Section 3: Anesthesia Management
Part B: Monitoring
Chapter 30: Cardiovascular Monitoring

Arterial Pressure Waveforms

Direct arterial pressure monitoring in anesthetized patients began more than 50 years ago. 79  In that early era, arterial pulse waveform analysis was noted to provide useful diagnostic information, but somewhat surprisingly, modern physicians pay little attention to the morphology and detail of the arterial pressure waveform. O‘Rourke and Gallagher 144  attribute this change in practice to the reliance on cuff sphygmomanometry, which provides “numbers which came to be linked in a simplistic way to cardiac strength (systolic pressure) and arteriolar tone (diastolic pressure). Pseudoscience had arrived with (these) numbers. Even when monitored directly in operating theatres and critical care areas, anesthetists and intensivists show little interest in the waveform and base their judgments on values of systolic, diastolic and mean pressure.”

Because clinicians today benefit from the widespread availability of high resolution, multicolored monitor displays, renewed interest in waveform analysis should expand clinical monitoring capabilities. 80  Appreciation of the diagnostic clues provided by the direct arterial pressure waveform requires full understanding of normal waveform components, their relation to the cardiac cycle, and differences in waveforms recorded from different sites in the body.

The systemic arterial pressure waveform results from ejection of blood from the left ventricle into the aorta during systole, followed by peripheral arterial runoff of this stroke volume during diastole (Fig. 30–19). The systolic components follow the ECG R wave and consist of a steep pressure upstroke, peak, and decline and correspond to the period of left ventricular systolic ejection. The downslope of the arterial pressure waveform is interrupted by the dicrotic notch, then continues its decline during diastole following the ECG T wave, reaching its nadir at end-diastole. The dicrotic notch recorded directly from the central aorta is termed the incisura (from the Latin, a cutting into). The incisura is sharply defined and undoubtedly is related to aortic valve closure. 145  In contrast, the peripheral arterial waveform generally displays a later, smoother dicrotic notch that only approximates timing of aortic valve closure and depends more on arterial wall properties. 146  Note that the systolic upstroke of the radial artery pressure trace does not appear for 120 to 180 milliseconds after inscription of the ECG R wave (see Fig. 30–19). This interval reflects the sum of times required for spread of electrical depolarization through the ventricular myocardium, isovolumic left ventricular contraction, aortic valve opening, left ventricular ejection, transmission of the aortic pressure wave to the radial artery, and finally, transmission of the pressure signal from the arterial catheter to the pressure transducer.

Click thumbnail to see full size image
FIGURE 30–19 Normal arterial blood pressure waveform and its relation to the electrocardiographic R wave. (1) Systolic upstroke, (2) systolic peak pressure, (3) systolic decline, (4) dicrotic notch, (5) diastolic runoff, and (6) end-diastolic pressure. (From Mark JB: Atlas of Cardiovascular Monitoring. New York, Churchill Livingstone, 1998: Fig. 8–1.)

The bedside monitor displays numeric values for the systolic peak and end-diastolic trough pressures. Measurement of mean pressure is more complicated and depends on the algorithm employed by the monitor. 147  In simplest terms, MAP is equal to the area beneath the arterial pressure curve divided by the beat period and averaged over a series of consecutive heartbeats. Although MAP is often estimated as diastolic pressure plus one-third times the pulse pressure, this estimation can be misleading. At equivalent heart rates, narrow or thin arterial pressure waveforms spend more time at lower pressures, resulting in a low MAP, whereas wide or full arterial pressure waveforms spend more time at higher pressures, resulting in a higher MAP.

One of the most important features of the arterial pressure waveform is the phenomenon of distal pulse amplification. Pressure waveforms recorded simultaneously from different arterial sites will have different morphologies owing to the physical characteristics of the vascular tree, namely impedance and harmonic resonance (Fig. 30–20). 48, 109, 119, 144, 148  As the arterial pressure wave travels from the central aorta to the periphery, several characteristic changes occur. The arterial upstroke becomes steeper, the systolic peak becomes higher, the dicrotic notch appears later, the diastolic wave becomes more prominent, and the end-diastolic pressure becomes lower. Thus, compared with central aortic pressure, peripheral arterial waveforms have higher systolic pressure, lower diastolic pressure, and wider pulse pressure. Furthermore, there is a delay in the arrival of the pressure pulse at peripheral sites, so that the systolic pressure upstroke begins approximately 60 milliseconds later in the radial artery than in the aorta. Despite morphologic and temporal differences between peripheral and central arterial waveforms, the MAP in the aorta is just slightly greater than that in the radial artery. 119, 148 

