Tlie differential pressure method of Womersley and McDonald was used to measure instantaneous blood flow in the main pulmonary artery in ten human subjects. Three subjects had normal pulmonary arterial pressures and flows, seven had mitral stenosis and pulmonary hypertension. The spectrum of input impedance versus frequency was similar to that previously reported for the dog and rabbit, with the modulus decreasing from relatively high values at zero frequency to a minimum between 2 and 5 cycles/sec. Characteristic impedance and phase velocity were lower in the normal subjects than in those with pulmonary hypertension (averages, 23 dyne sec crrr r ' and 1.68 m/sec in the normals; 46 dyne sec c n r ' and 4.77 m/sec in the hypertensives). Hydraulic energy dissipated per unit time by pulsations in the pulmonary bed was usually higher in the hypertensive than in the normal cases, because of the greater stiffness of the pulmonary arteries in the subjects with pulmonary hypertension. The elasticity of the pulmonary arterial tree appears to be as important as the state of the arterioles and capillaries in determining the energy required for pulsatile pulmonary blood flow. ADDITIONAL KEY WORDS differential pressure hydraulic energy vascular elasticity pulmonary hypertension mitral stenosis blood flow

To determine the systemic input impedance, pulsatile pressure and flow were measured in the ascending aorta in 16 human subjects who were undergoing diagnostic cardiac catheterization. Blood flow was measured with a catheter-tip electromagnetic velocity meter, and pressure with an external transducer connected with the fluid-filled lumen of the catheter. Five subjects were found to have no evidence of cardiovascular disease (group A, mean age 32 +/- 2 years, mean aortic pressure 97 +/- 4 mm Hg). Seven had clinical and angiographic signs of coronary arterial disease, and mean pressures less than 100 mm Hg (group B, mean age 48 +/- 2 years). Four subjects had signs of coronary disease and mean pressures greater than 100 mm Hg (group C, mean age 48 +/- 3 years). The frequency spectra of impedance were qualitatively similar in all three groups and resembled those previously observed in the canine aorta. Characteristic impedance was lower in the normal subjects (group A, average 53 dyn sec cm-5) than in the subjects with coronary artery disease (groups B and C, average 129 dyn sec cm-5). Among the subjects with coronary disease, characteristic impedance was higher in the hypertensive subjects (group C, average 202 dyn sec cm-5) than in those with lower mean pressures (group B, average 95 dyn sec cm-5). External left ventricular work per unit time (hydraulic power) averaged 1715 milliwatts (mW) in group A, 1120 mW in group B, and 2372 mW in group C. Cardiac outputs were within normal limits in all subjects, but tended to be lower in group B than in group C. These results suggest that the subjects of group C were better able to meet the increased energy demands imposed by an abnormally high aortic input impedance. Further investigation is needed to learn whether the high impedances in subjects with coronary disease represent an increase with age and transmural pressure alone, or whether some additional factor is involved. The data on relatively normal subjects permit a tentative definition of the normal limits for aortic input impedance in man: 26-80 dyn sec cm-5.

Pulmonary vascular input impedance and hydraulic power were measured at various heart rates in 29 anesthetized and 5 unanesthetized dogs. Hydraulic power at the pulmonary veno-atrial junction was measured in 5 dogs. The pulmonary vascular impedance spectrum in the unanesthetized dogs did not differ significantly from that in the anesthetized dogs. Average pulmonary arterial power in the anesthetized dogs was 157 milliwatts (mw), of which 108 mw was associated with mean pressure and flow, and 49 mw with the pulsations around these means. Seventy-eight per cent of this input power was dissipated in passage through the pulmonary bed. Kinetic energy accounted for 1% of the total input power.Because of a steep fall in impedance between zero and 3 cycles/sec, and a rate-dependent change in the harmonic structure of flow pulsations, there was an inverse relationship between heart rate and the input power for a given mean flow, up to 180 beats/ min. Pulmonary vascular dimensions and elasticity, which determine impedance, thus embody a mechanism whereby tachycardia can increase pulmonary blood flow by as much as 35% with an increase in pulmonary arterial input power of less than 5%, without the intervention of vasomotor activity.

