Non-reproducibility of serial vectorcardiograms obtained by the Frank system***

Summary Three vectorcardiograms were obtained at a few days' interval from each of 33 normal subjects. Special care was taken to avoid, as much as possible, the influence of factors known to cause significant variations in serial electrocardiograms and vectorcardiograms. These are: respiratory movements, shifts in the location of the electrodes, inaccuracies of the visual measurements, and the presence of low resistance paths. Comparisons among the means of the measurements obtained from the first, second and third tracings of the 33 subjects considered as a group disclosed only small differences. In contrast, important variations in some measurements were found in the vectorcardiograms of individual cases. This study demonstrates that differences in the position of the electrodes are not the only important cause of intra-individual day-to-day variations in the vectorcardiograms of normal subjects. Furthermore, it strongly suggests that one or more physiological factors exist, other than those mentioned above, that may induce the variations observed. Changes in the skin-electrode impedance, displacements of the electrical center of the heart by causes other than respiratory movements, and variations in the conductivity of the medium surrounding the heart, are some of the factors that deserve further investigation as possible sources of the variations observed. In the case of rounded loops or vector loops with some peculiar geometrical forms, large shifts of the maximal planar QRS vectors do occur which are most misleading when expressed in numerical terms and blindly accepted to be fed into the computer; in these instances, qualitative over-all inspection of the spatial vectorcardiogram becomes very helpful in placing the apparently wide quantitative variations in its proper perspective.