Click thumbnail to see full size image
FIGURE 30–20 Distal pulse wave amplification of the arterial pressure waveform. Compared with pressure in the aortic arch, the more peripherally recorded femoral artery pressure waveform demonstrates a wider pulse pressure (compare 1 and 2), a delayed upstroke (3), a delayed, slurred dicrotic notch (compare arrows), and a more prominent diastolic wave. See text for greater detail. (From Mark JB: Atlas of Cardiovascular Monitoring. New York, Churchill Livingstone, 1998: Fig. 8–4.)

Pressure wave reflection is the predominant factor that influences the shape of the arterial pressure waveform as it travels peripherally. 119, 144, 149, 150, 151  As blood flows from aorta to radial artery, mean pressure only decreases slightly, because there is little resistance to flow, but then falls markedly in the arterioles, owing to the dramatic increase in vascular resistance at this site. This high resistance to flow diminishes pressure pulsations in small downstream vessels but acts to augment upstream arterial pressure pulses owing to pressure wave reflection. 119, 152  Murgo et al 150, 151  and O‘Rourke et al 119, 144  provide detailed explanations of the arterial pressure wave, along with models of the circulation that provide more complete insight into these phenomena. These studies underscore the importance of wave reflection in determining the shape of the arterial pulse recorded from all sites in the body, in health and disease. For example, elderly patients have reduced arterial distensibility, which results in early return of reflected pressure waves, an increased pulse pressure, a late systolic pressure peak, and disappearance of the diastolic pressure wave (Fig. 30–21). 119, 144, 152 

Click thumbnail to see full size image
FIGURE 30–21 Impact of pressure wave reflection on arterial pressure waveforms. In elderly individuals with reduced arterial distensibility, early return of reflected waves increases pulse pressure, produces a late systolic pressure peak (arrow), and attenuates the diastolic pressure wave. Myocardial oxygen balance is more tenuous in the elderly because these changes in blood pressure cause an increased myocardial oxygen demand during systole and a reduced myocardial oxygen supply during diastole. See text for greater detail.

From these considerations, it becomes evident that the morphology of the arterial waveform and the precise values for systolic and diastolic blood pressure vary throughout the body under normal conditions in otherwise healthy individuals. Perhaps of even greater importance, the relation between central and peripheral arterial pressure varies with age and is altered by various physiologic changes, pathologic conditions, and pharmacologic interventions. (See the section Choosing the Site for Arterial Pressure Monitoring for more detail.)

Morphologic features of individual arterial pressure waveforms provide diagnostic clues to various pathologic conditions. Aortic stenosis produces a fixed obstruction to left ventricular ejection, resulting in a reduced stroke volume and an arterial pressure waveform that rises slowly (pulsus tardus) and peaks late in systole (Fig. 30–22). A distinct shoulder, termed the anacrotic notch, often distorts the pressure upstroke. 153  In addition, the dicrotic notch may not be discernible and the arterial pressure is small in amplitude (pulsus parvus). All of these features make the arterial pressure waveform appear overdamped, in contrast to the normal arterial pressure waveform, which has a sharp upstroke and distinct dicrotic notch.

Click thumbnail to see full size image Click thumbnail to see full size image
FIGURE 30–22 Influence of pathologic conditions on arterial pressure (ART) waveform morphology. (A) Normal ART and pulmonary artery pressure (PAP) waveform morphologies demonstrating the similar timing of these waveforms relative to the electrocardiographic R wave. (B) In aortic stenosis, the ART waveform is distorted and demonstrates a slurred upstroke and delayed systolic peak. These changes are particularly striking in comparison with the normal PAP waveform. Note the beat-to-beat respiratory variation in the PAP waveform. See text for greater detail. For (A) and (B), the ART scale is on the left and the PAP scale is on the right. (C) Aortic regurgitation produces a bisferiens pulse and a wide pulse pressure. See text for greater detail. (D) Arterial pressure waveform in hypertrophic cardiomyopathy shows a peculiar “spike-and-dome” configuration. Pressure waveform assumes a more normal morphology following surgical correction of this condition. See text for greater detail. (From Mark JB: Atlas of Cardiovascular Monitoring. New York, Churchill Livingstone, 1998: Figs. 3–3, 17–21, and 17–24.)

