 |
| The Clinical Case for Heart Rate Variability: |
|
The Most Powerful Single Measure of
Cardiac Dysfunction and Cardiac Risk in
the 40+years Male Population. |
| |
| Director of Research, Cardio-Thoracic
Centre - Liverpool, UK |
| |
| Introduction |
| The prognostic power of Heart Rate Variability
(HRV) measurement, although recognised at the top levels of
cardiology,
is not widely known or understood by the majority of medical
practitioners, due mainly to the impracticality of gathering
and analysing the data from the patient using conventional methods.
Until the recent development of a new and powerful computer
based remote telemetry ECG/HRV measurement technology (such
as VariaCardio® TF5 System), HRV measurement required insisted
on the use of costly and inefficient ambulatory ‘Holter’ recording
of 24 hour single channel ECG data using a portable tape or
digital recorder. The HRV parameters are produced by computer
analysis of the 24 hour ECG tape recording, an additional time
consuming process. Now that the TF5 System has enabled accurate
HRV data to be digitally collected and processed for the individual
patient using a simple, non-invasive, fast, reliable and cost
effective 15 minute test, accurate cardiac risk stratification
and prognosis4-5
by means of HRV can be better and easily carried out in both
the screening and primary care environments. |
| |
| Quantification of Heart Rate Variability |
HRV has become the conventionally accepted
term to describe variations in both instantaneous heart rate
and (preferably measured by changes in time series of RR intervals).
The RR interval is the time elapsing between two normal adjacent
R wave peaks in the ECG signal. Many methods of quantifying
HRV have been investigated1, including:-
(1) ‘Time Domain Methods’ (measurement of all
intervals between adjacent QRS complexes of the ECG signal resulting
from normal sinus node depolarisation, known as the NN /or better
RR/ interval, producing intuitively simple measures such as
mean NN interval, Minimum and Maximum NN intervals at certain
time points etc). Time domain analysis methods include, as well
‘Statistical Methods’:- (measurement of standard deviation
of NN intervals and standard deviation in sequential differences
in NN intervals).
‘Geometric Methods’
(applying simple mathematical transformations to NN or normal
RR intervals to produce geometrical/graphical ‘surfaces’ whose
deviation from the ‘ideal’ surface shape quantifies abnormality
in an individual’s HRV /such as St.George’s Index/).
(2) ‘Frequency Domain Methods’ (applying a
complex mathematical transformation known as Fast Fourier Transformation
(FFT) to the sequential RR intervals, producing an abstract
but complete representation of the data in the ‘Spectral Domain’).
(3) ‘Chaos Analysis’ -- recently proposed for
exploration of non-harmonic features of RR variations time series.
However, more research, controlled trials and clinical evidence
is necessary to better estimate its position in today’s non-invasive
cardiology risk assessment.
The investigations have shown that the ‘Frequency Domain Methods’
produce very important most clinically useful parameters. are
produced by the ‘Frequency Domain Methods’. These ‘Frequency’
or ‘Spectral Domain’ HRV parameters exhibit a close relationship
with the underlying physiological processes3
controlling heart rate (autonomic nerve traffic) and exhibit
good prognostic properties4-5
and risk stratification. The clinical relevance of an abstract
mathematical representation of the HRV data becomes less surprising
when it is recognised that Fourier Transformation and ‘Spectral
Domain’ representation of data is itself used by the body at
a number of physical/neurological interfaces. (For example,
the ear carries out a Fourier transform at the cochlea, enabling
cerebral interpretation of ‘Sound’ from the pressure wave stimulation
at the eardrum.) |
| |
| Spectral HRV Parameters and their
Clinical Significance |
If the series of sequential RR intervals
are accurately recorded, as shown in Figure 1, the data can
be represented in the ‘Spectral Domain’ using a mathematical/computer
technique called Fast Fourier Transformation (FFT). The spectral
data represents the frequency (Hz) with which the sequential
RR intervals change during the period of recording. In fact,
the data is represented in the ‘Spectral Domain’ by a range
of frequencies (x axis, Hz) and the ‘amount’ (spectral power,
y axis, ms2) of change in sequential RR intervals at each frequency,
shown in Figure 2.
Although an exact relationship does not exist, it is established
in a number of autonomic manipulation trials1-3
that the ‘low’ frequency components of the HRV ‘spectral’ data
relate preferably to the sympathetic nerve traffic speeding
up the heart and the ‘high’ frequency components relate predominantly
solely to the parasympathetic autonomic nerve traffic slowing
down the heart. Hence, the total area under the ‘spectral’ curve
(Total Frequency BandPower, 0.01-0.40 Hz) relates to the total
autonomic nerve traffic, the area under the low frequency portion
of the curve (Very Low and Low Frequency Band, 0.05-0.15 Hz
Power) relates predominantly to sympathetic nerve traffic and
the area under the high frequency portion of the curve (High
Frequency PowerBand, 0.15-0.40 Hz) relates to parasympathetic
autonomic nerve traffic. Relationship between spectral power
within the area of Very Low Frequency Band (0.01-0.05 Hz) and
bodily functions is still not fully elucidated. It is expected,
however, that a significant role play renin-angiotensin system,
thermoregulatory variations and/or very slow parts of sympathetic
autonomic control here.
