Download Clarkson University

Clarkson University
Determination of the Impact of T Wave Alternans on the Prediction of Sudden
Cardiac Death
A Thesis Proposal by
Vidoje Mihajlovik
Department of Electrical and Computer Engineering
March 2008
Signature
Professor Stephanie Schuckers
Date
Introduction
Sudden cardiac death (SCD) is a leading cause of death among Americans. According to
estimates of the American Heart Association over 300,000 adults in United States die
each year as a result of SCD, 95 percent of whom die before they reach a source of
emergency aid, such as defibrillation. (1) SCD commonly mistaken as a heart attack is
caused by two reasons: an arrhythmic disorder called ventricular fibrillation, which
makes the blood-pumping contraction and rhythm of the ventricles irregular, and
ventricular tachycardia which causes the ventricles to beat too fast affecting their ability
to pump enough blood which may lead to ventricular fibrillation from where death can
result in a few seconds.
There aren’t many effective measures to estimate the risk factors leading to a SCD. One
new approach, however, has suggested looking at the repolarization alternans (T-wave
alternans (TWA)) analyzed from implantable cardioverter defibrillators (ICD) or the
output of a digital electrocardiogram (EKG), as a predictor of ventricular
tachyarrhythmias. (2) (3) Obvious T-wave alternans from an output of an EKG are not
common and are difficult to be seen with a naked eye. (4) However, digital signal
processing techniques which can analyze data at the microvolt level, can analyze data
gathered from an implantable cardioverter defibrillator to detect subtle degree of T-wave
alternans, which may lead toward predicting ventricular tachyarrhythmias.
The T-wave alternans which show fluctuations in the heartbeat as a result might indicate
whether a person is at risk for sudden cardiac arrest. Previous research has shown that
this method is most effective to predict a SCD when the heart rate is elevated between
rates of 100bpm and 120bpm, and as a result TWA is usually measured during exercise
or any other activity which raises the heart rate. The implantable cardioverter defibrillator
is the primary means of stopping an SCD, or giving therapy, however it is still difficult to
clearly identify those patients who are to have a cardiac arrest. (5)(6) Furthermore the
defibrillators used in the current research at Clarkson do not incorporate any sort of TWA
algorithms or use TWA as a screening test for further electrophysiological (EPS)
examination and therapy. As a result the data produced by these defibrillators can be
analyzed to see if any TWA tests can be formed to better predict a cardiac death.
In addition, in the past TWA has been associated with several pathophysiologic
conditions such as myocardial ischemia, (7)(8)(9) the long QT syndrome, (10)(11)(12)
the Brugada syndrome, (13)(14) vasospastic angina, (15)(16) electrolyte abnormalities
(e.g. hypocalcemia, hypokalemia, hypomagnesemia), (17)(18) treatment with quinidine
or amiodarone, (19) hypertrophic cardiomyopathy, (20) alcoholic cardiomyopathy, (21)
and congestive heart failure, (22); furthermore T-wave alternans have also been reported
following cardiac resuscitation. In addition, the presence of TWA besides predicting the
above mentioned conditions and tachyarrhythmic events, such as sudden cardiac death,
sustained ventricular tachycardia, ventricular fibrillation, can also predict when
implantable cardioverter defibrillator (ICD) therapy for ventricular tachyarrhythmia, and
cardiac arrest is needed. (23)(24) The first microscopic TWA study with proposed
relationship between TWA and susceptibility to ventricular tachyarrhythmias was
reported in 1982, even though microscopic TWA were first reported in 1948, and various
studies have been done since done, either to prove a proposed relation or to establish new
ones. (25)(26)
Background
TWA is the variation in amplitude and width of the T wave which occurs on every other
beat, and was first described in 1908 in Munich, Germany by Hering who defined it as
the electrocardiographic expression of repolarization of the ventricular myocardium. (27)
It was further characterized by Sir Thomas Lewis in 1910 as occurring "either when the
heart muscle is normal but the heart rate is very fast or when there is serious heart disease
and the rate is normal." (28) The T wave is shown bellow as part of the full PQRST
cycle: (29)
Kalter and Schwartz later in 1948 identified the T-wave alternans on surface ECG. (25)
In their review they examined 5 patients which were identified with macroscopic TWA
with frequency of 0.08%. Among propositions that there might be a relation between the
TWA and SCD they showed that electrical alternans must be distinguished from
mechanical even though both may coexist if a relation is to be established.
