Reducing errors in parallel conductance measurement.

Abstract

Conductance (COND) measures left ventricle (LV) and right ventricle (RV) volume continuously during the cardiac cycle. COND measurement of the ventricle can be impaired by electrically conductive factors extrinsic to the heart that cause an artifactual increase in COND. This is known as parallel COND. A hypertonic saline injection has traditionally been used to measure parallel COND. The entry of hypertonic saline into the ventricle causes a rise (ascending region) in ventricular COND tracing, whereas its dissipation causes a fall (descending region). The hypothesis of this study is that parallel COND measurement can vary based on the region of COND tracing (ascending versus descending versus both) chosen for calculations. Parallel COND was measured in the LV (15 pigs and 5 sheep) and the RV (13 pigs and 5 sheep). In the LV, average +/- standard error of mean (SEM) parallel COND measured from the ascending region (55.4 +/- 9.2) was significantly different (p < 0.05) from the descending region and from both regions (72.2 +/- 10.3 and 66.4 +/- 9.2, respectively). Additionally, LV parallel COND measured from the descending region and from both regions were not different (p = NS; 72.2 +/- 10.3 and 66.4 +/- 9.2, respectively). In the RV, there was no significant difference (p = NS) among parallel COND calculated from ascending, descending, and both regions (102.9 +/- 8.1, 105.6 +/- 10.0, and 103.9 +/- 7.5, respectively). Average +/- SEM number of points used for parallel COND calculation (N) in the LV for each region (ascending versus descending versus both) were significantly different (p < 0.05) from one another (8 +/- 1 vs 11 +/- 1 vs 18 +/- 1). Similarly, N values used for the calculation of RV parallel COND in ascending versus descending versus both regions were significantly different (p < 0.05) from one another (6 +/- 1 vs 9 +/- 1 vs 14 +/- 1). In conclusion, there were significant differences in parallel COND calculation based on varying regions of LV COND. This was not true for the RV. To reduce errors that are caused by the differences cited here, one region should be used consistently to measure parallel COND. More study will be required to determine the optimal region of the COND tracing for the determination of parallel COND.