Medical Physiology, 3rd Edition

How to Measure Blood Pressure, Blood Flow, and Cardiac Volumes

Blood pressure can be measured directly by puncturing the vessel

One can record blood pressure anywhere along the circulation—inside a heart chamber, inside an artery, within a capillary, or within a vein. Clinicians are generally concerned with the intravascular pressure at a particular site (e.g., in a systemic artery) in reference to the barometric pressure outside the body and not with pressure differences between two sites.

The most direct approach for measurement of pressure is to introduce a needle or a catheter into a vessel and position the open tip at a particular site. In the first measurements of blood pressure ever performed, Stephen Hales in 1733 found that a column of blood from a presumably agitated horse rose to fill a brass pipe to a height of 3 m. It was Poiseuille who measured blood pressure for the first time by connecting a mercury-filled U-tube to arteries through a tube containing a solution of saturated NaHCO3. In modern times, a saline-filled transmission or conduit system connects the blood vessel to a pressure transducer. In the most primitive form of this system, a catheter was connected to a closed chamber, one wall of which was a deformable diaphragm. Nowadays, the pressure transducer is a stiff diaphragm bonded to a strain gauge that converts mechanical strain into a change in electrical resistance, capacitance, or inductance (Fig. 17-12). The opposite face of the diaphragm is open to the atmosphere, so that the blood pressure is referenced to barometric pressure. The overall performance of the system depends largely on the properties of the catheter and the strain gauge. The presence of air bubbles and a long or narrow catheter can decrease the displacement, velocity, and acceleration of the fluid in the catheter. Together, these properties determine overall performance characteristics such as sensitivity, linearity, damping of the pressure wave, and frequency response. To avoid problems with fluid transmission in the catheter, some high-fidelity devices employ a solid-state pressure transducer at the catheter tip.


FIGURE 17-12 Direct method for determining blood pressure.

In catheterizations of the right heart, the clinician begins by sliding a fluid-filled catheter into an antecubital vein and, while continuously recording pressure, advances the catheter tip into the superior vena cava, through the right atrium and the right ventricle, and past the pulmonary valve into the pulmonary artery. Eventually, the tip reaches and snugly fits into a smaller branch of the pulmonary artery, recording the pulmonary wedge pressure (see p. 519). The wedge pressure effectively measures the pressure downstream from the catheter tip, that is, the left atrial pressure.

In catheterizations of the left heart, the clinician slides a catheter into the brachial artery or femoral artery, obtaining the systemic arterial blood pressure. From there, the catheter is advanced into the aorta, the left ventricle, and finally the left atrium.

Clinical measurements of venous pressure are typically made by inserting a catheter into the jugular vein. Because of the low pressures, these venous measurements require very sensitive pressure transducers or water manometers.

In the research laboratory, one can measure capillary pressure in exposed capillary beds by inserting a micropipette that is pressurized just enough (with a known pressure) to keep fluid from entering or leaving the pipette.

Blood pressure can be measured indirectly by use of a sphygmomanometer

In clinical practice, one may measure arterial pressure indirectly by use of a manual sphygmomanometer (Fig. 17-13). An inextensible cuff containing an inflatable bag is wrapped around the arm (or occasionally, the thigh). Inflation of the bag by means of a rubber squeeze bulb to a pressure level above the expected systolic pressure occludes the underlying brachial artery and halts blood flow downstream. The pressure in the cuff, measured by means of a mercury or aneroid manometer, is then allowed to slowly decline (see Fig. 17-13, diagonal red line). The physician can use either of two methods to monitor the blood flow downstream of the slowly deflating cuff. In the palpatory method, the physician detects the pulse as an indicator of flow by feeling the radial artery at the wrist. In the auscultatory method, the physician detects flow by using a stethoscope to detect the changing character of Korotkoff sounds over the brachial artery in the antecubital space.


FIGURE 17-13 Sphygmomanometry. The clinician inflates the cuff to a pressure that is higher than the anticipated systolic pressure and then slowly releases the pressure in the cuff.