From the spectral representation of HRV, three parameters (Very
Low Frequency Power, Low Frequency Power and Total Power) have
been established4-5
using the Framingham Heart Survey data (using7364 + 25015 subjects
followed between 4 to 12 years) as exhibiting a ‘Hazard Ratio’
of approximately 1.4 for the prognosis of cardiac events. (The
‘Hazard Ratio’ is the definitive measure of ‘prognostic power’
of a measured parameter based on‘actual outcome’ data). The
value of 1.4 is significantly higher than the prognostic power
of any conventional ‘non HRV’ parameter for prognosis of cardiac
events, established earlier in the Framingham Heart Survey6-8.
(The conventional parameters considered were:- Systolic and
Diastolic Blood Pressure, Total Cholesterol, HDL Cholesterol,
Blood SugarGlucose, Smoking, Diabetes, Left Ventricular Hypertrophy,
Hypertensionve Treatment, Atrial Fibrillation, Personal History
of Cardiovascular Disease. Note that the Framingham data6-8
is used for the formulation of baseline absolute risk of cardiac
events in the UK population and routinely used for the calculation
(using the non-HRV parameters) of relative and absolute reduction
in risk expected in the individual by virtue of proposed pharmaceutical
interventions).
In summary, the spectral HRV parameters of Very Low Frequency
Power, Low Frequency Power, and Total Power are reported in
these studies to represent the best single prognostic indicators
of future cardiac events in the 40+ male population5. Further,
the Framingham Heart Study concludes that these parameters are
the best prognostic indicator of sudden death from all causes
in the elderly male population5,
and, according to the Zutphen Study, in the entire 40+ male
population9,10.
It is clear that a practical spectral analysis HRV test, e.g.,
as offered by the TF5 System, has might have might have mian
important role in ‘front line’ cardiac care.
Practical Initial Clinical Evaluation of Cardiac Risk Using
the TF5 System. It is established that the powerful prognostic
spectral HRV parameters4-5 can be measured accurately using
the sequential RR interval data collected over periods as short
as five minutes provided the data is collected and processed
in accordance with the specification published in the Special
Report on HRV produced by the European Society for of Cardiology
(ESC) and North American Society of Pacing and Electrophysiology
(NASPE) Task Force1.
The TF5 System exceeds the European Society for Cardiology ESC
& NASPE specification, providing unprecedented technical
accuracy in spectral HRV measurement in a simple 15 minute test
by:- |
| |
•
|
Recording the ECG data with a temporal resolution of
1000Hz (compared to 135Hz typical of for some 24 hour
Holter monitors) allowing accurate RR interval measurement
which, in turn, significantly extends both the range and
accuracy of the ‘Spectral Domain’ data. |
| |
|
•
|
Determining sequential RR intervals accurately and consistently
in all circumstance by means of patented specific ECG
pattern recognition. |
| |
|
•
|
Automatically rejecting and replacing ectopic beats,
which, if included in the RR interval data set, would
reduce the accuracy of the spectral data, produced by
the FFT. |
| |
|
| • |
Using ‘Course-Graining’10 FFT to automatically reject
‘chaotic’ data. |
|
| |
| The TF4 System provides clinical practicality
and simplicity by:- |
| |
•
|
Using high quality electrodes and signal processing
in a compact cordless chest belt worn under the clothing
during the test. Skin preparation is unnecessary. The
high resolution ECG data is transmitted up to 100 metres
from the chest belt to a highly compact combined radio
receiver and pre processor unit which is connected to
a ‘host’ computer. |
| |
|
•
|
Providing full portability when using a laptop host
computer. |
| |
|
•
|
Providing a dedicated and simple to use patient database,
installed on the host computer. |
| |
|
•
|
Presenting the HRV spectral parameters clearly both
numerically and graphically on the host computer such
that interpretation of results and stratification of the
individual’s risk of cardiac events is simple and straightforward.
|
| |
|
•
|
Quantifying and recording the ‘quality’ of data acquisition
to facilitate clinical audit. |
| |
|
•
|
Providing an option of a telemedicine consultation by
expert in the HRV field by sending of selected records
via Internet, when in doubt: Should ‘high risk’ stratification
or unusual ‘Spectral Domain’ patterns be produced by the
computer based TF4 System, the user has the option of
transmitting, at the push of a button, the complete patient
data set to a leading UK cardiologist via the Internet.
