetectable is no longer the case and is primarily due to the
advancements in non-invasive signal acquisition and processing
techniques such as signal averaging. The detection of late potentials
in patients has prognostic significance especially in patients with
acute myocardial infarction. The presence of ventricular late
potentials in patients with acute MI is an indicator of risk from
subsequent MI or sudden cardiac death. Earlier this classification and
subsequent diagnosis was possible only via invasive or minimally
invasive techniques. A SAECG study is accepted as a standard procedure
in the United States with due reimbursement from Medicare.
6. Conclusion
High Resolution Electrocardiography has serious research and
clinical significance. The techniques used in SAECG can be applied to
a variety of other research topics in order to better study the
mechanisms
of cardiac abnormalities that manifest on the body surface as abnormal
signals, for example: premature ventricular contractions.
References
1.
Breithardt G, Schwarzmaier J, Borggrefe M, Haerten K, Seipel L.
Prognostic significance of ventricular late potentials after acute
myocardial
infarction. European Heart J 4:487, 1983
2. Dennis
AR,
Cody DV, Russell PA, Young AA, Ross DL, Uther JB. Prognostic
significance
of inducible ventricular tachycardia after myocardial infarction. J
American College of Cardiology 3:610, 1984
3. Breithardt
G, Borggrefe M, Haerten K. Role of programmed ventricular stimulation
and noninvasive recording of ventricular late potentials for the
identification of patients at risk of ventricular tachyarrhythmias
after acute myocardial infarction. In Zipes DP, Jalife J (eds): Cardiac
Electrophysiology and Arrhythmias. Pp 553-61 1985.
4. Berbari EJ,
Scherlag BJ, Hope R, Lazzara R. Recordings from the Body Surface of
Arrhythmogenic Ventricular Activity during the ST segment. American
Journal of Cardiology, 41:697, 1978
5.
Narayanaswamy S. Analysis of Cardiac Late Potentials, T-Waves,
and Beat Intervals to Identify the Mechanisms of Premature Ventricular
Beats. Master's Thesis - University of Oklahoma, pp 28-30, 1993
6. SAECG Lead
Placement Chart. Rajal Biotechnologies.
7. Oppenheim
AV, Shafer RW. Discrete-time Signal Processing. Prentice Hall,
Englewood Cliffs, NJ, pp 411-425, 1989
8. Simson MB.
Use of Signals in the Terminal QRS complex to identify Patients with
Ventricular Tachycardia after Myocardial Infarction. Circulation
61:235, 1981
9. Breithardt
G, Cain ME, Flowers NC, Simson MB. Standards for Analysis of
Ventricular Late Potentials using High-Resolution or Signal-Averaged
Electrocardiography: A statement by a task force committee of the
European Society of Cardiology, the American Heart Association, and the
American College of Cardiology. Journal of the American College of
Cardiology
17: 999-1006,
1991