The palpatory method permits determination of the systolic pressure; that is, the pressure in the cuff below which it is just possible to detect a radial pulse. Because of limited sensitivity of the finger, palpation probably slightly underestimates systolic pressure. The auscultatory method permits the detection of both systolic and diastolic pressure. The sounds heard during the slow deflation of the cuff can be divided into five phases (see Fig. 17-13). During phase I, there is a sharp tapping sound, indicating that a spurt of blood is escaping under the cuff when cuff pressure is just below systolic pressure. The pressure at which these taps are first heard closely represents systolic pressure. In phase II, the sound becomes a blowing or swishing murmur. During phase III, the sound becomes a louder thumping. In phase IV, as the cuff pressure falls toward the diastolic level, the sound becomes muffled and softer. Finally, in phase V, the sound disappears. Although some debate persists about whether the point of muffling or the point of silence is the correct diastolic pressure, most favor the point of muffling as being more consistent. Actual diastolic pressure may be somewhat overestimated by the point of muffling but underestimated by the point of silence.

Practical problems arise when a sphygmomanometer is used with children or obese adults or when it is used to obtain a measurement on a thigh. Ideally, one would like to use a pressure cuff wide enough to ensure that the pressure inside the cuff is the same as that in the tissue surrounding the artery. In 1967, the American Heart Association recommended that the pneumatic bag within the cuff be 20% wider than the diameter of the limb, extend at least halfway around the limb, and be centered over the artery. More recent studies indicate that accuracy and reliability improve when the pneumatic bag completely encircles the limb, as long as the width of the pneumatic bag is at least the limb diameter.

Blood flow can be measured directly by electromagnetic and ultrasound flowmeters

The spectrum of blood flow measurements in the circulation ranges from determinations of total blood flow (cardiac output) to assessment of flow within an organ or a particular tissue within an organ. Moreover, one can average blood flow measurements over time or record continuously. Examples of continuous recording include recordings of the phasic blood flow that occurs during the cardiac cycle or any other periodic event (e.g., breathing). We discuss both invasive and noninvasive approaches.

Invasive Methods

Invasive approaches require direct access to the vessel under study and are thus useful only in research laboratories. The earliest measurements of blood flow involved collecting venous outflow into a graduated cylinder and timing the collection with a stopwatch. This direct approach was limited to short time intervals to minimize blood loss and the resulting changes in hemodynamics. Blood loss could be avoided by ingenious but now antiquated devices that returned the blood to the circulation, in either a manual or a semiautomated fashion.

The most frequently used modern instruments for measurement of blood flow in the research laboratory are electromagnetic flowmeters based on the electromagnetic induction principle (Fig. 17-14). The vessel is placed in a magnetic field. According to Faraday's induction law, moving any conductor (including an electrolyte solution, such as blood) at right angles to lines of the magnetic field generates a voltage difference between two points along an axis perpendicular to both the axis of the movement and the axis of the magnetic field. The induced voltage is

image (17-14)

where B is the density of magnetic flux, image is the average linear velocity, and D is the diameter of the moving column of blood.


FIGURE 17-14 Electromagnetic flowmeter.

Ultrasound flowmeters employ a pair of probes placed at two sites along a vessel. One probe emits an ultrasound signal, and the other records it. The linear velocity of blood in the vessel either induces a change in the frequency of the ultrasound signal (Doppler effect) or alters the transit time of the ultrasound signal. Both the electromagnetic and ultrasound methods measure linear velocity, not flow per se.

Noninvasive Methods

The electromagnetic or ultrasonic flowmeters require the surgical isolation of a vessel. However, ultrasonic methods are also widely used transcutaneously on surface vessels in humans. This method is based on recording of the backscattering of the ultrasound signal from moving red blood cells. To the extent that the red blood cells move, the reflected sound has a frequency different from that of the emitted sound (Doppler effect). This frequency difference may thus be calibrated to measure flow. Plethysmographic methods are noninvasive approaches for measurement of changes in the volume of a limb or even of a whole person (see p. 617). Inflation of a pressure cuff enough to occlude veins but not arteries allows blood to continue to flow into (but not out of) a limb or an organ, so that the volume increases with time. The record of this rise in volume, as recorded by the plethysmograph, is a measure of blood flow.

With the exception of transcutaneous ultrasonography, the direct methods discussed for measurement of blood flow are largely confined to research laboratories. The next two sections include discussions of two indirect methods that clinicians use to measure mean blood flow.