The first large clinical study of TWA included 83 patients which were monitored on both
EPS and TWA data during atrial pacing which lasted 20 months, and was conduced by
Rosenbaum. (30) In this time period ventricular tachyarrhythmic events occurred in 81%
of patients with a significant level of TWA, compared with only 6% of those without
significance level of TWA.
Gold et al. also reported a study of TWA which included 313 patients who had data
measured during bicycle exercise which was repeated periodically for 297 days. (31)
Electrocardiography (ESP) data was also measured during the time of TWA data
collection. The predictive value of TWA and EPS for arrhythmic events was similar to
the above mentioned study, and better than the SAECG (signal-averaged
electrocardiography) which was also performed. The combination of TWA with ESP
appeared to enhance the predictive value for arrhythmic events, even though TWA
appeared to show better results than ESP on individual level.
TWA in patients with cardiomyopathy
Adachi et al. shows a study of 58 patients with dilated cardiomyopathy (DCM)
who underwent a TWA testing. (32) Analysis of ventricular tachyarrhythmias, nonsustained and sustained VT, showed that ventricular tachyarrhythmias were more
common in patients with a significant level of TWA where the predictive accuracy,
specificity, and sensitivity rates of TWA to predict ventricular tachyarrhythmias were
77%, 72%, and 88% correspondingly.
Klingenheben et al. using Multivariate Cox regression analysis also showed that TWA
was the only independent predictor of arrhythmic events. (33) There were 107 patients
studied in this study with congestive heart failure which had a mean LVEF of 28±7%,
and no history of sustained ventricular tachyarrhythmias. During 18 months of follow-up
there were no patients in the TWA negative group that experienced an arrhythmic event
or SCD.
Another study of 104 patients with DCM, and with 12 arrhythmic events which lasted 7
to 35 months depending on individual patients, demonstrated that TWA in a group of
patients with an onset heart rate less than 100 beats per minute was the most significant
predictor of arrhythmia survival where the sensitivity, specificity, PPV, NPV, and the
relative risk were 75%, 78.9%, 37.5%, 94.9%, and 7.4, correspondingly. (34)
Furthermore Momiyama et al. studied 14 patients with HCM. (35) A significant level of
TWA was found in 71% of 7 patients which were at high risk of ventricular
tachyarrhythmias, compared with none of the other 7 patients who were at low risk of
ventricular tachyarrhythmias. The result suggests that TWA may be a useful indicator for
high risk of ventricular tachyarrhythmias in patients with HCM. However, this finding
was based on a small number of patients. The position of TWA for prediction in patients
with HCM needs to be further researched.
TWA in patients with prior myocardial infarction
There are only a few studies relating TWA and the prediction of myocardial infraction
(MI) besides several SAECG and LVEF studies. TWA was measured in 102 patients with
recent myocardial infraction, usually around 14 to 26 days after the occurrence of the
myocardial infraction. (36) TWA showed relative risk, highest sensitivity, negative
predictive value, but the lowest specificity and accuracy as compared to SAECG and
LVEF. However, a combination of TWA and SAECG showed as the most significant
predictor and much more accurate then SAECG alone. Which might also indicate that
sometimes combination of TWA with other methods might lead to better indicator then
the two methods alone.
Another study also testing TWA examined 836 patients 2 months after the occurrence of
the myocardial infraction relating TWA to SCD or ventricular fibrillation. (37) Since few
studies show TWA to be a predictive factor in SCD, using TWA for prediction of MI
needs to be further researched.
TWA in patients with the long QT and Brugada syndromes
TWA as a predictor for the long QT syndrome hasn’t been established yet, even though
TWA has been reported in patients with the long QT syndrome.(10)(11)(12)(13) It is
postulated that the prolongation and unstable state of the ventricular action could produce
the reported TWA and result in a polymorphic ventricular tachyarrhythmias state known
as torsade de pointers.
Inconclusive reports about the predictive value of TWA as related to the Brugada
syndrome have also been made. (13)(14) Some reports have shown that administration of
class IC anti-arrhythmic drugs stimulated TWA which resulted in ventricular fibrillation.
It has been proposed that these drugs produce an unstable state of repolarization and as a
result create TWA and induce ventricular fibrillation. Ikeda et al. on the other hand
reported a low predictive value of TWA as related to the Brugada syndrome. (38) More
research is needed in both the long QT and Brugada syndromes to establish a predictive
TWA relation.
