|
Presentation Start/End Time:
|
Tuesday, Apr 01, 2008, 10:00 AM -11:00 AM
|
|
Author Block:
|
Phyllis K. Stein, Peter P. Domitrovich, Devang
Sanghavi, Prakash Deedwania, Washington University School of Medicine,
St. Louis, MO
|
|
Background:
Microvolt T-wave alternans (TWA) measured from treadmill testing and using
spectral analysis techniques identifies patients at increased risk of sudden
death (SCD). TWA can be measured from 24-hour Holter recordings using a time
domain method called modified moving average (MMA). The relationship of
Holter TWA to SCD is unknown.
Methods:
The design and results of the EPHESUS study have been published previously.
EPHESUS enrolled hospitalized post-MI patients with heart failure and/or
diabetes with left ventricular dysfunction. Prior to randomization to drug
treatment, Holter recordings were obtained from 493 patients. Of these
patients, 48 died, 18 suddenly and the rest of cardiovascular causes
(nonSCD). A case control study was performed using 92 controls, alive at the
end of follow up and matched with non-survivors based on age, gender and
having diabetes. MMA TWA analysis was performed using a MARS system (GE
Healthcare, Milwaukee, WI) running software version 7.1.1. The software
provides a histogram of TWA values and the ECG strips associated with them.
The three highest values for TWA from artifact-free periods were identified
and averaged (AVMAXTWA) for each channel. The cutpoint that maximized the
separation between those who had SCD and those who did not was determined
using Cox regression analyses.
Results:
AVMAXTWA on either channel was higher for those who had SCD but not different
between nonSCD and survivors. For channel 1, AVMATWA was 38±12 µV for SCD vs.
29±11 µV for survivors and 26±8 µV for nonSCD (p=0.002). For channel 2,
AVMATWA was 45±19 µV for SCD vs. 33±19 µV for survivors and 33±13 µV for
nonSCD (p=0.05). The best cutpoint for prediction of SCD from AVMAXTWA on
channel 1 was 43 µV. AVMAXTWA above this cutpoint identified 14 patients with
a nearly six fold risk of sudden death of [RR=5.9 (95% CI= 2.2-15.8,
p<0.001)]. Seven of these 14 died, all but one suddenly. The channel 2
cutpoint was 47 µV with a relative risk of sudden death among the 19 patients
above this cutpoint of 5.5 (95% CI=2.2-13.8, p<0.001). Eight died
suddenly.
Conclusions:
Holter-based TWA measured using MMA appears to be a powerful predictor of SCD
in high-risk post-MI patients
|