Cardiac output can be measured indirectly by the Fick method, which is based on the conservation of mass

The Fick method requires that a substance be removed from or added to the blood during its flow through an organ. The rate at which X passes a checkpoint in the circulation (image) is simply the product of the rate at which blood volume passes the checkpoint (F) and the concentration of X in that blood:

image (17-15)

The Fick principle is a restatement of the law of conservation of mass. The amount of X per unit time that passes a downstream checkpoint (image) minus the amount of X that passes an upstream checkpoint (image) must equal the amount of X added or subtracted per unit time (image) between these two checkpoints (Fig. 17-15A):

image (17-16)


FIGURE 17-15 Fick method for determining cardiac output.

image is positive for the addition of X. If the volume flow is identical at both checkpoints, combining Equation 17-15 and Equation 17-16 yields the Fick equation:

image (17-17)

We can calculate flow from the amount of X added or subtracted and the concentrations of X at the two checkpoints:

image (17-18)

It is easiest to apply the Fick principle to the blood flow through the lungs, which is the cardiac output (see Fig. 17-15B). The quantity added to the bloodstream is the O2 uptake (image) by the lungs, which we obtain by measuring the subject's O2 consumption. This value is typically 250 mL of O2 gas per minute. The upstream checkpoint is the pulmonary artery (point A), where the O2 content ([O2]A) is typically 15 mL of O2 per deciliter of blood. The sample for this checkpoint must reflect the O2 content of mixed venous blood, obtained by means of a catheter within the right atrium or the right ventricle or pulmonary artery. The downstream checkpoint is a pulmonary vein (point B), where the O2 content ([O2]B) is typically 20 mL O2 per deciliter of blood. We can obtain the sample for this checkpoint from any systemic artery. Using these particular values, we calculate a cardiac output of 5 L/min:

image (17-19)

Cardiac output can be measured indirectly by dilution methods

The dye dilution method, inaugurated by G.N. Stewart in 1897 and extended by W.F. Hamilton in 1932, is a variation of the Fick procedure. One injects a known quantity of a substance (X) into a systemic vein (e.g., antecubital vein) at site A while simultaneously monitoring the concentration downstream at site B (Fig. 17-16A). It is important that the substance not leave the vascular circuit and that it be easy to follow the concentration, by either successive sampling or continuous monitoring. If we inject a single known amount (QX) of the indicator, an observer downstream at checkpoint B will see a rising concentration of X, which, after reaching its peak, falls off exponentially. Concentration measurements provide the interval (Δt) between the time the dye makes its first appearance at site B and the time the dye finally disappears there.


FIGURE 17-16 Dye dilution method for determining blood flow. In B, C, and D, the areas underneath the three red curves—as well as the three green areas—are all the same.

If site B is in the pulmonary artery, then the entire amount QX that we injected into the peripheral vein must pass site B during the interval Δt, carried by the entire cardiac output. We can deduce the average concentration [image] during the interval Δt from the concentration-versus-time curve in Figure 17-16B. From the conservation of mass, we know that

image (17-20)

Because the volume of blood (V) that flowed through the pulmonary artery during the interval Δt is, by definition, the product of cardiac output and the time interval (CO · Δt),

image (17-21)

Note that the product Δt · [image] is the area under the concentration-versus-time curve in Figure 17-16B. Solving for CO, we have

image (17-22)

In practice, cardiologists monitor [X] in the brachial artery. Obviously, only a fraction of the cardiac output passes through a brachial artery; however, this fraction is the same as the fraction of QX that passes through the brachial artery. If we were to re-derive Equation 17-22 for the brachial artery, we would end up multiplying both the CO and QX terms by this same fraction. Therefore, even though only a small portion of both cardiac output and injected dye passes through any single systemic artery, we can still use Equation 17-22 to compute cardiac output with data from that artery.

