T-Wave Alternans in Patients With Suspected Acute Coronary Syndrome

IRB Application

III: Title: T-Wave Alternans for Acute Coronary Syndrome Patients Objective: Chest Pain is
a common and high-risk presenting symptom in the Emergency Department. One of the central
challenges Emergency Physicians face is differentiating cardiac causes of chest pain from
non-cardiac causes. A rich array of mature technologies exists to aid in this process.
None, however, has shown adequate sensitivity within the first 12-18 hours of symptoms to
reliably exclude cardiac ischemia. As a result many patients are admitted to Coronary Care
Units for a period of observation which typically involves serial blood analysis and
continuous cardiac monitoring. An estimated $13 billion dollars per year are devoted to
this "Rule-out MI" process. The majority, typically 80%, of these patients are ultimately
found not to have cardiac disease. That is, their chest pain is ascribed to
gastro-esophageal reflux, or esophageal or chest wall muscle spasm as an explanation for
their symptoms.

While no test can reliably discriminate cardiac from non-cardiac symptoms in the emergency
department, a test that improves our ability to identify a subset of patients who are safe
to discharge would be of immense utility in reducing unnecessary hospital admissions.

Background: In the past 15 years T-Wave Alternans (TWA) testing has become accepted among
cardiac electrophysiologists as a means of identifying patients at high risk for ventricular
arrhythmias. TWA testing is typically performed in conjunction with a cardiac stress test
and involves computerized analysis of the EKG for subtle alternate beat variations in the
amplitude of the T-wave segment. The test is completely non-invasive, can return results
immediately, and can be performed with compact, portable equipment. An extensive body of
literature has emerged demonstrating the test's very high negative predictive value for
various groups of patients with regard to ventricular tachyarrhythmias and sudden cardiac
death. For example, Rosenbaum studied patients referred for electrophysiological testing
due to high clinical suspicion of risk of sudden cardiac death. His landmark paper
(Electrical Alternans and vulnerability to ventricular Arrhythmias, N Engl J Med
1994;330:235-41) found that a positive TWA test identified a subset of patients in this
group with 81% mortality over 20 months versus 6% for those in whom the test was negative.
The predictive value of TWA testing for cardiac arrhythmia in this group was equivalent to
invasive electrophysiological testing. Numerous subsequent studies have corroborated these
findings in specific patient populations such as Post-MI patients, Congestive Heart Failure
patients, and patients with Syncope. TWA testing was approved (Nov 2001) for Medicare
reimbursement at approximately $260 per test. The "HearTwave System" from Cambridge Heart
Inc (Boston MA) gained FDA approval in Apr 2001 for testing "any patient at risk for
ventricular arrhythmia." The test has been validated in the context of traditional cardiac
stress testing but there is not much clinical literature on the utility of the test in other
contexts. Specifically, there are no reports of the applicability of TWA testing to
Emergency Department patients with acute chest pain.

Verrier et al. (J Amer Coll Cardiol 2001; 37:1719-25) found in an animal study that
emotional stress had a profound multiplicative effect on the magnitude of TWA caused by
cardiac ischemia. This research supported the idea that the emotional stress of presenting
to the Emergency Department with acute chest pain or shortness of breath might be an
effective substitute for the stress that provocative cardiac stress tests (treadmill or
pharmacologic stress testing) induce. If true, then it would be of great interest to
Emergency Physicians to measure the predictive value of TWA for cardiac ischemia in patients
who are "stressed" by acute chest pain.