In aortic regurgitation, the arterial pressure wave rises rapidly, pulse pressure increases, and diastolic pressure is low, owing to the runoff of blood into the left ventricle as well as the periphery during diastole. Because of the large stroke volume ejected from the left ventricle in this condition, the arterial pressure pulse may have two systolic peaks (bisferiens pulse) (see Fig. 30–22). These two peaks represent separate percussion and tidal waves, with the former resulting from left ventricular ejection and the latter arising from the periphery as a reflected wave. 153  The bisferiens pulse is also described in patients with mixed aortic regurgitation and stenosis and in patients with hypertrophic cardiomyopathy, although the physiologic basis is different in the latter condition. In hypertrophic cardiomyopathy, the arterial pressure waveform assumes a peculiar bifid shape, termed a “spike-and-dome” configuration. After an initial sharp pressure upstroke that results from rapid left ventricular ejection in early systole, arterial pressure falls rapidly as dynamic left ventricular outflow obstruction develops during midsystole, and a late systolic reflected wave follows, thereby creating the characteristic double-peaked waveform (see Fig. 30–22). 153, 154 

Observation of arterial waveform patterns over consecutive heartbeats provides an additional set of diagnostic clues. Pulsus alternans is recognized by the alternating beats of larger and smaller pulse pressures (Fig. 30–23). In general, it is considered to be a sign of severe left ventricular systolic dysfunction, often noted in patients with advanced aortic stenosis. It may be seen occasionally during general anesthesia, presumably as a consequence of the anesthetic-induced reduction in sympathetic nervous system activity in patients with underlying impairment of left ventricular contractility. 155  Pulsus alternans should be distinguished from the bigeminal pulse that arises from a bigeminal rhythm, usually ventricular bigeminy. Both abnormalities create an alternating pulse pressure in the arterial pressure waveform, but the rhythm is regular in pulsus alternans.

Click thumbnail to see full size image Click thumbnail to see full size image Click thumbnail to see full size image
FIGURE 30–23 Beat-to-beat variability in arterial pressure waveform morphologies. (A) Pulsus alternans. (B) Pulsus paradoxus. The marked decline in systolic arterial pressure and pulse pressure during spontaneous inspiration (arrows) is characteristic of this condition. (C) Systolic pressure variation. Compared with systolic blood pressure recorded at end expiration (1) a small increase occurs during positive pressure inspiration (2, D up) followed by a decrease (3, D down). Normally, total systolic pressure variation does not exceed 10 mm Hg. In this instance, the large D down indicates hypovolemia even though systolic arterial pressure and heart rate are relatively normal. See text for greater detail. (From Mark JB: Atlas of Cardiovascular Monitoring. New York, Churchill Livingstone, 1998: Figs. 18–10C, and 16–16.)

Pulsus paradoxus is an exaggerated inspiratory fall in systolic arterial pressure, exceeding 10 to 12 mm Hg during quiet breathing (see Fig. 30–23). 153, 156, 157, 158  The term may be confusing because a small inspiratory reduction in blood pressure is a normal phenomenon, and pulsus paradoxus is not truly paradoxic, but rather an exaggeration of this normal inspiratory decline in blood pressure. Pulsus paradoxus is a characteristic, almost universal finding in cardiac tamponade and occurs in many patients with pericardial constriction. It is said to occur in patients with airway obstruction, bronchospasm, dyspnea, or any condition in which there are large swings in intrathoracic pressure. 153  However, in cardiac tamponade, the pulse pressure and left ventricular stroke volume fall during inspiration, in contrast to the blood pressure changes observed in patients with forced breathing patterns and exaggerated changes in intrathoracic pressure in which pulse pressure is relatively unchanged. 156 