Reason of existence and linkage of TWA
There are several reasons that explain the presence and linkage of TWA. There is
evidence that TWA is linked to alternations in cellular calcium homeostasis, which
influences the action potential duration, which in a failing heart leads to higher risk of
SCD. (39)(40) There is other research that suggests that potassium channels might play a
role in ischemia induced TWA. (41)(42)(43) It further explains that the sensitivity of
KATP channel activation during ischemia between endocardium and epicardium may be
linked to TWA at the cellular level
The principle behind TWA is based on the notion that increased dispersion of
repolarization and heterogeneous prolongation produce tachyarrhythmias. (44) The
dispersion of repolarization results in a 2:1 ration on the surface ECG and as a result
separates the wave front and cause reentry which is also created by the conduction area in
the prolongation and repolarization states.
Shimizu et al. did an experiment with QT syndrome models employing an arterially
wedge of canine on the left ventricular wall. (45) When paced at a critical fast rate, it was
found that distinct alternation of APD of mid-myocardial cells, results in a reversal of the
transmural repolarization creating TWA in the ECG .
Pastore et al. also investigated TWA in Langendorff-perfused guinea pig heart using
epicardial APD during pacing. (46) The critical pacing rate created harmonious TWA,
which further developed to conflicting alternans of APD and increased vulnerability to
ventricular tachyarrhythmias.
Measurement of TWA
TWA is measured by signal processing techniques with high-resolution electrodes as to
reduce noise. About 128 beats are sampled and a time series of amplitudes of multiple
corresponding points on the T wave are analyzed using a Fast Fourier Transform in order
to generate a power spectrum (see figure bellow). TWA measurements are usually
performed with atrial pacing to bring the heart to the desired target zone. (47) Either
bicycle or treadmill exercise can be used with high resolution electrodes and noise
reduction algorithms to perform the measurement. (48) The peaks on the picture bellow
correspond to respiratory variation, pedaling (or other exercise if bicycling is not
performed), and noise. The alternans are usually represented by a frequency peak at 0.5
cycles per beat. (49) By doing this analysis two measurements are generated: the
alternans ration and the magnitude. The alternans ratio represents the magnitude of the
alternation variation in the T wave structure as compared to the mean T wave magnitude
where a 1.9 µV threshold is used for significance. (50) The alternans ratio is further more
a statistical measure of the alternans with respect to the standard deviation of the
background noise, usually it should be greater than 3 so any significance can be attached.
In addition, in order to measure TWA, high heart rate has to be sustained for more than 1
minute. (50)
In order for TWA to be measured properly the heart rate should be high, but, there are
several things that must be kept in mind. Since TWA is rate dependent as a result it can
develop in normal conditions as large enough heart rate. (51) However, it has been shown
that the onset heart rate is low in patients with structural heart disease and history of
sustained ventricular tachyarrhythmia. (51) Furthermore Kavesh et al. showed that TWA
and false positive results increase with heart rate. Therefore, an onset heart rate of less
than 110 beats per minute is a usual requisite for positivity. (51)
.
T wave alternans measurement: spectral method.
Limitations
There are limitations to TWA both technical and electrophysiologic. The limitations are
that TWA cannot be measured in those patients with atrial fibrillation, a common
arrhythmia in patients with structural heart disease; and if there is a presence of atrial or
ventricular ectopy, excessive motion artifacts or the inability to achieve the target rate of
100pbm to 120pbm TWA would also not be able to be applied to give specific results.
Furthermore intermediate results are in about 25% of the published papers, which shows
that it needs further research. It has also been shown that TWA looses its prediction
capability within a month after the onset of a MI.
Current and previous research at CU
TWA has not been measured or modeled on data at Clarkson University. QT
measurement is the current research that analyzes the implantable cardioverter
defibrillator data from 50 subjects. The data available is from a baseline recording of the
side to side ECG data channel (lead V5), the front to back BCG data channel (anterior /
posterior), and from treated VT/VF recording of the side to side ECG data channel (lead
V5), and the front to back BCG data channel (anterior / posterior). The current research
focuses on QT analysis by performing manual measurement of QT interval using
computer software where the V5 channel is used as the standard for QT measurement.
The subjects chosen to be analyzed are those with both baseline and normal ECG
recorded in the onset of VT/VF, where a QTc factor is used for correction of the heart
rate.
