Real-time continuous measurement of right ventricular volume using a conductance catheter

The authors propose using a multi-electrode conductance catheter to measure continuous right ventricular volume. True ventricular volume measurements are affected by four main sources of error. 1) field non-uniformity, 2) catheter curvature, 3) blood conductivity changes, and 4) leakage of current through surrounding tissues. Three-dimensional finite-element models were developed to investigate the effects of these sources of error and to devise schemes for correcting them. The models include an axisymmetric cylindrical model, a rectangular block model, and a heart model with left and right ventricular chambers. The heart model is built from conical primitives, with major dimensions derived from the literature. Finite-element simulations showed that volume measurements were underestimated due to field nonuniformity to as much as 1/25th actual volume in segments near the exciting electrodes. The extent of underestimation in a segment decreased with increasing distance of the segment from the exciting electrodes and increased for larger segmental volumes. Catheter curvature overestimated measured volume by as much as 4.5 times when the curvature was increased from 0.0 to 1.25 (from a straight catheter to a very curved one). The leakage of current through surrounding tissues overestimated volume by nearly 30%. The sensitivity of volume measurement to blood resistivity changes was found to be very high, at 70%. Correction factors established with the computer models compensate for field nonuniformity. Mathematical mapping of the curved catheter onto a fictitious straight catheter corrects for the catheter curvature error. Correction for both nonuniform field and catheter curvature allowed measurement of total ventricular volume with an error of 7%. Leakage current is determined by using different frequencies to build the catheter electric field and to separate tissue and blood resistance paths. Using this scheme, the percentage overestimation in volume measurement due to leakage could be determined with an accuracy of 85%. The proposed correction scheme for blood conductivity changes involves the in-vivo measurement of blood conductivity with the catheter itself. It was found that blood conductivity could be determined with insignificant error (< 0.5%) so long as the blood volume around the exciting electrodes had a radius of more than the electrode spacing.