Comparison of ECG parameters using GE GETEMED continuous ECG Holter and MAC-1200 bedside ECG

[Speaker] Jorg Taubel:1,2
[Co-author] Dilshat Djumanov:1
1:Richmond Pharmacology, St George's University of London, UK, 2:St George's University of London, UK

Bedside ECG can be replaced by continuous Holter recordings to obtain ECGs at higher frequency time-points. Bedside ECG QT measurement algorithms are more accurate than Holter, although processing the Holter data with the same algorithm as used for bedside ECGs overcomes this limitation.

Two types of ECGs were performed during a randomised double blind 2-period crossover study: continuous ECGs and bedside 10s ECGs at 12 time-points. Abnormalities may be missed when using bedside ECGs due to limits to the number of ECGs that can be recorded on a dosing day and longer intervals between later recordings. Continuous Holter ECGs were recorded to cover these gaps, and HR and QTcF values extracted by averaging 3 consecutive beats. The following filtering methods to remove outlier values were compared:
- The median value was compared between 1 to 10 min intervals using one minute steps.
- Each value was compared to a moving average of 22 adjacent values and when the value differed by more than 5% from the average it was considered implausible and excluded from analysis.
The moving average was used in further analyses. Values of QTcF from Holter and adjudicated bedside ECG recordings were compared, using high frequency scatter plots.

The visual review of 308 graphs showed a good correlation, with approximately 10 ms difference in most cases. The variance could be attributed to using different algorithms for QT measurement which can be resolved by using 12SL algorithm (used in MAC-1200 devices) on Holter data, as it proved to be the most accurate for measuring the QT interval.
Furthermore, manual adjudication showed that sudden changes in Holter QTcF and implausible values were either due to lead disconnection or from movement artefacts. This could not have been detected by the bedside ECGs. Holter outputs also showed well defined meal-induced QTcF reductions between 6 and 12 hours that were not detected by bedside ECGs.

We advise that subjects should remain supine at required time points to avoid these artefacts. More frequent measurements provide a more precise daily time course of QTcF, since the recording is continuous and minimal intervention is required.
Advanced Search