The QT is measured from the beginning of the earliest onset of the QRS complex to the
end of the T wave. The R-R interval which is also measured is from the R peak to the
next consecutive R peak. The QT has been computed for both baseline and treated
recordings to provide comparisons between both recordings. The picture bellow
illustrates the various wave segments and the QTc factor used to normalize the heart rate:
There are two ways to measure the QT segment: manual and automated. The manual
method is done with a computer mouse measure of the length of the QT interval as shown
in the picture bellow:
The measurement is done both to the baseline recording from the V5 lead, and before the
onset of a VT/VF, and both are sampled at 100 Hz, where each block on the picture
above represents 10msec;. A sample measurement is illustrated bellow:
Baseline Recording:
Prior to Onset of VT/VF
QT=51*10=510 ms
RR=82*10=0.820 s
QTc=510/sqrt(0.82)= 563.20
QT=72*10=720 ms
RR=92*10=0.920 s
QTc=720/sqrt(0.92)=750.65
The automated method is performed by a Matlab algorithm, which finds QRS peaks
through peak-slope detection. The T and P peaks are detected between the S peak of the
current peak and Q-peak of the next beat by application of thresholds. The end of the T
peak is detected by location of the minimum value between T peak and P peak. The peak
results from the automated method were plotted versus the baseline to check for accuracy
for both the RR and QT interval:
The manual recordings plotted versus the automated recordings to check the accuracy of
the automated method for the QT segment are shown bellow.
My Research Objective
My research objective will be try to find some indication of TWA in the implantable
cardioverter defibrillator data at Clarkson University following the procedures outlined in
the above mentioned papers, and try to create a screening test either alone with TWA or
in combination with the QT analysis that would indicate potential risks of VT/VF.
Research Methodology
Task 1. Visual assessment of features compared to manually measured T waves
I will be using a written MATLAB program by graduate student Katherine Bellor to
extract the width and height of the wave and then to compare these measurements with
the baseline or with the measurement before the onset of a VT/VF to check for any
obvious discrepancy. I will also be using other programs available to manually calculate
the width and length of the T wave. Once I obtain the measurements from before the
onset of a VT/VF I will compare them to the base measurements with statistical methods
and graphs to try to find a pattern so a screening test can be developed.
Task 2. Development of screening tests
Once I have performed statistical and differential analysis on the data I will try to create a
model based on the ICD data to try to predict VT/VF. Then I will try to test the model on
other ICD data to see if the same results occur.
Task 3. Integration of TWA with QTc
The third task would be to combine the results of the TWA test with the QTc analysis.
This will be done in parallel with the graduate student currently working on the QTc
analysis to create one general model that could better predict a VT/VF.
Realistic Timetable
Date
April 30, 2008
May 2008-August 2008
September 2008 – December 2008
Description
Thesis Progress Report
Summer Research at Clarkson
 Task 1. Visual assessment of
features compared to manually
measured T waves.
 Task 2. Development of screening
tests
Task 3. Continue developing screening test,
work in cooperation with QT research to
form combined model.
December 2008
January 2009
Start Abstract
Finish Final Draft of Abstract
Final Thesis, work on presentation
References:
(1) American Health Assistance Foundation: About Heart Disease and Stroke: Treatment,
Risk Factors, Symptoms, News, Research, Resources. 2003, May 1
(1) AHA. American Heart Association: Heart and Stroke Statistical Update. Dallas, TX:
2001.
(2) Armoundas AA, Cohen RJ. Clinical utility of T-wave alternans. Card Electrophysiol
Rev 1997;1:390-4.
(3) Rosenbaum DS, Albrecht P, Cohen RJ. Prediction sudden cardiac death from T-wave
alternans of the surface electrocardiogram: promise and pitfalls. J Cardiovasc
Electrophysiol 1996;7:1095-111.
(4) Habbab MA, El-Sherif N. TU alternans, long QTU, and torsade de pointes: Clinical
and experimental observations. PACE 1992;15:916-31.
(5) Moss AJ, Hall WJ, Cannom DS, et al.: Improved survival with an implanted
defibrillator in patients with coronary disease at high risk for ventricular arrhythmia.
Multicenter Automatic Defibrillator Implantation trial investigators. N Engl J Med
1996;335:1933-40.
(6) Prystowsky EN. Screening and therapy for patients with nonsustained ventricular
tachycardia. Am J Cardiol 2000;86:K34-9.
