Non-Stationarity of Microvolt T Wave Alternans in Long QT Syndrome Patients
Laura Burattini, Wojciech Zareba, Jean P. Couderc, Jo A. Konecki, Arthur J. Moss. University of Rochester, Rochester, NY
Background. Detection of microvolt T wave alternans (TWA) is a non-invasive method to identify pts at risk for sudden cardiac death. ECG tracings with visible TWA often show non-stationary pattern of this phenomenon. The purpose of the study was to evaluate stationarity of TWA in long QT syndrome (LQTS) pts, using our new correlation method (CM) for microvolt TWA detection.
Method and Results. Differently from accepted spectral method (SM), CM is able to identify TWA in as few as seven beats, and to detect which beats are alternating. In a group of 32 LQTS pts, 128-beat ECG recordings were performed to detect TWA using both CM and SM. TWA was identified by CM in 14 (44%) pts, and in 4 (13%) pts using SM. The features of TWA detected by CM in relation to the number of alternating beats (N) are shown in the following table (ACM=alternans correlation amplitude; NS_TWA=non-stationary TWA; SNS_TWA=strongly NS_TWA; S_TWA=stationary TWA).
|
|
SNS_TWA N<38 |
NS_TWA 38£ N£ 64 |
S_TWA N>64 |
p** |
|
# pts |
8 |
4 |
2 |
|
|
N |
20± 9 |
45± 10 |
78± 15 |
|
|
ACM (m V) |
83± 51 |
35± 14* |
44± 5* |
0.094 |
|
RR (ms) |
957± 203 |
1115± 55* |
1264± 22 |
0.061 |
* p<0.05 when comparing SNS_TWA vs. NS_TWA and S_TWA.
** Kruskall-Wallis Test
Significant correlations between ACM and RR (r=.70: p=.005) and between N and RR (r=-.57: p=.033) were observed.
Conclusions. 1)LQTS pts show non-stationary TWA more frequently than stationary TWA. Usually, 30-40 beats out of 128 were alternating. 2) Our correlative method was more effective than SM in non-stationary TWA detection. 3) Non-stationary TWA is associated with higher heart rate.