Left ventricular volume measurement by conductance catheter in intact dogs. Parallel conductance volume depends on left ventricular size.

The conductance catheter is a promising new instrument for continuously measuring left ventricular (LV) volume. Absolute LV volume (Vt) is related to uncorrected conductance volume, B(t), according to the equation: V(t) = (1/alpha)(B(t) - alpha Vc). The alpha Vc factor represents parallel-conductance volume due to conducting material outside the LV blood pool, and may be estimated by transiently changing blood conductivity using a bolus injection of hypertonic saline. alpha is the slope in the relation between B(t) and true LV volume. We tested the assumption that alpha Vc and alpha are constant over a range of hemodynamic conditions. We performed multiple hypertonic saline alpha Vc determinations in seven intact dogs during control conditions and subsequent temporary balloon occlusions of inferior vena cava (IVCO), aorta (AO), and pulmonary artery (PAO). We also compared B(t) with simultaneous biplane angiographic LV volume during similar control and intervention conditions. The saline-derived alpha Vc was 76 +/- 2 ml during control and fell significantly by -7 +/- 2 ml during IVCO (p less than 0.001) but not during AO or PAO. According to multiple linear regression analyses, the strongest predictor of saline-derived alpha Vc was uncorrected end-systolic Bes, with a sensitivity coefficient of 0.60 +/- 0.06 ml/ml (p less than 0.001). Angiographically derived alpha Vc showed a similar dependence on Bes, with a coefficient of 0.77 +/- 0.14 ml/ml (p less than 0.001). Angiographically determined alpha also showed significant variation with hemodynamic interventions, largely reflecting an underlying dependence on alpha Vc. The variation in alpha Vc and alpha with LV size may stem from nonlinearity in the B(t)-V(t) relation. Although the conductance catheter provides a useful measure of relative LV volume, measurement of absolute LV volume over a wide hemodynamic range using constant alpha Vc and alpha factors is unrealistic. This result calls into question the current use of this technique for the measurement of the absolute end-systolic--pressure-volume relation.