Left ventricular volume determination in dogs: a comparison between conductance technique and angiocardiography

Left ventricular (LV) volume was determined simultaneously by monoplane cineangiocardiography and conductivity using a multielectrode conductance catheter at rest and during pressure loading in seven mongrel dogs (mean body weight 22 kg). LV volumes were calculated frame-by-frame (75 frames s-1) by angiocardiography and matched with instantaneous volumes obtained by conductivity. There was an excellent correlation between the two techniques at rest (correlation coefficient, r = 0.96) and during pressure loading (r = 0.92) when the data of each dog were pooled. The standard error of estimate of the mean angiographic volume was 4%. The slope of the regression analysis showed a small but significant (P less than 0.01) decrease from 0.365 at rest to 0.289 during pressure loading, whereas the intercept remained unchanged (24 versus 26 ml). Since no calibration for parallel conductivity of the surrounding tissue was performed, LV end-systolic volume was significantly over- and LV ejection fraction significantly underestimated whereas LV end-diastolic volume was estimated correctly by the conductance technique. It is concluded that LV end-diastolic volume can be determined accurately by the conductance technique in dogs. However, LV end-systolic volume is significantly over- and ejection fraction significantly underestimated. Since there is a good correlation between angiocardiography and conductivity, exact determination of LV volumes and ejection fraction is feasible using a correction factor. The change in slope of the regression equation between angiocardiography and conductivity suggests a change in conductivity of the surrounding tissue during pressure loading which limits the application of the conductance catheter to stable haemodynamic situations or calls for repeated calibrations by an independent technique during acute interventions.