Compared with the [X] profile in the pulmonary artery, the [X] profile in the brachial artery is not as tall and is more spread out, so that [image] is smaller and Δt is longer. However, the product [image] · Δt in the brachial artery—or any other systemic artery—is the same as that in the pulmonary artery. Indocyanine green dye (Cardiogreen) is the most common dye employed. Because the liver removes this dye from the circulation, it is possible to repeat the injections, after a sufficient wait, without progressive accumulation of dye in the plasma. Imagine that after we inject 5 mg of the dye, [image] under the curve is 2 mg/L and Δt is 0.5 min. Thus,

image (17-23)

A practical problem is that after we inject a marker into a systemic vein, blood moves more quickly through some pulmonary beds than others, so that the marker arrives at checkpoint B at different times. This process, known as dispersion, is the main cause of the flattening of the [X] profile in the brachial artery (see Fig. 17-16C) versus the pulmonary artery (see Fig. 17-16B). If we injected the dye into the left atrium and monitored it in the systemic veins, the dispersion would be far worse because of longer and more varied path lengths in the systemic circulation compared with the pulmonary circulation. In fact, the concentration curve would be so flattened that it would be difficult to resolve the area underneath the [X] profile.

A second practical problem with a closed circulatory system is that before the initial [X] wave has waned, recirculation causes the injected indicator to appear for a second time in front of the sensor at checkpoint B (see Fig. 17-16D). Extrapolation of the exponential decay of the first wave can correct for this problem.

The thermodilution technique is a convenient alternative approach to the dye technique. In this method, one injects a bolus of cold saline and an indwelling thermistor is used to follow the dilution of these “negative calories” as a change of temperature at the downstream site. In the thermodilution technique, a temperature-versus-time profile replaces the concentration-versus-time profile. During cardiac catheterization, the cardiologist injects a bolus of cold saline into the right atrium and records the temperature change in the pulmonary artery. The distance between upstream injection and downstream recording site is kept short to avoid heat exchange in the pulmonary capillary bed. The advantages of this method are that (1) the injection of cold saline can be repeated without harm, (2) a single venous (versus venous and arterial) puncture allows access to both the upstream and the downstream sites, (3) less dispersion occurs because no capillary beds are involved, and (4) less recirculation occurs because of adequate temperature equilibration in the pulmonary and systemic capillary beds. A potential drawback is incomplete mixing, which may result from the proximity between injection and detection sites.

Regional blood flow can be measured indirectly by “clearance” methods

The methods used to measure regional blood flow are often called clearance methods, although the term here has a meaning somewhat different from its meaning in kidney physiology. Clearance methods are another application of the Fick principle, using the rate of uptake or elimination of a substance by an organ together with a determination of the difference in concentration of the indicator between the arterial inflow and venous outflow (i.e., the a-v difference). By analogy with Equation 17-18, we can compute the blood flow through an organ (F) from the rate at which the organ removes the test substance X from the blood (image) and the concentrations of the substance in arterial blood ([X]a) and venous blood ([X]v):

image (17-24)

One can determine hepatic blood flow with the use of BSP (bromsulphthalein), a dye that the liver almost completely clears and excretes into the bile (see p. 951). Here, image is the rate of removal of BSP from the blood, estimated as the rate at which BSP appears in the bile. [X]a is the concentration of BSP in a systemic artery, and [X]v is the concentration of BSP in the hepatic vein.

In a similar manner, one can determine renal blood flow with the use of PAH (para-aminohippurate). The kidneys almost completely remove this compound from the blood and secrete it into the urine (see pp. 749–750).

It is possible to determine coronary blood flow or regional blood flow through skeletal muscle from the tissue clearance of rapidly diffusing inert gases, such as the radioisotopes 133Xe and 85Kr.

Finally, one can use the rate of disappearance of nitrous oxide (N2O), a gas that is historically important as the first anesthetic, to compute cerebral blood flow.

A similar although qualitative approach is thallium scanning to assess coronary blood flow. Here one measures the uptake of an isotope by the heart muscle, rather than its clearance (Box 17-2).

Box 17-2

Thallium Scanning for Assessment of Coronary Blood Flow

Thallium is an ion that acts as a potassium analog and enters cells through the same channels or transporters as K+ does. Active cardiac muscle takes up injected 201Tl, provided there is adequate blood flow. Therefore, the rate of uptake of the 201Tl isotope by the heart is a useful qualitative measure of coronary blood flow. Complete 201Tl myocardial imaging is possible by two-dimensional scanning of the emitted gamma rays or by computed tomography for a three-dimensional image. Thus, in those portions of myocardial tissue supplied by stenotic coronary vessels, the uptake is slower, and these areas appear as defects on a thallium scan. Thallium scans are used to detect coronary artery disease during exercise stress tests.