In 2003, we performed a preliminary study in the Loma Linda University ED to test this
research question. We monitored 100 acute chest pain ED patients using the HearTwave TWA
monitor. We determined that TWA in this setting was relatively common, and that the
amplitude of TWA seemed to correlate highly with the presence of cardiac ischemia. The
results of this work were presented at several local and one national meeting of academic
Emergency Physicians and were received with enthusiasm. A problem with the results was
related to a limitation of the HearTwave device. The HearTwave is designed for use in
conjunction with an exercise stress test and requires that heart rate increases over a
threshold (105 beats per minute) as the test proceeds. Because our recordings were done
"passively" in patients at rest, the data was inappropriate for the proprietary algorithm
used by the HearTwave and so our analysis required a different processing technique that was
qualitative rather than quantitative. Even so, we were able to demonstrate that TWA
activity in patients with heart rates above a much lower threshold (60 beats/minute) was
100% sensitive for cardiac ischemia at a level of specificity of approximately 50%.

In the present proposal, we seek IRB approval for a protocol very similar to what we used
before. The main difference will be that we will use a device that gives us access to the
raw data and enable us to apply our own numerical processing algorithms to measure TWA.

The study will look at TWA testing in an observational study of approximately 500 emergency
department patients presenting with symptoms suggestive of cardiac ischemia. The study will
be conducted at Loma Linda University Medical Center (LLUMC) an academic institution with
over 44,000 Emergency Department visits per year. All emergency Department patients age 18
or older who are being evaluated for possible Acute Coronary Syndrome will be eligible.
Actual testing will be performed by medical students who will have completed a 1-hour
training course in the use of the electrocardiographic instrument.

Methods: The device to be employed in the test is a an FDA approved ECG probe that enables
capture of the filtered ECG signal to digital storage. The testing procedure is essentially
identical to standard ECG recording. Ten electrodes are applied to the patient's chest in a
configuration that does not interfere with standard lead placement. The patient is
connected via electrical leads to the device and the operator presses a button to initiate
the measurement cycle. The complete test process typically takes 5-15 minutes. No action
is required by the patient or anyone else after the test is initiated. Results are stored
automatically for later review. The results of the test will not be made available to the
treating physicians and so will not influence treatment decisions. Performing the test will
not interfere in any way with delivery of standard acute care to the patient. Testing is
non-invasive - no stimulating current is delivered to the patient. The instrument is
designed for use with standard electrodes.

Following testing, patients' data will be processed in several ways. There are two standard
algorithms described in this literature to measure TWA - the "Spectral Analysis" method, and
the "Modified Moving Average" method. In addition to these two methods, we will use Neural
Network Techniques to further analyze the data. These methods allow quantitative
determination of the degree of TWA present in the data. We will examine the predictive
value of TWA testing for various measures of cardiac morbidity based on discharge diagnosis,
results of further cardiac testing, presence of ventricular arrhythmia, myocardial
infarction, need for revascularization. Follow-up will be made by telephone to the patient
or their family at 3 months post testing to assess for the occurrence of death, myocardial
infarction, or revascularization.

The central question guiding our research is to what extent can TWA testing aid in the
process of early risk stratification of the acute chest pain patient. We are particularly
interested in the negative predictive value of the test in clinically low-risk patients.

Risks: Minimal risk to the patient. Benefits: Individual participants will not benefit
directly from their participation. If TWA testing shows utility in identifying patients at
very low risk for ischemic complications then the results would be of immense value in
guiding the appropriate use of clinical resources in future patients in this context.

Compensation: Patients will not be compensated for participation. Consent: An investigator
will discuss the risks, benefits and objectives of the study. Written informed consent will
be obtained prior to participation on all patients in the study. Because the logistics of
the study are substantially the same as in our preliminary IRB approved study, we will use a
consent form that is substantially equivalent as that used in our preliminary study.
Subjects will be a given a written copy of the T-Wave Alternans in Acute Chest Pain informed
consent sheet.

Confidentiality: TWA test results will be used for research purposes only and will remain
confidential.

Inclusion Criteria:

- Adults with Acute Chest Pain

Exclusion Criteria:

- Patients with underlying myoclonic tremors making measurement artifact unreliable

- Less than 18 years of age

- Skin disorders that do not allow the Cardiac Monitor Leads to adhere to the skin