Computational vectorelectrocardiography: comparison of representations according to McPhee-Parungao and Frank

Abstract

Aim: to compare amplitude-temporal and vectorcardiographic (VCG) parameters calculated through most commonly used matrix transformations.Materials and methods. Preliminary literature search revealed Kors and inverse Dower transformation to be the most commonly used matrices for converting ECG‑12 into Frank VCG. Trunov-Aidu matrix appeared to be the only one for deriving VCG in the McFee-Parungao lead system.The study included 1250 ECGs of patients at the National Medical Research Center of Cardiology with various cardiovascular diseases: arterial hypertension, coronary heart disease, post-infarction cardiosclerosis, aortic stenosis. For each patient, using the above transformations, three VCGs with an intact and three VCGs with an inverted direction of the Z axis (according to Ozol) were obtained. The comparison was carried out based on the main amplitude-temporal indicators of P-QRS-T, as well as on VCG parameters: spatial QRS-T angle, QRS loop area, QRS integral vector, fractional integrals P and QRS. A total of 9 pairwise comparisons of 134 indicators were carried out for each synthesized VCGResults. The strongest (r > 0.9 or r < -0.9) significant (p < 0.001) linear correlation was found between the spatial metrics: P, QRS, T wave loop lengths, QRS loop area, QRS-T angle, ventricular gradient, and QRS shared integrals. Significant relationship between projection parameters and P-QRS-T amplitude and duration was observed in some cases. The largest number of correlations was noted between parameters calculated in McFee-Parungao and Frank lead system derived by Kors.Conclusions. Strong linear relationship was found between the VCG parameters determined through Kors, Dower and Trunov-Aidu matrix transformations, which indicates the equivalence of the results obtained through these matrices.