The accuracy of the EASI-derived spatial QRS-T angle

There has been recent interest in whether the spatial QRS-T angle (SA) can be used in Thorough QT studies to serve as a marker for increased risk of torsades de pointes. The determination of the SA requires vectorcardiographic data. Such data is however seldom recorded in monitoring applications. Specifically the number and the location of the electrodes, that are required when recording the Frank VCG, complicate the recording of vectorcardiographic data in monitoring applications. An alternative and more practical way for obtaining vectorcardiographic data in monitoring applications is the utilization of the EASI lead system. A previously published set of linear lead transformations allows for the derivation of the Frank VCG from the EASI lead system. This EASI-derived VCG can be used for the determination of an EASI-derived SA (ESA). The accuracy of the ESA has, however, not been reported in the literature. The aim of this research was the quantification of the differences between the ESA and the SA. This was achieved using electrocardiographic data recorded from 220 healthy subjects. To this end, the difference (ESA-SA) between the ESA and the SA was calculated for all 220 subjects. This difference was subsequently analyzed in order to determine the systematic error (mean difference) and the random error (span of the Bland- Altman 95% limits of agreement) that is made when determining the ESA. The systematic error between the SA and the ESA was found to be 11.6° [95% confidence interval: 9.8°; 13.40°]. The random error was found to be 52.9° [95% confidence interval: 48.44°; 58.45°]. The findings of this research suggest that both systematic and random error can not be overlooked when using the ESA as a substitute for the SA.