Ventricular dimensions, ventricular volumes, and volume changes can be measured by angiography and echocardiography

Clinicians can use a variety of approaches to examine the cardiac chambers. Gated radionuclide imaging employs compounds of the gamma-emitting isotope 99mTc, imageN17-6 which has a half-life of 6 hours. After 99mTc is injected, a gamma camera provides imaging of the cardiac chambers. Electrocardiogram (ECG) gating (i.e., synchronization to a particular spot on the ECG) allows the apparatus to snap a picture at a specific part of the cardiac cycle and to sum these pictures over many cycles. Because this method does not provide a high-resolution image, it yields only a relative ventricular volume. From the difference between the count at the maximally filled state (end-diastolic volume) and at its minimally filled state (end-systolic volume), the cardiologist can estimate the fraction of ventricular blood that is ejected during systole—the ejection fraction—which is an important measure of cardiac function.


99mTc Scanning

Contributed by Emile Boulpaep

Several compounds labeled with 99mTc—for instance, technetium Tc 99m sestamibi and technetium Tc 99m tetrofosmin—have been introduced for imaging myocardial perfusion. The 99mTc label emits gamma radiation at 140 keV by an isomeric transition (indicated by the m in 99m); it has a half-life of 6 hours. Following injection, the initial distribution of these agents in the myocardium is proportional to the relative distribution of myocardial blood flow. The radiochemical enters cardiac myocytes passively in such a way that about 30% to 40% of the chemical is extracted by the myocardium. Extraction may be enhanced by administering nitrates prior to injection. Because the radiochemical leaves the myocyte rather slowly (over several hours), one can perform the imaging with the gamma camera over a time period of hours. Note that absolute measurements of myocardial blood flow would require positron-emission tomography (PET), which can quantitate counts per unit volume of tissue.

It is possible to use 99mTc-labeled compounds not only for assessing myocardial perfusion but also for assessing myocardial function. In single-photon emission computed tomography (SPECT), the computer acquires imaging data synchronized with the R wave of the ECG (see Fig. 21-7). This gated imaging allows one to display end-diastolic and end-systolic images along various axes of the heart. These end-diastolic and end-systolic dimensions can then be compared to assess ejection fraction, stroke volume, regional wall motion, and regional wall thickening.

Angiography can accurately provide the linear dimensions of the ventricle, allowing the cardiologist to calculate absolute ventricular volumes. A catheter is threaded into either the left or the right ventricle, and saline containing a contrast substance (i.e., a chemical opaque to x-rays) is injected into the ventricle. This approach provides a two-dimensional projection of the ventricular volume as a function of time. In magnetic resonance imaging, the physician obtains a nuclear magnetic resonance (NMR) image of the protons in the water of the heart muscle and blood. However, because standard NMR requires long data-acquisition times, it does not provide good time resolution.

Echocardiography, which exploits ultrasonic waves to visualize the heart and great vessels, can be used in two modes. In M-mode echocardiography (M is for motion), the technician places a single transducer in a fixed position on the chest wall and obtains a one-dimensional view of heart components. As shown in the upper portion of Figure 17-17A, the ultrasonic beam transects the anterior wall of the right ventricle, the right ventricle, the septum, the left ventricle, the leaflets of the mitral valve, and the posterior wall of the left ventricle. The lower portion of Figure 17-17A shows the positions of the borders between these structures (x-axis) during a single cardiac cycle (y-axis) and thus how the size of the left ventricle—along the axis of the beam—changes with time. Of course, the technician can obtain other views by changing the orientation of the beam.


FIGURE 17-17 M-mode and two-dimensional echocardiography. In A, the tracing on the bottom shows the result of an M-mode echocardiogram (i.e., transducer in a single position) during one cardiac cycle. The waves represent motion (M) of heart boundaries transected by a stationary ultrasonic beam. In two-dimensional echocardiography (upper panel), the probe rapidly rotates between the two extremes (broken lines), producing an image of a slice through the heart at one instant in time.

