One problem faced by intracardiac conductance volumetry is the non-uniform distribution of the injected current. Salo, in 1989, proposed a method to correct this undesirable effect. The objective here is to test Salo's method in known volumes of simple geometry by obtaining volume profiles. A plastic rod with 15 metallic rings simulated the conductance catheter. Five sections were used for the resistance measurements employing the upper electrode as fixed current source and the lowest one as the shifting source. This is part of Salo's procedure. The source-to-section distance was measured from the moving source to the section (linear definition) or using the equivalent distance concept (Salo's). Thereafter, each sectional resistance set of values was plotted as a function of the inverse of the source-to-section distance (either definition) elevated to an empirical exponent k to obtain the corrected sectional resistance by extrapolation back to zero of the regression line, i.e., a value produced by a source theoretically placed at infinity. In addition, a mathematical analysis was attempted, searching for an optimum k based on minimum volume error. The best volume profiles for two cylinders and a frustum were obtained with k = 2 using the linear definition of distance (errors of -3.49%, -1.25% and -3.65% respectively). Moreover, the frustum angle was determined within 0.4-degrees (2.7%) of the real value. The theoretical analysis led to an inverse logarithmic relationship between the exponent k and the source-to-section distance. In conclusion: (1) The linear definition of source-to-section distance applying Salo's correction with k = 2 produced the smallest errors, both in volume and angle estimations; (2) there is no optimum k; (3) for very large distances, k tends to a low value (about 0.8); (4) for heart sizes, k = 2.1 can be suggested for all sections.
