T-wave alternans test's prognostic horizon exceeds two years=
in
analysis
July 22, 2008 | Steve Stil=
es (The Heart.org)
![](3D"Chan_file/image001.gif")
Kansas City, MO and Cincinnati, OH =
- T=
he power
of the microvolt T-wave alternans (mTWA) test in stratifying patients with ischemic cardiomyopathy for risk of sudden cardiac death as pa=
rt of
the screening of candidates for an implantable cardiov=
erter-defibrillator
(ICD) remains significant over a horizon of up to three years, according to=
an
analysis from one of the technique's most experienced groups [1].
The finding sheds light on a persisting
question about the mTWA test: how often should =
it be
performed in patients with a negative result but who remain eligible for
"Although it makes intuitive sense th=
at
the predictive value of a screening test in such a high-risk population wan=
es
over time, our findings suggest that mortality and arrhythmic risk with a n=
onnegative
mTWA test result remain robust through the firs=
t two
to three years," write the authors in the August 1, 2008 issue of the =
American
Journal of Cardiology.
By convention, nonnegative mTWA
tests include those that either show mTWA or are
indeterminate.
"At least in this cohort, there didn't
seem to be any huge fall-off in the test's discriminatory power at a follow=
-up
of two to three years," lead author Dr Paul S Chan (St Luke's M=
id
America Heart Institute,
The post hoc analysis looked at a cohort in
which a baseline mTWA test had previously been =
shown
to stratify sudden-death risk over an average of 18 months [2], as reported by heartwire
at the time. It included 768 patients with ischemic heart disease and an LV=
EF <35%
but no history of sustained ventricular arrhythmias.
About 60% of the patients were implanted w=
ith ICDs; the analysis controlled for that as well as
demographics, LVEF, ECG parameters, medications, and c=
omorbidities.
A nonnegative baseline mTWA
test was associated with more than double the adjusted cumulative risk of
all-cause death or appropriate ICD discharge six months later and at interv=
als
of six months out to three years.
Hazard ratio for all-cause mortality and appropriate ICD shock=
s by
six-month intervals after a nonnegative mTWA te=
st
|
Months follow-up |
HR (95% CI) |
p <= /span> |
|
6 <= /span> |
2.49 (0.93-6.66) |
0.07 |
|
12 = |
2.21 (1.11-4.40) |
0.02 |
|
18 = |
2.22 (1.24-3.98) |
0.008 |
|
24 = |
2.60 (1.48-4.56) |
<0.001 |
|
30 = |
2.43 (1.45-4.07) |
<0.001 |
|
36 = |
2.37 (1.47-3.84) |
<0.001 |
With the test showing no apparent decrease=
in
predictive power over three years, the group looked at risk at intervals of
one, two, and three years, with hazard ratios reflecting noncumulative
risk during each specified 12-month period. The risk of all-cause death or
appropriate shock was doubled in the first year, more than tripled in the
second, and trended higher in the third year.
"We didn't expect it to have signific=
ant
statistical power year by year," Chan said. A more important observati=
on,
he emphasized, is that statistical tests for interaction among the year
intervals showed no significant variations, suggesting there is little
meaningful change in the test's prognostic power at one, two, and three yea=
rs.
Hazard ratio for all-cause mortality and appropriate ICD shock=
s by
year intervals after a nonnegative mTWA test
|
Follow-up year |
HR (95% CI) |
p <= /span> |
|
0-12 mo |
2.19 (1.10-4.34) |
0.03 |
|
>12 to 24 mo |
3.36 (1.28-8.83) |
0.01 |
|
>24 to 36 mo |
2.06 (0.81-5.22) |
0.13 |
To download table=
s as
slides, click on slide logo above
With the analysis confined to a specific
patient cohort, Chan observed, questions remain about the mTWA's
predictive power in different populations. He pointed, in particular, to th=
e MASTER
1 trial, which—as previously covered by h=
eartwire—questioned
the test as a risk stratifier in a population of
patients with ischemic heart disease and an LVEF <30%, all of whom had b=
een
implanted with ICDs.
|
The study was partly funded by Me= dtronic. |