In two-dimensional echocardiography, the probe automatically and rapidly pivots, scanning the heart in a single anatomical slice or plane (see Fig. 17-17A, area between the two broken lines) and providing a true cross section. This approach is therefore superior to angiography, which provides only a two-dimensional projection. Because cardiac output is the product of heart rate and stroke volume, one can calculate cardiac output from echocardiographic measurements of ventricular end-diastolic and end-systolic volume.

A problem common to angiography and M-mode echocardiography is that it is impossible to compute ventricular volume from a single dimension because the ventricle is not a simple sphere. As is shown in Figure 17-17B, the left ventricle is often assumed to be a prolate ellipse, with a long axis L and two short axes D1 and D2. To simplify the calculation and to allow ventricular volume to be computed from a single measurement, it is sometimes assumed that D1 and D2 are identical and that D1 is half of L. Unfortunately, use of this algorithm and just a single dimension, as provided by M-mode echocardiography, imageN17-7 often yields grossly erroneous volumes. Use of two-dimensional echocardiography to sum information from several parallel slices through the ventricle, or from planes that are at a known angle to one another, can yield more accurate volumes.


Ventricular Volume from M-Mode Echocardiography

Contributed by Emile Boulpaep

As shown in Figure 17-17B, the left ventricle is often assumed to be a prolate ellipse, with a long axis L and two short axes D1 and D2. To simplify the calculation, and to allow ventricular volume to be computed from a single measurement, it is sometimes assumed that D1 and D2 are identical, and that D1 is half of L. Unfortunately, use of this algorithm and just a single dimension, as provided by M-mode echocardiography, often yields grossly erroneous volumes.

One can obtain a more accurate estimate of ventricular volume by including an independent measurement of a second dimension, as is done in two-dimensional echocardiography. For example, one could obtain the long axis (L) in addition to the short axes (D1 and D2, which are assumed to be the same in the simple calculation). However, the ventricle often does not resemble a prolate ellipse, certainly not in pathological states. Thus, cardiologists have used more complex geometric models (e.g., bullet shape).

In addition to ultrasound methods and angiography, the technique of magnetic resonance angiography, an application of magnetic resonance tomography, is used to obtain two-dimensional images of slices of ventricular volumes or of blood vessels.

In contrast to standard echocardiography, Doppler echocardiography provides information on the velocity, direction, and character of blood flow, just as police radar monitors traffic. In Doppler echocardiography (as with police radar), most information is obtained with the beam parallel to the flow of blood. In the simplest application of Doppler flow measurements, one can continuously monitor the velocity of flowing blood in a blood vessel or part of the heart. On such a record, the x-axis represents time, and the y-axis represents the spectrum of velocities of the moving red blood cells (i.e., different cells can be moving at different velocities). Flow toward the transducer appears above baseline, whereas flow away from the transducer appears below baseline. The intensity of the record at a single point on the y-axis (encoded by a gray scale or false color) represents the strength of the returning signal, which depends on the number of red blood cells moving at that velocity. Thus, Doppler echocardiography is able to distinguish the character of flow: laminar versus turbulent. Alternatively, at one instant in time, the Doppler technician can scan a region of a vessel or the heart, obtaining a two-dimensional, color-encoded map of blood velocities. If we overlay such two-dimensional Doppler data on a two-dimensional echocardiogram, which shows the position of the vessel or cardiac structures, the result is a color flow Doppler echocardiogram (Fig. 17-18).


FIGURE 17-18 The colors, which encode the velocity of blood flow, are superimposed on a two-dimensional echocardiogram, which is shown in a gray scale. A, Blood moves through the mitral valve and into the left ventricle during diastole. Because blood is flowing toward the transducer, its velocity is encoded as red. B, Blood moves out of the ventricle and toward the aortic valve during systole. Because blood is flowing away from the transducer, its velocity is encoded as blue. (From Feigenbaum H: Echocardiography. In Braunwald E [ed]: Heart Disease: A Textbook of Cardiovascular Medicine, 5th ed. Philadelphia, WB Saunders, 1997.)

Finally, a magnetic resonance scanner can also be used in two-dimensional phase-contrast mapping to yield quantitative measurements of blood flow velocity.