Pulsus paradoxus is a phenomenon described during spontaneous ventilation, not during positive-pressure mechanical ventilation in which large cyclic changes in systolic blood pressure occur in many patients. 156, 159, 160  These large arterial pressure changes during mechanical ventilation have been noted as an incidental observation on the bedside monitor for a long time, but only recently have they been shown to have clear diagnostic implications. 160, 161, 162, 163, 164, 165, 166, 167  Systolic pressure variation is the difference between maximal and minimal values of systolic arterial pressure recorded over the mechanical positive-pressure respiratory cycle. 164, 168  Using end-expiration as the equilibrium period for pressure measurement, the total systolic pressure variation is divided into an early inspiratory increase in pressure, D up, and a later decrease in pressure, D down (see Fig. 30–23). Delta up reflects the inspiratory augmentation in left ventricular output, and D down reflects the impairment in systemic venous return that becomes manifest in the arterial pressure trace shortly thereafter. Normally, mechanically ventilated patients will have D up and D down of about 5 mm Hg each and total systolic pressure variation of approximately 10 mm Hg. 160 

The greatest clinical use of systolic pressure variation has been in the early diagnosis of hypovolemia. 161, 163, 164, 168  Both in experimental animals and patients, hypovolemia causes a dramatic increase in systolic pressure variation, particularly D down. Some authors have suggested that the increase in systolic pressure variation and D down may herald hypovolemia, even in patients in whom the arterial blood pressure is maintained near normal by compensatory arterial vasoconstriction. 168  In a heterogeneous group of intensive care patients, Marik 162  demonstrated that a large systolic pressure variation (>15 mm Hg) was highly predictive of a low PAWP (<10 mm Hg). Using echocardiography to measure left ventricular cross-sectional area as a surrogate for preload, Coriat et al 161  found D down to be an even better predictor of left ventricular preload than wedge pressure. Other uses of systolic pressure variation focus on changes in the D up portion of the arterial pressure trace. Just as D down may reveal changes in cardiac preload, the D up portion of the arterial pressure trace may provide clues to the afterload dependence of the left ventricle. Preliminary evidence suggests that a marked increase in D up during positive-pressure inspiration occurs when the increased pleural pressure reduces transmural left ventricular pressure sufficiently that left ventricular stroke output increases in the failing, afterload-dependent left ventricle. 164, 166, 168, 169 

The arterial pressure waveform provides diagnostic clues in other more unusual physiologic states. Proper time of intra-aortic balloon counterpulsation mandates a detailed interpretation of the arterial waveform (Fig. 30–24). 170  Even during nonpulsatile cardiopulmonary bypass, the minor variations in blood pressure created by the arterial roller head allow calculation and confirmation of the adequacy of systemic blood flow (see Fig. 30–24). 171 

Click thumbnail to see full size image Click thumbnail to see full size image
FIGURE 30–24 Unusual arterial pressure waveforms. (A) Intraaortic balloon counterpulsation with a 1:2 balloon-assist ratio produces a characteristic change in the arterial pressure waveform. Four cardiac cycles are shown, two with balloon assistance and two without. (0) Unassisted enddiastolic pressure, (1) unassisted systolic pressure, (2) dicrotic notch, (3) assisted or augmented diastolic pressure, (4) end-diastolic or presystolic dip, (5) assisted systolic pressure. Effective afterload reduction by the intraaortic balloon is evidenced by the presystolic dip pressure (4) lower than the unassisted end-diastolic pressure (0) and the assisted systolic pressure peak (5) lower than the unassisted systolic pressure peak (1). (B) Arterial pressure waveform during cardiopulmonary bypass. Small phasic pressure variations (arrows) result from the mechanical action of the bypass roller pump. Bypass pump flow rate may be estimated by measuring these pulsations. Nineteen pulsations are recorded in a 3-sec time interval. In this case, the pump configured with 3/8-in tubing has an effective stroke volume of 27 mL. Pump flow rate may be calculated as follows: (19 pulsations/3 sec) × (1 pump revolution/two pulsations) × (27 mL/revolution) × (60 sec/min) = 5130 mL/minute. This calculated pump flow rate should equal the flow rate displayed on the pump console (i.e., 5.2 L/minute). (From Mark JB: Atlas of Cardiovascular Monitoring. New York, Churchill Livingstone, 1998: Figs. 20–3 and 19–8.)