No abstract

Pulmonary vascular hydraulic input impedance was measured in 13 anesthetized openchest dogs with normal sinus rhythm, and 2 dogs with surgically induced atrioventricular block, by means of electromagnetic flowmeters and strain gauge manometers of known frequency response. The linearity of the pulmonary bed was evaluated by measuring impedance while the heart rate, and hence the pulsatile input to the bed, was varied. The pulmonary bed behaved as a quasilinear system, within the limits of accuracy of the methods employed and the range of frequencies tested. The use of input impedance, or oscillatory pressure/flow ratio, to describe some characteristics of the bed is therefore justifiable, and analogies with linear models like the simple transmission line are not unreasonable. The characteristic input impedance averaged 3,094 dyne sec cm -5 kg, or about one-third the magnitude of the pulmonary vascular resistance, and was therefore a significant part of the total opposition that must be overcome in moving blood through the lungs. This impedance to pulsatile flow is not included in calculations of resistance from mean pressure and flow measurements. The pattern of the impedance spectrum suggested that reflections originating from arteries 1.0 mm and less in diameter play a large role in determining input impedance and its variations with frequency. Pulmonary vasoconstriction by 5-hydroxytryptamine (serotonin) altered the impedance pattern in a manner consistent with increased wave reflection and displacement of the dominant reflecting sites to positions nearer the main pulmonary artery. Capillary compression by increased intra-alveolar pressure also increased reflection, but did not alter the sites of reflection significantly, providing further evidence that the normal impedance pattern and its modification by serotonin are both determined by the characteristics of the arterial part of the bed.

Supported in part by grants from the American Heart Association, and from the National Heart Institute (H-328), U.S. Public Health Service. Patient Selection and MethodsThe patients studied were undergoing diagnostic catheterization of the left heart by the transbronchial4 or percutaneous dorsal5 route, and simultaneous right heart catheterization via an antecubital vein. The majority had rheumatic lesions of the mitral or aortic valve; individual diagnoses are listed in table 1. All gave a history of cardiopulmonary symptoms. None showed signs of congestive failure at the time of this study.Patients were studied after fasting at least 4 hours, and were usually given 10 mg. of morphine sulfate intramuscularly, and in some cases 100 mg. of pentobarbital, 1 hour prior to the procedure. Measurements were made with the patient supine.A no.-7 cardiac catheter was positioned in the pulmonary artery just beyond the valve by standard methods of intravenous right heart catheterization. The left atrium was catheterized with polyvinyl (I.D. 0.51 mm.) or polyethylene (I.D. 0.58 mm.) tubing passed through the puncture needle. Pressures were recorded by Statham P23-D or P23-A manometers* and a model DR-8 recorder.t Mean pressures were measured at end-expiration by electronic integration. The zero reference level for pressure measurement was a plane parallel to the top of the catheterization table, at a distance above it equal to two thirds the sagittal diameter of the thorax at the sternal angle.Dilution curves from the brachial or femoral artery following injection of indocyanine green ("Cardiogreen")t were recorded by continuous samnpling through a euvette densitometer (Model 103).* The methods used for injection, sampling, recording, and calibrating have been described in detail elsewhere.6When the catheters were in place and control blood samples were drawn, a dye injection was made into the left atrium and the arterial dilution curve was recorded. As soon as the primary curve was recorded, the sampling syringe was replaced, a new control blood sample was taken, and a pulmonary artery injection was made, with less than 5 minutes intervening between the 2 injections. Since the dye leaves the circulation rapidly, the

True propagation coefficients of pulse wave harmonics in an artery can be determined in vivo by measuring pulsatile blood pressure and flow at each of two points along the length of the vessel. These coefficients, which are complex numbers that describe the attenuation and the phase shift imposed on a traveling wave, are independent of the reflected waves in the circulation and thus provide information about the viscoelastic state and other properties of an artery. The equations involved are implicit in standard transmission-line theory, but they have not previously been applied in this particular way to blood vessels. The femoral artery, exposed in situ, was studied in 11 anesthetized dogs. At 1.5 Hz, true attenuation constants averaged 0.0151 nepers/cm, and true phase constants averaged 0.0155 radians/cm. As frequency increased, the apparent phase velocity of flow, in contrast, was relatively low at the first harmonic and rose as frequency increased. True phase velocities lay between the apparent pressure and flow values. Characteristic impedance at 1.5 Hz had an average modulus of 1.76 times 10-4 dyne sec/cm5 and a phase of minus 0.31 radians. The modulus diminished as frequency increased, and the phase became less negative. These results show that true phase constants and characteristic impedances determined by this method are consistent with data reported by others and provide information not previously available about flow wave propagation.

No abstract