A report on the test data and recommendation, if necessary,
for further specified investigations is returned within
two working days. |
|
| |
The TF4 System maximises clinical information
and the repeatability of results by using a specific measurement
protocol. In brief:-
The patient is measured for a total of 15 minutes, the first
5 minutes prone, the second 5 minutes standing and the third
5 minutes prone. The patient is required to be fully relaxed
(such that the heart rate is under the control of the autonomic
nervous system) for each consecutive 5 minute period and be
free from the effects of smoking, alcohol, caffeine or similar
pharmaceuticals. (Note that the TF4 System can be also used
to determine the effectiveness of Beta Blockerstreatment, e.g.
by beta-blockers). Of course no useful HRV data can be acquired
from a patient fitted with a pacemaker. In addition to determining
the spectral HRV parameters established by the Framingham Heart
Studies4-5,
further clinical insight is gained by graphical visualisation
of the changes in the spectral representation of the data through
the complete test as the patient changes position. (The autonomic
nervous system’s control of heart rate responds to both air
pressure and blood pressure receptors in the lungs and blood
pressure receptors in the arterial system).
Patterns of change in the spectral data have been identified
which indicate a range of clinical conditions, including, e.g.: |
| |
•
|
Risk of dangerous arrhythmia or even of Sudden Cardiac
Death (SCD) syndrome particularly in patients after acute
myocardial infarction. |
| |
|
•
|
Development of coronary artery disease including early
stages of heart failure |
| |
|
•
|
Congestive Heart Disease (CHD) |
| |
|
•
|
Cardiac Autonomic Neuropathy in diabetes (CAN), |
| |
|
•
|
Early Renal Failure Specific neurological conditions
like autonomic failure etc. |
| |
|
•
|
Multiple metabolic syndrome with overweight, hypertension
and/or hyperlipidemia |
|
| |
| In summary, the TF4 System enables a
practical fast and accurate ‘front line’ test of cardiac risk
and autonomic [cardiac?] dysfunction. It determines the most
powerful single risk factors for the prognoses of cardiac events
as established by the Framingham Heart Study4-5
using methods that meet the specifications produced by the European
Society for Cardiology ESC & NASPE Task Force1. Further,
results are displayed clearly, both numerically and graphically,
allowing straightforward interpretation of the results. Interpretation
is fully supported, when necessary, by an Internet data connection
to one of the UK’s leading cardiologists who will return a report
on the test data within two working days. |
| |
|
1. Malik M., [rest of committee from Appendix C which I do not
possess!] Heart Rate Variability: Standards of Measurement,
Physiological Interpretation and Clinical Use, a Special Report.
Task Force of the European Society for Cardiology and the North
American Society for Pacing and Electrophysiology. Circulation
Vol 93 No5 March1 1996, 93, 5: 1043-1065.
2. Berntson G.G., Bigger J.T., Eckberg D.L., Grossman P., Kaufmann
P.G., Malik M., Nagaraja H.N., Porges S.W., Saul J.P., Stone
P.H., Van Der Molen M.W.Committee Report. Heart Rate Variability:
Origins, Methods and Interpretative Caveats. Psychophysiology
84 1997, 84: pp 623-648.
3. Schwartz P.J. The Autonomic Nervous System and Sudden Death.
Eur Heart J., Vol 19, Suppl F 1998: F72-F80.
4. Hisako Tsuji H et al: ‘Reduced Heart Rate Variability and
Mortality Risk in an Elderly Cohort.’ Circulation 1994; 90:
878-883.
5. Hisako Tsuji H et al: ‘Impact of Reduced Heart Rate Variability
on Risk for Cardiac Events’ . The Framingham Heart Study. Circulation
1996; 94: 2850-56.,
6. Framingham heart studies, Anderson KM et al: ‘Cardiovascular
Disease Risk Profiles’. Framingham heart studies. Am Heart J
1990; 121: 293-8.
7. Anderson KM et al:‘An Updated Coronary Risk Profile, A Statement
for Health Professionals’. Circulation 1991; 83: 356-62.
8. Wolf PA et al: ‘Probability of a Stroke, A Risk Profile from
the Framingham Study’. Stroke 1991;22:312-8.
9.Dekker JM et al The Zutphen Study, ‘Heart Rate Variability
from Short Term Electrocardiographic Recordings Predicts Mortality
from all Causes in middle-aged and Elderly Men’. Am J Epidemiol
1997, 145, 10: 899-908.
10. Dekker C et al: Low heart rate variability in a 2-minute
rhythm strip predicts risk of coronary heart disease and mortality
from several causes. Circulation, 2000, 102: 1239-1244.
10. Yamamoto Y., Hughson R.L. Course Graining Spectral Analysis:-
New Method for Studying Heart Rate Variability.[rest of ref
please]. |
|
All material contained
within this website is copyrighted to Advanced Medical
Diagnostic Group Limited.
VariaCardio® and VariaPulse are registered trademarks
of AMDG Ltd. This product is protected by patents owned
by AMDG Ltd.
Manufactured for AMDG by MIE Medical Research Ltd (an
ISO13488 accredited company). AMDG and MIE Research Ltd
reserve the right to amend specifications without notice
in accordance to their policy of continual product improvement. |
| |
| © 2001-5 Advanced Medical Diagnostics Group Ltd. All Rights Reserved. |
| |
| AMDG Ltd,
6 Wortley Moor Road, Leeds, LS12 4JF
Tel: +44 (0) 113
279 1010 Fax: +44
(0) 113 231 0820
 |
| |
|
|
|
|