(7) Nearing BD, Huang AH, Verrier RL. Dynamic tracking of cardiac vulnerability by
complex demodulation of the T wave. Science 1991;252:437-40.
(8) Konta T, Ikeda K, Yamaki M, et al: Significance of discordant ST alternans in
ventricular fibrillation. Circulation 1990;82:2175-9.
(9) Puletti M, Curione M, Righetti G, et al.: Alternans of the ST segment and T wave in
acute myocardial infarction. J Electrocardiol 1980;13:297-300.
(10) Schwartz P, Malliani A. Electrical alternation of the T-wave: clinical and
experimental evidence of its relationship with the sympathetic nervous system and with
the long QT syndrome. Am Heart J 1975;89:45-50.
(11) Zareba W, Moss AJ, Le Cessie S, et al: T wave alternans in idiopathic long QT
syndrome. J Am Coll Cardiol 1994;23:1541-6.
(12) Platt SB, Vijgen JM, Albrecht P, et al: Occult T-wave alternans in long QT
syndrome. J Cardiovasc Electrophysiol 1996;7:144-8.
(13) Takagi M, Doi A, Takeuchi K, et al: Pilsicanide-induced marked T wave alternans
and ventricular fibrillation in a patient with Brugada syndrome. J Cardiovasc
Electrophysiol 2002;13:837.
(14) Chinushi M, Washizuka T, Okumura H, et al.: Intravenous administration of class Ic
antiarrhythmic drugs induced T wave alternans in a patient with Brugada syndrome. J
Cardiovasc Electrophysiol 2001;12:493-5.
(15) Kleinfeld MJ, Rozanski JJ. Alternans of the ST segment in Prinzmetal's angina.
Circulation 1977;55:574-7.
(16) Cheng TC. Electrical alternans. An association with coronary artery spasm. Arch
Intern Med 1983;143:1052-3.
(17) Reddy CV, Kiok JP, Khan RG, et al.: Repolarization alternans associated with
alcoholism and hypomagnesemia. Am J Cardiol 1984;53:390-1.
(18) Shimoni Z, Flatau E, Schiller D, et al: Electrical alternans of giant U waves with
multiple electrolyte deficits. Am J Cardiol 1984;54:920-1.
(19) Bardaji A, Vidal F, Richart C: T wave alternans associated with amiodarone. J
Electrocardiol 1993 Apr; 26(2): 155-7.
(20) Momiyama Y, Hartikainen J, Nagayoshi H, et al: Exercise-induced T-wave alternans
as a marker of high risk in patients with hypertrophic cardiomyopathy. Jpn Circ J
1997;61:650-6.
(21) Luomanmaki K, Heikkila J, Hartikainen M: T-wave alternans associated with heart
failure and hypomagnesemia in alcoholic cardiomyopathy. Eur J Cardiol 1975 Oct; 3(3):
167-70
(22) Klingenheben T, Zabel M, D'Agostino RB, et al.: Predictive value of T wave
alternans for arrhythmic events in patients with congestive heart failure. Lancet
2000;356:651-2.
(23) Adam DR, Smith JM, Akselrod S, et al: Fluctuations in T-wave morphology and
susceptibility to ventricular fibrillation. J Electrocardiol 1984;17:209-18.
(24) Smith JM, Clancy EA, Valeri CR, et al: Electrical alternans and cardiac electrical
instability. Circulation 1988;77:110-21.
(25) Kalter HH, Schwartz ML. Electrical alternans. N Y State J Med 1948;1:1164-6.
(26) Adam DR, Powell AO, Gordon H, et al: Ventricular fibrillation and fluctuations in
the magnitude of the reporalization vector. IEEE Comput Cardiol 1982;241-4.
(27) Hering HE. Das Wesen des Herzalternans. Munchen Med Wochenshr 1908;4:141721.
(28) Hearing H: Experimentelle Studien an Saugentieren uber das Electrocardiogram. Z
Exper Med 1909; 7: 363.
(29) http://www.vetgo.com/cardio/concepts/concsect.php?conceptkey=166
(30) Rosenbaum D, Jackson LE, Smith JM, et al.: Electrical alternans and vulnerability to
ventricular arrhythmias. N Engl J Med 1994;330:235-41.
(31) Gold MR, Bloomfield DM, Anderson KP, et al.: A comparison of T wave alternans,
signal averaged electrocardiography and programmed ventricular stimulation for
arrhythmia risk stratification. J Am Coll Cardiol 2000;36:2247-53.
