The influence of the pericardium on ventricular loading during respiration was studied in 17 acutely instrumented anesthetized dogs. Changes in intrapericardial surface pressures (Ppe) on the ventricles were measured by use of air-filled flat latex balloons during acute changes in ventricular loading with the chest open or during negative intrathoracic pressure (NITP) produced by phrenic nerve stimulation with the chest closed. Ppe always demonstrated a phasic change within a cardiac cycle, with its maximum near end diastole and minimum near end systole, and a waveform similar to ventricular dimensions measured by sonomicrometer crystals. With the chest open we found that 1) inferior vena caval constriction decreased Ppe on both ventricles at end diastole (P less than 0.01), 2) aortic constriction increased Ppe on both ventricles at end systole and end diastole (P less than 0.05), and 3) pulmonary artery constriction increased Ppe on the right ventricle (RV) (P less than 0.01) while decreasing Ppe on the left ventricle (LV) at end diastole (P less than 0.05). Thus regional Ppe over a ventricle is influenced by changes in ventricular loading conditions. During NITP with lung volume either constant or increased, Ppe over the anterolateral LV decreased less than two independent extrapericardial measures of intrathoracic pressure, and this resulted in an increased transpericardial pressure at end systole (P less than 0.05) and end diastole (P less than 0.01). During NITP with increased transpericardial pressure, Ppe over the anterior LV, lateral LV, and RV inflow showed small regional differences, but all decreased less than esophageal pressure (P less than 0.01). These results sugges that the increase in transpericardial pressure during late diastole to early systole, produced by increases in ventricular volume during NITP, could effectively attenuate the increases in ventricular preload and afterload caused by respiration, analogous to a negative feedback loop.

The authors studied the effects of thiopental on baseline airway tone in intact guinea pig tracheas using a preparation where the epithelial (inside) and serosal (outside) surfaces were isolated. Whole tracheas were excised, cannulated, and mounted in 50-ml tissue baths. The serosal and epithelial surfaces were perfused via separate circuits with Krebs-Henseleit solution. All data were expressed as a percent of constriction produced by 2 X 10(-6) M carbachol (a concentration that elicited a 90 + % of maximal constriction). Thiopental elicited a dose-dependent constriction in all 25 tracheas. Increases in tone were first seen at 10(-5) M (14.3 +/- 1.84%; mean +/- SEM) and reached a peak at 10(-3) M (29 +/- 3.16%; P less than .0001). Responses to thiopental were similar when the epithelium was removed, when thiopental was added to the inner perfusate, and when tracheas were pretreated with 10(-5) M pyrilamine. Constriction was entirely inhibited by pretreatment with indomethacin 10(-5) M. The authors conclude that thiopental, at concentrations in the clinical range, causes a reproducible dose-dependent constriction of guinea pig trachea. This effect is mediated by constrictor prostaglandins.

We present a theoretical model of gas transport through the dead space during high-frequency ventilation (HFV) with volumes less than dead space volume. The analysis is based on the axial distribution of transit times of gas moving through the dead space. The model predicts that for tidal volumes (V) much less than dead space (Vd), gas exchange will be proportional to the product of frequency (f) and V2. If gas transport is analyzed in terms of Fick's law, then the effective diffusion coefficient (Deff) can be shown to be equal to fV2 times a constant, whose value equals the square of the coefficient of dispersion of axial transit times through the dead space\((\sigma _t /\bar t)^2 \). Experimental results in straight tubes fit the predictions of this model quite well. A\((\sigma _t /\bar t)\) through the entire dead space of about 30% is more than sufficient to account for gas exchange during HFV in physical models or in intact animals. An axial dispersion of this magnitude can be measured directly from a typical Fowler dead space determination in healthy subjects.

The degree of distortion in dye curves due to sampling and recording systems was estimated and appropriate correction factors were determined. These factors were used in recording corrected dye curves in circulatory models and during left heart catheterization. Calculations of ventricular ejection fractions and volumes were made from the corrected curves.