(32) Adachi K, Ohnishi Y, Shima T, et al.: Determinant of microvolt-level T wave
alternans in patients with dilated cardiomyopathy. J Am Coll Cardiol 1999;34:374-80.
(33) Klingenheben T, Zabel M, D'Agostino RB, et al.: Predictive value of T wave
alternans for arrhythmic events in patients with congestive heart failure. Lancet
2000;356:651-2.
(34) Kitamura H, Ohnishi Y, Okajima K, et al.: Onset heart rate of microvolt-level T
wave alternans provides clinical and prognostic value in nonischemic dilated
cardiomyopathy. J Am Coll Cardiol 2002;39:295-300.
(35) Momiyama Y, Hartikainen J, Nagayoshi H, et al: Exercise-induced T-wave
alternans as a marker of high risk in patients with hypertrophic cardiomyopathy. Jpn Circ
J 1997;61:650-6.
(36) Ikeda T, Sakata T, Takami M, et al.: Combined assessment of T wave alternans and
late potentials used to predict arrhythmic events after myocardial infarction. A
prospective study. J Am Coll Cardiol 2000;35:722-30.
(37) Ikeda T, Saito H, Tanno K, et al.: T wave alternans as a predictor for sudden cardiac
death after nyocardial infarction. Am J Cardiol 2002;89:79-82.
(38) Ikeda T, Sakurada H, Sakabe K, et al.: Assessment of noninvasive markers in
identifying patients at risk in the Brugada syndrome: insight into risk stratification. J Am
Coll Cardiol 2001;37:1628-34.
(39) Choi BR, Salama G. Simultaneous maps of optical action potentials and calcium
transients in guinea-pig hearts: mechanisms underlying concordant alternans. J Physiol
2000;529:171-88.
(40) Cohn JN, Archibald DG, Ziesche S, et al: Effect of vasodilator therapy on mortality
in chronic congestive heart failure. Results of a Veterans Administraction Cooperative
Study. N Engl J Med 1986;314:1547-52.
(41) Miyoshi S, Miyazaki T, Moritani K, et al: Different responses of epicardium and
endocardium to KATP channel modulators during regional ischemia. Am J Physiol
1996;271:H140-7.
(42) Kleber AG, Janse MJ, van Capelle FJ, et al: Mechanism and time course of ST and
TQ segment changes during acute regional myocardial ischemia in the pig heart
determined by extracellular and intracellular recordings. Circ Res 1978;42:603-13.
(43) Dilly SG, Lab MJ. Electrophysiological alternans and restitution during acute
regional ischemia in myocardium of anaesthetized pig. J Physiol 1988;402:315-33.
(44) Kuo CS, Amlie JP, Munakata K, et al. Dispersion of monophasic action potential
durations and activation times during atrial pacing, ventricular pacing, and ventricular
premature stimulation in canine ventricles. Cardiovasc Res 1983;17:152-61.
(45) Shimizu W, Anzelevitch C. Cellular and ionic basis for T-wave alternans under long
QT conditions. Circulation 1999;99:1499-507.
(46) Pastore JM, Girouard SD, Laurita KR, et al.: Mechanism linking T-wave alternans
to the genesis of cardiac fibrillation. Circulation 1999;99:1385-94.
(47) Rosenbaum D, Jackson LE, Smith JM, et al.: Electrical alternans and vulnerability to
ventricular arrhythmias. N Engl J Med 1994;330:235-41.
(48) Rosenbaum DS, Albrecht P, Cohen RJ. Prediction sudden cardiac death from Twave alternans of the surface electrocardiogram: promise and pitfalls. J Cardiovasc
Electrophysiol 1996;7:1095-111.
(49) Cohen RJ: In Zipes DJ, Jalife J (eds.): Cardiac Electrophysiology: from cell to
bedside. 3rd edition. Philadelphia, WB Saunders, 2000.
(50) Bloomfield DM, Cohen RJ. Repolarization alternans. In: Malik M, editor. Risk of
Arrhythmia and Sudden Death. London: BNJ Books, 2001,256-65.
(51) Tanno K, Kobayashi Y, Adachi T, et al.: Onset heart rate on microvolt T wave
alternans during atrial pacing. Am J Cardiol 2000;86:877-80.
(52) Kavesh NG, Shorofsky SR, Sarang SE, et al.: Effect of heart rate on T wave
alternans. J Cardiovasc Electrophysiol 1997;8:987-93.